Fungal

Virology

Kenneth William Buck, Ph.D., D.Sc.

Reader In Fungal and Plani Virology
Department of Pure and Applied Biology
Imperiai College of Science and Technology
University of London
Bugiami

CRC Press
Taylor & Francis Group
Boca Raton London New York

C RC Press is an im print of thè

Taylor & Francis G roup, an inform a business
First published 1986 by CRC Press
Taylor & Francis Group
6000 Broken Sound Parkway NW, Suite 300
Boca Raton, FL 33487-2742

Reissued 2018 by CRC Press

© 1986 by CRC Press, Inc.

CRC Press is an imprint of Taylor & Francis Group, an Informa business

No claim to original U.S. Government works

This book contains information obtained from authentic and highly regarded sources. Reasonable efforts have been made to publish
reliable data and information, but the author and publisher cannot assume responsibility for the validity of all materials or the
consequences of their use. The authors and publishers have attempted to trace the copyright holders of all material reproduced in this
publication and apologize to copyright holders if permission to publish in this form has not been obtained. If any copyright material has
not been acknowledged please write and let us know so we may rectify in any future reprint.

Except as permitted under U.S. Copyright Law, no part of this book may be reprinted, reproduced, transmitted, or utilized in any form
by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying, microfilming, and recording,
or in any information storage or retrieval system, without written permission from the publishers.

For permission to photocopy or use material electronically from this work, please access www.copyright.com (http://www.copyright.
com/) or contact the Copyright Clearance Center, Inc. (CCC), 222 Rosewood Drive, Danvers, MA 01923, 978-750-8400. CCC is a not-
for-profit organization that provides licenses and registration for a variety of users. For organizations that have been granted a photocopy
license by the CCC, a separate system of payment has been arranged.

Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and
explanation without intent to infringe.

Library of Congress Cataloging-in-Publication Data

Main entry under title:

Fungal virology

Bibliography: p.
Includes index.
1. Fungal viruses. I. Buck, Kenneth William.
QR343.F86 1986 589.2’04234 85-28059
ISBN 0-8493-6228-8

A Library of Congress record exists under LC control number: 85028059

Publisher’s Note
The publisher has gone to great lengths to ensure the quality of this reprint but points out that some imperfections in the original copies
may be apparent.

Disclaimer
The publisher has made every effort to trace copyright holders and welcomes correspondence from those they have been unable to
contact.

ISBN 13: 978-1-315-89310-5 (hbk)

ISBN 13: 978-1-351-07220-5 (ebk)

Visit the Taylor & Francis Web site at http://www.taylorandfrancis.com and the
CRC Press Web site at http://www.crcpress.com
 PREFACE

Since their discovery 2S years ago, fungal viruses have created a new field of study
in mycology and virology. Of common occurrence in fungal populations, in some of
their properties they resemble traditional viruses, whereas in others their genomes have
similarities to plasmids. In addition, some recently described virus-like particles have
been shown to be intermediates in the movement of transposable elements (Ty ele-
ments) around the yeast genome, and Ty elements may be regarded as members of a
group of retrotransposons which include the copia element in Drosophila as well as
animal retroviruses. As our knowledge of the molecular biology of the cell increases,
traditional boundaries between viruses and other cellular genetic elements are breaking
down, and instead of attempting to precisely define a virus it is possibly more useful
to incorporate the phenomenon of virus-host interactions into a more comprehensive
picture of heredity.
The purpose of this book is not only to serve as a useful reference work but also to
provide reviews of the important advances which have taken place since the last books
on fungal viruses appeared. An introductory chapter gives a critical overview of fungal
virology in the context of virology as a whole and of recent developments in molecular
biology. Specialist chapters follow, all written by experts who are currently active in
fungal virus research, and cover ongoing research areas. The first of these deals with
double-stranded RNA-encoded killer proteins of Saccharomyces cerevisiae, about
which a great deal of knowledge has accumulated in the last few years, particularly
regarding the molecular biology of killer protein synthesis and the genetics of virus-
host interactions. For comparison, the killer protein encoded by a linear DNA plasmid
of Kluyveromyces lactis is also discussed. As well as being valuable as model systems
for studying virus/plasmid-host interactions, the S. cerevisiae and K. lactis systems are
both of current interest in the development of secretion vectors with biotechnological
applications. Another chapter covers the killer system of the corn smut pathogen, Us-
tilago maydis, which has additional interest in its potential for producing smut-resist-
ant plants by incorporating the gene for killer toxin into the plant genome.
One of the current major interests in fungal virology lies in the area of plant pathol-
ogy, namely, the potential of viruses and dsRNA as biological control agents for plant
pathogens. One chapter deals with a model system, namely, a transmissible disease in
the oat pathogen, HeIminthosporium victoriae, for which there is now good evidence
for a viral etiology. Other chapters cover plant pathogens of current economic impor-
tance in forestry and agriculture, namely, Endothia parasitica (the chestnut blight fun-
gus), Ceratocystis (Ophiostoma) ulmi (the Dutch elm disease fungus), and Gaeuman-
nomyces graminis (the wheat take-all fungus), and discuss critically the roles of
cytoplasmic elements associated with viruses and/or dsRNA in the natural and artifi-
cial control of these diseases.
To complete the picture and to put fungal virology in the context of extrachromo-
somal genetic elements in general, a final chapter comprehensively reviews nonviral
extrachromosomal elements in fungi, i.e., mitochondrial DNA and DNA plasmids,
abnormal phenotypes caused by defective mitochondria, and the uses of extrachro-
mosomal DNA in the genetic engineering of fungi.
It is hoped that the book will be of value to both students and researchers, not only
in microbiology (mycology and virology) but also in the related areas of genetics, mo-
lecular biology, and plant biology.

K. W. Buck
March 1986
 THE EDITOR

Kenneth W. Buck, Ph.D, D.Sc. is Reader in Fungal and Plant Virology at the Im-
perial College of Science and Technology, University of London, England.
Dr. Buck was graduated from the University of Birmingham with a First Class B.Sc.
(Hons) degree in Chemistry in 1959 and a Ph.D. degree in 1962. After periods as
Cancer Research Campaign Fellow and I.e.I. Research Fellow he joined the Imperial
College of Science and Technology as Lecturer in Biochemistry in 1965 and moved to
the Department of Pure and Applied Biology at Imperial College in 1981. In 1983 he
was awarded the degree of D.Sc. for his research on viruses, nucleic acids, and carbo-
hydrates.
Dr. Buck serves on the Editorial Boards of the Journal of General Virology and
Intervirology. He is a member of the Society for General Microbiology and the British
Society for Plant Pathology.
Dr. Buck has been a member of the International Committee of Virus Taxonomy
since 1976 and Chairman of its Fungal Virus Subcommittee since 1981. In 1985 he was
elected Secretary of its Executive Committee.
Dr. Buck has published over 100 research papers and articles in scientific journals
and books. His current research interests include: the molecular biology of plant DNA
and RNA viruses, including the development of transient expression vectors based on
geminivirus replicons and the molecular basis of host range, symptom development,
and single gene resistance; the molecular basis of race specificity of plant pathogenic
fungi; and viruses of plant pathogenic fungi and their potential as biological control
agents.
 CONTRIBUTORS

Bernhard Bockelmann Yigal Koltin, Ph.D.

Lehrstuhl fur Allgemeine Botanik Professor in Genetics
Ruhr Universitat Bochum Department of Microbiology
Bochum Tel Aviv University
West Germany Ramat Aviv
Israel
C. M. Brasier, Ph.D., D.Sc.
Pathology Branch
Heinz Dieter Osiewaez
Forestry Commission Research Station
Lehrstuhl fur Allgemeine Botanik
Farnham, Surrey
Ruhr-Universitat Bochum
England
Bochum
West Germany
Jeremy Bruenn, Ph.D.
Associate Professor
Division of Cell and Molecular Biology Hilary Joan Rogers, B.Sc.
State University of New York at Department of Pure and Applied
Buffalo Biology
Buffalo, New York Imperial College of Science and
Technology
London
Kenneth W. Buck, Ph.D., D.Se.
England
Reader in Fungal and Plant Virology
Department of Pure and Applied
Biology
Frank Rainer Schmidt
Imperial College of Science and
Hoechst AG
Technology
Frankfurt
London
England West Germany

S. A. Ghabrial, Ph.D. Erika Schulte

Associate Professor Lehrstuhl fur Allgemeine Botanik
Department of Plant Pathology Ruhr-Universitat Bochum
University of Kentucky Bochum
Lexington, Kentucky West Germany

Neal K. Van Alfen, Ph.D.

Professor
Department of Biology
Utah State University
Logan, Utah
 TABLE OF CONTENTS

Chapter 1
Fungal Virology-An Overview .................................................................... 1
Kenneth W. Buck

Chapter 2
The Killer Systems of Saccharomyces cerevisiae and Other yeasts ....................... 85
1eremy Bruenn

Chapter 3
The Killer Systems of Ustilago maydis........ ................................................. 109
Yigal Koltin

Chapter 4
Hypovirulence of Endothia (Cryphonectria) parasitica and Rhizoctonia solani ..... 143
Neal K. Van Alfen

Chapter 5
A Transmissible Disease of Helminthosporium victoriae - Evidence for a Viral
Etiology ............................................................................................... 163
S. A. Ghabriel

Chapter 6
The d-Factor in Ceratocystis ulmi - Its Biological Characteristics and
Implications for Dutch Elm Disease ............................................................ 177
C. M. Brasier

Chapter 7
The Molecular Nature of the d-Factor in Ceratocystis ulmi .............................. 209
Hilary 1. Rogers, Kenneth W. Buck, and C. M. Brasier

Chapter 8
Viruses of the Wheat Take-All Fungus, Gaeumannomyces graminis var. tritici..... 221
Kenneth W. Buck

Chapter 9
Extrachromosomal DNA in Fungi - Organization and Function ...................... 237
Bernhard Bockelmann, Heinz D. Osiewacz, Frank R. Schmidt, and Erika Schulte

Index ................................................................................................... 285

 Chapter 1

FUNGAL VIROLOGY - AN OVERVIEW

K. W. Buck

TABLE OF CONTENTS

I. Introduction .................................................................................... 2
A. The Discovery and Nature of Viruses ........................................... 2
B. The Discovery of Viruses of Fungi. ............................................. 6
1. Fungi as Vectors of Plant Viruses ....................................... 6
2. Viruses Which Replicate in Fungi. ...................................... 6
a. General Considerations ........................................... 6
b. Viruses of Cultivated Mushrooms ............................. 7
c. Interferon Inducers from Fungi - Double-Stranded
RNA and Viruses in Pencillium and Aspergillus spp ...... 7
d. Screening for Fungal Viruses .................................... 9

II. Morphological Types of Viruses and Virus-Like Particles in Fungi ............. 11

A. Rigid Rods ............................................................................ 21
B. Flexuous Rods ....................................................................... 25
C. Bacilliform and Bullet-Shaped Rods ........................................... 28
D. Club-Shaped Particles ............................................................. 31
E. Enveloped Pleomorphic Particles .............................................. .31
1. Neurospora crassa VLPs ................................................. 31
2. Saccharomyces sp. VLPs ................................................. 33
F. Particles Similar to Herpesviruses .............................................. 33
G. Particles with Heads and Tails ................................................... 34
1. VLPs from Yeasts ......................................................... 34
2. Bacteriophages from Cultures of Penicillium spp. (PB
Viruses) ....................................................................... 35
H. Geminate Particles .................................................................. 36
I. Isometric Particles .................................................................. 37
1. Double-Stranded RNA Viruses ......................................... 37
2. Single-Stranded RNA Viruses from Sc1erophthora
macrospora .................................................................. 37
3. Double-Stranded DNA Viruses from Rhizidiomycessp ......... .37
4. VLPs from the Lower Fungi with Diameters in the Range
40 to 200 nm ................................................................ 40
J. Retrovirus-Like Particles and Transposition in Yeast. .................... .40

III. The Biology and Biochemistry of Isometric Double-Stranded RNA

Mycoviruses ................................................................................... 42
A. Transmission ......................................................................... 42
1. Transmission during Hyphal Growth ................................. 42
2. Transmission via Asexual Spores ..................................... .43
3. Transmission via Sexual Spores ....................................... .43
4. Transmission via Heterokaryons and Heteroplasmons .......... .44
5. Transmission with Cell-Free Virus Preparations .................. .45
6. Host Range .................................................................. 46
2 Fungal Virology

B. Structure, Genome Organization, and Taxonomy ......................... .46

C. Virus-Host Interactions .......................................................... .49
1. Replication .................................................................. 49
a. Replication in Relation to Host Growth - Virus
Latency .............................................................. 49
b. Ultrastructural Studies ........................................... 53
c. The Virus Replication Cycle - Virion-Associated RNA
Polymerases ........................................................ 54
d. Mixed Infections - Compatibility and Incompat-
ibility ................................................................. 55
2. Virus Infections and the Host Phenotype ............................ 56
a. Secondary Metabolites ........................................... 57
b. Killer Proteins ..................................................... 57
c. Transmissible Diseases - Lytic Plaques in Penicillium
chrysogenum; Cold Sensitivity in Saccharomyces
cerevisiae; Die-Back Disease of Mushrooms ............... 58
d. Phytopathogenicity ............................................... 61
D. Evolution ............................................................................. 61

IV. Infection of Fungi with Alien Viruses ................................................... 62

A. Animal Viruses ...................................................................... 62
B. Plant Viruses ......................................................................... 63
C. Conclusions .......................................................................... 63

V. Outlook ........................................................................................ 64

References .............................................................................................. 64

I. INTRODUCTION

A. The Discovery and Nature of Viruses

Diseases of animals and plants now known to have a viral etiology have been rec-
ognized for thousands of years. A disease resembling smallpox was described in China
in the tenth century BC, and yellow fever, long known in tropical Africa, may have
been responsible for the legends of such cursed ships as the Flying Dutchman and that
of "The Ancient Mariner".' A description of the yellow leaf of Eupatorium, now
known to be caused by a geminivirus, appears in "Manyoshu", a classic Japanese
anthology of the 8th century, and this may be the earliest recorded plant virus disease. 2
The ornamental variegation of tulips, caused by tulip breaking virus, a potyvirus, has
been known for centuries and infected tulips were once prized as distinct varieties. 3
The word virus itself (Latin virus = evil smelling or poisonous liquid), since ancient
times has been equated with a poison. However, by the 19th century when the micro-
bial theory of infectious disease was gaining in popularity, the term virus was used
generally to refer to an infectious agent of disease and was not distinguished from other
pathogenic agents such as bacteria.' It was not until the end of the 19th century that
the word acquired a meaning similar to its present day usage. Iwanowski s and Beijer-
inck 6 showed that the infectious agent, capable of causing mosaic disease of tobacco
by sap inoculation, was able to pass through a bacteria-proof filter. Similarly, Loeffler
and Frosch 7 showed that foot and mouth disease of cattle was caused by an agent that
 3

also could pass through a bacteria-proof filter and it was soon realized that these "ul-
trafilterable viruses" represented a new class of infectious agents quite distinct from
bacteria. In fact, soon afterwards Twort 8 and d'Herelle 9 discovered that bacteria them-
selves could be infected with viruses which the latter investigator termed "bacterio-
phages." However, direct visualization of viruses and determination of their dimen-
sions had to await the development and refinement of the techniques of X-ray
crystallography and electron microscopy in the 1930s and later. Viruses have now been
found in nearly all the major groups of prokaryotes and eukaryotes.
Even today there is still no generally accepted definition of a virus, but the following
characteristics serve to distinguish viruses from cellular organisms, subcellular organ-
elles and plasmids.

1. Viruses are infectious subcellular particles which are able to enter cells from with-
out and to promote their own replication within the infected cells. After replica-
tion and release from their host cells, viruses are able to survive, at least for a
short time, and sometimes for long periods, in the extracellular environment,
before infecting a new host cell.
2. Virus particles are composed of a genome of either DNA or RNA, surrounded
by a virus-encoded protein coat or capsid made up of many polypeptide chains.
A principal function of the virus capsid is to protect the genome when the virus
is in an extracellular environment. Some viruses also have an outer lipoprotein
envelope, derived from host cell membranes and into which virus-encoded poly-
peptides are inserted. The outer surface of the virus particle contains protein
which is responsible for the attachment and, in some cases, penetration of the
virus or virus genome into the host cell. This protein may be part of the capsid
or envelope, or may comprise the whole or part of special appendages or tails
attached to the "head" of the virus particle.
3. Viruses are obligate parasites, not essential to their host. They multiply only in
host cells (in a few instances viral genomes have been replicated in cell-free sys-
tems with components derived from host cells). In particular they require ribo-
somes, precursors for nucleic acid and protein synthesis, host membranes, and a
source of energy. Other requirements vary with the virus. Some enveloped viruses
(arenaviruses) contain host ribosomes, but such ribosomes cannot function
within the virus particle.
4. Viruses replicate by producing multiple copies of their components and assem-
bling the particles from a pool. This is in contrast to cells, which replicate by
doubling their contents, then dividing by binary fission with the cell membrane
always intact. In the case of those viruses which are surrounded by an outer
lipoprotein envelope, this membrane is lost after adsorption of the virus to the
cell surface and only the viral nucleocapsid enters the cell. After replication and
assembly of the nucleocapsid from a pool of components, the membrane is re-
formed by maturation of the virus from intracellular membranes of "budding"
(exocytosis) from the cytoplasmic membrane.
S. Most viruses are able to cause disease in at least some hosts; and indeed, most
viruses were first recognized as a result of the diseases which they produced in
their hosts. The ability to cause disease in at least one host has been considered
to be an essential property of a virus.lO However, pathogenicity is a property of
a virus-host interaction, and a virus which causes disease in one host may cause
an inapparent infection in another. Several single-stranded DNA filamentous
bacteriophages (e.g., fl, fd, M13) are released from their hosts without cell death
or lysis, and infected cells are almost normal. Some modern definitions of virus
emphasize only the possession of specific genetic materials that utilize the cellular
4 Fungal Virology

synthetic machinery and the possession of an extracellular infective phase repre-

sented by the mature virus particle or virion which serves as a vehicle to introduce
the viral genome into other cells.",12
6. Virus genomes vary in size from about 3.5 kb for the small RNA bacteriophages
to over 350 kb for some poxviruses. Virus particles (virions) vary considerably
both in size and shape. Some of the small icosahedral viruses are only 20 to 25
nm in diameter whereas the large pox viruses have diameters up to 450 x 250 nm,
as large as small cellular organisms sllch as mycoplasmas and the infectious ele-
mentary bodies of chlamydiae.

From the foregoing properties it is clear that viruses are distinguished from cells by
their lack of ability to produce energy and substrates for protein and nucleic acid syn-
thesis, lack of a protein-synthesizing system, and by their mode of replication.
Viruses are distinguished from genome-containing subcellular organelles (mitochon-
dria, chloroplasts, nuclei) by their mode of replication, ability to survive outside the
cell and to reinfect other cells, and their nonrequirement by the host. In the latter
context it should be noted that petite yeasts, which lack functional mitochondria, when
grown on glucose can still survive by making A TP anaerobically by glycolysis; this
process is, however, much less efficient than oxidative phosphorylation, so that petite
yeasts are at a severe disadvantage compared to those with functional mitochondria.
Viruses are distinguished from plasmids by their possession of a capsid and an or-
ganized particle which can survive outside the cell and reinfect other cells. Plasm ids are
essentially naked nucleic acid molecules (usually DNA) which encode proteins not es-
sential, but generally beneficial, to their hosts. Transmission of plasmids in nature is
intracellular and they often encode proteins which enable their transfer from one host
to another. The similarity of some self-transmissible bacterial plasmids to some bacte-
riophages has been noted. '3 F sex factors (plasmids) in male bacteria encode proteins
(pilins) which accumulate in the cytoplasmic membrane and form the connections (F-
pili) between male and female bacteria, through which the DNA is transferred. The
single-stranded DNA filamentous bacteriophages (e.g., fd, M13) encode coat protein
which accumulates in the cytoplasmic membrane, and this assembles around the DNA
in a left-handed helix during maturation and release of the phage from the cell. Phage
A with amber mutations in the N gene can persist indefinitely as an independent repli-
cating unit, a plasmid, in the cytoplasm of host Escherichia coli cells. 10 Both viruses
and plasmids have probably evolved from nucleic acids of their hosts and it is not
surprising that they have similar features. Viruses have evolved an extracellular mecha-
mism of transmission and hence need to be protected by protein coats, whereas plas-
mids have evolved an intracellular mechanism of transmission and can exist as naked
DNA.
Some viruses are able to insert their genome, or a DNA copy in the case of certain
RNA viruses (retroviruses), into that of the host cell. This may occur as part of the
normal virus replication cycle (retroviruses) or as an alternative pathway in which the
"vegetative" viral functions (those involved in the production of virus particles) are
partly or wholly repressed, and the viral DNA replicates stably, in a provirus form, as
an integral part of the host genome (endogenous retroviruses, some groups of animal
DNA viruses, temperate bacteriophages). In some cases host DNA may also be trans-
duced along with the viral DNA. After suitable stimuli the integrated viral DNA may
become excised, resulting (if a complete virus genome has been inserted) in a normal
replication cycle and production of mature virus particles.
The integration of viral nucleic acid into host DNA may alter the properties of the
host. For example, strains of Salmonella typhimurium contain prophage DNA which
encodes enzymes which modify the cell wall and alter the serological properties of the
 5

bacterial cell (lysogenic conversion).ls Such changes may actually be beneficial to the
bacteria (themselves pathogenic to man) in enabling them to evade their hosts' immune
systems. Similarly, virulent, but not avirulent, strains of Corynebacterium diphtheriae
contain, integrated into their chromosomes, phage DNA, which encodes the diphtheria
protein toxin!6 Insertion of the whole or part of viral DNA (or DNA copies of viral
RNA) into the chromosomes of animal cells, can, in some instances, lead to the for-
mation of transformed cells which, when introduced into animals, can form tu-
mors. 17 • IS Interestingly a region of the DNA (T-DNA) of Ti plasmids of Agrobacterium
tumefaciens can integrate into the chromosomal DNA of plants which are hosts for
this bacterium; genes within the T-DNA, when expressed in the host plant, result in
tumor (crown gall) formation. 19
Viral or plasmid DNAs that can be integrated into (and excised from) host genomes
may be considered as members of a general class of mobile genetic elements 20 that
includes the following: prokaryotic IS elements and transposons; transposable ele-
ments in yeast, Drosophila, maize, and other eukaryotes; invertible DNA segments
causing phase variation in Salmonella and host range variation in phage Mu; transpos-
able DNA involved in yeast mating-type switches and antigenic variation in trypano-
somes; and DNA rearrangements involved in the generation of antibody diversity. To
be added to this list are the DNA segments located in mitochondria which can be
integrated into mitochondrial DNA or exist as independent replicons. DNAs of this
type include those associated with cytoplasmically transmitted male-sterility (cms-S) in
maize21 and those (senDNAs) associated with senescence in the fungus Podospora an-
serina. 22 . 23 In the latter case senDNAs can also be incorporated into nuclear DNA.24
Indeed, there are now several examples of apparent transfer of DNA between organ-
elles (mitochondria, chloroplasts, nuclei) in plants and fungi. 25.26 Also relevant in this
section is the existence of intervening sequences (introns) in genes in eukaryotes and
the detection of intronless pseudogenes and "Alu" - type sequences which are often
flanked by direct repeats and concluded at one end by oligo (dA).27.28 This suggests
possible reverse transcription of RNA into DNA and integration involving transposa-
ble elements. Discussion of the various examples of DNA transposition and rearrange-
ments does not imply a unique mechanism. Several different mechanisms are involved
in different cases and have been reviewed. 20
Another class of self-replicating agent (found in plants) which can be distinguished
from viruses and plasmids is the viroids. 29 Like viruses, viroids are highly infectious
pathogenic agents which can spread in nature from plant to plant, but like plasm ids
they are composed of naked nucleic acid. The high degree of base-pairing of the single-
stranded RNA molecules (ca. 360 nucleotides) which constitude viroids may contribute
to their stability. Unlike viruses or plasmids, viroids apparently do not encode any
protein product. They are replicated entirely by cellular enzymes via RNA intermedi-
ates (possibly by a rolling circle mechanism). There is no evidence that a cDNA copy
of viroid RNA is either produced or integrated into host DNA during the viroid repli-
cation cycle. 30 Based on comparative nucleotide sequence analyses of host DNA and
viroid RNA it was suggested that viroids may be escaped introns, i.e., they may have
originated by splicing out and circularization of intervening sequences during the proc-
essing of host mRNA.31.32 More recently, striking sequence similarities with the ends
of transposable genetic elements (those of retroviral proviruses in particular) have sug-
gested that viroids may have originated from such elements. 33 Possible similarities of
viroids to the transmissible agents of animal neurological diseases, the subacute spon-
giform encephalopathies, which include the agent causing scrapie in sheep and goats,
have been discussed. 29
The overall picture that emerges is superimposed on the stable genome of an organ-
ism, a considerable degree of plasticity mediated by DNA transpositions and rear-
6 Fungal Virology

Table 1
PLANT VIRUSES TRANSMITTED BY FUNGAL VECTORS

Fungal Vector Virus or virus·like agent transmitted Refs.

Olpidium brassicae Tobacco necrosis virus, tobacco necrosis satellite virus, 35-38, 295-298
tobacco stunt virus, lettuce big vein virus
Olpidium radicale Cucumber necrosis virus, red clover necrotic mosaic vi- 39,40
rus
Polymyxa graminis Soil-borne wheat mosaic virus, peanut clump virus, bar- 41-48
ley yellow mosaic viru" wheat yellow mosaic virus,
rice necrosis virus, oat mosaic virus, wheat spindle
streak virus, brome mosaic virus
Polymyxa betae Beet necrotic yellow vein virus 49
Spongospora subterranae Potato mop top virus 50

rangements, plasmids, and viruses. Such plasticity is important in some instances in

the control of gene expression, in others in conferring selective advantages (e.g., anti-
biotic resistance) on an organism, and in yet others can lead to a diseased state of the
celL Although it is unlikely that all viruses have evolved by a common mechanism, at
least some viruses could have evolved from cellular genetic elements that were origi-
nally beneficial to the celL Perhaps we should not attempt to define a virus precisely
but to incorporate the phenomenon of virus-host interactions into a more comprehen-
sive picture of heredity. It is against this background that I will describe the discovery
and properties of viruses which infect fungi.

R The Discovery of Viruses of Fungi

1. Fungi as Vectors of Plant Viruses
Although it has been known since the beginning of the present century that a number
of plant viruses are transmitted through soil, it has been only within the last 25 years
that their vectors, fungi and nematodes, have been identified. The fungal vectors are
all obligate parasites which inhabit the roots of plants; some are pathogens in their
own right, whereas others are avirulent root parasites. All the proven vectors are lower
fungi, Olpidium spp, in the Chytridiales and Polymyxa and Spongospora spp. in the
Plasm odioph orales. Examples are given in Table 1. Each virus has a specific vector,
but there is no evidence that the virus mutiplies in the fungus. These are therefore
plant, not fungal, viruses. The virus-vector specificity probably lies in a specific inter-
action between the virus coat protein and a component of the surface or interior of the
fungal spore or thallus. The mode of interaction of individual viruses with their vectors
has been reviewed by CampbelL 34 Interestingly, tobacco stunt and lettuce big vein vi-
ruses have genomes of double-stranded RNA29s.298 which is also common in fungi;
however, tobacco stunt and lettuce big vein viruses are rod-shaped, whereas all the
known double-stranded RNA viruses of fungi are isometric. An early report of the
transmission of potato virus X by the chytrid fungus Synchytrium endobioticum'99 has
not been confirmed by subsequent investigators. 153

2. Viruses Which Replicate in Fungi

a. General Considerations
True fungal viruses (mycoviruses), i.e., those which replicate inside fungal cells, were
discovered less than 25 years ago. There are four main reasons for the somewhat be-
lated discovery of mycoviruses compared to viruses of animals, higher plants, and
bacteria.

L Many virus-infected fungi do not show symptoms, i.e., the viruses are latent or
cryptic in these hosts.
 7

2. Diseased colonies of filamentous fungi may be difficult to detect or isolate from

their natural source. Mycologists may discard isolates which do not grow well
and such isolates are unlikely to find their way into national culture collections.
3. Cell-free transmission with purified viruses is difficult and must be carried out
under carefully controlled conditions to exclude contamination of the virus in-
oculum or host culture with air-borne virus-infected fungal spores.
4. Mycologists who were aware of transmissible diseases of fungi, such as the dis-
ease of Heiminthosporium victoriae which was suspected to be caused by a vi-
rus,51 were not familiar with virological techniques. It is perhaps significant that
the first discovery of a fungal virus, in cultivated mushrooms, was made by an
experienced plant virologist, M. Hollings.56

b. Viruses of Cultivated Mushrooms

The edible mushroom, Agaricus bisporus, has been cultivated in Europe for about
300 years and in the U.S. for about a century. However, it was not until 1948 that a
disease of mushrooms, characterized by thin long stipes and globular thin-fleshed caps
of the fruiting bodies, was first observed in a mushroom house owned by La France
Brothers in Pennsylvania, and termed "La France disease". 52 Mushroom diseases with
similar properties were subsequently discovered in several other parts of the U.S., in
Europe (England, France, The Netherlands, Italy, Denmark), in Japan, and in Aus-
tralia. Several different names, such as "X-disease", "brown disease" , "watery stipe"
and "die-back" were coined for what was considered by Sinden S3 to be essentially the
same disease.
Although the infectious nature of this mushroom disease had been recognized from
an early stage and it was suspected that it might be caused by a virus, it was not until
1959 that Gandy S4,ss showed conclusively that the disease could be transmitted by hy-
phal anatomosis, thereby implicating a cytoplasmic factor as its cause. A few years
later Hollings56 discovered the first fungal virus by extracting three types of particles
from the fruit bodies of diseased mushrooms in England. These were spherical particles
of two diameters, 25 and 29 nm, subsequently named mushroom viruses 1 and 2 (MV 1
and MV2), and bacilliform particles, with a length of 50 nm and a width of 19 nm,
subsequently named mushroom virus 3 (MV3). Two other types of spherical particle
with diameters of 34 to 35 nm (MV 4) and 50 nm (MV5) were discovered later. 57,58
Similar particles were observed in diseased mushrooms in several countries, the most
common being the bacilliform particles and the spherical particles of diameters 25 and
34 to 35 nm. The characteristics of these particles and their relationship to mushroom
virus disease will be discussed in later sections. Several review articles are devoted either
entirely or partially to mushroom virus disease and its control. 5'-66

c. Interferon Inducers from Fungi - Double-Stranded RNA and Viruses in

Penicillium and Aspergillus spp.
The next major discoveries of fungal viruses were made, not by searching for my-
coviruses, but as results of screening programs for substances with antiviral activity in
animals. The antiviral protein, interferon, which is produced as an early defense re-
sponse to virus infections in animals, attracted much attention after its discovery in
1957 by Isaacs and Lindenmann. 67 However, interferon proved both difficult to purify
and to obtain in the amounts needed for clinical tests of its antiviral efficacy. Attention
switched in the 1960s to interferon inducers: substances which when administered to
animals stimulated them to produce their own interferon and hence protected them
from subsequent virus infections. A number of such substances were obtained from
fungi.
The first of these 68 had been obtained by R. E. Shope from an isolate of Penicii-
8 Fungal Virology

lium funiculosum which he had found growing on the isinglass cover of a photograph
of his wife, Helen; the antiviral extract from this fungus was appropriately called
"helenine"69 and later shown to exert its antiviral effects in animals by inducing the
synthesis of interferon. Purification and fractionation of helenine by Hilleman and co-
workers 7l showed that the ingredient with antiviral and interferon-inducing activities
was double-stranded RNA (dsRNA), and these investigators postulated that the
dsRNA could have arisen from a fungal virus infection. Subsequently, Banks et al. 72
isolated isometric virus particles, 25 to 30 nm in diameter, from P. funiculosum. The
purified virus particles were active in inducing interferon formation in mice and con-
tained dsRNA. 73
Second, culture filtrates of Pencillium stoloniferum had been found to have antiviral
activity in mice by Powell et al. 7. and subsequently a partially purified preparation of
this antiviral agent was called "Statolon" by Probst and Kleinschmidt,75 presumably a
hybrid name from its virostatic properties and its origin from P. stoloniferum. Like
helenine, statolon was found to be a potent interferon inducer. 76,77 Originally the active
ingredient of statolon was believed to be a polysaccharide!8 However, examination of
statolon preparations by Ellis and Kleinschmidt'9 revealed the presence of spherical
virus-like particles and fractionation of statolon by sucrose density gradient centrifu-
gation showed that a substantial proportion of its interferon-inducing activity was as-
sociated with a band formed in the gradient by these particles. Furthermore, Banks et
al. 72 isolated two polysaccharides from statolon, one derived from the culture medium
in which P. stoloniferum was grown, and the other from autolysis of the fungal cell
wall,80 and found both of them to be completely inactive as interferon inducers. They
then went on to isolate virus particles, 25 to 30 nm in diameter, from P. stoloniferum,
to show that the viral nucleic acid was an active interferon-inducer in mice and to
obtain evidence that the viral nucleic acid was dsRNA; this was subsequently con-
firmed by Kleinschmidt et al. 81
Further reports of fungal antiviral agents, from Penicillium chrysogenum 82 and
Penicillium cyaneo-fulvum,83 led to the discovery of isometric, dsRNA-containing, vi-
rus particles in these two fungi 8• 89 and this was soon followed by a similar report of
virus particles from Aspergillus foetidus!O From all three fungi, both virus particles
and isolated dsRNA were effective interferon inducers both in tissue culture and in
intact animals.
DsRNA has proved to be the most potent interferon inducer so far discovered and
is probably the active inducer of interferon formation during virus infections of ani-
mals. As little as one molecule of dsRNA per cell has been estimated as the threshold
for interferon induction!1 Interestingly, dsRNA is also required to activate two en-
zymes, an oligoisoadenylate synthetase and a protein kinase, which contribute to the
antiviral action of interferon!2 DsRNA from a variety of sources has been found to
be active, e.g., synthetic polyriboinosinic acid:polyribocytidylic acid,.3 the replicative
form dsRNA of bacteriophage MS2 9• and reovirus dsRNA.95 The structural require-
ments of dsRNA interferon inducers have been reviewed. 96
Mycoviral dsRNA has been a useful source of dsRNA of defined size for studying
the clinical potential of dsRNA; quantities of 1 kg of dsRNA from Penicillium chry-
sogenum have been prepared by a pilot plant extraction process. 97 DsRNA has a wide
variety of biological activities, including antiviral and antitumor activities, immuno-
genicity, adjuvant properties, immunosuppression, and cytotoxicity!7,98,579 The toxic-
ity of dsRNA which has been compared to that of bacterial endotoxins,99,100 has pre-
cluded the use of dsRNA in clinical medicine. However, dsRNA is still used in
laboratory studies of the mechanism and control of interferon induction.
In recent years attention has again turned to the direct uses of interferons as antiviral
and antitumor agents, especially now that genes for the three types of human inter-
 9

feron, a,(J and y, have been cloned and expressed in bacteria and yeast, and interferons
can be purified by affinity chromatography with monoclonal antibodies. '02 Sufficient
interferon can now be obtained by recombinant DNA technology for extensive clinical
trials. However, like dsRNA, interferons have a wide range of biological activities,
including complex effects on the immune system and toxicity.'03 Their enormous clin-
ical potential has yet to be realized. Two books providing a comprehensive coverage
of current interferon research have been published.'°'·'os

d. Screening for Fungal Viruses.

The discovery of fungal dsRNA interferon inducers and viruses of Penicillium and
Aspergillus spp. had a major impact on fungal virology. It was realized that fungal
viruses were probably much more common than had been understood hitherto and
many investigators were stimulated to look for viruses in the fungi in which they were
interested. This led to discoveries of dsRNA and viruses in fungal pathogens of
plants.,0•. ,07 and humans, 108.S80 in "killer" strains of fungi 10. 112 and in diseased
fungi. 113 115
Others were stimulated to carry out random searches for viruses in extracts of fungi
by electron microscopy., From the results of such investigations Bozarth"· concluded
that 10 to 15070 of randomly sampled fungal isolates contained virus-like particles. This
is probably a conservative estimate since (1) some fungi may contain particles below
the level of detection by electron microscopy (for small isometric viruses in Gaeuman-
nomyces graminis, the sensitivity for crude preparations, in which virus from 10 g wet
weight of mycelium was resuspended in 1 m! of buffer, was about 10 particles per
hyphal compartment;'17 the sensitivity could be increased using immunospecific elec-
tron microscopy (ISEM) or enzyme-linked immunosorbent assay (ELISA), but virus-
specific sera are of limited use for random screening); and (2) the fractionation proce-
dure used to prepare fungal extracts for electron microscopy may select certain classes
of particles while rejecting others. 118
Another random screening procedure involved the use of antisera to dsRNA to de-
tect dsRNA in fungal extracts. DsRNA-positive fungi could then be screened by con-
ventional means for virus particles. Usually the synthetic double-stranded polyribonu-
cleotides, polyadenylic acid:polyuridylic acid or polyinosinic acid:polycytidylic acid,
conjugated to bovine serum albumin, have been used to raise the antibodies in
rabbits"' Although the antisera produced in this way have only moderate titers (up to
1 :32 in gel immunodiffusion tests or 1 :256 in ring precipitin tests), sera can be obtained
which are reasonably specific for dsRNA. Responses of heterologous dsRNAs, how-
ever, are often different from those of the homologous dsRNA. The use of antisera
prepared to fungal viral dsRNA, rather than synthetic double-stranded polyribonu-
cleotides, could be advantageous in this respect. Monoclonal antibodies prepared to
rice dwarf virus (RDV) dsRNA were completely specific for dsRNA, no reaction being
obtained with ssRNS, ssDNA, or dsDNA.300 When RDV dsRNA, poly A:poly U and
poly I:poly C were tested against the monoclonal antibodies in gel double-diffusion
assays the precipitin lines fused completely, suggesting that the anti-RDV dsRNA mon-
oclonal antibody is specific to the double-helical structure (A form) common to these
three dsRNAs. Several immunological techniques have been described for detecting
dsRNA with such antisera; 120.121 these include fluorescent antibody assays, two-dimen-
sional immunodiffusion, immune electron microscopy, immunoelectrophoresis,
counterimmunoelectrophoresis, rocket immunoelectrophoresis, and indirect agglutin-
ation. Recently the sensitivity of this method has been increased by the use of the
enzyme-linked immunosorbent assay (ELISA); 122 the double antibody sandwich
method was particularly suited for the detection of heterologous dsRNA antigens. 123
Most of these test procedures have been applied to relatively few mycovirus systems
and their suitability for large scale screening programs has yet to be determined.
10 Fungal Virology

Using antisera to poly U and poly I:poly C and detection by gel immunodiffusion,
Moffitt and Lister '24 found dsRNA in 29(\'0 (20 out of 70) of fungal isolates tested.
This is again probably a conservative estimate for the proportion of virus-infected
isolates for two reasons. First, the method may not be capable of detecting low levels
of dsRNA in fungi. Lister '21 estimated the minimal concentration of synthetic dsRNA
detectable in gel diffusion tests to be 4 /-Ig/m1. If nucleic acid from 10 g wet weight of
fungal mycelium is resuspended in 1 ml of buffer, this would correspond to 0.4 /-Ig
dsRNA/g mycelium. For dsRNA in Gaeumannomyces graminis this equates to about
40 molecules of dsRNA per hyphal compartment. The method is less sensitive than
electron microscopy, but in this case the sensitivity could be increased using the ELISA
technique. '23 Second, the assay detects dsRNA viruses, and possibly the dsRNA repli-
cative form from infections with single-stranded (ss) RNA viruses, but is probably not
sensitive enough to detect the very small amounts of dsRNA which have been shown
to be produced in infections with some DNA viruses. '25 When 18 of the 20 dsRNA-
positive fungi '24 were examined by electron microscopy, virus-like particles were found
in only 5. This could be due to extraction problems, the occurrence of naked dsRNA,
or particles not visible with the straining procedures used.
Morris and Dodds '26 have described another method for detecting dsRNA in plant
and fungal tissue which is an adaptation of that of Franklin '27 for separating dsRNA,
ssRNA, and DNA from extracts of phage-infected bacteria. The basis of the method
is that, in a buffer containing 15070 ethanol, cellulose will bind dsRNA, but not ssRNA
or DNA. DsRNA can subsequently be eluted from the cellulose with ethanol-free
buffer. The method could be combined with the dsRNA serological method described
above to give a 10- to 20-fold increase in sensitivity, since a much smaller volume of
buffer is required to resuspend the dsRNA fraction (after ethanol precipitation) com-
pared with that required for total nucleic acid. However, the dsRNA, prepared by the
cellulose method, is pure enough to analyze by agarose gel electrophoresis or polya-
crylamide gel electrophoresis (PAGE) which has the advantage of providing informa-
tion on the size and number of dsRNA components present. For a single dsRNA com-
ponent the sensitivity of detection with ethidium bromide straining is about 0.I/-1g/m1
and is therefore more sensitive than the dsRNA serological method used alone (or of
comparable sensitivity if the serological and cellulose methods are combined and the
dsRNA consists of several components). The method has been used for detecting
dsRNA components in species of Gaeumannomyces and Phialophora. 128
Another sensitive method, which has been used to isolate replicative form dsRNA
from plants infected with tomato bushy stunt virus,129 involves making a total nucleic
acid preparation, precipitating high molecular weight ssRNA with 2M-LiCl, 130 and
removal of DNA by sedimenting the RNA through a CsCl cushion. The pellet which
contains dsRNA and some tRNA can be taken up in a small volume of buffer for
analysis by PAGE. Both this and the cellulose methods should be useful for detecting
dsRNA viruses and the replicative form dsRNA from ssRNA viruses in fungi.
From the fairly limited screening programs that have so far been carried out, we can
guess that viruses will be present in at least 30% of fungal species. Ainsworth '31 has
estimated that there are at least 50,000 fungal species, so the majority of fungal viruses
almost certainly await discovery. It is difficult to extrapolate existing data to estimate
the total number of fungal viruses, since some viruses are known which infect more
than one species, while infections of some species with several different viruses are
known.
The incidence of viruses in fungi is compared with that in other phyla in Table 2.
Only well characterized viruses have been included. To some extent the figures will
reflect the number of species in a phylum and their importance. Most virologists are
concerned with viruses that cause disease in man, domesticated animals, and plants,
 11

Table 2
INCIDENCE OF VIRUSES IN
DIFFERENT HOST PHYLA"

Phylurn Nurnber of species in which

viruses have been described'

Eukaryotes
Fungi ++
Algae ++
Pteridophytes +
Gyrnnosperrns +
Angiosperrns +++
Protozoa +
Nernatodes +
Arthropods +++
Molluscs +
Vertebrates +++
Prokaryotes
Bacteria and +++
blue-green
algae

• Table adapted frorn Gibbs and Harrison.1J2

+, I to 10 species; ++, 10 to 100 species; +++,
100 to 1000 species.

viruses that have potential for the control of pests of animals or plants, viruses of
microorganisms of economic and medical importance, and viruses which are valuable
as model systems for probing the molecular biology of cellular organisms. The absence
of reports of viruses from lower plants, e.g., diatoms, bryophytes and cycads, and
from many phyla of lower animals, e.g., sponges, coelenterates, platyhelminths, roti-
fers, polyzoa, brachiopods, annelids, and echinoderms, may be a reflection of the
absence of careful searches for viruses in these organisms.

II. MORPHOLOGICAL TYPES OF VIRUSES AND VIRUS-LIKE

PARTICLES IN FUNGI

Several morphological types of virus-like particle (VLP) have been detected in the
fungi (Table 3). Few of these VLPs satisfy the criteria for virus described in Section
I.A and only a small proportion of them havtJ been isolated, so that their composition
is unknown. In many cases the particles have been characterized only by electron mi-
croscopy. In this article the term virus will be used for particles which morphologically
resemble viruses and which have been purified and shown to contain a nucleic acid
genome enclosed in a protein coat, even in cases when infectivity has not been estab-
lished. The term VLP will be used for particles which morphologically resemble vi-
ruses, but for which no information on particle composition is available. Table 3 lists
over 200 VLPs from more than 100 fungal species. However the number of true viruses
may be less for two reasons.
First, some of the VLPs may be duplicated, e.g., isometric VLPs of apparently dif-
ferent diameters measured in different laboratories may in fact be the same. This is
apparent from values of particle diameters of the same virus measured in different
laboratories, e.g., values for the diameter of particles from the same strain of Penicil-
lium chrysogenum vary from 35 nm 85 to 40 nm. 250 There are many variables which may
affect particle dimensions measured by electron microscopy, such as the stain used,
purity of the virus preparation, whether or not particles are close packed, and stand-
 Table 3
MORPHOLOGICAL TYPES OF VIRUS-LIKE PARTICLES DETECTED IN FUNGI
-
IV

Fungi in which detected

~
Particle diameter (nm) I::
::s
Division or or dimensions
~
Morphology subdivision Class Genus and species (width x length, nm) Ref. '-

Rigid rod Ascomycotina Discomycetes Peziza ostracoderma (= Pli- 17 x 350 139,140

.,-
$
0
'-
caria fulva 0
Pyrenomycetes Erysiphe graminis TMV-like 141 ~
Sphaerotheca lanestris TMV-like 141,142
Basidimycotina Hymenomycetes Agaricus bisporus (= A. bru- 17 x 350 63,139
nescens = A_ hortensis
Lentinus edodes 25-28 x 280-310 64,136,150,208
Teliomycetes Coleosporium asterum TMV-like 141
C. madiae TMV-like 141
Puccinia helianthi 12 x 260 143
P_ sorghi 12 x 260 143
Uromyces fabae TMV-like 141
U. phaseoli 12 x 260 143
TMV-like 141
Deuteromycotina Hyphomycetes Ostracoderma (Chromolos- 15-17 x 350 144-146
porium)sp. (anamorph of
Peziza ostracoderma)
Mycogone perniciosa 18 x 120 147,148

Flexuous rod Ascomycotina Pyrenomycetes Erysiphe polygoni 15-17 x 1500 136

Basidiomycotina Hymenomycetes Boletus edulis 13 x 500 149
Collybia peronata 17 x 1500 136
Len tin us edodes 15-17 x 1500 64,136,150-152
L. lipideus 17 x 1500 136
Teliomycetes Puccinia helianthi 10 x 660 143
16 x 740 143
Uromyces phaseoli 10 x 660 143
16 x 740 143
Deuteromycotina Hyphomycetes Helminthosporium sacchari 17 x 1500 152
Mastigomycotina Chytridiomycetes Synchytrium endobioticum 19 x 2000 153
Acilliform and Ascomycotina Pyrenomycetes Microsphaera mougeotti 19 x 48 152
bullet-shaped
rods
Basidiomycotina Hymenomycetes Agaricus bisporus 19 x 50 56,152,154-164
A. campestris 19 x 50 60
Armillaria (Armillariella) 22-28 x 119 165
mellea
Inocybe dulcamara 22-28 x 119 165
Laccaria laccata 22-28 x 119 165
Deuteromycotina Hyphomycetes Verticillium fungicola 17 x 35 145
Mastigomycotina Oomycetes Phytophthora infestans 50 (width), length indetermi- 166,167
nate
Zygomycotina Zygomycetes Strongwellsea magna 100 x 390 347

Club-shaped Ascomycotina Pyrenomycetes Endothia (Cryophonectria) "Head" 50-90, "tail" 168,169

parasitica length 25-300
Basidiomycotina Hymenomycetes Agaricus bisporus "Head" 70, "tail" 25 x 170-172
150-200
Pleomorphic with Ascomycotina Pyrenomycetes Neurospora crassa 250-400 (nucleoid, 120- 173,174
membrane enve- 170
lope
Herpes-like Mastigomycotina Oomycetes Phytophthora parasitica 130-170 (nucleocapsid) 167
Schizochytrium aggregatum 100-11 0 (nucleocapsid) 175
Thraustochytrium sp. 130 x 280 (110, nucleo- 176,177
capsid)

Head and tail Ascomycotina Hemiascomycetes Saccharomyces carlsbergen- 70-80 (head), 178
(bacteriophage sis 70 (tail length)
type)
Deuteromycotina Blastomycetes Candida tropicalis 50 x 60 (head) 179,180
Rhodotorula glutinis 20 x 30; 50 x 60 (head) 179
Hyphomycetes Penicillium brevicompactum Head, 45 nm with long non- 181,182
contractile tail
Head, 53 nm with short tail 181,182

-
Isometric, gemi- Ascomycotina Pyrenomycetes Neurospora crassa 20 (monomer),20 x 30 (di- 118,183
nate mer)
{ ;.l
 Table 3 (continued)
MORPHOLOGICAL TYPES OF VIRUS-LIKE PARTICLES DETECTED IN FUNGI
-
.j:>.

Fungi in which detected

~
i
Particle diameter (nm)
Division or or dimensions
Morphology subdivision Class Genus and species (width x length, nm) Ref.
..,.~
Isometric Ascomycotina Discomycetes Diplocarpon rosae 32-34 184 2-
Peziza ostracoderma 25 185 ~
Hemiascomycetes Sat;charomyces cerevisiae 35-40 110,186-189 ~

S.ludwigii 60 179
Saccharomycessp. 100 190-192
Yarrowia (Saccharomycop- 50 193
sis) lipolytica
Loculoasco- Cochliobolus miyabeanus 30 152
mycetes
Pyrenomycetes Erysiphe graminis 40 152
Gaeumannomyces graminis 27,29 194,195
35 128,196-200
40 198-200,201
Hypoxylon multiforme 36 202
Microsphaera mougeotti 32 152
M. poligoni NR 152
Neurospora crassa 20 183
30 203
60 173
80 204
120-130 173
Sphaerotheca fuliginea NR 152
Basidiomycotina Hymenomycetes Agaricus bisporus 19 63,156-158,205
25 56,63,154-158,160
29 56,63,154-158,160
34 63,154-158,160
50 60,61,63
A. campesms 25 60
29 60
50 60
 Armillarea mellea 30 105
Boletus edulis 28 149
32 149
42 149
Boletus sp. 50 60
Coprinus lagopus 130 206
Corticium roIfsii 28 152
43 152
Inocybe dulcamara 30 165
Laccaria amethystina 28 207
L. laccata 28-30 165,207
Len tin us edodes 25 64,208
30 64,150,152,208
36 64,136,150,152,208
39 208-210,216
45 136,152
L. lipideus 32 136
Schizophyllum commune 130 211
Teliomycetes HemiIeia vastatrix NR 143,214
Puccinia aIIi 40 152
P. coronata NR 152
P. graminis 38 215
P. helianthi 35 143,217
P. horiana 40 152
P. malvacearum 34 146,218
P. miscanthi 40 152
P. recondita 40 152
P. sorghi 35 143
P. striiformis 34 146,218
P. suaveolans 34 146
40 152
P. triticina 40 152
TiIIetiopsis sp. 40 116
Uromyces duras 40 152
U.lopecuri 40 152

-
U. phaseoli 35 143
Ustilago maydis 41 111,219
VI
 Table 3 (continued) ..-
MORPHOLOGICAL TYPES OF VIRUS-LIKE PARTICLES DETECTED IN FUNGI 0'1

Fungi in which detected

Particle diameter (nm) ~
Division or or dimensions ::J
Morphology subdivision Class Genus and species (width x length, nm) Ref. ~....
$
Isometric (conL) Deuteromycotina Blastomycetes Candida albicans 12
18
220
220 e:
C. tropicalis
28-30
150
220
221-223
~
C. utilis 50 179
Coelomycetes Colletotrichum atramentar- 26 152
ium
C. lindemuthianum 224,225
Hyphomycetes Arthrobotrys sp. 25-30 180
Alternaria ten uis 30-40 226,227
Aspergillus awamori 25 180
A. flavus 30 228-230,310
A. foetidus 40-42 90,231-233
A. glaucus 25 60
A. niger 40-42 90,234
A. ochraceous 34 235
Cephalosporium acremon- 30 236
ium (= C. chrysogenum =
Acremonium chrysogenum)
Coremiella (Briosia) cubis- 30 237
pora
Fusarium oxysporum f. sp. NR 146,238
lini
F. moniliforme 40 116
F. roseum f. sp. culmoreum 25 124,146
Geotrichum candidum 40 238,239
(anamorph of Endomyces 40 332
geotrichum)
Gonatobotrys sp. 30 240
Helminthosporium car- 39 241
bonum
H. maydis 39 242
48 243
H. oryzae 25 244
H. sacchari 18 152
30 152
45 152
H. victoriae 40 113,116,245,146
Histoplasma capsula tum 60-66,80,115 108,580
Mycogone perniciosa 42 148,247
Ostracoderma (Chromelos- 25 144
porium) sp (anamorph of 28 144
Peziza ostracoderma) 36 144
40 145
Paecilomyces sp. 37 180
Penicillium brevicompactum 40 248,262
30 302

P. chrysogenum 23 249
35-40 85-88,152,250-262
P. citrinum 20 264,265
30 264,265
P. cJaviforme 25-30 266
50-70 266
P. cyaneo-fulvum 35 89,254
P. funiculosum 25-30 72
P. multicolor 32-34 228
P. notatum 25 60
P. purpurogenum NR 267
P. stoloniferum 25-30 72,79,235,268-271
P. varia bile 45-50 263
Periconia circinata 32 241,272
Phialophora graminicola 30 128,309
(anamorph of Gaeumanno-
myces cylindrosporus)
Phialophora sp. (lobed hy- 27 128,309
phopodia) (anamorph of 35 128,308,309
Gaeumannomyces graminis 40 128,309
var. graminis) -..l
Pyricularia grisea 36-45 273
 Table 3 (continued)
MORPHOLOGICAL TYPES OF VIRUS-LIKE PARTICLES DETECTED IN FUNGI -
00

Fungi in which detected

Particle diameter (nm)
~
::s
Division or or dimensions
Morphology subdivision Class Genus and species (width x length, nm) Ref. ~
$
Isometric (cont.) Deuteromycotina Blastomycetes
(cont.) (cont.) P.oryzae 25
30
274
276,303
~
~
35-36 275,277
45 276,303
Rhizoctonia solani(ana- 33 212
morph of Thanatephorus 25-31 213,554
cucumeris) 30 172,278,279,306
Sclerotium cepivorum 45 172,278,279,306
Stemphylium botryosum 25 60
Thielaviopsis basicola (= 40 152,280
Chalara elegans)
Tricothecium roseum 45 393
VerticiIIium dahlia NR 307
V. fungicola 35 145,146
48 145,146
V. malthousei NR 307
Mastigomycotina Chytridiomycetes Allomyces arbuscula 40 281-284,304,305
Rhizidiomyces sp. 60 285,286
Oomycetes Albugo candida 200 373
Phytophthora infestans 60-95 166,167,180
Sclerophthora macrospora 30-32 287,288
32-35 287,288
Thraustochytrium aureum 150 289
Plasmodiophoro- Plasmodiophora brassicae 43 290
mycetes
Myxomycota Acrasiomycetes Guttulinopsis vulgaris 60 291
Labyrinthulales Labyrinthomyxa marina (= 46-53 292
Dermocystidium marinum)
Zygomycotina Trichomycetes Paramoebidium arcuatum 105-110 293
 Uncertain Aphelidium sp. 200 294
taxonomy
Retrovirus-like Ascomycotina Hemiasco- Saccharomyces cerevisiae 60 584
mycetes
Unspecified mor- Ascomycotina Pyrenomycetes Daldinia sp. NR 116
phology
Hypoxylon sp. NR 116
Basidiomycetes Hymenomycetes Hypholoma sp. NR 116
Polyporus sp. NR 116
Thanatephorus cucumeris NR 116
Deuteromycotina Blastomycetes Kloekera sp. NR 116
Hyphomycetes Botrytis sp. NR 116
Chrysosporium sp. NR 116
Gliocladium sp. NR 116
Gliomastic sp. NR 116
Paecilomyces sp. NR 116
Scopulariopsis sp. NR 116
Tricothecium sp. NR 116
Double-stranded Ascomycotina Hemiascomycetes Saccharomyces capensis 331
RNA, but not
VLPs, reported S. diastaticus 331
S. uvarum 331
Basidiomycotina Teliomycetes Puccinia hordei 594
P. recondita 594
P. striiformis 594
Deuteromycotina Coelomycetes Colletotrichum faIcatum 124
C. gramicola 124
Hyphomycetes Aspergillus [umigatus 580
Helminthosporium turcicum 124
Penicillium cyc10pium 124
P. funiculosum 124
Torulopsis glabrata 580
Mastigomycotina Oomycetes Pythium butleri 124

Note: TMV = tobacco mosaic virus; NR = not reported.

-
\0
20 Fungal Virology

ards used for calibration. Hence when particles of several different sizes have been
reported from a single fungal species, it is often not clear precisely how many different
VLPs they represent. For example, in Pyricularia oryzae, Spire et al. 303 and
Yarakhiya 276 both describe two sizes of particles with diameters of 30 and 45 nm,
whereas others 27 4.275,277 describe particles with diameters of 25 and 35 to 36 nm. It has
yet to be shown whether these reports refer to four distinct VLPs or whether the two
sizes of particles described in the various laboratories refer to the same two VLPs.
Second, some of the VLPs may not be viruses, but merely normal subcellular con-
stituents. Two examples of subcellular particles or organelles, which may be mistaken
for VLPs in thin section electron microscopy, namely, glycogen granules and gamma
particles, will be discussed.
Some years ago I undertook a study, in collaboration with M. J. Carlile and D. J.
Border,133 to investigate the possibility that viruses might be involved in somatic incom-
patibility reactions between strains of the slime mold, Physarum polycephalum.'34
Plasmodia were extracted by standard virological procedures. A pellet, obtained after
ultracentrifugation, was resuspended and subjected to sucrose density gradient centrif-
ugation. A light-scattering band, sedimenting at ca. 150 to 200S, was obtained. Ex-
amination of this band by electron microscopy revealed numerous ovoid VLPs ca. 50
x 100 nm. Much to our surprise, all the strains of P. polycephalum which we examined,
irrespective of their behavior in incompatibility tests, contained copious amounts of
these particles. However, spectrophotometry revealed no absorption peak at 260 or
280 nm. suggesting the absence of nucleic acid or protein, and we suspected that the
particles might be glycogen granules, a view strengthened when we found that they
were susceptible to a-amylase action. A search through the literature revealed that
glycogen granules of similar morphology to the ones we had found had been described
by Goodman and Rusch. 135 Similar conclusions were reached by Mori 136 with regard
to rosette-type particles from Cylindrocladium scoparium and Lentinus edodes.
Koltin 301 now considers that VLPs found in Schizophyllum commune 211 could have
been glycogen granules.
Another intracellular entity which has been considered as a possible virus is the
gamma-particle of Blastoc1adiella emersonii.3l1 - 317 Although the gamma-particle was
first described nearly 30 years ago by Cantino and Horenstein 311 it is only recently that
its function has been elucidated.316,317 Gamma-particles are formed, during sporoge-
nesis, from electron-dense granules, about 40 nm in diameter, which appear within
cisternae of rough endoplasmic reticulum. These coalesce to form aggregates about
100 nm in diameter which in turn are converted to gamma-particles. In thin section
gamma-particles measure 400 x 550 nm and are enclosed by a trilaminar, outer lipid
membrane. On average the B. emersonii zoospore contains 12 gamma-particles. Par-
ticles resembling gamma-particles have also been detected in other Chytridiomycetes,
namely, Allomyces arbuscula,318 Allomyces macrogynus,319,320 Catenaria anguillu-
lae,321 Coelomomyces punctatus,l22 Coelomycidium simulii,'23 Olpidium brassicae,'24
Phlyctochytrium irregulare,l25 and Rozella allomycis.326
Gamma-particles are composed principally of lipid (570/0) and protein (41 %); how-
ever, small amounts of RNA (0.9 to 1.8%) and DNA (0.2 to 0.3%) have also been
detected. The RNA sediments at 4S and is possibly tRNA. The DNA has a mol wt of
7 to 8 X 10 7, a buoyant density in caesium chloride of 1.687 g m£-l, and a G + C content
of 27 to 29%; it is one of three satellite DNAs that have been detected in the zoospore
of B. emersonii. 312 ,313
Gamma-particles resemble poxviruses in their size and morphology and in the size
and base composition of their DNA.315 This superficial resemblance is probably for-
tuitous. There is no evidence that gamma-particles are self-replicating or infectious.
Hence they are unlikely candidates for viruses.
 21

There is now good evidence that gamma-particles are the progenitors of chitosomes.
Chi to somes are spheroidal vesicles which contain a chitin synthetase complex capable
of forming chitin microfibriis. 327 ,328 Their function is to transport chitin synthetase to
the sites where micro fibrils are assembled at the surface of fungal cells. They are prob-
ably ubiquitous among chitinous fungi, having been isolated from Alloymces, Mucor,
Neurospora, Saccharomyces, and Agaricus Spp.327 Zoospores of B. emersonii lack a
cell wall, but during the period of encystment synthesize a chitinous wall in the absence
of RNA and protein synthesis. 329 ,330 During zoospore encystment gamma-particles alter
in appearance and release numerous vesicles about 80 nm in diameter. These gamma-
particle vesicles migrate to the surface of the spore and fuse with the plasma mem-
brane, an event which coincides with the appearance of the initial cyst wall. In vitro
studies have shown that isolated gamma-particles contain chitin synthetase activity.'14
When gamma-particles are incubated in conditions which induce them to vesiculate
into chitosome-like particles, 70 to 120 nm in diameter, the chitin synthetase activity
increased three-fold.316 There seems little doubt that gamma-particles are the progeni-
tors of chitosomes and that their function is to store chitin synthetase in an inactive or
zymogen state which, following fusion of gamma-particle vesicles (chitosomes) with
the plasma membrane, can be activated to allow encystment of the zoospore to occur
in the absence of de novo RNA and protein synthesis.
The gamma-particle is most properly classified as a subcellular organelle. The func-
tion and mode of replication of its DNA are unknown and remain interesting problems
for future investigation.
Other structures which may be mistaken for VLPs in thin sections of cells include
crystalline aggregates of proteins, e.g., alcohol dehydrogenase in Saccharomyces carls-
bergensis: 13 polyphosphate granules,326 cross-sections of tubular structures of the
Golgi apparatus,180 and membranous structures generated by destruction of mitochon-
dria in sporulating yeast cells. 180
These examples emphasize the need to isolate and characterize VLPs, detected by
electron microscopy, before they can be considered seriously as viruses. Methods for
the extraction and purification of mycoviruses have been reviewed thoroughly by Holl-
ings. 61
In Table 3 the fungal classification of Ainsworth l37 has been employed. This divides
the fungi into two divisions, the Myxomycota (slime molds and related organisms) and
the Eumycota (true fungi).
The Eumycota are those fungi that (with few exceptions) do not possess plasmodia
or pseudopiasmodia and whose assimilative phase is filamentous or unicellular. They
are divided into five subdivisions, according to their mode of reproduction and the
type of sexual spore or spore-bearing structure which they produce «Mastigomycotina,
Zygomycotina, Ascomycotina [Ascomycetes 1, Basidiomycotina [Basidiomycetes]) or
the lack of a sexual state (Deuteromycotina or "Fungi Imperfecti"). In some classifi-
cation schemes the two subdivisions of the lower fungi (Mastigomycotina and Zygo-
mycotina) are grouped together as the Phycomycetes. The Deuteromycotina are prob-
ably mutants which have lost the ability to reproduce sexually. For this reason many
of these fungi have been found to have affinities with sexually reproducing fungi in
other subdivisions. Indeed in many cases sexually reproducing forms (teleomorphs)
and non-sexually reproducing forms (anamorphs) of the same fungus have been de-
scribed. The classification of imperfect fungi under the general umbrella of the Deu-
teromycotina is therefore unsatisfactory, lumping together many unrelated organisms,
but no better system is available at present.

A. Rigid Rods
Rigid rod-shaped VLPs have been described in several species in the Ascomycotina,
22 Fungal Virology

Basidiomycotina, and Deuteromycotina (Table 3) and are probably widespread in the

fungi. Rigid rod-shaped viruses have not been described in animals and are rare in
prokaryotes, only one group (Plectrovirus) having been described. Phages in the Plec-
trovirus genus infect mycoplasmas, consist of short, straight rods of about 84 x 14 nm
with one rounded end and have a genome of circular ssDNA (4500 kbp).341 They do
not appear to be similar to any of the rigid rod VLPs in Table 3 and are considered
later, along with the bullet-shaped and bacilliform VLPs.
In contrast, rigid rod-shaped viruses are common in the higher plants (Angios-
perms). Five groups have been described Cfable 4), four of which have viruses with
genomes of ssRNA and the fifth viruses with dsRNA genomes. The ssRNA viruses
have monopartite (Tobamovirus group), bipartite (Furovirus and Tobravirus group),
and tripartite (Hordeivirus group) genomes. Those with monopartite genomes have
one length of particles, whereas those with divided genomes have particle lengths which
correlate with the sizes of the genome RNA species.
The rigid rod VLPs found in fungi each appear to consist of one particle size and
hence might be expected to have an undivided genome. Some of these are similar in
dimensions to tobacco mosaic virus (TMV). In fact, Yarwood and Hecht-Poinar '41
claimed that TMV -like rods detected in several species of rusts and mildews were
strains of TMV. When conidia from, or extracts of plants infected with, these fungi,
and several additional rusts and mildews in which TMV -like rods were not detected
(Table 5), were inoculated onto Chenopodium quinoa indicator plants, hypersensitive-
like and/or virus-like lesions of variable size were obtained. Some of these lesions,
especially large ones which appeared on the petiole or which ran down the petiole from
the lamina, contained VLPs visible in the electron microscope and produced virus-like
lesions on reassay. Most consistent results were obtained with Erysiphe graminis (bar-
ley powdery mildew), Sphaerotheca lanestris (powdery mildew from oak), and Uro-
myces phaseoli (bean rust), but even with these fungi positive results were obtained in
only 7"10 of tests. Positive serological reactions were obtained with standard TMV anti-
serum and purified virus from rusted bean tissue and leaf tissue of Chenopodium
amaranticolor, C. quinoa, Phaseolus vulgaris, Vigna sinensis, and Nicotiana tabacum
infected with the virus from bean rust and these were correlated with the electron
microscopic observation of large numbers of TMV-like rods. Nienhaus 34 ' reported de-
velopment of local lesions on C. amaranticolor and C. quinoa inoculated with Sepha-
dex G100 fractions of homogenates of conidial suspensions of powdery mildews from
oak and barley. In some cases transmission from the inoculated Chenopodium plants
to other plants resulted in 600 to 1000 lesions per leaf. The infective agent was deduced
to be a strain of TMV by differential host indexing, cross protection tests, and heat
inactivation tests.
One of the problems with the above reports is that contamination with strains of
TMV could not be completely excluded, especially in view of the low proportion of
successful transmissions. Indeed Yarwood and Hecht-Poinar '41 noted that "ordinary"
TMV was propagated in the same greenhouse as their tests and that some contamina-
tions did occur. They also pointed out that the biological properties, e.g., host range
and heat sensitivity, of their fungal TMV-like infective agent were different from those
of "ordinary" TMV. However it has been shown that even single lesion isolates of
TMV consist of a number of different strains and variants 343 and these could be se-
lected on different hosts. Hence contamination could not be completely excluded by
the isolation of strains which differ from common TMV.
Another problem is that the TMV -like rods from rusts and mildews were inade-
quately characterized. Electron microscopy was carried out by a leaf dip method with
palladium shadowing. Precise particle dimensions were difficult to determine and no
direct comparisons with authentic TMV were carried out. If the claims of Yarwood
 Table 4
PROPERTIES OF RIGID ROD-SHAPED PLANT VIRUSES

Pitch of Mol wt of capsid

Length x width helix polypeptide Size of genome
Genus or group Virus (nm) (nm) (X 10-') Genome components (kb) Ref.

Tobamovirus Tobacco mosaic virus 300 x 18 2.3 17.5 ssRNA 6.4 333
Tobravirus Tobacco rattle virus 180-215 x 22 2.5 22 ssRNA 7.2 334
46-114 x 22 1.8-3.2
Hordeivirus Barley stripe mosaic 100-150 x 20 2.5 21 ssRNA 4.2-3.3 335, 336
virus (tripartite
genome)
Furovirus Soil-borne wheat mos- 300 x 20 NO 20 ssRNA 6.5 337, 338
aic virus 100-160 x 20 2.5-3.5
Furovirus Beet necrotic yellow 390 x 20 2.6 21 ssRNA 7.1" 339, 340
vein virus 265 x 20 4.8
100 x 20 1.8
85 x 20 1.5
Tobacco stunt group Tobacco stunt virus 300-360 x 18 5 48-52 ssRNA 6.7" 295-
(2 bands) 298, 346

Note: NO, not determined.

Number of RNA components required for virus replication is not known for these viruses.

N
t.;J
24 Fungal Virology

Table 5
DETECTION OF TOBACCO MOSAIC VIRUS-LIKE PARTICLES AND
INFECTIOUS AGENTS FROM RUSTS AND POWDERY MILDEWS

Passage of
infective agent through
TMV-like Chenopodium
Fungus Plant host of fungus rods detected quinoa

Coleosporium asterum Aster chilensis + +

C. madiae Madia sativa + +
Erysiphe graminis hordei Hordeum vulgare + +
E. graminis avenae A vena sativa + NO
E. polygoni Phaseolus vulgaris +
Frommea obtusa duchesneae Duchesnea indica NO +
Kunkelia nitens Rubus vitifoIia +
Phragmidium sp. Rosasp. +
PhyJIactinia corylea Plantanus acerifolia +
Puccinia iridis Iris xiphioides NO +
P.oxaIidis Oxalis pes-caprae +
P. pelargonii-zonalis Pelargonium domesticum +
Sphaerotheca lanestris Quercus agrifoIia + +
Uromyces phaseoli typica Phaseolus vulgaris + +
U. phaseolis vignae Vigna sinensis + +
U. fabae Vida faba + +
U. polygoni Polygonum aviculare +

Note; NO, not determined.

Table adapted from Yarwood and Hecht-Poinar.'4'

and Hecht-Poinar '41 and Nienhaus 343 are to be substantiated the VLPs from the rusts
and mildews will need to be directly extracted and purified and compared with similarly
purified virus recovered after inoculation of plants with purified fungal VLPs. Such
comparisons would require precise particle size measurements, determinations of the
mol wt of their capsid polypeptides and nUcleic acids, and serological comparisons, as
well as bioassays.
There is now little doubt that rigid rods exist in at least some of the fungi examined
by Harwood and Hecht-Poinar. 141 McDonald and Heath 143 detected rigid rod-shaped
particles not only in extracts of uredospores, germ tubes, and pustules of the cowpea
rust fungus Uromyces phaseoli var. vignae but also in serial thin sections of vacuolate
intercellular hyphae (from pustules at the "green-fleck" stage and from sporulating
pustules). These rods usually had one end butted to spherical vesicles and were ran-
domly distributed on their surface, so that in thin section they appeared to project in a
radial manner from the outside of the vesicle. They were never observed in germ tubes
or during the early stages of the formation of infection structures (appressoria, substo-
matal vesicles, and infection hyphae). These rods are probably the same as those ob-
served in the same isolate of cowpea rust fungus by Yarwood and Hecht-Poinar. 141
MacDonald and Heath 143 showed clearly that they are not strains of TMV by direct
comparison of the two on the same electron microscope grid. The rigid rods from
cowpea rust fungus measured 260 x 12 nm and were considerably shorter and narrower
than those of TMV (300 x 18 nm) and indeed narrower than any of the rigid rod viruses
isolated from plants (Table 4). Rods of identical morphology to those from cowpea
rust were also found in sunflower rust, Puccinia helianthi and corn rust, P. sorghi;
these are also clearly not strains of TMV.
More recently Yarwood 344 reported that four plant pathogenic fungi, Pseudoperon-
 25

ospora cubensis, Sphaerotheca fuliginea, Thielaviopsis basicola, and Uromyces phas-

eoli, when inoculated onto cucumber cotyledons (Cucumus sativus), gave rise to mos-
aic symptoms remarkably similar to those caused in the same plants by tobacco
necrosis virus and tomato bushy stunt virus. Whether the symptoms were caused by
the fungus per se or by a virus originating from the fungus is not known. However,
the whole subject of virus-like symptoms produced by fungal pathogens would merit
further careful study.
Dieleman-van Zaayen l39 detected rigid rod-shaped VLPs (350 x 17 nm) in Peziza
ostracoderma (Plicaria fulva), which appeared superficially similar to TMV in mor-
phology. In thin sections of apothecia crystalline arrangements of the particles were
found in vacuoles and sometimes the cytoplasm in cells just below the asci. Sectioning
of the crystalline aggregates at different angles and at different levels showed rods
alternately in transverse and in longitudinal array, as well as cross-hatched patterns,
similar to those seen in plant cells infected with the aucuba strain of TMV. 140 However,
the P. ostracoderma rods did not react with antiserum to TMV and optical diffraction
patterns showed that the particle structure is helical with a basic pitch of 2.7 nm which
is larger than the 2.3 nm pitch of TMV. Furthermore, it was found that the structure
did not repeat after three turns as precisely as that of TMV. Since the Peziza rods
appeared to resemble the aucuba strain of TMV in its intracellular aggregates, optical
diffraction and X-ray diffraction studies were carried out on the latter. However, it
was found that the aucuba strain of TMV is very like that of the common strain of
TMV both in pitch and helical parameters (3-turn repeat).345 Attempts to infect a range
of plants, known to be hosts to strains of TMV, i.e., Nicotiana glutinosa, N. tabacum,
Phaseolus vulgaris, Gomphrena globosa, and Chenopodium amaranticolor did not in-
duce any visible reaction in these plants. Hence apart from a superficial resemblance,
the Peziza rods differed distinctly from TMV.
There is no information available at present on the chemical composition of any of
the fungal rigid rod-shaped particles. It is not even known if any of them contain
nucleic acid and there are no reports of reinfection of VLP-free fungi with them. The
rigid rods found in rusts, mildews, and Peziza were not associated with any abnormal-
ities in the host fungus. Low concentrations of rigid rods have been detected in diseased
mushrooms (Agaricus bisporus). It has not been possible to reinfect healthy mush-
rooms with them and their role in mushroom virus disease is unknown. In view of the
likely widespread occurrence of rigid rods in fungi further research into their proper-
ties, particularly to see if they contain nucleic acid, would be justified.
Reports that some fungi can be infected experimentally with TMV are considered in
Section IV.

B. Flexuous Rods
Flexuous rod-shaped particles have been found in species of fungi in the Ascomy-
cotina, Basidiomycotina, Deuteromycotina, and Mastigomycotina (Table 4) and like
the rigid rods are probably of widespread occurrence in the fungi. There are no recog-
nized groups of flexuous rod-shaped animal viruses and only one in bacteria (the Ino-
virus genus of the family Inoviridae) members of which have genomes of circular
ssDNA (Table 6). Filamentous viruses are common in higher plants, there being five
recognized groups (Table 6), all with genomes of ssRNA. Four of the groups contain
viruses with undivided genomes. The fifth group (rice stripe virus group) has four RNA
species, although it has been suggested 354 that only the largest may be needed for infec-
tivity. Viruses in the rice stripe virus group are unusual in that the infective component
consists of filamentous particles, 8 nm in width, of indeterminate length and occasion-
ally branched. These are composed of supercoiled ribonucleoproteins, 3 nm in width.
McDonald and Heath 143 distinguished two types of flexuous rods in extracts of the
26 Fungal Virology

cowpea rust fungus, Uromyces phaseolis var. vignae; long narrow-diameter flexuous
rods (NFR), 660 x 10 nm, and long wide-diameter flexuous rods (WFR), 740 x 16 nm.
The NFRs were only moderately flexuous and usually a central canal was visible.
Sometimes small aggregates in parallel array were observed. In thin section, NFRs were
most commonly detected in germ tubes and during the early stages of the formation of
infection structures (appressoria, substomatal vesicles, and infection hyphae) and were
always found in tight parallel arrays. The dimensions of the NFRs are fairly close to
those of members of the Carla virus and Potyvirus groups of plant viruses (Table 6).
The WFRs were more flexuous than the NFRs. They were rather wider and had a
higher sedimentation coefficient (ca. 1905) than any of the flexuous plant viruses in
Table 6. No central canal was detected but a relatively loose structure with a large pitch
was suggested by the ease with which the stain penetrated between the rows of subunits.
Parallel arrays of WFRs were not found, but sometimes tangled masses were observed.
WFRs were unstable in phosphotungstic acid which caused unwinding of the helix at
one end, both ends and internally. They were more stable in uranyl acetate, but broken
particles were seen frequently; sometimes the broken pieces seemed to be attached by
a thin strand, suggesting the presence of nucleic acid. In thin section, WFRs were
found scattered in the more vacuolate regions of the cytoplasm and in senescing cells
with broken tonoplasts. Neither NFRs nor WFRs were associated with any abnormal-
ities in the fungus.
Huttinga et al. 149 isolated flexuous rods from both healthy and diseased isolates of
the wild-growing, but edible, mushroom Boletus edulis. The particles were morpholog-
ically very similar to those of potato virus X, but further information is required to
determine if the two are related.
Filamentous particles 1500 x 15 to 17 nm with a clear axial canal have been detected
in extracts of the shiitake mushroom, Len tin us edodes, by several investigators in Ja-
pan.64.136.150-IS2 In thin section the particles were observed in the cytoplasm and in vac-
uoles. The rods were shown to contain protein with one major polypeptide species of
mol wt 23,000. Hence in length and polypeptide mol wt the particles resemble closter-
oviruses. However, they are significantly wider than closteroviruses and indeed than
any of the flexible filamentous plant viruses (Table 6). Flexuous rods of dimensions
similar to those of the L. edodes particles have been detected in extracts of Erysiphe
polygoni, Collybia peronata, L. lipideus, and Helminthosporium sacchari (Table 3).
Interestingly the H. sacchari and L. edodes particles have been shown to be serologi-
cally related. ls2 This represents a relationship between VLPs in two fungi of very dif-
ferent taxonomic position since H. sacchari (Deuteromycotina) is the anamorph of
Cochliobolus miyabeanus (Ascomycotina), whereas L. edodes is classified in the Basi-
diomycotina.
Very long tubular particles (2000 x 19 nm) have been detected in the chytrid fungus
Synchytrium endobioticum, the infectious agent of potato wart disease. These helical
particles have an electrontransparent core, 14 nm in diameter, with ca. 17 turns per
100 nm. Although similar in length to the plant closteroviruses they are much wider
and do not resemble virions of any known virus. Some similarity to the helical nucleo-
capsids of the animal paramyxoviruses 357 is evident, but the latter although similar in
width are only about half as long as the Synchytrium particles. In thin sections the
particles are seen as small aggregates or as larger inclusions of tangled tubules. A close
association between the particles and disintegrated mitochondria and lipid bodies was
discerned. The particles apparently migrate through both the nuclear envelope and the
bounding plasmalemma of very young naked sporangia. No VLPs could be detected
in the surrounding cells of the host potato plant.
It is clear that several different types of flexuous rod-shaped particles exist in the
fungi and, like the rigid rods, they are probably widely distributed. It may well, in the
 Table 6
PROPERTIES OF FLEXUOUS ROD-SHAPED PLANT AND BACTERIAL VIRUSES

Mol wtof
Pitch of capsid
Length x width helix polypeptide Size of genome
Genus or group Virus (nm) (nm) (x 10-') Genome components (kb) Refs.

Potexvirus Potato virus X 470-580 x 13 3.4 18-23 ssRNA 6.4 348

Carlavirus Carnation latent 600-700 x 13 3.4 32 ssRNA 8.1 349
virus
Potyvirus Potato virus Y 680-900 xiI 3.4 32-36 ssRNA 9.0-10.5 350
Closterovirus Sugar beet yellows 600-2000 x 12 3.4-3.7 23-27 ssRNA 6.6-14.1 351
virus
Rice stripe virus Rice stripe virus Variable ND 32 ssRNA 5.7,4,2, 1.0,0.9 352-
group length x 8 355
Inovirus Phage fd 760-1950 x 6 1.6 5 ssDNA 5.7-8.1 356

IV
...:J
28 Fungal Virology

future, be possible to place some of these into established groups of plant viruses, while
new groups will be required for others. However, all of these VLPs are in urgent need
of much more detailed characterization, particularly with regard to their putative nu-
cleic acid components.

c. Bacilliform and Bullet-Shaped Rods

A number of rod-shaped fungal VLPs have been described which have straight par-
allel sides, but which differ from the rigid-rod shaped particles described in Section
II.A in having rounded ends (Table 3) and probably also differ in their mode of con-
struction. Whereas the rigid and flexuous rod-shaped viruses have protein subunits
arranged in helical symmetry, it is likely that most tubular viruses with rounded ends
have structures based on icosahedral end-caps. The geometrical principles involved in
the construction of such tubes have been reviewed by Hull. 358 To facilitate comparisons
properties of tubular particles with one rounded end (bullet-shaped) and two rounded
ends (bacilliform) from other host taxa are summarized in Table 7. Some icosahedral
viruses also give rise to tubular structures in infected cells. 358
Hull 358 has proposed a model for the structure of alfalfa mosaic virus (AMV) based
on a T = 1 sphere. He pointed out that a 12 morphological subunit sphere with subunits
of about 9.6 nm would have a diameter of 18 nm, as observed for AMV particles. A
series of models was envisaged starting with a sphere of 12 pentamer morphological
subunits (60 polypeptides) and increasing in steps of three hexamer morphological sub-
units (18 polypeptides). From this series and from their physical data possible struc-
tures of the various components of AMV were suggested. The structure of mycoplasma
virus MVL51 may also be based on a T = 1 icosahedron?·5
The structures of enveloped tubular viruses are more complex, since the enveloped
particles are bacilliform or bullet-shaped but the nucleocapsids may have different
symmetry. For example in the Rhabdoviridaethe nucleocapsids are helical (700 to 1000
x 20 nm when uncoiled) and linked to the envelope, in which the G (glycoprotein)
subunits are embedded, by the M (matrix) protein. Optical diffraction patterns suggest
that the M and G subunits are hexagonally arranged. Possible model structures have
been discussed by Hull. 358 In a DNA virus isolated from the honey-bee, Apis meJli-
[era,'·2 the nucleocapsid could be uncoiled to form long filaments 3000 x 85 nm. In the
Baculoviridae one or more nucleocapsids may be enclosed in a single envelope. The
covalently closed circular dsDNA is of low superhelical density; its packaging into rod-
shaped nucleocapsids reflects its interaction with an argine-rich, lysine-free protein,
probably encoded by the virus DNA?·· Viruses in the "Polydnaviridae" may be simi-
lar, although they have several DNA segments, and the nucleocapsid may be enclosed
by two envelopes, one acquired de novo in the nucleus, and the second, outer envelope
acquired by budding through the plasmalemma.
The best studied of the bacilliform particles from fungi is mushroom virus 3 (MV3)
from Agaricus bisporus (Table 3). This has non-enveloped particles, 50 x 19 nm. Op-
tical diffraction patterns of electron micrographs of such particles are similar to those
obtained from micrographs of alfalfa mosaic virus (AMV).358 The lattice spacing from
these diffraction patterns suggested a morphological subunit diameter of ca. 10 nm
which is consistent with a tubular structure based on a T = 1 sphere. Furthermore both
viruses have a single capsid polypeptide species of mol wt 24,000. ,.3 Because of these
structural similarities, Hollings·' suggested that MV3 should be placed in the AMV
group. However AMV and MV3 are serologically unrelated'·2 and there is still no
evidence that MV 3 has a tripartite genome or more than one length of particle, as does
AMV. Lapierre and co-workers,·,,3.7 detected two species of ssRNA (ca 8.7 kb and 2.6
kb) in an impure preparation of MV3. However, using a highly purified preparation
of MV3 from a caesium sulfate density gradient, Tavantzis et al. ,.3 showed that MV3
 Table 7
PROPERTIES OF BULLET-SHAPED AND BACILLIFORM PARTICLES FROM ANIMALS, BACTERIA AND
PLANTS

Property Enveloped Non-Enveloped

Virus family Baculoviridae Rhabdoviridae Polydnaviridae Inoviridae "Tricornaviridae"

Morphology Bacilliform" Bacilliform or Bacilliform" Bacilliform Bullet-shaped Bacilliform •

bullet-shaped

Dimensions 110-320 x 30-50 130-380 x 50-90 330 x 84 450 x 150 84 x 14 58 x 18

(length x width, nm) 48 x 18
36 x 18
28 x 18 (ellipsoidal)
Genome Circular dsDNA Linear (-)ssRNA Circular dsDNA Linear dsDNA Circular ssDNA Linear (+)ssRNA
Size of genome Single component, Single component, Polydisperse, Single component, Single component, Tripartite genome
components (kb) 100-150 10.5-14.0 2.3-12.0 18.0 4.5 3.3
2.4
2.1
0.9 (subgenomic)
Hosts Arthropods (insects, Vertebrates, inverte- Parasitoid Hyme- Honey bees Mycoplasmas Plants
arachnids, crusta- brates, plants noptera
ceans)
Examples Autographa califor- Vesicular stomatis vi- Hyposoter exiguae Apis filamentous MVL51 Alfalfa mosaic virus
nica nuclear poly- rus, lettuce necrotic virus virus
hedrosis virus, Tri- yellows virus
choplusia ni-
granulosis virus
References 359 360 361 362 341 358, 363, 364

Refers to single nucleocapsids with a single envelope. Virions may be much larger due to possession of two or more nucleocapsids or a double envelope.
Not all members of the "Tricornaviridae" are bacilliform.

N
\0
30 Fungal Virology

contains a single major ssRNA species of ca. 4 kb; this probably corresponds to the
2.6-kb RNA detected by Lapierre et al.!3!,367 When translated in vitro in a reticulocyte
lysate system, the 4-kb RNA gave rise to a major polypeptide of mol wt 77 ,000 and
several minor polypeptides, confirming th(: RNA to be the positive strand. !64 Since the
77K polypeptide accounts for only about half of the coding capacity of the 4-kb RNA,
this could represent a significant difference from AMV RNAs 1 and 2 which both give
rise to in vitro translation products corresponding to the total genetic information
present in these RNAs. Infection experiments would be required to confirm that MV3
has only one ssRNA component. If confirmed this would exclude MV3 from the alfalfa
mosaic virus group and the proposed "Tricornaviridae" family. 364
MV3 has never been found to occur alone in mushrooms. It is usually accompanied
by two isometric dsRNA viruses, MVI (diameter 25 nm), and MV4 (diameter 34 to 35
nm). Although MVI and MV4 are serologically unrelated to MV3,!62 it is noteworthy
that the capsid polypeptide molecular weights of MV I and MV3 are the same (24,000).
In some dsRNA mycoviruses, and possibly all, (see Section II for further details) full
length ssRNA transcripts are encapsidated as a stage in the virus replication cycle and
can be isolated from virus preparations. MVI is reported to contain two dsRNAs each
about 2 kbp in length. !62 Estimates for the size of MV3 ssRNA are ca. 4 kb (by gel
electrophoresis)!63 and ca. 8.7 kb and 2.6 kb (by sedimentation analysis). !6!,367 Al-
though both estimates were made under nondenaturing conditions and are therefore
not precise they are both larger than would be expected for transcripts of MV 1 ds-
RNAs. However, further comparisons between MVI and MV3 by nucleic acid hybrid-
ization analysis would be worth while. A number of investigators have also noted iso-
metric particles of diameter 19 nm in mushroom virus preparations. 63 ,1S6 !58,205 These
are of interest because they have the same diameter as MV3 and it has been
suggested 60 ,6! that they are merely fragments of MV3 viewed end on. Until the 19 nm
diameter particles have been isolated and characterized this question will not be re-
solved. Possible roles of the various particles in mushroom virus disease are considered
in Section III.
Bacilliform particles with dimensions similar to those of MV3 have been detected in
Microsphaera mougeotti, !52 but it is not known if the two VLPs are related. Bacilli-
form particles found in the field mushroom, Agaricus campestris,60 are probably re-
lated to MV3. Bacilliform particles, 35 x 17 nm, isolated from Verticillium fungicola,
unlike AMV and MV3, were penetrated by phosphotungstic acid and were considered
to be stabilized by protein-protein interactions. !45 They always occurred in association
with one or both of two isometric particles, diameters 35 and 48 nm, but it is not
known if they are related to either of these. Bacilliform rods, 119 x 22 to 28 nm,
isolated from fruiting bodies of Armillaria mellea and other Hymenomycetes'65 bear-
ing hymenoform proliferations, are clearly of a different type; whether they are the
cause of the abnormalities in these fungi is unknown.
In addition to the non-enveloped particles described above, two types of enveloped
bacilliform VLPs have been described in fungi. Particles similar in appearance to the
invertebrate baculoviruses were detected in large numbers in hyphae of the entomopar-
asitic fungus, Strongwellsea magna, growing in the fly Fannia canicularis. 347 The VLPs
(390 x 100 nm) consisted of a densely staining core (350 x 50 nm) within a poorly
staining envelope. Apparent stages of core envelopment were observed and occurred
adjacent to the sporophore cell wall. The plasma membrane proliferated in these areas
and either directly enveloped cores or formed vesiculate structures which subsequently
enveloped cores. The particles were more similar to baculoviruses than to rhabdovi-
ruses, but the difficulties of working with the S. magna - F. canicularis system have
precluded their more extensive examination including isolation, biochemical character-
ization, and experimental transmission.
 31

Tubular particles of indeterminate length, approximately 50 nm in diameter, were

detected in nuclei of Phytophthora infestans by Corbett and Styer. [66,[67 Some
obliquely sectioned particles were seen to have a closed-rounded end. In cross section
the particles were strikingly similar to rhabdoviruses, but they have never been isolated
to enable further comparisons to be made. The VLPs occurred in interphase and mi-
totic nuclei of hyphae, sporangia, and zoospores, but were not observed in the cyto-
plasm or perinuclear space. Their presence was not correlated with growth rate, race
specificity, or pathogenicity of isolates and they were not detected in seven other spe-
cies of Phytophthora.

D. Club-Shaped Particles
Club-shaped particles were isolated from severely diseased mushrooms (Agaricus
bisporus) by Lesemann and Koening l1l and by Atkey and Barton'77 and are similar in
appearance to VLPs observed earlier in thin sections of diseased mushrooms by AI-
bouy et al.170, [72,368 Such particles contained a double-membrane envelope and an elec-
tron-dense core. Particles of similar morphology have been isolated from some hypo-
virulent strains of the chestnut blight fungus, Endothia parasitica, [68, [69 and shown to
contain dsRNA. They have a high lipid content and seem to lack a normal virus capsid.
None of the isometric particles, characteristic of dsRNA viruses in fungi, have been
detected in any of the isolates containing the club-shaped VLPs. It is well established
that replication of nucleic acids occurs in close association with membranes. 369 Poly-
merases involved in the replication of both animal and plant RNA viruses are mem-
brane-bound 370 ,J7[ and RNA replication is often accompanied by the proliferation of
membranes. 372 These club-shaped particles may therefore be membranous structures
involved in the replication of naked dsRNA. For further discussion of the origin of
these particles, see Chapter 4 of this book.

E. Enveloped Pleomorphic Particles

1. Neurospora crassa VLPs
Several types of VLP have been isolated from slow-growing mutants of Neurospora
crassa. These included isometric particles of diameters 20 nm, 30 nm 60 nm, and 120
to 130 nm, [73,[83,203 geminate particles, 20 x 30 nm, [[8,[83 and pleomorphic particles with
diameters up to 400 nm. m ,[74 The small isometric particles were considered to be un-
related to the slow-growing phenotype 2OJ and the geminate particles are discussed in
section II.H. Attention in this section is focussed on the membrane-bound VLPs found
in the respiratory deficient, extranuclear mutants, "abn-1" and "poky".
Kuntzel et al.[74 isolated polymorPhic particles, 250 to 400 nm in diameter, from the
cytoplasm and from lysed mitochondria of strain "abn-1" and these are probably
similar to the polymorphic particles isolated from the same strain by Tuveson and
Peterson. [73 In thin section electron microscopy the particles were seen to have an
electron-dense "nucleoid" of diameter 120 to 170 nm bounded by one, or in some
cases, two or more envelopes that had the trilamellar "unit membrane" appearance of
most biological membranes. The particles contained 8 to 100/0 RNA, 83 to 85% pro-
tein, and 7% phospholipids. The RNA consisted of one major species which sedi-
mented at 33S but after heating in SDS was converted to a heterodisperse RNA with
two main species sedimenting at 9S and 7S. This suggests that the 33S RNA either
contained hidden breaks or was an aggregate of smaller species. The protein contained
only two major polypeptide species, one of mol wt 95,000 probably located in the
"nucleoid" and another of mol wt 15,000 probably located in the phospholipid mem-
brane. The phospholipid was composed of phosphatidylethanolamine (67.5%), cardi-
olipin (9.8%), phosphatidylinositol (1.8%), phosphatidylserine (9.1 %) and phospha-
tidylcholine (11.8%). Comparison with the constituents of the mitochondrial inner and
32 Fungal Virology

outer membranes suggested that the VLPs did not acquire mitochondrial membranes
by budding, but rather used components of the inner membrane to construct a new
membrane with a simpler protein structure. VLPs containing a 35S RNA have also
been isolated from the "poky" mutant of N. crassa. 390 Their density and sensitivity to
a non-ionic detergent suggested that, like the VLPs from abn-l, they contained lipid.
Furthermore, the VLP RNA had the same electrophoretic mobility as a 35S RNA
which accumulated in the mitochondria, suggesting a mitochondrial origin also for the
"poky" VLPs.
The VLPs found in "poky" and "abn-l" mutants were not detected in wild-type
strains of N. crassa. Although several families of animal viruses, e.g., Orthomyxovir-
idae, Paramyxoviridae, Arenaviridae, and Bunyaviridae,'O have enveloped pleo-
morphic particles and ssRNA genomes, and a superficial resemblance of the poly-
morphic particles from "abn-l" to the swollen form of potato yellow dwarf virus (a
plant rhabdovirus 'O ) was noted,'73 none of these viruses is specifically associated with
mitochondria. Viruses capable of replicating within the mitochondria of a eukaryotic
cell would represent a new and unique group.
The possible relationships of VLPs to the "poky" and "abn-l" phenotypes are of
interest. Clearly the VLPs could be the cause, or the result, of these phenotypes.
"Poky" mutants belong to the group I extranuclear mutants of N. crassa 374 which are
characterized by initially slow and progressively faster growth. They are female fertile
and the trait has been shown to be maternally inherited. 374 The "abn-l" mutant be-
longs to the group III extranuclear mutants of N. crassa 374 which are characterized by
a start and stop growth (stopper mutants). They are female sterile but association of
the "abn-l" mutation (as well as that of "poky') with mitochondria was shown by
transmission of the phenotypes to normal cells by injection of hyphae with mitochon-
dria from the mutant strains. 375 ,415 In heteroplasmons and heterokaryons both "poky"
and "abn-l" mutants are dominant over wild-type strains. 375 ,38I,382 In addition both
mutants exhibit a deficiency of cytochromes aa3 and b. 376 ,377 It is unlikely that VLPs
are the cause of either the "poky" or "abn-l" phenotypes because group I and group
III mutants complement each other in heteroplasmons. 416 They could, however, result
from the mutations.
"Poky" mutants exhibit a deficiency of mitochondrial small ribosomal subunits and
19S ribosomal (r)RNA.378,385 The resulting deficiency of mitochondrial protein
synthesis 379 is sufficient to account for many aspects of the' 'poky" phenotype, includ-
ing the cytochrome deficiency. Pulse-labeling experiments suggested that the deficiency
of small ribosomal subunits could be due to impaired processing and/or instability of
the 19S rRNA.387 Also the residual small subunits in "poky" mutants are deficient in
a number of ribosomal proteins, including S5 (formerly called S4a) which is synthe-
sized in the mitochondria. 38o ,384 Because of these observations 19S rRNA and protein
S5 have been considered as candidates for the primary site of the mutation, but no
alterations in them were detected by gel electrophoresis or fingerprinting. 380 ,383,384,387
Furthermore, although "poky" strains with variations in their mitochondrial DNA
have been described,391 these alterations are not characteristic of all "poky" strains
and their presence has no apparent effect on the "poky" phenotype. Gel electropho-
retic patterns of restriction endonuclease digestions of mitochondrial DNA from most
"poky" strains and wild-type strains are indistinguishable. 382 However, using Sl map-
ping and nucleotide sequencing techniques it has now been shown 417 that "poky" and
other group I mutants contain a 4-bp deletion in the coding sequence for the mitochon-
drial 19S rRNA, just downstream from what would normally be the 5' end of this
RNA. It was proposed that this 4-bp deletion, which apparently results in synthesis of
aberrant 19S rRNAs that are missing 38 to 45 nUcleotides from their 5' ends, is the
primary defect in "poky" and other group I mutants.
 33

The assertion by Turna and Grones 390 that "poky" mutants belong to the class of
mutants with defects in splicing the 35S RNA precursor of mitochondrial 25S
rRNA,386,389 and that the occurrence of VLPs (which contain a 35S RNA) depends on
the accumulation of this RNA is obviously incorrect, since mitochondrial 25S rRNA
and large subunits are not deficient in "poky" mutants. However, the possibility re-
mains that the VLPs accumulate as a result of failure to correctly process a 35S pre-
cursor of 19S rRNA. Akins and Lambowitz 417 have shown that precursors of 19S
rRNA with sizes up to 5.6 kbp accumulate in "poky" mutants. Hence the VLP 35S
RNA could either be the 5.6 kbp precursor of 19S rRNA or an aggregate of smaller
precursor RNAs with each other and/or 25S rRNA. It is noteworthy that Grimm and
Lambowitz 386 showed that the small amount of 35S RNA obtained from wild-type
mitochondria consisted of an aggregate of 19 and 25S RNAs, together with smaller
amounts of separate precursors of 19 and 25S RNAs. More critical experiments are
required to investigate possible relationships of VLP 35S RNA to 25 and 19S RNAs.
Such experiments would include electrophoresis of the 35S RNA in denaturing gels
followed by Northern hybridization with (separate) labeled cDNA probes to 19 and
25S RNAs and, if homology is found, high resolution Sl mapping of the 35S RNA on
the mitochondrial DNA genome.
The "abn-l" mutation differs from "poky" in that alterations in the ratios of mi-
tochondrial small and large ribosomal subunits or 19 and 25S ribosomal RNAs were
not observed. 378 Analysis of mitochondrial DNA from several "stopper" mutants of
N. crassa 392 revealed deletions of up to 24 megadaltons. It was proposed that the
"stop-start" growth resulted from competition between certain defective mitochon-
drial DNAs which have a tendency to predominate and low concentrations of less
defective mitochondrial DNA species which must be retained to sustain growth since
N. crassa is an obligate aerobe. Interestingly the region of DNA retained in the mutants
contained both mitochondrial ribosomal RNA genes and most tRNA genes. It is very
unlikely that the VLPs in "abn-l" and "poky" mutants are identical, since they arise
in mutants with different genetic bases. It is feasible, however, that both are derived
from abberrant mitochondria and that their RNA is derived from mitochondrial RNA.
For further discussion of "poly" and "stopper" mutants see Chapter 9.

2. Saccharomyces sp. VLPs

Membrane-bound VLPs of various sizes with electron-dense cores about 100 nm in
diameter were observed by Lindegren and co-workersl90 192 by electron microscopy of
thin sections of isolates of Saccharomyces (the offspring of a series of hybrids devel-
oped by interbreeding S. microellipsoideus, S. cerevisiae, S. diastaticus, S. carlsbergen-
sis, and S. chodati) which developed lytic plaques and multiple buds. Although poorly
characterized, the particles have some similarity to retroviruses. The possibility that
the abnormal phenotype might have been due to rampant transposition should be con-
sidered (see Section II.J).

F. Particles Similar to Herpesviruses

VLPs which morphologically resemble members of the Herpesviridae familylO have
been detected in the nuclei of three lower fungi (Oomycetes), namely, Phytophthora
parasitica vaL parasitica,'67 Schizochytrium aggregatum, l7S and a Thraustochytrium
Sp.176,177,394 The VLPs from Thraustochytrium sp. have been studied the most. The
extracellular particles (280 x 130 nm) were bounded by a membrane (6.8 nm thick) and
contained, eccentrically placed, an isometric capsid (110 nm in diameter) with an elec-
tron-opaque core. Treatment of the particles with DNase, but not with RNase, re-
moved this electron-opaque core, suggesting that the core consisted of DNA.
The life cycle of Thraustochytrium sp. is relatively simple. 395 Sporangia release zoo-
 34 Fungal Virology

spores which then encyst and develop into progeny sporangia. All zoospores in a single
sporangium are therefore the progeny of one parent zoospore. Ability to produce
VLPs was stable over many subcultures and single spore isolations over a period of
several years. Kazama 394 and Kazama and Schornstein 177 made single-spore isolations
of Thraustochytrium sp. for ten consecutive generations in such a way that continuity
through each generation was maintained through a single spore. At the tenth genera-
tion 75 cultures derived from single spores were all found to be virus productive. This
suggested that the stability of the ability to produce virus in cultures is due to vertical
(intracellular) transmission in all viable zoospores and not to the existence of carrier
cells in a population of noninfected cells.
The VLPs of Thraustochytrium sp. are normally latent in the sense that they have
not been detected in numerous serially sectioned zoospores and sporangia under non-
permissive (i.e., non-VLP-producing) conditions. They are found only when cultures
are subjected to permissive (i.e., VLP-producing) conditions (starvation me-
dium).176,177,394 To achieve this, cultures were grown at 20°C for 36 hr, then flooded
with estuarine water and kept at room temperature. VLPs were observed between 5
and 20 hr after flooding. The primary site of virus replication appears to be the nucleus
in which partially formed particles, as well as mature nucleocapsids (110 nm in diam-
eter) were observed. Nucleocapsids were also observed budding from the nucleus into
the cytoplasm. The cytoplasmic particles were at first surrounded by the two unit mem-
branes of the nuclear envelope, but this envelope was later replaced by a coat of elec-
tron-opaque material. The final envelope of the VLPs appeared to be acquired by
budding into various cytoplasmic organelles, envelopment by Golgi-related vesicles or
during egress from the cell. Virus-productive cells were invariably uninucleate, and
VLP production was accompanied by formation of fibrous inclusions in the nuclei,
margination of the nucleolus, disorganization of mitochondria, inhibition of ecto-
plasmic net formation (thereby preventing adhesion of the cells to a substrate), and
ultimately cell death and lysis. VLPs were found only in about 16070 of the cells (out of
500 examined) under permissive conditions. Induction of the VLP replication cycle is
therefore relatively inefficient, since all single-spore cultures have the ability to produce
VLPs, as discussed above.
Aspects of the morphological development of the Thraustochytrium sp. VLPs and
their inducible latent phase show resemblances to members of the Herpesviri-
dae!O,396,397 However, an unusual feature of the Thraustochytrium sp. VLPs is the
temporary acquisition of the two unit membranes of the nucleus which are later re-
moved in the cytoplasm where the final envelope is acquired by budding through cy-
toplasmic membranes. Herpesviruses characteristically acquire their envelope by bud-
ding of the nucleocapsid through the inner lamella of the nuclear membrane; virus
particles accumulate in the space between the inner and outer lamellae of the nuclear
membrane and in the cisternae of the endoplasmic reticulum and are released by trans-
port to the surface through the modified endoplasmic reticulum. Envelopment of her-
pesvirus nucleocapsids at the cytoplasmic and plasma membranes, however, has also
been observed. 397 It is also noteworthy that the Schizochytrium aggregatum VLPs
which morphologically resemble those of the closely related Thraustochytrium sp. ap-
parently do acquire their envelope by budding through the inner nuclear mem-
brane. 175 ,394 Ultrastructural evidence alone is insufficient for classification of these
VLPs as herpesviruses and further progress will require their isolation and detailed
characterization of their envelope, capsid, and DNA genome.

G. Particles with Heads and Tails

1. VLPs from Yeasts
VLPs with heads and tails have been detected by electron microscopy of thin sections
 35

of several yeasts (Table 3). In Saccharomyces carlsbergensis such particles were found
in abnormal cells which produced multiple buds and asci containing more than four
spores,17B but there is no evidence that these abnormalities were the result of VLP
infection. The only recognized viruses with heads and tails are all bacteriophages with
genomes of dsDNA. lo These have isometric or elongated heads (40 to 180 nm in di-
ameter), based on icosahedral structures, and are grouped into three families based on
their type of tail: Myoviridae, long, contractile tails (80 to 455 nm), e.g., phage T2;
Styloviridae, long, noncontractile tails (64 to 539 nm), e.g., phage A; Podoviridae,
short, noncontractile tail (about 20 nm), e.g., phage T7. The VLPs from Candida
tropicalis,174.lBO and S. carlsbergensis,17B and the larger of the two VLPs detected in
Rhodotorula glutinisJ79 fall within the size ranges of these bacteriophages but none of
the yeast head and tail VLPs has been isolated to enable more detailed comparisons to
be made.

2. Bacteriophages from Cultures oiPenicillium sp. (PB Viruses)

Tikchonenko and co_workers lBl ,lB2 have propagated bacteriophages, obtained from
cultures of several Penicillium spp. (P. brevicompactum, P. chrysogenum, P. cyc1o-
pium, P. nigricans, and P. stoloniierum) and Cephalosporium acremonium, in strains
of Escherichia coli. Phages originating from cultures of P. brevicompactum (PBV-I,
PBV -2, and PBV -3) and P. chrysogenum (PBV -5) were isolated and purified (after
propagation in E. coli) PBV -1, PBV -2, and PBV -3 were the head and tail type of phage
with genomes of dsDNA; PBV-I and PBV-3 had long noncontractile tails and were
similar to phages of the Styloviridae family, whereas PBV-2 had a short tail and was
similar to phages of the Podoviridae family. PBV -5 was a small, isometric phage with
a genome of ssDNA, similar to phages of the ~XI74 type (Microviridae familylO).
Phages were only detected after disruption of fungal mycelium or spores and infectivity
was abolished by incubation with PBV -specific antisera. Contaminating bacteria could
not be detected in the fungal cultures and phage titers (on E. coli) were unchanged
when the fungi were grown in the presence of neomycin. This reduces, but does not
completely eliminate, the possibility that the phage particles were the result of a low-
level contamination by phage-infected bacteria, since the contaminants could be neo-
mycin resistant. Furthermore the contaminant bacteria might be specifically parasitic
on individual Penicillium spp. and both bacteria and phage particles might bind
strongly to the fungal cell walls. This would explain both the failure to culture free-
living bacteria and the inability to detect phage infectivity unless the fungal cells were
disrupted.
The numbers of phage particles obtained from the P. brevicompactum cultures (es-
timated by titration on E. coli) were extremely small, corresponding to only 1 particle
per 10' to 1010 cells (hyphaI compartments).lB2 However, reassociation kinetics of
PBV-I and PBV-3 DNA in the presence and absence of P. brevicompactum DNA
showed about four copies of the PBV-3 genome and about 40 copies of the PBV-I
genome per fungal genome. 398 ,399 On the basis that the total amount of virus genomes
found in the fungus appeared to be 109 to 10 10 times that of virus genomes in the form
of infectious virus, Tikchonenko 182 suggested that the genomes of the PB viruses might
be integrated into the fungal genome and might be spontaneously excised at a low
frequency to give infectious virus, in a similar way to lysogenic phages, such as phage
A. However no evidence has been presented that either PBV -1 or PBV -3 can replicate
in fungal cells. Indeed current knowledge of the differences in prokaryotic and eukar-
yo tic molecular biology suggests that it is unlikely that a prokaryotic virus could
undergo a complete multiplication cycle in a eukaryotic cell (except within a prokar-
yotic endosymbiont; see below). The only authenticated examples in nature where
DNA is transferred from prokaryotic cells to become integrated into the chromosomal
36 Fungal Virology

DNA of eukaryotic cells are the Agrobacterium Ti and Ri plasmid/plant cell sys-
tems.400.401 However the DNA which is transferred from the Agrobacterium Ti plasmid
(the T -DNA) and integrated into the plant chromosomal DNA, is typically eukaryotic
DNA in structure and is expressed in the transformed plant cells but not in Agrobac-
terium. Virulence genes on the Ti plasm ids are typically prokaryotic DNA in structure
and are expressed in Agrobacterium as a response to exudates from plant cells. 402
A more likely possibility is that the PB viruses originate from bacterial endosym-
bionts in the fungus, by analogy with the Paramecium-endosymbiont-bacteriophage
and Hydra viridis/Chlorella/virus systems.40J.411.412 This was considered unlikely by
Tikchonenko l82 on the basis that a fungal cell could not contain as many as 20 copies
of a bacterial en do symbiont (which would be required in the case of PBV -1, assuming
that each symbiont contained two copies of integrated PBV -1 DNA). However a more
likely explanation would be that each fungal cell contained only one copy of a bacterial
endosymbiont. Assuming integration of PB virus DNA into endosymbiont DNA, in-
duction of the virus vegetative replication cycle in a proportion of cells would give rise
to multiple phage particles (an average of 40 per cell in the case of PBV-I). This would
imply that the number of phage particles (as opposed to viral genomic DNA) in fungal
homogenates was grossly underestimated by Tikchonenko and co-workers. 182 This is
not unlikely, since the fungal extracts contained an inhibitor which had to be removed
by chloroform extraction before phage infectivity could be demonstrated and some
inhibition could have remained. Furthermore it was observed that the PB viruses
bound strongly and, apparently selectively, to the cell walls of their "host" fungi.
Hence a majority of phage particles might have been bound to fungal cell wall frag-
ments and not available to infect E. coli. Electron microscopy could resolve this ques-
tion.
The origin of the PB viruses will remain speculative until more definitive experiments
are carried out. It would be comparatively easy to obtain a library of cloned genomic
DNA fragments from P. brevicompactum and to locate clones containing PBV se-
quences by probing with labeled DNA obtained from a library of cloned genomic frag-
ment from purified PBV DNA. Chromosome walking techniques 404 and DNA se-
quence analysis 40S could then be employed to determine whether or not PBV DNA
sequences in P. brevicompactum were integrated into chromosomal DNA.

H. Geminate Particles
Geminate particles, 20 x 30 nm, have been isolated from slow-growing strains of
Neurospora crassa (see Section II.E), although there was no evidence that these VLPs
were connected with the slow-growth phenotype. Bozarth" 8 considered that these par-
ticles were dimers of isometric particles, 20 nm in diameter, which have been shown to
contain RNA.20J However, their appearance suggests that they may in fact be geminate
particles similar to the plant geminiviruses. lo Geminiviruses are of two types, those
transmitted by leafhoppers, e.g., maize streak virus, which apparently have a genome
of one circular ssDNA component of ca. 2.7 kb406.407 and those which are transmitted
by whiteflies, e.g., cassava latent virus and tomato golden mosaic virus, which have
genomes of two circular ssDNA components, each of ca. 2.5 to 2.7 kb.408.409 Both types
of geminivirus have particles of similar morphology (isometric, geminate, with overall
dimensions ca. 20 x 30 nm) which, in the case of Chloris striate mosaic virus,4lO consist
of two incomplete icosahedra with a T = 1 surface lattice and a total of 22 capsomers.
The instability of the Neurospora geminate particles has precluded their complete pu-
rification and hence more detailed comparisons with members of the geminivirus group
have not been made.
 37

I. Isometric Particles
Particles with isometric morphology are by far the most common type of VLP to be
detected in fungi (Table 3). They by no means constitute a homogeneous group, how-
ever, and several types can be distinguished.

1. Double-Stranded RNA Viruses

Isometric particles with genomes of dsRNA are very common in fungi. Particle di-
ameters are most usually within the range 25 to 50 nm although exceptionally values as
low as 20 nm (Penicillium citrinum) 264,265 and as high as 60 to 66 nm (Histoplasma
capsulatum)'08 have been recorded. Mehta et al. 220 isolated spherical particles as small
as 12 nm in diameter (as well as those with diameters of 18 nm and 28 to 30 nm) from
Candida albicans but it was not established that they were complete particles contain-
ing dsRNA. Most of the known isometric dsRNA mycoviruses are from the higher
fungi (Ascomycotina, Basidiomycotina, Deuteromycotina) but one such virus has been
found in the chytridiomycete, Allomyces arbuscula. 281 - 284 ,304,305 The properties of iso-
metric dsRNA mycoviruses are discussed in Section III. An electron micrograph of one
such virus, Penicillium stoloniferum virus, is shown in Figure 1A.

2. Single-Stranded RNA Viruses from Sclerophthora macrospora

Isometric mycoviruses with genomes of ssRNA have so far been found only from a
lower fungus, namely, the rice downy mildew fungus, Sc1erophthora macrospora, an
oomycete. Two serologically unrelated types of virus particles, designated A and B,
were purified from S. macrospora-infected rice plants. 287 ,288 Out of 80 samples col-
lected throughout Japan, 27 samples contained both types, 14 samples contained only
type A, 25 samples contained only type B, and 14 samples contained neither type A
nor B. No particles were detected in healthy rice plants and ultrastructural studies
showed that the virus particles were localized in the fungal mycelia and oospores but
not in plant cells.
Virus A was 32 nm in diameter with characteristic spikes of 4 nm on its periphery
(Figure 1B). It had two capsid polypeptide species of mol wt 39,000 and 43,000 and
three segments of ssRNA with mol wt 1.1 x 106 (3.2 kb), 0.7 x 106 (2.0 kb), and 0.33 x
10 6 (1.0 kb). The amount of the smallest segment varied among different isolates,
suggesting separate encapsidation. Virus B was 35 nm in diameter without spikes (Fig-
ure lC). It had a single capsid polypeptide species of mol wt 41,000 and one ssRNA of
mol wt 1.8 x 10 6 (5.2 kb).
Basic properties of the recognized groups of isometric ssRNA viruses are given in
Table 8. In comparing virus A with these viruses, it has to be borne in mind that no
infectivity studies have yet been carried out and the number of genome RNA compo-
nents is not known. Either one or both of the two smallest RNA components could be
subgenomic or satellite RNAs. lO Similarly, proof that the capsid contains two unigue
polypeptide species, as opposed to one being derived from the other, would require
proteolytic fingerprinting studies. Even with these provisos the only virus group with
any similarities to virus A is the Nodaviridae family which includes viruses infecting
insects (Diptera, Coleoptera, and Lepidoptera). However it is distinguished from no-
daviruses by its characteristic spikes, and hence appears to be distinct from any known
virus. The properties of virus B are similar to those of the plant tombusviruses, al-
though its RNA appears to be slightly larger. Further investigations will be required to
detennine the extent of the affinities, if any, with this group.

3. A Double-Stranded DNA Virus from Rhizidiomyces sp.

Only one non-enveloped isometric mycovirus with a genome of dsDNA has so far
been isolated, i.e., a virus of 60 nm diameter obtained from isolates of the chytridi-
 w
00

~
~......
z·
<::::

..~
~
.'
. '~". ..
. '1;

... I' t
;~
,~.;~
clo4'C ( "
.~. .... . .rl£~'.. •
~L
~~ '~~I

FIGURE I. Electron micrographs of different types of isometric mycovlrus particles and VLPs. (A) PenicIllium stoloniferum
virus S. (B) Sclerophthora macrospora virus A; courtesy of Dr. Y. Shirako. (C) S. macrospora virus B; courtesy of Dr. Y.
Shlrako (D) Thin section of a mature sporangium of Albugo candida shoWIng numerous hexagonal VLPs in the cytoplasm (N
= nucleus); courtesy of Dr. J. L. Gay. (E) Enlargement of a group of a A. candida VLPs showing the densely stained outer layer
of the capsid; courtesy of Dr. J. L. Gay. (F) "Double-shell" virus particles from Lentinus edodes;courtesy of Dr. R. Ushiyama.
The micrographs were stained with phosphotungstate (A) or uranyl acetate (8 to F). The bar represents 50 nm in A, 8, C, E, and
F, and 500 nm in D.
 39

Table 8
PROPERTIES OF ISOMETRIC ssRNA VIRUSES FROM ANIMALS, PLANTS
AND BACTERIA

Capsid polypeptide
ssRNA species species
Virion
diameter Approximate
Family or group Host" (nm) Number size (kb) Number

Maize chlorotic P 30 9.3

dwarf virus group
Picornaviridae A 22-30 7.3 4 5.5-41
Caliciviridae A 35-39 7.8 I 60-71
Tymovirus P 29 6.0 20
Luteovirus P 25-30 6.0 24
Nudaurelia virus A 35 5.2 60-70
group
Tombusvirus P 30 4.4 41
Sobemovirus P 30 4.1 30
Necrovirus P 28 4.1 23
L.eviviridae B 23 3.5 I major 12-14
I minor 35-44
Nepovirus P 28 2 8.1,3.8-7.0 55-56
Comovirus P 28 2 7.0,4.1 2 22, 42
Pea enation mosaic P 28 2 4.9,3.8 I major 22
virus group I minor 28
Dian th 0 virus P 31-34 2 4.4, 1.5 40
Nodaviridae A 29 2 3.3, 1.4 40
Cucumovirus P 29 3 3.7,3.3,2.4 24
Brom 0 virus P 26 3 3.2,2.8, 2.1 20
Ilarvirus P 26-35 3 3.2,2.6,2.0 25

A, animal; B, bacterium; P, plant.

Data compiled from Reference 575.

omycete Rhizidiomyces, parasitic on Oomycetes.28S·286.414 Virus particles were found

only after thermal shock treatment, after which the Rhizidiomyces zoospores would
no longer readily infect the oogonia of two of their normal hosts, Achlya and Sapro-
legnia. Pathogenesis was apparent in all fungal developmental stages from zoospores
to mature sporangia. Within uninucleate encysted zoospores virus particles were first
observed in the nucleus in small clusters, many of which were close to the nuclear
envelope. As the number of particles increased, the mitochondria and other cell organ-
elles became disorganized and the nucleus became very lobed. Eventually most of the
subcellular organelles were destroyed, leaving predominantly virus particles both free
and attached to membranes and fibrillar structures, but the cell wall of the encysted
zoospore remained intact. Some zoospores were able to develop into multinucleate
sporangia before disease symptoms became evident. In these sporangia were paracrys-
talline structures associated with mitochrondria, followed by the appearance of parti-
cles in the nuclei, organelle destruction, replacement of nucleus and cytoplasm by virus
particles, cell wall breakdown, and release of virus particles into the medium. About
50070 of each isolate was destroyed, but sufficient numbers of zoospores developed into
sporangia which produced progeny zoospores to maintain the isolates in culture. The
appearance of the virus particles under conditions of stress (heat, low nutrition, aging)
and their vertical transmission through 10 single-spore generations resemble the prop-
erties of the herpes-like VLPs in Thraustochytrium sp. described in Section II.F.
40 Fungal Virology

The Rhizidiomyces virus has been isolated from culture filtrates and purified by
sucrose density gradient centrifugation. 286 The isolated particles were very similar to
those observed in thin sections of infected cells. They were isometric, 60 nm in diame-
ter, with no detectable membranes or projt!ctions, sedimented at 625S and had a buoy-
ant density in CsC1 of 1.314 g/ml. The viral nucleic acid was shown to be dsDNA
with a mol wt ca. 17 x 106 (26 kbp). Its Tm in standard saline citrate buffer was 86.5°C
corresponding to a G + C content of 42070. Although the topology of the DNA was not
determined, it migrated as a single band in gel electrophoresis, suggesting a linear mol-
ecule. Fourteen capsid polypeptide species were detected with mol wt in the range
84,500 to 26,000, the largest of which accounted for at least 50% of the total protein
of the virions. Basic properties of the recognized groups of icosahedral dsDNA viruses
of animals, plants, and bacteria are given in Table 9. Of these the Rhizidiomyces virus
resembles the animal adenoviruses the most closely. Its particles are of a similar size
and its DNA is only slightly smaller. However, the fibrous projections, characteristic
of adenoviruses, were not observed and the G + C content of the DNA was lower than
the range for adenovirus DNA (48-55%). The Rhizidiomycesvirus therefore probably
represents a new group of isometric dsDNA viruses.

4. VLPs from the Lower Fungi with Diameters in the Range 40 to 200 nm
Apart from the well-characterized viruses of ScJerophthora macrospora and Rhizi-
diomyces sp. described in the preceding sections, several isometric VLPs, covering a
range of sizes, have been detected by electron microscopy in the cytoplasm and/or
nucleus of cells of several lower fungi (Mastigomycotina, Myxomycota, Zygomyco-
tina; Table 3.): e.g., nucleus: Guttulinopsis vulgaris, Paramoebidium arcuatum, Phy-
tophthora infestans; cytoplasm: Albugo candida, Aphelidium sp., Thraustochytrium
aureum; nucleus and cytoplasm: Labyrinthomyxa marina, Plasmodiophora brassicae.
None of these particles have been isolated, so that their viral nature remains in doubt.
However, particles of diameter ca. 200 nm found in the cytoplasm of lysing protoplasts
of Aphelidium sp., an intracellular parasite of the green alga, Scenedesmus armatus,'94
showed some resemblance to some members of the Iridoviridae family, which includes
viruses of vertebrates and invertebrates and which are also assembled in the cyto-
plasm. 10 The finding of similar particles in Albugo candida (Figures 1D and E) suggests
that Aphelidium sp., an organism of doubtful taxonomic position, might have affini-
ties with the Oomycetes. Isometric particles of diameter 40 nm to >200 nm are also
common in eukaryotic algae and protozoa. 417 Few of these have been characterized,
but it is tempting to speculate that affinities could exist between some of the VLPs of
lower fungi and those of eukaryotic algae and protozoa.

J. Retrovirus-Like Particles and Transposition in Yeast

The yeast, Saccharomyces cerevisiae, contains about 30 transposable elements (Ty
elements) which move about its genome by both homologous, i.e., recombinational
events (frequency 10-4 to 10-5) and nonhomologous, i.e., transpositional events (fre-
quency 10-7 to 10-8).582 Insertion of a Ty element into the flanking region 5' to a gene
may result in activation or inactivation of the gene; insertion into the coding region
inactivates the gene. Transposition occurs through RNA intermediates 583 which are
enclosed in virus-like particles (Ty-VLPs) containing reverse transcriptase activity. 584
In the sequence DNA-RNA-DNA, Ty elements resemble retroviral proviruses,585
hepatitis B virus,586 and cauliflower mosaic virus. 587 With Ty elements and retroviruses,
the RNA stage is encapsidated and RNA is transcribed from DNA copies inserted into
the host chromosome. With hepatitis B and cauliflower mosaic virus, the DNA stage
is encapsidated and RNA is transcribed from extrachromosomal DNA copies. Trans-
 Table 9
PROPERTIES OF ICOSAHEDRAL DOUBLE-STRANDED DNA VIRUSES FROM ANIMALS, PLANTS
AND BACTERIA

Virion DNA Polypeptide species

Sedimentation Buoyant density Molwt Replication

Family or group Host" coefficient (S) Diam (nm) in CsCl(g/ml) Size (kbp) Topology' Number (xlO-J) site<

Iridoviridae' A 1300-4450 125-300 l.l6-1.35' 150-375 f 13-25 10-250 N/C

Herpesviridae' A 100-110 120-225 At least 20 12-220 N
Adenoviridae' A 70-90 1.32-1.35 30-31 At least 10 5-120 N
Tectiviridae B 390 65 1.28 12 16-18
Corticoviridae< B 230 60 1.28 9 ccc 4 5-43
Caulimovirus P 208 50 1.37 8 oc 42 N/C
Papovaviridae A 240-300 44-55 1.32 4.5-7.5 ccc 5-7 10-75' N

A, Animal; B, bacterium; P, plant.

£, Linear; ccc, covalently closed circular; oc, open circular.
N, DNA replication, and virus assembly occur in the nucleus; N/C, nucleus is required for DNA replication, but virus assembly takes
place in the cytoplasm.
Some members of this family possess an outer envelope and hence have a lower buoyant density than the non-enveloped members.
Data are given for the icosahedral nucleocapsids, not the virions which are enveloped.
Virions possess characteristic fibers projecting from the 12 vertices of the icosahedron.
Virions have brush-like spikes on the 12 vertices of the icosahedron.
Low mol wt species are cellular histones.

Data from Reference 575.

.f>,
42 Fungal Virology

position via RNA intermediates probably also occurs with copia-like elements in Dro-
sophila 588 and may be common in eukaryotes.
Structurally and functionally, Ty elements are similar to retrovirus proviral DNA.
They consist of a 5.3 kb internal region, epsilon, flanked by direct repeats of ca. 335
bp called delta sequences or long terminal repeats (L TRs). Transcription takes place
from L TR to L TR to produce an RNA that is terminally redundant for 45 nucleo-
tides. Regeneration of Ty elements containing the complete L TRs probably takes place
by reverse transcription in an analogous fashion to synthesis of retrovirus proviral
DNA.585 Adjacent to the 5' LTR there is a sequence that could serve as a tRNA primer
binding site and adjacent to the 3' L TR there is an oligopurine tract that could prime
second strand synthesis. The internal region of Ty elements contains two overlapping
open reading frames: tya encodes a protein with homology to DNA binding proteins
and is probably equivalent to the retroviral gag region; tyb specifies a protein with
homology to the protease, integrase, and reverse transcriptase regions of the retroviral
pol gene. The products of the tyb and pol genes are thought to be synthesized as tya-
tyb and gag-pol fusion proteins resulting from specific translational frameshifts. 589-591
By coupling a genetically tagged Ty element to a yeast {3-galactosidase promoter and
cloning it into a replicating CEN plasmid upon galactose induction the frequency of
transposition is dramatically increased, so that virtually every cell in the population
has multiple transpositions of the marked element from the plasmid onto the chromo-
somes. 583 Such cells grow very slowly on galactose, probably because of an intolerably
high mutation frequency resulting from increased transposition frequency. In these
conditions RNA transcribed from the tagged element constitutes about 5 to 10070 of the
total yeast mRNA and Ty-VLPs are readily visible in electron micrographs of thin
sections of the yeast cells. 584 The VLPs, up to 1000 per cell, are spherical to ovoid in
shape, approximately 60 nm in diameter, and are confined almost entirely to the cyto-
plasm. They are similar morphologically and functionally to mammalian retrovirus
type A particles59 2.593 and to the copia particles found in Drosophila tissue culture
cells. 588 One of the functions of Ty-VLPS is probably to isolate the reverse transcrip-
tase from cellular mRNAs by sequestering it in a particle. Unlike retroviruses there is
no evidence that Ty-VLPs are released from cells or that they can re-infect from with-
out. However, presumably Ty-VLPs could be transmitted between different yeast
strains by cytoduction or mating, leading to synthesis of Ty elements and transportion
in the acceptor strain.

III. THE BIOLOGY AND BIOCHEMISTRY OF ISOMETRIC

DOUBLE-STRANDED RNA MYCOVIRUSES

A. Transmission
DsRNA mycoviruses are unusual in that they do not lyse their hosts and are appar-
ently transmitted only by intracellular routes, within an individual in hyphal growth or
in asexual or sexual spores, and between individuals via heterokaryosis. Unlike most
viruses, there is no evidence for an extracellular phase to their life cycle (but see Section
III. A.4) and there are no known transmission vectors. Furthermore, there is no evi-
dence, as yet, for DNA proviruses in the replication or transmission of any dsRNA
mycovirus,"7.431.432 (however, see Section III. C.l.b). The intracellular mode of trans-
mission and absence of lysis result, not only in the common occurrence of mixed infec-
tions with two or more viruses, but also in the accumulation of satellite and defective
dsRNAs.

1. Transmission during Hyphal Growth

Hyphae of the higher fungi (Ascomycotina, Basidiomycotina, and their anamorphs)
 43

in which most isometric dsRNA mycoviruses occur, are septate and grow by apical tip
extension, largely as a result of fusion of vesicles, carrying cell wall precursors and
enzymes, with the apical plasmalemma 327 The growing tip is only a small proportion
of the apical hyphal compartment, but protoplasm in much of the region in which the
septal pores remain unplugged (the peripheral growth region) contributes to hyphal
growth. 41B Since hyphal tips appear to be either virus-free or to contain very low levels
of virus particles,61.117.419 it appears that virus replication takes place in the distal part
of the peripheral growth zone and virus particles are carried forward towards the tip
in the net flow of protoplasm that occurs during hyphal growth. S~ptal pores, which
allow the transport of subcellular organelles as large as nuclei, 420 will not be a barrier
to the movement of virus particles.
If the septal pore in the apical hyphal compartment became blocked before virus
transmission had occurred, subsequent growth could be virus-free. This could explain
the occasional "self-curing" of some fungi on repeated subculturing. The ease with
which this could happen would depend on the timing of septal pore occlusion. In some
fungi, e.g., Geotrichum lactis, this occurs soon after septa are laid down, whereas in
others septa remain unplugged for some time after their formation so that a number
of hyphal compartments (several in Aspergillus and Penicillium spp., over 150 in Neu-
rospora crassa) remain unplugged 4lB

2. Transmission via Asexual Spores

Transmission of viruses through conidiospores appears to be both general and effi-
cient and has been reported for species of Aspergillus,229 Colletotrichum: 22 Fusar-
ium: 23 Gaeumannomyces,'94.195 Penicillium,72.B9.424.425 and Pyricularia. 274 .426 Virus par-
ticles have been detected by electron microscopy in thin sections of conidia of
Penicillium brevicompactum and P. stoloniferum 427 and have been extracted from
conidia of P. brevicompactum, P. chrysogenum, and P. stoloniferum. 42B Similar amounts
of virus were found in conidia and mycelium. Efficiency of virus transmission into
single conidial isolates of Penicillium and Pyricularia spp. was 90 to 100070.269.426 .• 29.430
In the case of a strain of P. stoloniferum which was infected with two viruseS', Sand
F, conidia contained both viruses (93%), only virus S (2070), or no virus (5%). Levels
of virus in the single conidial cultures varied from much less than to much more than
that of the parent culture. 430 This could explain why mass conidial inoculations re-
tained a constant level of infection. The efficiency of conidiospore infection, compared
with that in hyphal tips, could depend on the initiation of conidiophores from hyphal
regions containing high levels of virus particles and the relatively slow process of con-
idiogenesis, allowing time for virus replication and protoplasmic mixing before the
conidial septal pore is blocked. Virus infection appears to have little effect on conidial
viabiity422 .• 30 At Imperial College, conidia of several Aspergillus and Penicillium sp.
stored in liquid nitrogen have remained viable and virus infected for more than 15
years. Virus particles have also been detected in other kinds of vegetative spore, such
as chlamydospores of Mycogone perniciosa, 147 uredospores of Puccinia graminis f. sp.
tritici,'15 and sclerotia of Sclerotium cepivorum.279

3. Transmission via Sexual Spores

Virus transmission occurs through sexual spores in the Basidiomycotina, e.g., Agar-
icus bisporus,'55,433 Len tin us edodes,'OB and Ustilago maydis.'12 In the latter case, anal-
ysis of single spores indicated that transmission was very efficient. Spore viability is
not reduced as a result of virus infection. Indeed, it has been reported that basidios-
pores produced from infected carpophores of Agaricus bisporus germinated more rap-
idly and abundantly than those from healthy carpophores.62.434 Such spores were still
viable and capable of initiating mushroom virus disease after storing for 3 years at
44 Fungal Virology

room temperature or 5 years at 4°C.435.436 Spore transmission is probably the most

important method of spread of mushroom virus disease. 61
In the only known example of an isometric dsRNA virus in a lower fungus, i.e., in
Allomyces arbuscula (Mastigomycotina), efficient transmission is ensured by the loca-
tion of virus particles in the nuclear cap of gametes, diploid mitospores, and probably
also in haploid meiospores. 283
In Saccharomyces cerevisiae, transmission of dsRNA virus particles into ascospores
is very efficient, possibly even more so than would be expected from the volume of
cytoplasm in the spores, and it has been speculated that some selection mechanism,
perhaps an association with the nucleus, could be operative. 437 In filamentous asco-
mycetes, such as Gaeumannomyces graminis, a positive exclusion mechanism appears
to operate during the sexual stage, so that many (but not all) ascospores are virus-free
or have much reduced levels of virus particles. 438 (See also Chapters 4 and 7 for discus-
sion of transmission of dsRNA, not known to be associated with virus particles, in
Endothia parasitica and Ceratocystis ulmi).

4. Transmission via Heterokaryons and Heteroplasmons

Transmission of mushroom disease by hyphal anastomosis was discussed in Section
I. B.2.b. The disease, and its associated viruses, are spread as a result of anastomoses
between hyphae of a healthy colony and either germinating infected spores (often be-
tween the tip of a newly emerged germ tube and the growing tip of an uninfected
hypha)62 or hyphae of infected mycelium remaining in a mushroom tray from a pre-
vious crop. Once infected, further spread will occur by anastomoses within the newly
infected mycelium.
Transmission of several dsRNA mycoviruses by heterokaryosis using auxotrophic,
colored, or fungicide-tolerant mutants as recipients has been demonstrated for several
fungi, e.g., Aspergillus niger, Colletotrichum lindemuthianum,422 Penicillium chryso-
genum: 40 and P. stoloniferum441 and probably occurs generally in the fungi. However,
hyphal anastomosis is generally limited to individuals within a species. 422 Even within
a species, transmission of cytoplasmic elements, including viruses, is likely to be re-
stricted by vegetative incompatibility. This could take the form either of fusion incom-
patibility (inability of hyphae to fuse), for example individuals in different anastomosis
groups of Rhizoctonia solani,433 in which case transmission would not be possible, or
of postfusion incompatibility, in which case transmission would be restricted. 444
Anagnostakis 445 concluded that efficiency of transmission of dsRNA in Endothia par-
asitica was inversely dependent on both the number of vegetative compatibility (v-c)
gene differences and on the "strength" of individual v-c genes. Similar conclusions
were reached by Brasier446 who found that the frequency of transmission of the
d 2-factor, a cytoplasmic genetic element in Ceratocystis ulmi, was 40/0 when all v-c
genes were different, 50OJo when only one v-c gene was different, and 100010 when all
v-c genes were the same. Use of strains of Agaricus bisporus from different v-c groups
for successive crops has been employed as a method of reducing the spread of mush-
room virus disease. Agaricus bitorquis has been regarded as immune to mushroom
viruses because cultures did not become diseased when inoculated with infected spores
or mycelial fragments of A. bisporuS. 447 However, vegetative incompatibility between
the two species could have prevented virus transmission. Further information on the
effect of vegetative incompatibility on transmission dsRNA and viruses is given in
Chapters 4,6, and 8.
For fungi in which anastomoses between hyphae of different strains are rarely ob-
tained, virus particles can be transmitted from one strain to another by protoplast
fusion. This was achieved for pyricularia oryzae by Boissonet-Menes and Lecoq426 by
fusing protoplasts of a virus-free, benomyl-resistant, glutamic acid-requiring strain
 45

with those of a virus-infected, benomyl-sensitive, nicotinic acid-requiring strain. After

regeneration, prototrophic thalli were never obtained: hence, heterokaryons were rare
or unstable. Nevertheless virus infection was achieved through plasmogeny as demon-
strated by the isolation of virus-infected, benomyl-resistant, glutamic acid-requiring,
single conidial cultures; such cultures were still infected 2 years after the original ex-
periment.

5. Transmission with Cell-Free Virus Preparations

Numerous attempts to infect fungal mycelium with purified virus preparations have
been unsuccessful and it is generally considered that the cell wall is a barrier to virus
penetration. This contrasts with bacteria, which contain specific cell-wall receptors for
attachment of viruses l2 and probably reflects the completely intracellular existence of
dsRNA mycoviruses.
It has been reported that Agaricus bisporus can be infected by injection of partially
purified virus preparations from diseased mushrooms into developing sporop-
hores. 56,59,154 However, the process was very inefficient and difficult to reproduce and
it is hard to be sure that chance contamination from air-borne spores or spores in the
virus preparations did not occur. Apparently as few as one to ten spores are sufficient
to initiate an infection.62 Also, the finding that mushroom spawns can contain at least
low levels of virus particles (see Section III. C.2.b) introduces the possibility that a
latent infection may have been reactivated in the very few cases when infection was
detected. Similar considerations apply to the low frequency of infection of mushrooms
(three out of 168 cultures) using phorid flies which had been allowed to feed on purified
virus preparations. 448
To overcome the cell wall barrier, many attempts have been made to infect fungal
protoplasts with cell-free virus preparations; many of these have not been success-
ful. 426 449,450 Lhoas 451 reported that when protoplasts of Penicillium stoloniferum were
incubated with a mixture of viruses Ps V -S and Ps V -F, 12 out of 20 colonies regenerated
from single protoplasts were infected with PsV-S at a level of about 10070 of that of the
donor strain from which the virus inoculum was obtained. This level was apparently
maintained after two further single conidial subcultures and could be a property of the
white-spored recipient strain, since similar levels were obtained in this strain after virus
was transmitted to it by heterokaryosis. 441 Similar results were obtained by Pallett 576
who also reported infections of Penicillium chrysogenum protoplasts with viruses from
P. chrysogenum and P. stoloniferum and infection of Marasmius androsaceous and
Mucor hiemalis protoplasts with viruses of P. chrysogenum. However, the level of
virus in the newly infected cultures was low (1 % or less of that of the donor strains)
and the infection was apparently unstable, virus becoming undetectable after subcul-
turing for 3 years. The high frequencies of infection reported by Lhoas 451 and Pallett 576
are surprising since no reagents,such as polY-L-ornithine (PLO) or polyethylene glycol
(PEG), were added to increase uptake.
More recently Ghabrial and co-workers 246 have transmitted a disease of Helmintho-
sporium victoriae by inoculating protoplasts with purified virus in the presence of
PEG, although the frequency of infection was low and isolation and characterization
of virus particles from the newly diseased tissue has not yet been achieved. Stanway
and Buck438 " 52 achieved infection of 10% of protoplasts of Gaeumannomyces graminis
incubated with purfied virus particles in the presence of PEG. (No infection was ob-
tained without PEG.) This result was unequivocal because (a) the recipient was known
to be completely free from virus particles, but known to be susceptible to virus infec-
tion, (b) the virus preparations were filter sterilized and completely free from fungal
propagules, (c) inoculations were carried out under asceptic conditions, (d) the viruses
in the newly infected cultures were isolated and thoroughly characterized, and (e) the
46 Fungal Virology

levels of viruses in the newly infected cultures were similar to those in the parent cul-
tures and remained stable over three successive subcultures. Demonstration of the in-
fectivity of a dsRNA mycovirus (together with other properties, see Section III.B and
III. C.l.b) fully justifies the use of the term virus, despite the absence of an extracel-
lular phase in the life cycle and, in many cases, absence of associated disease (see
Section III. C). It is likely that infection of protoplasts with all, or most, isometric
dsRNA mycoviruses could be achieved given the right conditions. When attempting
protoplast infection it is important to select virus-free recipient strains that are sucep-
tible to virus infection. In this context it is noteworthy that several naturally occurring,
virus-free strains of Penicillium brevicompactum and P. stoloniferum could owe their
absence of virus to the fact that they secrete the antiviral agent mycophenolic acid. 453
Other fungal metabolites with antiviral activity include patulin and gliotoxin.454456
An ingenious way of overcoming the cell wall barrier without the need to produce
protoplasts was suggested by Lhoas,457 who incubated mating pairs of Saccharomyces
cerevisiae in the presence of viruses from Aspergillus niger and Penicillium stoloni-
ferum. He argued that the cell wall would be broken down and that virus might be
taken up when the two cells fused. Unfortunately, although infection with these viruses
was claimed,4S7,4S. later examination of the putatively infected cultures revealed only
an endogenous virus present in the original yeast strains, ••6 with no trace of A. niger
or P. stoloniferum viruses. 4S9,460

6. Host Range
If transmission in nature occurs only via plasmogamy it might be expected that the
natural host range of dsRNA mycoviruses would be limited to individuals within a
species or within closely related species. Because a reliable method of protoplast infec-
tion has only recently been developed,452 no unequivocal results on experimental virus
host ranges are available as yet. Information on host ranges therefore comes from the
identification of identical or closely related viruses in different fungal species, e.g.,
viruses in Agaricus bisporus and A. campestris;60 Aspergillus foetidus and A. ni-
ger; 232,234 Aspergillus ochraceous, Diplocarpon rosae, and Penicillium stoloni-
ferum;'·4,235 Fusarium roseum and Sclerotium cepivorum;239 Gaeumannomyces gra-
minis var. tritici and Phialophora sp. (lobed hyphopodia);299 Gaeumannomyces
graminis var. tritici and Phialophora graminicola;46' Penicillium brevicompactum, P.
chrysogenum, and P. cyaneo_fulvum;248.250,254 and Penicillium funiculosum and P.
purpurogenum!67 Because of their intracellular modes of transmission it might be ex-
pected that many fungi would remain persistently infected, perhaps indefinitely. Hence
dsRNA mycoviruses may have evolved along with their hosts. 462 The finding of similar
viruses in species of the same fungal genus suggests that infection could have arisen
early in their phylogeny before the species diverged. This explanation is also possible,
but perhaps less feasible, for species of widely divergent genera, such as Diplocarpon
rosae and Penicillium stoloniferum, in which virus divergence would have been ex-
pected also. Nevertheless such occurrences do indicate that dsRNA mycoviruses can
have a wide host range. Alternative explanations would involve transmission of viruses
between unrelated hosts, for example, by (1) anastomoses of very young hyphae, e.g.,
germ tubes emerging from spores, which may be sufficiently compatible to allow brief
plasmogamy; or (2) infection of anastomosing hyphae within an individual mycelium
by extracellular virus released from an autolyzing unrelated species. Relatively few
comparative studies of viruses from different fungi have as yet been carried out and a
broad survey, using nucleic acid hybridization as well as serological tests, may well
reveal the widespread occurrence of related viruses in a wide range of fungi.

B. Structure, Genome Organization, and Taxonomy

Isometric dsRNA mycoviruses are among the simplest known viruses. Most of them
 47

have single-shelled capsids composed of one major polypeptide species. Analyses, in

the few cases where they have been carried out, are consistent with icosahedral struc-
tures, comprised of 60 structure units (T = 1) each of which consists of one or two
polypeptide subunits, giving a total of 60 subunits, e.g., Penicillium chrysogenum vi-
rus/ 54 P. cyaneo-fulvum virus/ 54 Saccharomyces cerevisiae virus L1 (LA);463 or 120
subunits, e.g, Aspergillus foetidus viruses Sand F,23Z Helminthosporium maydis vi-
ruS/ 43 Penicillium stoloniferum virus S.271 Their genomes (3.5 to 10 kb) are just large
enough to encode the virus capsid polypeptide and one or a few other polypeptides.
Proof that the virus genome encodes the capsid polypeptide has been obtained by in
vitro translation of denatured dsRNA or mRNA from the following viruses: Gaeuman-
nomyces graminisviruses 019/6-A and 38_4_A;464.465 Penicillium chrysogenum virus;466
P. stoloniferum virus S;467 Saccharomyces cerevisiae viruses L1(LA)468 and La(LB/
C)463.469 It is generally assumed that the dsRNA-dependent RNA polymerases, which
have been found to be present in the virions of all dsRNA mycoviruses so far exam-
ined,.67 are also virus-encoded, but in no case has this been proved. However, although
host polymerases able to transcribe ssRNA 470 and dsRNA 471 have been described, there
is good evidence that complete replication of viral RNA in bacteria, plants, and ani-
mals requires at least one virus-encoded polypeptide. 472 479
Despite their simplicity there appears to be considerable diversity among the isomet-
ric dsRNA mycoviruses, e.g., dsRNA segments ranging in size from 0.4 kbp to >10
kbp and capsid polypeptide mol wts ranging from 18,000 to 125,000, perhaps even
greater than the diversity shown by the small isometric plant viruses with genomes of
ssRNA in a similar size range (Table 8). There have been considerable problems in
their taxonomy,.80.481 particularly in establishing the minimum number of genome seg-
ments required for virus replication. This is because many isolates apparently contain
satellite RNAs and/or defective RNAs. A satellite482 is a virus or nucleic acid that is
unable to mUltiply in cells without the assistance of a specific "helper" virus, is not
necessary for the multiplication of the helper virus, and has no appreciable sequence
homology with the helper virus genome. Satellite viruses encode their own capsid poly-
peptides, whereas satellite RNAs are encapsidated in capsids encoded by their helper
virus. Defective RNAs483 are derived from the helper virus genome and depend on it
for replication but are not needed by it for infection. The distinction between satellite
and defective RNAs is therefore whether or not they are related to their helper virus
genome. Since dsRNA is not infective452 and since for many mycoviruses it is difficult
to separate particles containing dsRNAs of closely similar size, indirect methods have
been used to deduce the number of genome segments. For example, virus isolates from
Ustilago maydis may contain up to seven dsRNA segments, but deletion mutants with
only one dsRNA segment which still produce virus particles are readily obtained. 484
Hence only one dsRNA segment is required for virus multiplication. In another ex-
ample, protoplasts of Gaeumannomyces graminis were incubated with a mixture of
viruses 3bla-(A, BI, B2, and C).452 Out of 30 cultures regenerated from single proto-
plasts, only three were infected, two with virus BI and one with virus B2 (see Chapter
8 for further details). Each of the infected cultures retained the two dsRNA segments
characteristic of their viruses which, since the overall frequency of infection was low
and only one virus was found in each of the infected cultures, is good evidence that
both RNAs are required for virus multiplication. Similar arguments apply to the re-
quirement of two dsRNA segments in Penicillium stoloniferum virus S (PsV-S). When
conidia of P. stoloniferum infected with PsV-S and another virus, PsV-F, were UV
irradiated, survivors were found to contain either no virus or only PsV-S; the latter all
contained two dsRNA segments. 485 Since the UV treatment resulted in complete dele-
tion of PsV-F, and in some isolates of both viruses, it would have been expected that
48 Fungal Virology

if PsV-S required only one dsRNA segment, some isolates containing only one dsRNA
segment would have been obtained.
A similar problem relates to the capsid polypeptide species. Although many dsRNA
mycoviruses appear to have a capsid composed of one major polypeptide species, when
two or more polypeptides are present, it is often uncertain whether the extra polypep-
tides are (a) degradation products, (b) impurities, (c) RNA polymerase molecules, (d)
aggregates, (e) additional structural polypeptides, or (f) due to a mixture of viruses or
virus variants. Examples of most of these categories have been recorded. Preparations
of Aspergillus foetidusvirus S have two polypeptides, but the smaller was shown to be
derived from the larger by degradation in vitro. 232 Putative RNA polymerase molecules
have been reported for Aspergillus foetidus viruses Sand p32.486 and Penicillium sto-
loniferum viruses Sand F, 271 while aggregation is common in polypeptides from Gaeu-
mannomyces graminis viruses. 198 Polypeptide heterogeneity in virus preparations from
Saccharomyces cerevisiae has been shown to be due to the presence of two viruses,
ScV-LI (LA) and ScV-La (LB/C).463.469
Initially six groups of isometric dsRNA mycoviruses were proposed 487 based on par-
ticle size and sedimentation rate, size and numbers of dsRNA segments and capsid
polypeptide species, serology, and nucleic acid hybridization. Further consideration
has led to the establishment of two new virus families based primarily on genome
organization. 488 The first of these, the Totiviridae, comprises viruses with an undivided
genome in the size range 4.7 to 6.3 kbp (Table 10). Less than half the genome is re-
quired to encode the capsid polypeptide species, so the dsRNA is assumed to be at least
dicistronic, but this has not been proved for any member. The capsid polypeptide genes
in Saccharomyces cerevisiae viruses LI(LA) and La(LB/C) are located on their ds-
RNAs near their 5'termini,>69 but no comparable information is available for other
members as yet. The second family, the Partitiviridae, comprises viruses with genomes
of two monocistronic dsRNA components, usually of similar size, in the range 1.4 to
2.2 kbp (Table 11). One dsRNA segment encodes the capsid polypeptide and the other
segment encodes an unrelated polypeptide (shown so far only for Gaeumannomyces
graminis viruses OI9/6-A and 38_4_A).464.465 Another group, the Penicillium chryso-
genum virus group, comprises viruses with three or four monocistronic dsRNA seg-
ments (Table 12), but it is uncertain how many dsRNA segments are required for virus
replication. If it is only two, this group could become a genus of the Partitiviridae; if
it is three, it could become a new family.
A number of dsRNA mycoviruses remain unclassified pending further information
on their genome organization (Table 13). Some of these will probably ultimately be
placed in one of the existing families, whereas new families or genera may be needed
for others. For example, Helminthosporium maydis virus which has only one dsRNA
segment (8.3 kbp) probably at least dicistronic, could become a genus of the Totiviri-
dae, unlike Gaeumannomyces graminis virus 45/1OI-C whose single 1.8 kbp dsRNA
has only sufficient coding capacity for its coat protein. Virus 45/101-C could be a
satellite virus, but caution is needed because of the possibility that it could contain two
distinct dsRNAs of the same size. A virus of 39 nm diameter from Lentinus edodes is
unusual in that it appears from electron microscopy to have a double-shelled capsid
(Figure IF) the outer shell of which is removed after chymotrypsin treatment, giving
"core" particles of 34 nm diameter. Although somewhat smaller, the structure of the
virus shows some resemblance to members of Phytoreovirus, a genus of the family
Reoviridae. However, its diameter and single dsRNA segment of 6.5 kbp would oth-
erwise place it in the Totiviridae. Information on the capsid polypeptide composition
of this virus is needed to confirm its double-shell structure.
All of the known isometric dsRNA mycoviruses are completely different from
dsRNA viruses in the Birnaviridae, Cystoviridae, and Reoviridae families, the rod-
 49

Table 10
MEMBERS AND POSSIBLE MEMBERS OF THE TOTIVIRIDAE

Capsid
polypeptide
Diam S20 .... dsRNA species species
(mm) (S units) (kbp) (Mol wt (XlO-l) Refs.

Members
Saccharomyces cerevisiae 40 161 4.7 88 463,468
virus Ll (LA), type spe-
cies
Gaeumannomyces gra- 40 N.D. 6.1 84 199
minis virus 87-I-H
Mycogone perniciosa vi- 42 N.D. 6.3 69 247
rus
Saccharomyces cerevisiae 40 161 4.7 82 463,468
virus La (LB/C)
Ustilago maydis virus 41-43 172 6.1 73 219
PI-HI
Yarrowia Iipolytica virus 50· N.D. 5.5 76 193
Possible members·
Aspergillus foetidus 40 176 6.0 83 232,233
virus S (+3.8,0.4)
Aspergillus niger virus S 40 N.D. 6.0 N.D. 234
(+3.9,3.8,0.4)
Geotrichum candidum 40 N.D. 5.2 94 (+71) 332
virus
Helminthosporium victo- 35-40 190 4.4 88 (+83) 113, 246
riae 190 virus
Pyricularia oryzaevirus 36 N.D. 4.7 N.D. 277
Thielaviopsis basicola 40 150 6.6 N.D. 280
viruses c

Note: N.D., Not determined.

The particle diameter of this virus appears to be larger than that of the other members, but this could be
a measurement artifact. Significant variations in diameters of the same virus from different laboratories
have been reported. 6 ' 4 ••
DsRNA and polypeptide species of uncertain status in the possible members are enclosed in parentheses.
This fungus probably contains a mixture of five viruses, each with one dsRNA segment; only the fastest
sedimenting component and largest dsRNA are included here.

shaped dsRNA plant viruses and the VLPs containing dsRNA associated with male
sterility in some higher plants. However, there appears to be a close resemblance be-
tween dsRNA mycoviruses in the Partitiviridae family and the small isometric "cryp-
tic" or "temperate" viruses which have now been found in low concentrations in a
wide range of symptomless plants. Although the plant cryptic viruses are seed trans-
missible with high efficiency they are not mechanically, aphid, or graft transmissible.
Although these viruses have two to five segments of dsRNA, their intracellular mode
of transmission, like that of mycoviruses, could lead to the accumulation of satellite
or defective RNAs. Hence, like the Partitiviridae, two could be the minimum number
of segments required for replication. Properties of dsRNA viruses in different families
and groups, and different host taxa, are compared in Table 14.

C. Virus-Host Interactions
1. Replication
a. Replication in Relation to Host Growth - Virus Latency
The majority of isometric dsRNA mycoviruses appear to have no overt effect on
50 Fungal Virology

Table 11
MEMBERS AND POSSIBLE MEMBERS OF THE PARTITIVIRIDAE

Capsid
dsRNA polypeptide
Diam S020.", species species
(nm) (S units) (kbp) (mol wt x to- 3 ) Refs.

Members
Gaeumannomyces 35 126 1.8, I. 7 60 198,464,
graminisvirus 019!6-A 465
(type species)
Agaricus bisporus virus 4 35 140-145 2.2,2.0 69 162
Gaeumannomyces gra- 35 133 2.2,2.1 73 198
minis virus T1-A
Penicillium stoloniferum 30-34 tOl 1.6,1.4 56 235,270,
virus S' 42< 271
Possible members'
Phialophora virus 2-2-A 35 116 1.9, 1.8 60 308
(+1.5)
Penicillium stoloniferum 30-34 104 1.5, 1.3 59 235,269,
virus F (+0.67) 47< 271,489

Viruses from Aspergillus ochraceous and Diplocarpon rosae are serologically related to P. stoloniferum
virus S. 184 2J5
DsRNA species of uncertain status in the possible members are shown in parentheses.
Possibly formed by proteolytic degradation of the larger species. m

Table 12
MEMBERS AND POSSIBLE MEMBER OF THE PENICILLIUM
CHRYSOGENUMVIRUS GROUP

Capsid
dsRNA polypeptide
Diam. S,o .• species species
(nm) (S units) (kbp) (mol wt x to-3 ) Refs.

Members
Penicillium chrysogenum 35-40 145--150 3.2,3.0, 125 250,254,
virus' (type species) 2.9 256
Penicillium brevicompactum 35-40 147 3.2,3.0, N.D. 248
virus 2.9
Penicillium cyaneo-fulvum 35-40 147 3.2,3.0, 125 254
virus 2.8
Possible member
Helminthosporium victoriae 35-40 145 3.5,3.2, t06 (+97, 92) 113
virusb 3.1,
2.9

Some isolates contain additional dsRNA species. 256 . " ,

Polypeptide species of uncertain status are shown in parentheses.

their hosts. Such infections have been termed latent, even though the virus may repli-
cate to high titers. There are probably two reasons for latency: (1) Most of these viruses
do not code for lytic enzymes or for any products which directly inhibit host DNA
replication, transcription or translation. (2) Virus and dsRNA replication are probably
under strict host control in the actively dividing cells, so that amounts of virus and
dsRNA produced are insufficient to interfere significantly with host macromolecule
 51

Table 13
SOME UNCLASSIFIED ISOMETRIC DSRNA MYCOVIRUSES

Capsid
Diam S,o.w dsRNA species polypeptide
Virus (nm) (S units) (kbp) species (X 10-3) Refs.

Agaricus bisporus virus I 25 90-100 2.0,2.0 25 162

Allomyces arbuscula virus 40 75,67 3.6, 2.0, 1.6 38, 34, 28, 21 282
Aspergillus ffavus "virus'" 27-30 49 18 229
Aspergillus foetidus virus F 40-42 164, 145 3.8,2.7,2.5,2.1, 87 (major) 232,233
1.8 125, 100 (minor)'
Colletotrichum Iindemuth- 30 110,85 3.6, 1.6, 1.5 52 (major) 224
ianumvirus 45 (minor)
Gaeumannomyces graminis 29 127 1.8 66 195
virus 45/101-C
Gaeumannomyces graminis 40 N.D. 9.2,7.1 125 490
virus F3-A
Gaeumannomyces graminis 40 N.D. 9.2 94 490
virus FIO-A
Helminthosporium maydis virus 48 283 8.3 121 243
Lentinus edodes virus 39 N.D. 6.5 N.D. 210,491
Periconia circinata virus 32 150,140 2.5, 2.0, 1.8, 1.6, N.D. 272
0.7,0.6
Rhizoctonia soIanivirus 33 161 2.3, 2.1, 1.8 46 554

Data refer to particles without nucleic acid. However, dsRNA has recently been detected in this strain. 230
Possibly associated with the virion RNA polymerase.

biosynthesis by direct competition. Continued virus replication after host cell division
had ceased would lead to much higher virus levels in the older resting cells. In a batch
culture of a filamentous fungus this would lead to an overall lag of virus replication
behind that of its host.
Direct evidence that virus and dsRNA replication lagged behind host growth was
obtained by Still et al. 48S who studied shaken batch liquid cultures of Penicillium sto-
loniferum infected with PsV-S and PsV-F. Maximum rates of dsRNA synthesis were
not achieved until host growth had slowed down. Further, as the fungal growth moved
towards the stationary phase, PsV-F levels increased three-fold while biomass in-
creased by only 40070. Further evidence came from the observation that when resting
cells of P. stoloniferum, previously grown for 48 hr in a yeast extract-sucrose medium,
were suspended in distilled water, PsV-F levels increased three-fold over a further pe-
riod of 48 hours, while fungal biomass remained constant and little or no lysis was
observed.45s It has been caIculated 4S9 that, in P. stoloniferum grown in shaken liquid
culture, virus particles account for only 0.5% of the fungal biomass, and only about
20% of the particles were synthesized during exponential fungal growth, the remainder
being produced during the deceleration or stationary phases of growth. In most fungi
virus levels are much lower than in P. stoloniferum. Hence competition between virus
and host for precursors of nucleic acid and protein synthesis would be insignificant.
Possible ways in which virus replication could be controlled in actively replicating
fungal cells include (1) control by a limited supply of host factors required for dsRNA
replication, and (2) negative control by a regulator molecule. The work of Wickner
and co-workers (see Chapter 2) suggests that for viruses of Saccharomyces cerevisiae,
both these controls could be operative. They have shown that host genes are required
for the replication both of virus dsRNAs and of their satellites. 498 Recently they have
also provided evidence that dsRNA copy number is under the control of negative reg-
ulators, namely, the products of various host ski genes.'"
 Table 14 VI
N
COMPARATIVE PROPERTIES OF DSRNA VIRUSES
No. dsRNA Size range Range of mol
Diam. or segments of dsRNA No. capsid wt of capsid
Family or dimensions (nm) (mode of en- segments Mode of dsRNA polypeptide polypeptides ~
::J
group (hosts)" and morphology capsidation)b (kbp) Type of dsRNA replication species (X 10-3) Refs. ~
"-
Birnaviridae :;.
(V,I)
57-74, icosahedral,
single shell, no en-
2
(T)
3.4-3.6 Component A,
polycistronic;
Probably semi-
conservative
4 29-105 488,492
..,'0
"-
velope component B, 0
monocistronic ~
Cystoviridae 75, enveloped, ico- 3 3.4-7.3 Polycistronic Semi-conservative 11 6-82 488
(B) sahedral core (T)
Partitiviridae 30-35, isometric, 2 1.3-2.3 Monocistronic Semi-conservative 42-73 488
(F) single shell, no en- (S)
velope
Penicillium 35-40, isometric, 3 or 4 2.8-3.2 Monocistronic N.D. 125 488
chrysogenum single shell, no en- (S)
virus group vel ope
(F)
Reoviridae 60-80, icosahedral, 10-\2 0.3-4.4 Monocistronic Conservative 6-10 15-155 488,493
(V,I,P) double shell, no en- (T)
velope
Totiviridae 40-43, isometric, 4.7-6.3 Probably at least Probably con- 69-88 488
(F) single shell, no en- dicistronic servative
velope
Tobacco stunt 300-360 x 18, rigid 2 6.7 N.D. N.D. 48-52 295-298,
virus group rod (N.D.) 346
(P)
Plant cryptic 30-35, isometric, 2-5 0.6-2.2 Probably mono- N.D. 1-2 50-70 494-496
viruses (P) single shell, no en- (N.D.) cistronic
velope
Plant CMS' 70, spherical to el- 15-19 N.D. N.D. N.D. N.D. 497
VLPs (P) liptical, enveloped

Hosts: B, bacterium; F, fungus; I, invertebrate; P, higher plant; V, vertebrate.

S, Each dsRNA segment encapsidated in a separate particle; T, all dsRNA segments encapsidated together in one particle.
CMS, cytoplasmic male sterility; these VLPs are similar to the dsRNA-containing spherical bodies associated with replication of citrus tristeza clos-
tevovirus, a ssRNA virus.lSI
 53

Replication of mycovirus dsRNA has been compared 50o to that of the "relaxed"
mode of replication of certain bacterial plasmids, which are produced at a controlled
copy number during exponential bacterial growth, when cells are actively replicating,
but which increase significantly in numbers in the stationary phase, when cells are in a
resting state. 50. Control of copy number of many bacterial plasm ids is determined by
interaction between positive regulators, i.e., preprimer RNA molecules and negative
regulators, e.g., small RNA molecules which interact with preprimer RNA to prevent
its processing to form primer. 502,503 For both bacterial DNA plasmids and dsRNA my-
coviruses, runaway replication, if allowed to occur in actively replicating cells, would
probably be lethal.
There is very little information on the timing of virus replication within the mitotic
cell cycle. Because of their hyphal growth habit the growth of cells in cultures of fila-
mentous fungi, even starting from germinating spores, soon becomes asynchronous
and cell cycle studies of virus replication are largely precluded. Synchronous cultures
of yeast (Saccharomyces cerevisiae) cells can readily be obtained, but there are conflict-
ing reports as to whether dsRNA synthesis occurs throughout the cell cycle or occurs
throughout the cell cycle except for S phase when DNA synthesis occurs (see Chapter
2). By analogy with filamentous fungi, it is possible that virus replication could con-
tinue in resting yeast cells under certain conditions.

b. Ultrastructural Studies
Ultrastructural studies carried out on a range of fungi' 72 ,'8o,237,258,394.427 429,504,505 have
shown that virus particles accumulate in the cytoplasm of hyphal compartments, yeast
cells and buds, and sexual and asexual spores (see also Section III. A.1, 2, and 3).
Particles occur free in the cytoplasm scattered as individuals or in loose clusters, and
also enclosed in single- or double-membrane-bound vesicles or vacuoles when they
often form circular, linear, or crystalline aggregates. As many as 105 particles were
estimated to be present in some hyphal compartments of Penicillium chrysogenum,258
but numbers in most fungi are much lower. In some fungi release of fibrous material
from particles has been observed; 172,258 this could be ssRNA or dsRNA released from
transcribing or replicating particles (see Section III. C.1.b).
One of the most sytematic ultrastructural studies was carried out by Border, 25',4'9,458
who made serial thin sections of virus-infected and noninfected hyphae of colonies of
Penicillium stoloniferum, P. chrysogenum, and P. funiculosum grown on solid me-
dium, starting from the tip and progressing inwards towards the older cells in the center
of the colonies. No virus particles were found in the terminal 0.1 mm of the apex.
Virus particles were first detected between 0.1 and 2 mm from the hyphal tip in aggre-
gates of up to several hundred free in the cytoplasm. Most of the hyphal compartments
in this region were still unplugged. In the region 2 to 3 mm from the tip, aggregates of
virus particles were still seen free in the cytoplasm but many were also now enclosed in
vesicles. In this region and further back the septal pores were plugged. Further from
the tip (3 to 5 mm) a lumen formed between the clusters of virus particles and their
surrounding membrane forming small vacuoles. From 5 mm inwards, it appeared that
the small vacuoles had fused to form a large virus-containing vacuole which filled
almost the entire cell. Cells at the center of the colony were almost completely auto-
lyzed and the plamalemma appeared to be withdrawn from the cell wall, forming an
envelope surrounding the virus particles. Apart from the presence or absence of virus
particles, no differences in cell ultrastructure were found between virus-infected and
virus-free strains; in both cases cell degeneration in the older parts of the hyphae was
observed to start about 3 mm from the hyphal tip. This progression of events, together
with the investigations described in Section III. C.l.a, suggests the following model:
54 Fungal Virology

I. Little or no virus replication occurs in the apical hyphal compartment.

2. Replication occurs in the unplugged hyphal compartments of the pheripheral
growth zone but at relatively low levels controlled by a limited supply of host
replication factors and/or negative regulators.
3. Virus replication continues in the older plugged hyphal compartments which do
not contribute to host growth.
4. Virus replication finally ceases when the particles become enclosed in mem-
branes.

c. The Virus Replication Cycle - Virion-Associated RNA Polymerases

No direct studies on the virus replication cycle have been carried out, but it has been
suggested that the recently described protoplast infection system could be adapted to
obtain synchronous infection of protoplasts in liquid suspension, so that studies com-
parable to those with viruses of higher plants could be carried out. 452 Most of the
information presently available on the replication cycle comes from studies of virion-
associated RNA polymerases. 461 Several dsRNA mycoviruses, including those in the
Totiviridae and the Partitiviridae, have an RNA-dependent RNA polymerase that ca-
talyzes the end-to-end transcription of dsRNA to produce single-stranded virus mRNA
which is released from the particles, e.g., Allomyces arbuscula virus,506 Aspergillus
foetidusvirus S (AfV_S),501 Gaeumannomyces graminisviruses 019/6-A, 38-4-A, 3bla-
(B and C), F6-(B and C),s08 Penicillium stoloniferum virus S (PsV -S), 461 Phialophora
sp. (lobed hyphopodia) virus 2_2_A,509 Saccharomyces cerevisiae cerevisiae virus
Ll(LA),s10.511 and Ustilago maydisviruses.432 The isolation of particles containing only
ssRNA, e.g., from PsV -S, 210 or partially double-stranded RNA, e.g., in S. cerevisiae
virus Ll (LA)512 and the detection in the latter of a ss-'dsRNA polymerase activity,
suggests that replication in these viruses is asynchronous, as in reovirus 493 (i.e., first
transcription of dsRNA occurs to form (+) ssRNA, followed sometime later, probably
within subviral particles, by synthesis, of (-) RNA on the (+) RNA template to form
dsRNA).
PsV-S has an additional activity in vitro which catalyzes the complete replication of
dsRNA to form particles containing two molecules of dsRNA.513 This is probably the
result of the first transcript being retained in the particles where it acts as a template
for second-strand synthesis. The isolation from infected mycelium of particles of PsV-
S containing both dsRNA and ssRNA210 suggests that this reaction also occurs in vivo.
Therefore, in addition to being able to undergo asynchronous replication, PsV-S is
probably also able to undergo synchronous replication. Presumably dsRNA can be
released from the diploid virions in vivo and become encapsidated in newly synthesized
particles to initiate a new round of replication. Similar activities may be present in
PSV-pI4 and dsRNA virus particles from Agaricus bisporuS.S15 Based on the replicase
activity in PsV-S, Buck and Ratti 516 proposed a model for the replication of dsRNA
mycoviruses possessing this activity by doubling in synchrony with cell division. In the
absence of evidence to support it and in view of the ability of the viruses to replicate
in resting cells455 and also to undergo asynchronous replication, the model is probably
too simplistic. Buck461 suggested that the choice between synchronous and asynchron-
ous pathways for PsV-S may be governed by the physiological state of the host (e.g.,
replicating or resting cells) or by early/late switches in the virus replication cycle.
The replication of PsV-S dsRNA in vitro has been shown, by density labeling, to be
semiconservative. 513 The substantial amount of [3HJ-UMP incorporated into dsRNA
during transcription in vitro of Gaeumannomyces graminis viruses 38-4-A and 3bla-(B
and C) dsRNA suggested that transcription occurred by a semiconservative strand-
displacement mechanism. 508 It is likely, therefore, that transcription and replication of
dsRNA by viruses in the Partitiviridae family are semiconservatitive. Interestingly, ex-
 55

amination of transcribing particles of G. graminis virus particles revealed the presence

of particles showing the release of looped ssRNA molecules, both ends of which were
attached to the particle. This suggested that a loop of ssRNA, rather than the 5' end,
may initiate the extrusion of the transcript from the virion and that the 5' end of the
transcript may remain associated with the RNA polymerase, so that the first end to be
set free outside the particle might be the 3' OH end. Repeated cycles of transcription
would be facilitated if the dsRNA were held within the virions in a circular conforma-
tion.
There is still uncertainty about whether transcription of Saccharomyces cerevisiae
and Ustilago maydis viruses, two members of the Totiviridae family, occur by conser-
vative or semiconservative mechanisms, although the evidence favors a conservative
model for ScV (see Chapters 2 and 3). Density labeling and pulse chase experiments
have shown that RNA 2 of AfV-S, a possible member of the Totiviridae, is transcribed
by a semiconservative displacement mechanism,514 but there is no evidence for this in
RNA I. If RNA 2 is a satellite RNA, it is conceivable that it could replicate by a
mechanism different from that of the genomic RNA (RNA I). Although dsRNAs of
the viruses in the Totiviridae are considered to be at least dicistronic, no subgenomic
mRNAs are formed in vitro. Subgenomic RNAs have been found in vivo for ScV (see
Chapter 2), but no information is available on their translation products, if any.
There is no evidence for any capping activity associated with a virion RNA polymer-
ase of any dsRNA mycovirus and it is not known whether virus mRNAs in vivo are
capped. Similarly, no DNA polymerase activity has been found in particles of dsRNA
mycoviruses and there is no evidence for dsRNA-derived DNA proviruses in the gen-
omes of Gaeumannomyces graminis,'17 Saccharomyces cerevisiae,431 or Ustilago may-
dis.432 However, it is noteworthy that DNA polymerase b, the only DNA polymerase
produced during meiotic prophase in Coprinus cinereus, is able to polymerize dNTPs,
not only on DNA, but also on dsRNA templates. 519 The same enzyme activity has been
identified in Coprinus comatus and Agaricus bisporus. Further investigations as to
whether dsRNA might be copied in vivo would be justified. Ross 520 reported that a
"pale mushroom" phenotype in Coprinus congregatus, in which mushrooms do not
form the usual number of black spores, was associated with an infectious, cytoplasm-
ically transferable element, possibly dsRNA, that inhibited meiosis in the presynapto-
nemal complex stage. However, after several years, although the pale phenotype was
retained, infectivity and cytoplasmic transmissibility were lost, and the pale phenotype
was inherited as a nuclear gene. 521 This phenomenon suggests the transfer of genetic
information from the cytoplasm (on dsRNA?) to the chromosomal DNA in the nu-
cleus.

d. Mixed Infections - Compatibility and Incompatibility

Mixed infections of fungi with two or more unrelated dsRNA viruses are common,
probably due to their intracellular modes of transmission (Section III. A). Examples
include Agaricus bisporus viruses 1 and 4,162 Aspergillus foetidus viruses Sand F, 232
Gaeumannomyces graminis viruses F6-A, F6-B, and F6-C, 198 Penicillium stoloniferum
viruses Sand F,>69 and Saccharomyces cerevisiae viruses LI(LA) and La(LB/C).469
Faithful transmission of these viruses in single conidial isolates (see Section III. A.2
and Chapter 8), proves that mixture of viruses replicate in a single cell. The stability of
such mixed infections is probably due to the absence of competition between unrelated
viruses for essential replication factors. Wickner and co-workers (see Chapter 2) have
shown that host gene requirements for virus replication in Saccharomyces cerevisiae
differ with different viruses, e.g., LI(LA) dsRNA replication requires the host MAK3,
MAKlO, and PETl8 gene products, but La(LB/C) dsRNA replication does not. Simi-
lar considerations will probably apply to other dsRNA mycoviruses. Further evidence
56 Fungal Virology

for lack of interaction between unrelated viruses is provided by the absence of pheno-
typic and genotypic mixing in mixed infections, for example, AfV -S and AfV _F,>32
ScV-Ll (LA), and ScV-La (LB/C).469 Furthermore, satellites of ScV-Ll (LA), e.g.,
Ml-dsRNA, are specifically encapsidated in capsids encoded by their helper virus (i.e.,
ScV-LlILA]), but not by capsids encoded by ScV-La ILB/C).469
An unusual situation has been reported 23S for Penicillium stoloniferum viruses Sand
F, two serologically unrelated viruses which replicate together in the same fungal
strain!69,271 PsV-S has two dsRNA components (1.6 kbp and 1.4 kbp), whereas PsV-F
has three dsRNA components (1.5 kbp, 1.3 kbp, and 0.67 kbp). Until recently there
was no evidence for any interaction between the two viruses. However, Kim and
Bozarth 23S have now shown that a labeled probe, synthesized from a template of the
two PsV-S dsRNAs, hybridized with the 0.67 kbp dsRNA but not with the two larger
dsRNAs of PsV-F. This unexpected result suggests that either PsV-S and PsV-F have
evolved from a common ancestor and the 0.67 kbp dsRNA has retained sequences in
common to both viruses, or that recombination of PsV-S and PsV-F RNA segments
has occurred, possibly as a result of strand switching during replication, to produce a
mosaic structure consisting of sequences from both viruses. The latter mechanism has
been invoked to explain the formation of an RNA with a mosaic structure arising from
two RNA segments of influenza virus. 579 Whether the 0.67 kbp dsRNA of PsV-F en-
codes a protein or is simply a defective dsRNA is not know. Presumably the 0.67 kbp
dsRNA has retained the capsid assembly and RNA polymerase recognition sites of
PsV-F dsRNA (the extent of sequence homology between the 0.67 kbp dsRNA and the
two larger dsRNAs of PsV -F was not determined) and hence would not be expected to
compete with PsV -S for essential replication factors. This could explain the stable co-
existence of PsV-S and PsV-F in the same cell.
When two related viruses are introduced into the same cell they are often incompat-
ible with the result that one is eliminated, e.g., the cross of a strain of Saccharomyces
cerevisiae containing virus Ll(LA) with one containing virus L2 usually results in the
elimination of virus L2 (see Chapter 2). Such incompatibility could be due simply to
competition between related viruses for the same host factors essential to their repli-
cation; e.g., Ll and L2 dsRNAs both require the host MAKIO gene product. Similarly,
incompatibility between satellite dsRNAs and between satellite and defective dsRNAs
in S. cerevisiaeand Ustilago maydis(see Chapters 2 and 3) could be due to competition
for host or virally encoded replication factors or capsid polypeptide. It is noteworthy,
however, that related dsRNAs (M and L segments) stably replicate together in U. may-
dis. These may be exceptional in that L dsRNA appears to be a subgenomic dsRNA
derived from an M dsRNA segment, and replication of L dsRNA may be dependent
on the M dsRNA segment (see Chapter 3).
Bacterial DNA plasm ids are also usually compatible when they are unrelated and
incompatible when they are related; this is the basis for classification of plasmids into
different incompatibility groups. Incompatibility of many bacterial plasmids is deter-
mined by interaction between positive and negative regulators of plasmid DNA repli-
cation (see Section III. C.l.a).502,503 Since dsRNA copy number may also be at least
partly controlled by negative regulators (see III. C.1.a), this suggests an additional
mechanism for dsRNA incompatibility.

2. Virus Infection and the Host Phenotype

Notwithstanding the latent nature of many dsRNA mycovirus infections, there have
been many reports of the association of viruses with specific fungal phenotypes. Many
of these are of little value because isogenic strains with and without virus have not been
compared, and often there have not even been attempts to establish cytoplasmic trans-
missibility. In some other cases, when cytoplasmic transmission has been established
 57

(e.g., vegetative death and the "ragged" phenotype in Aspergillus spp., the "poky"
and "stopper" phenotypes of Neurospora crassa, and senescence in Podospora anser-
ina), the defects have been shown to lie in mitochondrial DNA. These nonviral, extra-
chromosomal elements are discussed in Chapter 9. Ability to secrete killer protein is
the only fungal phenotype so far shown unequivocally to be encoded by dsRNA (Sec-
tion III. C.2.b). Other possible associations between dsRNA and repression of afla-
toxin biosynthesis, transmissible diseases, hypovirulence and pathogenicity of phyto-
pathogenic fungi are discussed in Section III. C.2. (a, c, and d).

a. Secondary Metabolites
There are no examples where a dsRNA virus has been shown unequivocally to have
a direct effect on the production of a fungal secondary metabolite. 456 Lemke and co-
workers 260 treated a virus-infected, penicillin-producing strain of Penicillium chryso-
genum with heat and mutagenic agents. All of 30 resultant strains retained the ability
to produce penicillin, even though their viral contents varied from zero up to the levels
in the original strain. Similarly, strains of other antibiotic-producing fungi, Penicillium
notatum and Cephalosporium chrysogenum, mayor may not contain virus parti-
cles.60.249.522 In other cases where correlations have been claimed, based on rather few
samples, e.g., production of the toxin pyriculol by Pyricularia oryza&23 and high cell
wall galactosamine content in Penicillium stoloniierum,"o the observed differences
could equally well have been due to host chromosomal differences. It remains possible
that production of a metabolite or toxin could be affected in a strain in which virus
causes a degenerative disease, e.g., in Helminthosporium Yictoriae(see Chapter 5). The
inhibitory action of fungal toxins on virus replication was considered in Section III.
A.5.
The absence of aflatoxin production in a strain of Aspergillus flayus (NRRL 5565)
was originally correlated with the presence of nucleic acid-free VLPs (Table 13), but a
more extensive investigation failed to detect virus in other strains, whether or not they
produced aflatoxin. 228,229456 Recently dsRNA has been detected in strain NRRL 5565230
and it has been shown that cycloheximide induced aflatoxin synthesis in this strain, but
not in other nontoxigenic strains. 310 Furthermore, the ability to synthesize aflatoxin
was retained during subsequent subcultures in the absence of cycloheximide, indicating
that the strain could have been cured of a genetic determinant for control of aflatoxin
biosynthesis. Since cycloheximide is known to inhibit virus dsRNA synthesis selectively
in both fungaJ260454.524 and mammalian 493 hosts, it was suggested that this genetic ele-
ment might be dsRNA. Satellite dsRNAs may be more sensitive to cycloheximide than
their helpers and a specific loss of satellite dsRNA can lead to increased helper virus
dsRNA copy number. 499 The importance of aflatoxin, produced by A. flayus and A.
parasiticus contaminants, in the production of foods and feeds is well recognized,52'
and it was suggested 310 that dsRNA determinants, effectively transmitted to field iso-
lates of these fungi, might provide a mechanism for biological control of aflatoxin
production. However, further investigations are needed to demonstrate that derepres-
sion of aflatoxin synthesis has a genetic rather than a epigenetic basis. These would
include proof that the derepressed isolate had actually lost all, or specific segments, of
its dsRNA and demonstration of the cytoplasmic transmissibility of the ability to re-
press aflatoxin biosynthesis along with the dsRNA.

b. Killer Proteins
Killer strains of Saccharomyces cereyisiae, of several other species of different yeast
genera and of Ustilago maydis, secrete proteins which have the ability to kill sensitive
strains of the same or closely related species. For killer strains of Saccharomyces (with
one exception)4S0.526 and Ustilago spp., it has been shown unequivocally that specific
58 Fungal Virology

dsRNA segments, which replicate as satellites of helper viruses, encode the killer pro-
teins. In killer strains of species in some genera of yeasts, dsRNA appears to be absent
and in Kluyveromyces lactis killer protein is encoded by a linear dsDNA plasmid. Sat-
ellite dsRNAs in fungi may be analogous to DNA plasmids in bacteria. Such plasmids
generally encode products which are beneficial, but not essential, to the host. Specifi-
cally fungal killer proteins have been compared 480 to plasmid-encoded bacteriocins
which are of common occurrence in both Gram-positive and Gram-negative bacte-
ria.527.528
Except in certain mutant strains (see Section III. C.2.c), the presence of virus and
satellite dsRNA in killer yeasts do not appear to have any deleterious effects on their
hosts. Indeed, it is possible that ability to produce killer toxins by immune yeasts could
confer an advantage over sensitive organisms in a crowded environment. It has been
shown that in mixed cultures of a killer and a sensitive strain of Saccharomyces cere-
visiae, maintained at the pH optimum of the killer protein, the killer strain displaced
the sensitive strain 529 and this property has uses in the brewing industry, e.g., in pro-
ducing "killer" brewing strains capable of resisting contamination by adventitious sen-
sitive strains. 53O Although the pH optimum for toxin activity is narrow (pH 4.2 to 4.7)
this pH may be similar to that of some natural habitats for yeasts, e.g., rotting grapes.
The significance of killer strains of the corn smut pathogen Ustilago maydis is much
less clear. Propagation of this fungus in nature depends on mating of two compatible
strains within the host and, since matings between killer and sensitive strains in vivo
occur readily, apparently the toxin is not produced, or it is rapidly inactivated, in the
plant hosts. In fact, killer strains of U. maydis appear to be of rather rare occurrence
in nature 531 .'32 compared to those of yeasts, 331.533538 although it is not known what
proportion of the latter are associated with dsRNA. Whether killer strains occur gen-
erally in filamentous fungi is not known.
Killer-related dsRNAs and viruses have been very actively studied over the past 10
years in several laboratories and are now the most thoroughly investigated of all
dsRNA mycoviruses. The results of these investigations are described in detail in Chap-
ters 2 and 3.

c. Transmissible Diseases - Lytic Plaques in Penicillium chrysogenum; Cold Sensitiv-

ity in Saccharomyces cerevisiae; Die-Back Disease of Mushrooms
There are several transmissible diseases of fungi, which appear to be associated with
dsRNA viruses, or in some cases possibly naked dsRNA. Transmissible degenerative
diseases of Helminthosporium victoriae, Rhizoctonia solani, and Ceratocystis ulmiare
discussed in Chapters 4, 5, 6, and 7. Here three deleterious effects will be described,
namely, formation of lytic plaques in Penicillium chrysogenum, cold sensitivity in Sac-
charomyces cerevisiae, and die-back disease of mushrooms. The first two of these oc-
cur only in hosts with nuclear mutations which upset the normal balance between virus
and host. In the third, disease may possibly be caused by mutant viruses.
Erumpent patches of sterile white mycelia followed eventually by localized lytic
plaques are formed by certain mutant strains of Penicillium chrysogenum grown on an
unbuffered solid medium containing a high lactose concentration. 260 Experiments in-
volving heterokaryosis between genetically marked strains!61.424 have shown that
plaque formation depends on a host mutation (wild-type strains carry a nuclear gene,
dominant in diploids, for resistance to lysis) and a cytoplasmic genetic determinant
which is transmitted with, and could be, a dsRNA virus. Similar lytic plaques are
formed by virus-infected strains of P. citrinum:63.265 P. variable: 63 and Candida al-
bicans. 22o It has been suggested 45 ' that the mutant P. chrysogenum strain could have a
defect in cell wall structure, unable to withstand the increased vacuolation and result-
ant turgor pressure in the older hyphal compartments caused by virus infection. Alter-
 59

natively, the mutation could result in defective negative regulation of dsRNA synthesis
(see Section III. C.l.a) allowing lethal amounts of dsRNA to be produced in some
hyphal compartments.
Several super killer (ski) mutations in Saccharomyces cerevisiae cause elevated levels
of M dsRNA (see Chapter 2), apparently as a result of failure to produce a negative
regulator of M and Ll (LA) dsRNA synthesis (see Section III. C.La). The ski mutants
also conferred cold sensitivity for growth, but only if the cell carried M dsRNA, and
the degree of cold sensitivity correlated with the M dsRNA copy number. 539 The prod-
uct of the host L TS5 (=MAK6) gene is required for maintenance or replication of M
dsRNA and also for low temperature growth. It was therefore suggested S39 that cold
sensitivity in ski mutants is caused by a pleiotropic effect of the elevated levels of M
dsRNA which sequester the host L TS5 product. Ll(LA) dsRNA does not require the
product of the L TS 5 gene and, as predicted from the above model, the elevated levels
of Ll(LA) dsRNA in ski mutants which lack M dsRNA do not cause cold sensitivity.
The discovery and transmission of a die-back disease of the cultivated mushroom,
Agaricus bisporus, the characterization of viruses associated with the disease, and at-
tempts to obtain infection with cell-free virus preparations were described in Sections
LB. 2.b, II.C, lILA (1,3, and 5), and III.B. Here possible relationships between virus
infection and disease will be discussed.
It has not yet proved possible to associate die-back disease with anyone particle
type. In the original study of Hollings in 1962, spherical particles of 25 nm (MV1) and
29 nm (MV2) diameter and bacilliform particles (MV3) were detected. 56 Further studies
revealed a significant inverse correlation between concentrations of MV 1 and MV2 in
sporophores and amounts of mycelial growth on malt agar of isolates taken from
them. 540 In diseased mushrooms from several farms in England between 1966 and
1968,25- and 29-nm particles remained common, bacilliform particles and particles of
50 nm diameter (MV 5) were less common, and particles of 35 nm (MV 4) were rare.
However, in the 2 years 1969 and 1970, this pattern changed, with 35 and 25 nm being
common, 29- and 50-nm particles less common, and bacilliform particles rare. 60 This
apparent change could, at least in part, have been caused by a change in the extraction
procedure. 6J By 1979 Barton and Hollings. '62 again found 25- and 35-nm particles to
be the most prevalent, and this applied to diseased mushrooms from other countries
also. 6J Crops containing high concentrations of 25-nm particles and only low concen-
trations of 35-nm particles and vice versa have been found, implying that possibly
particles of either size could cause disease. More recently in the U.S., Koons et aL 205
and Wach and Romaine S41 have indicated that spherical particles of 19,25, and 34 nm
diameters are more consistently associated with disease than the baciIIiform particles.
However, the particles of 19 nm diameter have never been isolated, and it is still un-
certain whether they could be derived from the 19 x 50 nm bacilliform particles (see
Section II.C). There is, of course, no guarantee that particles of the same diameter
detected on different occasions or in different countries are, in fact, the same virus.
The occurrence of unrelated viruses of the same diameter in the same host is well
documented (see Section III. C.l.d). Serological methods of detection, such as ELISA
or ISEM (see Section I.B.2.d), overcome this problem 542 544 but are specific for the
antisera used and hence may not detect other viruses present.
The discovery of virus particles in apparently healthy mushrooms has added a new
dimension to the mushroom virus problem, since it appears that, like many mycovi-
ruses, mushroom viruses can give rise to latent infections. Nair in 1972 appears to be
the first to have detected virus particles, 19 and 25 nm in diameter, in a commercial
"spawn" line showing good in vitro growth and giving rise to high-yielding mushroom
crops in Australia. S4S There is now general agreement that virus particles can occur in
mushroom spawns, but reports vary from infrequent occurrence at very low
60 Fungal Virology

concentrations 542 544.546547, to ubiquitous occurrence, sometimes in high concentra-

tions. 156- 158548 Passmore and Frost,'56 158 detected spherical (19, 25, and 35 nm) and
bacilliform particles, sometimes in abundant amounts, by electron microscopy in more
than 500 different sporophore samples, whether healthy or diseased, from many dif-
ferent farms and also in all of 26 different mushroom strains from commercial spawns.
Barton and Atkey542 noted that the acid precipitation method used by Passmore and
Frost'56-158 often generates nonspecific virus-like artifacts in the mushroom virus size
range and that in the gel immunodiffusion procedure employed by Tavantzis and
Smith,548 nonspecific reactions from potent mushroom cell antigens such as tyrosinase
are possible. They considered that some of the results of the authors may have been
artifactual.
The key to the mushroom virus story may lie in the virus dsRNA segments rather
than in the particle sizes. Nearly all strains of Saccharomyces cerevisise and Ustilago
maydis are infected with spherical virus particles ca. 40 nm diameter, but only a pro-
portion (those containing specific satellite dsRNA segments) are killers (see Chapters 2
and 3). Similarly, only a proportion of mushroom viruses with particular RNA seg-
ments may be pathogenic, and different mushroom strains could exhibit different de-
grees of resistance or tolerance to different virus genotypes. Buck 500 suggested that
satellite dsRNAs encoding killer toxins might be present in some mushroom virus iso-
lates. If the immunity gene on the dsRNA became inactivated by mutation, infected
mushrooms could then become sensitive to the dsRNA-encoded killer toxin that they
produce; this would be analogous to the suicide mutants associated with killer yeasts
(see Chapter 2). Alternatively, mutation to runaway replication could be involved, but
this seems less likely in view of reports that latently infected mushrooms can contain
as much virus as diseased mushrooms;l56.157 the level of virus in diseased mushrooms
seems not to be any higher than in latent virus infections in many other fungi.
Support for dsRNA variation comes from the few dsRNA analyses reported to date.
Barton and Hollings '62 reported two dsRNA segments each ca. 2.0 kbp for MVI (25-
nm particles) and two dsRNA segments of 2.2 and 2.0 kbp for MV4 (35-nm particles).
Marino et al. 550 extracted six dsRNA segments (3.2, 2.8, 2.6, 2.5, 2.3, and 1.0 kbp)
from diseased mushrooms containing 25- and 34-nm particles (and bacilliform particles
which presumably contained ssRNA). Wach and Romaine 541 detected nine dsRNA seg-
ments in diseased mushroom sporophores; two of these (6.3 and 1.9 kbp) were appar-
ently associated with 19 and 25-nm particles, whereas the other seven (2.5, 2.2, 2.1,
1.3, 1.1,0.73, and 0.68 kbp) were apparently associated with 34-nm particles. In an-
other study they detected dsRNA in 19 of 65 sporophore isolates, which was associated
with decreased yields and deformed sporophores.550.551 Isolates contained from 3 to 12
dsRNA segments in the range 6.3 to 0.31 kbp. It is clear that there is likely to be
considerable variation in the numbers and sizes of dsRNA segments associated with
different isolates containing particles of a given size. Recently dsRNA was detected in
41 "10 of mushroom spawn strains from eight national and international stock cen-
ters,552 confirming previous reports of detection of particles in spawns by electron mi-
croscopy. None of the isolates showed aberrant growth. It would be very interesting to
compare dsRNA segment patterns from such latent infections with those from diseased
mushrooms.
Although mushrooms viruses were the first to be discovered, we still know very
little about their molecular biology and relationships between latency and pathogenic-
ity. Studies comparable to the excellent biochemical and genetical investigations of the
killer system of Saccharomyces and Ustilago (Chapters 2 and 3) are urgently needed to
obtain solutions to these problems. The virus-mushroom system is acknowledged to be
difficult to work with and virus yields from infected mycelium grown under sterile
conditions are reported to be low. 63 However, a molecular genetical approach is prob-
 61

ably the only way forward. In view of the uncertainty concerning the reports of cell-
free transmission of mushroom viruses (Section III. A.5) the possibility that die-back
disease might be caused by another cytoplasmic element (plasmid, defective mitochon-
drial DNA), the diseased mushroom merely being a favorable vehicle for virus repli-
cation, should perhaps not be completely ruled out at the present time.

d. Phytopathogenicity
The debilitation of strains of phytopathogenic fungi suffering from a transmissible
degenerative disease, such as those of HeJminthosporium victoriae (Chapter 5), Rhi-
zoctonia solani (Chapter 4), or Ceratocytis ulmi (Chapters 6 and 7), will perhaps inev-
itably result in a decrease in pathogenicity. However, in Endothia parasitica (Chapter
4) it appears that transmissible hypovirulence does not always affect the saprophytic
ability of the fungus, i.e., a specific effect on pathogenicity is involved. In Gaeuman-
nomyces graminis var. tritid, dsRNA viruses are common and predominantly latent.
Evidence that dsRNA might cause a reduction of pathogenicity in a small propor-
tion of the population, again without effects on saprophytic ability, is discussed in
Chapter 8.
Finkler et al. 55' have recently shown that a cytoplasmic element, probably a specific
segment of dsRNA, is required for pathogenicity in Rhizoctonia solani. This exciting
discovery, which is discussed further in Chapter 4, could create a new dimension both
in fungal virus research and in the genetics of plant pathogenic fungi.

D. Evolution
The origin and evolution of dsRNA mycoviruses, as with that of other vi-
ruses,555,55.,581 must be speculative. The intracellular modes of transmission of dsRNA
mycoviruses distinguish them from most other viruses. In particular for the many
dsRNA mycoviruses which give rise to latent infections, the viruses have probably
evolved along with their hosts'·2 as permanent subcellular particles. The killer system
of Saccharomyces cerevisiae is unique among eukaryotic viruses in the detail in which
interactions of virus and host components have been explored. These studies have re-
vealed the involvement of an unexpectedly large number of host genes and a tightly
balanced control, during host growth, of virus dsRNA replication and host DNA rep-
lication. It is tempting to speculate that dsRNA mycoviruses, like DNA plasmids,
might have evolved to be beneficial to their hosts, e.g., by producing killer proteins or
playing a positive role in pathogenicity. The few reported cases in which dsRNA ap-
pears to have a deleterious effect could be due to virus or host mutation. Indeed it has
been shown in S. cerevisiae that overproduction of a dsRNA segment, due to a host
mutation, results in host pathology (cold sensitivity; see Section III. C.2.c), apparently
due to overutilization by the virus of a specific, essential host component. Suicide killer
mutants in S. cerevisiae are an example of a mutation in a dsRNA which converts a
potentially beneficial product into one harmful to the host.
Alternatively, following the arguments of Doolittle and Sapienza 5S7 and Orgel and
Crick 5S8 on selfish DNA, it is conceivable that mycoviruses are simply convenient ve-
hicles for the replication of "selfish" dsRNA. They have evolved in such a way that
the load on the host cells is small, and selection pressures against their maintenance
may be insufficient to counteract virus mutations which act in the opposite direction
to help to ensure virus survival. The rate of mutation in RNA replication is known to
be much higher than that of DNA because of the absence of proof-reading activities in
RNA replicases. 55 •
Eukaryotic RNA viruses may have originated from a pre-DNA era or, more likely,
from mature mRNAs or from transcripts of individual introns, exons or transposable
elements. In this regard it is noteworthy that transcripts of the transposable element
62 Fungal Virology

copia in Drosophila are enclosed in retrovirus-like particles. 559 The small size of the
genomes of RNA viruses compared to many, but not all, DNA viruses may reflect this
origin and the RNA replication machinery, as it has evolved with its relatively high
error rate, is clearly unsuitable for the faithful copying of very large molecules. Like
other small viruses, dsRNA mycoviruses, at least those in yeast, appear to make effi-
cient use of host proteins for their replication. Replication of (+) ssRNA involves the
formation of a (-) ssRNA intermediate, and it is possible that dsRNA viruses may
have arisen from ssRNA viruses. Alternatively, cellular translational control mecha-
nisms, involving the production of antisense RNA,560 could also lead to the formation
of dsRNA, which might become a self-replicating entity.
DsRNA might be a particularly suitable genome for a persistently intracellular virus
and this could explain why dsRNA is so common in fungi compared to other organ-
isms. A dsDNA virus would probably be too competitive for this role, unless integrated
into host DNA, whereas dsRNA might be able to adapt some of the host dsDNA
replication machinery for its own purposes. DNA-binding proteins have been described
which also bind strongly to dsRNA.561 Furthermore, DNA topoisomerases I and II are
apparently essential for M dsRNA replication in Saccharomyces cerevisiae(see Chapter
2), although their function for this purpose is unknown. It can be speculated that they
might playa role in the packaging and replication of dsRNA which takes place within
particles. If the dsRNA were arranged in a circular conformation with its ends fixed
(which would aid repeated cycles of transcription), it might conceivably adopt a super-
coiled configuration within the particles. Although no DNA copies of virus dsRNA
have been found in three organisms where a search has been made (see Section III.A),
the finding that the Coprinus DNA polymerase b can polymerize dNTPs using a
dsRNA template would justify a wider search. Ability to integrate a DNA copy into
host DNA would be an additional survival mechanism for viruses which have evolved
a completely intracellular existence.
The widespread occurrence of dsRNA viruses in fungi, including at least one in a
lower fungus, the chytridiomycete Allomyces arbuscula, together with their apparently
restricted transmission between species, suggests that these viruses evolved at a very
early stage in the phylogeny of their hosts.

IV. INFECTION OF FUNGI WITH ALIEN VIRUSES

The role of fungi as vectors of plant viruses is well established but there is no evi-
dence that these viruses replicate in the fungus (see Section I. B.l). Reports that fungi
may be naturally infected with plant and bacterial viruses which replicate within fungal
cells were discussed in Sections II. A and II. 0.2, and the conclusion was reached that
there was no unquivocal evidence for such infections. In this section reports that fungi
can be experimentally infected with viruses from nonfungal hosts are considered.

A. Animal Viruses
There have been several reports of the infection of the yeasts, Saccharomyces cere-
visiaeand Candida albicans, with mammalian viruses. Examples include a DNA virus,
polyoma virus (Papovaviridae),"62 and a number of RNA viruses, poliovirus and en-
cephalomyocarditis virus (Picornaviridae),"63 565 Newcastle Disease virus (NDV) (Para-
myxoviridae),"66 and influenza virus (Myxoviridae).566
Inoculations were carried out using exponentially growing yeast cells and virus par-
ticles and additionally, for some viruses, with viral DNA 567 or RNA.564 Virus replica-
tion was assessed mainly by hemagglutination (HA) and infectivity assays. For po-
lyoma virus a 60-fold increase in HA activity and a 10 4 increase in infectivity was
achieved after 72 hr with concomitant decrease in the viability of the yeast cells of up
 63

to 100070. For NDV a 40-fold increase in HA activity and a 10 3 increase in infectivity

was obtained after 24 hr; most of this was in the extracellular fluid, despite the fact
that only 1 to 10070 decrease in cell viabiity occurred in this instance.
There are problems in accepting that all these results imply virus replication in the
fungal hosts. First, the factors which govern the host ranges of animal viruses will be
considered.

1. The first event in infection of an animal cell by a virus is adsorption of the virus
to the surface of the cell; this is a specific process involving binding of an attach-
ment site on the surface of the virion to a specific receptor on the cell surface.
This specific binding determines the host range and also the tissue tropisms within
a host, of the virus. It is highly unlikely that yeast cells would have receptors on
their cell walls for a range of unrelated animal viruses. Even if nonspecific bind-
ing occurred, it is difficult to see how the virus particles could traverse the cell
wall barrier. In the case of the picornaviruses and polyoma virus, it is possible
that only the viral nucleic acid was taken up, and indeed in some of the experi-
ments viral nucleic acids were used as inoculum. However, for NDV and influ-
enza virus, uptake of the viral core particles, which contain the RNA-dependent
RNA polymerase essential for infectivity, would be required.
2. Assuming that the uptake problem had somehow been overcome, host range
would then be determined by the ability of the virus to utilize and adapt the host
cell replication and transcriptional and translational machinery for its own pur-
poses. For polyoma virus this poses an immediate problem.

An early event in the replication of polyoma DNA and that of another papovavirus,
SV40, is formation of the mRNA for large T antigen, a viral protein essential for viral
DNA replication. Formation of this mRNA requires splicing of a precursor transcript.
Recently it has been shown, by introducing the gene for large T antigen into Saccha-
romyces cerevisiaeby means of a DNA vector, that, although polyoma and SV40 pro-
moters are functional in yeast, aberrant splicing occurs and no large T antigen is pro-
duced. 568 Similarly, virus coat proteins, which also require the formation of spliced
mRNAs, were not produced when the coat protein genes were introduced into yeast by
a DNA vector. A different problem arises for NDV. This virus is normally released
from cells by budding and this would be consistent with the increase in extracellular
infectivity with little decrease in yeast cell viability. However, it is difficult to see how
the virus particles could be released through the cell wall barrier. An alternative expla-
nation for at least some of the reports of replication of mammalian viruses in yeast
cells is that inoculum binds nonspecifically to the yeast cell surface and that a substance
produced in the culture medium enhances the infectivity of the inoculum.

B. Plant Viruses
A number of reports indicate that tobacco mosaic virus (TMV) and tobacco necrosis
virus (TN V) might replicate in certain species of Pythium. Increases in infectivity of
up to 70-fold were noted when liquid cultures of P. arrhenomonas and P. sylvaticum
were inoculated with TMV or TNV, 569 571 although decreases in infectivity were re-
corded for similar experiments with P. debaryanum and P. ultimum. TMV was also
detected 2 years after inoculation of a Pythium culture grown on solid medium,570 and
when radioactive TMV was used inoculum (3-tracks were observed originating from
virus material accumulated within the hyphal cell walls.572 Coutts et a1. 573 reported an
increase in infectivity after inoculation of yeast protoplasts with TMV.

C. Conclusions
Although infection by artificial inoculation of fungi with animal and plant viruses
64 Fungal Virology

may have occurred in some cases, unequivocal proof has not been obtained because
assays were based largely on infectivity measurements which could have been enhanced
by other materials produced in the cultures. For some of the viruses, e.g., polyoma
virus, there is difficulty in understanding how infection could have occurred in view of
current knowledge of the molecular biology of the virus and the behavior of its genes
when introduced into yeast. However, yeast would be a very convenient vehicle for
virus propagation and critical reexaminations of the above reports, using assays which
measure the actual amount of virus replication rather than just its infectivity, would
be justified.

V.OUTLOOK
The recognition of the widespread occurrence of dsRNA viruses in fungi, which
followed the discovery of their association with interferon-inducing activities of Peni-
cillium Spp.,71 72.79.81 was followed by a period of great activity worldwide in which
viruses were invoked as the possible cause of many hitherto unexplained phenomena
in mycology. Nearly 20 years later it is known that viruses fulfill only some of these
roles, but nevertheless they do contribute an important new facet to be considered in
fungal biology. Their role as the genetic determinants of killer proteins is well estab-
lished, and, as well as providing unique systems for understanding virus-host relation-
ships at the molecular level, such determinants also have valuable applications, e.g., in
brewing yeasts,530 and potential in producing disease-resistant plants by genetic engi-
neering. 574 Degenerative diseases and hypovirulence associated with dsRNA in plant
pathogenic fungi offer considerable scope for biological control, and cytoplasmically
transmissible determinants of phytopathogenicity, possibly associated with specific
dsRNA segments, offer an exciting new challenge of fundamental importance to plant
pathology. All these subjects, which are active areas of research at the present time,
are described in detail in Chapters 2 to 8. Finally, it is constructive to consider myco-
viruses in the context of extrachromosomal genetic elements in fungi in general; and
Chapter 9 comprises a comprehensive review of current knowledge in what is also
presently a very active field of research.

REFERENCES

1. Burnet, F. M., Virus as Organism, Harvard University Press, Cambridge, 1945.

2. Inouye, T. and Osaki, T., The earliest record of a possible plant virus disease, Abstr. Fifth Int.
Congr. Virology, Strasbourg, France, August 2nd to 7th, 1981,238.
3. Bawden, F. C., Plant Viruses and Virus Diseases, 4th ed., Ronald Press, New York, 1964.
4. Hughes, S. S., The Virus: a History of the Concept, Heinemann, London, 1977, 111.
5. lwanowsky, D. I., Uber die Mosaikkrankheit der Tabakspflanze, Isv. Imp. Akad. Nauk, 3,67,1892.
6. Beijerinck, M. W., Ueber ein Contagium vivum Fluidum als Ursache der Fleckendrankheit der Ta-
baksbhitter, Verh. K. Akad. Wet., 6, 3, 1898.
7. Loeffler, F. and Frosch, P., Berichte der Kommission zur Erforschung der Maul- und Klauenseuche
bei dem Institut fur Infektionskrankheiten in Berlin, CentralbJ. Bakt. Parasitkde., Abt. I, 23, 371,
1898.
8. Twort, F. W., An investigation on the nature of ultramicroscopic viruses, Lancet, 2, 1241, 1915.
9. d'Herelle, F. H., Sur un microbe invisible antagoniste des bacilles dysenteriques, C. R. hebd. Seanc.
Acad. Sci., Paris, 165,373,1917.
10. Matthews, R. E. F., Plant Virology, 2nd ed .. Academic Press, New York, 1981, II.
II. Hughes, S. S., The Virus: a History of the Concept, Heinemann, London, 1977, 106.
12. Luria, S. E., Darnell, J. E., Baltimore, D., and Campbell, A., General Virology, 3rd ed., John Wiley
& Sons, New York, 1978,2.
 65

13. Jacob, F. and Wollman, E. L., Viruses and genes, Sci. Am., 204(6),92, 1961.
14. Signer, E. R., Plasmid formation: a new mode of lysogeny by phage A, Nature, (London), 223, 158,
1969.
15. Robbins, P. W. and Uchida, T., Chemical and macromolecular structure of O-antigen conversion by
bacteriophage e", f. Biol.Chem., 240,384, 1965.
16. Ratti, G., Rappuoli, R., and Giannini, G., The complete nucleotide sequence of the gene coding for
diphtheria toxin in the corynephage omega (tox·) genome, Nucleic Acids Res., II, 6589, 1983.
17. Tooze, J., Ed., DNA Tumor Viruses, 2nd ed., Cold Spring Harbor Laboratory, Cold Spring Harbor,
New York, 1981.
18. Weiss, R., Teich. N., Varmus, H., and Coffin, J., Eds, RNA Tumor Viruses, 2nd ed., Cold Spring
Harbor Laboratory, Cold Spring Harbor, New York, 1982.
19. Bevan, M. W. and Chilton, M.-D., T-DNA of the Agrobacterium Ti and Ri plasmids, Annu. Rev.
Genet., 16, 357, 1982.
20. Shapiro, J. A., Ed., Mobile Genetic Elements, Academic Press, New York, 1983.
21. Levings, C. S., III, The plant mitochondrial genome and its mutants, Cell, 32,659, 1983.
22. Cummings, D. J. and Wright, R. M., DNA sequence of the excision sites of a mitochondrial plasmid
from seresceut Podospora anserina, Nucleic Acids Res., II, 2111, 1983.
23. Bockelmann, H. D., Osiewacz, F. R., Schmidt, F. R., and Schulte, E., Extrachromasomal DNA in
fungi - Organization and Function, in Fungal Virology, Buck, K. W., Ed., CRC Press, Boca Raton,
Fla., 1986, Chap. 9.
24. Wright, R. M. and Cummings, D. J., Integration of mitochondrial gene sequences within the nuclear
genome during senescence in a fungus, Nature (London), 302,86, 1983.
25. Stern, D. B. and Lonsdale, D. M., Mitochondrial and chloroplast genomes of maize have a 12-
kilobase DNA sequence in common, Nature, (London), 299,698, 1982.
26. Farrelly, F. and Butow, R. A., Rearranged mitochondrial genes in the yeast nuclear genome, Nature
(London), 301,296,1983.
27. Van Arsdell, S. W., Denison, R. A., Bernstein, L. B., Weiner, A. M., Manser, T., and Gestland, R.
F., Direct repeats flank three small nuclear RNA psuedogenes in the human genome, Cell,26, II,
1981.
28. Jagadeeswaran, P., Forget, B. G., and Weissman, S. M., Short interspersed repetitive DNA elements
in eucaryotes: transposable DNA elements generated by reverse transcription of RNA pol III tran-
scripts, Cell,26, 141, 1981.
29. Diener, T. 0., Portraits of viruses: the viroid, Intervirology,22, I, 1984.
30. Hadidi, A., Cress, D. E., and Diener, T. 0., Nuclear DNA from uninfected or potato spindle tubor
viroid-infected tomato plants contains no detectable sequences co,mplementary to cloned double-
stranded viroid cDNA, Proc. Natl. Acad. Sci. U.S.A., 78,6932, 1981.
31. Diener, T. 0., Are viroids escaped introns?, Proc. Natl. Acad. Sci. U.S.A., 78, 5014, 1981.
32. Gross, H. J., Krupp, G., Domdey, H., Raba, M.,Jank, P., Lossow, C., Alberty, H., Ramm, K., and
Sanger, H. L., Eur. f. Biochem., 121, 249, 1982.
33. Kiefer, M. C., Ownes, R. A., and Diener, T. 0., Structural similarities between viroids and trans-
posable genetic elements, Proc. Nat1. Acad. Sci. U.S.A., 80,6238, 1983.
34. Campbell, R. N., Fungal vectors of plant viruses in Fungal Viruses, Molitoris, H. P., Hollings, M.,
and Wood, H. A., Eds, Springer-Verlag, Berlin, 1979,8.
35. Temmink, J. H. M., Campbell, R. N., and Smith, P. R., Specificity and site of in vitro acquisition
of tobacco necrosis virus by zoospores of Olpidium brassicae, f. Gen. Viro1., 9,201, 1970.
36. Kassanis, B. and MacFarlane, I., The transmission of satellite viruses of tobacco necrosis by Olpi-
dium brassicae, f. Gen. Virol., 3,227, 1968.
37. Hiruki, C., Host specificity in transmission of tobacco stunt virus by Olpidium brassicae, Virology
33,131,1967.
38. Lin, M. T., Campbell, R. N., Smith, P. R., and Temmink, J. H. M., Lettuce big-vein virus trans-
mission by single-sporangium isolates of Olpidium brassicae, Phytopathology, 60, 1630, 1970.
39. Dias, H. F., Transmission of cucumber necrosis virus by Olpidium cucurbitacearum Barr & Dias,
Virology, 40, 828, 1970.
40. MacFarlane, I., Virus transmission by fungi. Red clover necrotic mosaic virus, Rothamsted Exp. Stn.
Rep. 1981, Part I, 190, 1982.
41. Rao, A. S. and Brakke, M. K., Relation of soil-borne wheat mosiac virus and its fungal vector,
Polymyxa graminis, Phytopathology, 59,581,1969.
42. Thouvenel, J. C. and Fauquet, C., Further properties of peanut clump virus and studies on its natural
transmission, Ann. AppJ. BioI. 97,99, 1981.
43. Inouye, T. and Saito, Y., Barley yellow mosaic virus, CMI/AAB Descriptions of Plant Viruses, No.
143, 1975.
44. Inouye, T. and Fujii S., Rice necrosis mosaic virus, CMI/AAB Descriptions of Plant Viruses, No.
172,1977.
66 Fungal Virology

45. Hebert, T. T. and Panizo, C. H., Oat mosaic virus, CMI/AAB Descriptions of Plant Virus, No. 145,
1975.
46. Toyama, A. and Kusaba, T., Transmission of soil-borne barley yellow mosaic virus. 2. Polymyxa
graminis Led. as vector, Ann. Phytopathol. Soc Jpn., 36,223, 1970.
47. Slykhuis, J. T. and Barr, D. 1. S., Confirmation of Polymyxa graminisas a vector of wheat spindle
streak mosaic virus, Phytopathology, 68,639, 1978.
48. Von Wechmar, M. B., Transmission of brame mosaic virus by Puccinia graminis tritici, Phytopathol.
Z., 99,289, 1981.
49. Tamada, T., Abe, H., and Baba, T., Beet necrotic yellow vein virus and its relation to the fungus,
Polymyxa betae, Proc. 1st Intersect. Congr. Int. Assoc. Microbial. Soc., 3,313, 1975.
50. Jones, R. A. C. and Harrison, B. D., The behavior of potato mop-top virus in soil and evidence for
its transmission by Spongospora subterranae(Wallr.) Lagerh., Ann. Appl. BioI., 63, I, 1969.
51. Lindberg, G. D., A transmissible disease of Helminthosporium victoriae, Phytopathology, 49,29,
1959.
52. Sinden, J. W. and Hauser, E., Report on two new mushroom diseases, Mushroom Sci., I, 98, 1950.
53. Sinden, J. W., The present state of knowledge roncerning viruses on mushrooms, Mushroom Grow-
er's Assoc. Bull., 216,608, 1967.
54. Gandy, D. G., "Watery Stipe" of cultivated mushrooms, Nature (London), 185,482, 1960.
55. Gandy, D. G., A transmissible disease of cultivated mushrooms ("Watery Stipe"), Ann. Appl. BioI.,
48, 427, 1960.
56. Hollings, M., Viruses associated with a die-back disease of cultivated mushroom, Nature (London),
196, 962, 1962.
57. Dieleman-van Zaayen, A. and Temmink, J. H M., A virus disease of cultivated mushrooms in the
Netherlands, Neth. J. Pl. Path. 74,48, 1968.
58. Hollings, M. Stone, O. M. and Atkey, P. T., Mushroom viruses, Rep. Glasshouse Crops Res. lnst.
1967, !O1, 1968.
59. Hollings, M., Gandy, D. G., and Last, F. T., A virus disease of fungus: die-back of cultivated
mushroom, Endeavour, 22, 112, 1963.
60. Hollings, M. and Stone, O. M., Viruses that infect fungi, Annu. Rev. Phytopathol., 9,93,1971.
61. Hollings, M., Mycoviruses: viruses that infect fungi, Adv. Virus Res., 22, I, 1978.
62. Schisler, L. C., Sinden, 1. W., and Sigel, E. M., Etiology, symptomatology and epidemiology of a
virus disease of cultivated mushrooms, Phytopathology, 57,519, 1967.
63. Van Zaayen, A., Mushroom viruses, in Viruses and Plasm ids in Fungi, Lemke, P. A., Ed., Marcel
Dekker, New York, 1979,239.
64. Ushiyama, R., Fungal viruses in edible fungi, in Fungal Viruses, Molitoris, H. P., Hollings, M., and
Wood, H. A., Eds., Springer-Verlag, Berlin, 1979,25.
65. Sinden, J. W., Pathogenicity of mushroom viruses, Dev. Ind. Microbial., 16, 123, 1975.
66. Hollings, M., Mycoviruses and plant pathology, Pant. Dis., 66, 1106, 1982.
67. Isaacs, A. and Lindenmann, J., Virus interference. I. The interferon, Proc. R. Soc. Ser. B, 147, 258,
1957.
68. Shope, R. E., The therapeutic activity of a substance from Penicillium funiculosum Thorn against
swine influenza virus infection of mice, Am. J. Bot., 35,803,1948.
69. Shope, R. E., An antiviral substance from Penicillium funiculosum. I. Effect upon infection in mice
with swine influenza virus and Columbia SK encephalomyelitis virus, J. Exp. Med., 97,601, 1953.
70. Rytel, M. W., Shope, R. E., and Kilbourne, E. D., An antiviral substance from Penicillium funicu-
losum. V. Induction of interferon by helenine. J. Exp. Med., 123, 577, 1966.
71. Lampson, G. P., Tytell, A. A., Field, A. K., Nemes, M. M., and Hilleman, M. R., Inducers of
interferon and host resistance. I. Double-stranded RNA from extracts of Penicillium funiculosum,
Proc. Natl. Acad. Sci. U.S.A., 58,782, 1967.
72. Banks, G. T., Buck, K. W., Chain, E. B., Himmelweit, F., Marks, J. E., Tyler, J. M., Hollings, M.,
Last, F. T. and Stone, O. M., Viruses in fungi and interferon stimulation, Nature (London), 218,
542, 1968.,
73. Buck, K. W. and Blaze, E., umpublished data, 1970.
74. Powell, H. M., Culbertson, C. G., McGuire, J. M., Hoehn, M. M., and Baker, L. A., A filtrate with
chemoprophylactic and chemotherapeutic action against MM and Semliki Forest viruses in mice,
Antibiot. Chemother., 2,432, 1952.
75. Probst, G. W. and Kleinschmidt, W. J., The nature of statolon, an antiviral agent, Fed. Proc., 20,
441,1961.
76. Kleinschmidt, W. J., Cline, J. c., and Murphy, E. 8., Interferon production induced by statolon,
Proc. Natl. Acad. Sci. U.S.A., 52,741, 1964.
77. Kleinschmidt, W. J. and Murphy, E. B., Interferon induction with statolon in the intact animal,
Bacterial. Rev., 31, 132, 1967.
 67

78. Kleinschmidt. W. J. and Probst. G. W .• The nature of statolon. an antiviral agent. Antibiot. Chem-
other.• 12. 298. 1962.
79. Ellis. L. F. and Kleinschmidt. W. J .• Virus-like particles of a fraction of statolon. a mould product.
Nature (London). 215.649. 1967.
80. Buck. K. W .• Chain. E. B. and Darbyshire. J. E .• High cell wall galactosamine content and virus
particles in Penicillium stoloniferum, Nature. (London), 223. 1273, 1969.
81. Kleinschmidt. W. J .• Ellis. L. F .• van Frank. R. M .• and Murphy, E. B .• Interferon stimulation by a
double-stranded RNA of a mycophage in statolon preparations. Nature (London). 220, 167. 1968.
82. Belgian Patent No. 691. 881. Glaxo Laboratories Ltd .• 1968.
83. Cooke. P. M. and Stevenson, J. W .• An antiviral substance from Penicillium cyaneo-fulvum Biourge.
I. Production and partial purification. Can. J. Microbiol .• II. 913, 1965.
84. David-West. T. S .• Cooke. P. M .• and Stevenson, J. W .• New methods of production and partial
purification of an antiviral agent from Penicillium cyaneo-fulvum, Can. J. Microbial., 14. 189. 1968.
85. Banks. G. T., Buck. K. W .• Chain. E. B.. Darbyshire. J. E .• and Himmelweit, F., Virus-like particles
in penicillin producing strains of Penicillium chrysogenum. Nature (London), 222,89, 1969.
86. Cox. R. A., Kanagaiingham. K .• and Sutherland, E. S .• Double-helical character of ribonucleic acid
from virus-like particles found in Penicillium chrysogenum, Biochem. J., 120, 549, 1970.
87. Lemke, P. A. and Ness. T. M., Isolation and characterisation of a double-stranded ribonucleic acid
from Penicillium chrysogenum, J. Virol.,6, 813. 1970.
88. Buck. K. W., Chain. E. B., and Himmelweit. F .• Comparison of interferon induction in mice by
purified Penicillium chrysogenum virus and derived double-stranded RNA. J. Gen. Virol., 12. 131,
1971.
89. Banks. G. T .• Buck, K. W .• Chain. E. B., Darbyshire. J. E., and Himmelweit. F .• Penicillium cy-
aneo-fulvum virus and interferon stimulation. Nature (London). 223. 155. 1969.
90. Banks. G. T .• Buck, K. W .• Chain. E. B., Darbyshire. J. E .• Himmelweit. F .• Ratti. G .• Sharpe. T.
J .• and Planterose. D. N .• Antiviral activity of double-stranded RNA from a virus isolated from
Aspergillus foetidus, Nature (London), 227,505,1970.
91. Marcus, P. I.. Interferon induction by viruses: one molecule of dsRNA as the threshold for interferon
induction. in Interferon. VoL 5. Gresser. I.. Ed., Academic Press. Inc .• London. 1984. 116.
92. Baglioni, C. and Nilsen, T. W .• Mechanisms of antiviral action of interferon in Interferon, Vol. 5.
Gresser. I.. Ed .• Academic Press. Inc .• London. 1984. 23.
93. Field, A. K .• Tytell. A. A .• Lampson. G. P., and Hilleman. M. R .• Inducers of interferon and host
resistance. II. Multistranded synthetic polynucleotide complexes. Proc. NatJ. Acad. Sci. U.S.A., 58,
1004, 1967.
94. Field. A. K .• Lampson, G. P .• Tytell. A. A .• Nemes. M. M .• and Hilleman, M. R., Inducers of
interferon and host resistance. IV. Double-stranded replicative form RNA (MS2-RNA) from E. coli
infected with MS2 coliphage. Proc. Natl. Acad. Sci. U.S.A. 58.2102, 1967.
95. Tytell, A. A., Lampson. G. P., Field, A. K .• and Hilleman, M. R., Inducers of interferon and host
resistance. III. Double-stranded RNA from reovirus type 3 virions (reo 3-RNA), Proc. Natl. Acad.
U.S.A., 58. 1719. 1967.
96. Buck. K. W .• Fungal viruses. double-stranded RNA and interferon induction, in Viruses and Plas-
mids in Fungi, Lemke. P. A .• Ed., Marcel Dekker. New York. 1979, I.
97. Dewdney. J. M., Clinical and veterinary applications of double-stranded ribonucleic acid from fungal
viruses. in Biologically Active Substances - Exploration and Exploitation, Hems, D. A., Ed., John
Wiley & Sons. Chichester. 1977. 149.
98. Hewitt. C. W. and Adler. J. P., Murine immunosuppression with mycoviral dsRNA, lmmunophar-
macology, 5. 103.1982.
99. Alexander. P. and Evans, R .• Endotoxin and double-stranded RNA render macrohages cytotoxic,
Nature (London) New Biology. 232,76,1971.
100. Stinebring. W. and Absher. M .• The double-bitted axe: A study of toxicity of interferon releasers, in
Biological Effects of PolynucIeotides. Proc. Symp. Molec. BioI., Beers, R. F. and Braun, W., Eds.,
Springer-Verlag. New York, 1971.249.
101. Kingsman, S. M. and Kingsman, A. J .• The production of interferon in bacteria and yeast, in Inter-
ferons: from Molecular Biology to Clinical Application, Society for General Microbiology Symp.
No. 35. Burke. D. C and Morris. A. G., Eds .• Cambridge University Press, Cambridge, 1983,211.
102. Slocombe, P .• Easton, A .• Boseley, P. A. and Burke. D. C. High level expression of an interferon
a2 gene cloned in phage M13 mp7 and subsequent purification with a monoclonal antibody, Proc.
Nat/. Acad. Sci. U.S.A .• 79.5455, 1982.
103. Scott. G. M., The toxic effects of interferon in man, in Interferon, Vol. 5, Gresser, L, Ed., Academic
Press, Inc .• London. 1984.
104. Burke. D. C. and Morris, A. G .• Eds .• Interferons: from Molecular Biology to Clinical Application,
Society for General Microbiology Symp. No. 35, Cambridge University Press, Cambridge, 1983.
105. Finter, N. B .• Ed., Interferon. VoL I to 4, Elsevier Science Publishers, Amsterdam, 1985.
68 Fungal Virology

106. Lemke, P. A., Fungal viruses and agriculture, in Virology in Agriculture, Allanhald Osmun and Co.,
Montclair, New Jersey, 1977, 159.
107. Day, P. R. and Dodds, J. A., Viruses of plant pathogenic fungi, in Viruses and Plasmids in Fungi,
Lemke, P. A., Ed., Marcel Dekker, New York, 1979,201.
108. Adler, J. P., Screening for viruses in human pathogenic fungi, m Fungal Viruses, Mohtoris, H. P.,
Hollings, M., and Wood, H. A., Eds., Springer-Verlag, 1979, 129.
109. Bevan, E. A., Herring, A. J., and Mitchell, D. J., Preliminary characterisation of two species of
dsRNA in yeast and their relationship to the killer character, Nature (London), 245,81,1973.
110. Herring, A. J. and Bevan, E. A., Virus-like particles associated with the double-stranded RNA species
found in killer and sensitive strains of the yeast Saccharomyces cerevisiae, 1. Gen. Virol., 22, 387,
1974.
111. Wood, H. A. and Bozarth, R. F., Heterokaryon transfer of virus-like particles associated with a
cytoplasmically mherited determinant in Ustilago maydis, Phytopathology, 63, 1019, 1973.
112. Koltin, Y. and Day, P. R., Inheritance of killer phenotypes and double-stranded RNA in Ustilago
maydis, Proc. Nat!. Acad. Sci. U.S.A., 73,594, 1976.
113. Sanderlin, R. S. and Ghabrial, S. A., Physicochemical properties of two distinct types of virus-like
particles from Helminthosporium victoriae, Virology, 87, 142, 1978.
114. Castanho, B. and Butler, E. E., Rhizoctonia decline: a degenerative disease of Rhizoctonia solani,
Phytopathology, 68, 1505, 1978.
115. Castanho, B., Butler, E. E. and Shepherd, R. J., The association of double-stranded RNA with
Rhizoctonia decline, 68, 1515, 1978.
116. Bozarth, R. F., Mycoviruses: A new dimension in microbiology, in Environmental Health Perspec-
tives, U.S. Department of Health, Education and Welfare, Washington, D.C., 1972,23.
117. McFadden, J. J. P., Buck, K. W., and Rawlinson, C. J., Infrequent transmission of double-stranded
RNA virus particles but absence of DNA proviruses in single ascospore cultures of Gaeumannomyces
graminis, J. Gen. Vlrol., 64, 927,1983.
118. Bozarth, R. F., The physico-chemical properties of mycoviruses, in Viruses and Plasm ids in Fungi,
Lemke, P. A., Ed., Marcel Dekker, New York, 1979,43.
119. Moffitt, E. M. and Lister, R. M., Detection of mycoviruses using antiserum specific for dsRNA,
Virology, 52,301, 1973.
120. Adler, J. P. and Del Vecchio, V. G., Specialised assays for detection of fungal viruses and double-
stranded RNA, in Viruses and Plasm ids in Fungi, Lemke, P. A., Ed., Marcel Dekker, New York,
1979, 325.
121. Lister, R. M., Serological screening for fungal viruses, in Fungal Viruses, Molitoris, H. P., Hollings,
M., and Wood, H. A., Springer-Verlag, 1979, 150.
122. Zanzinger, D. H. and Tavantzis, S. M., Detection of double-stranded ribonucleic acid (dsRNA) by
the enzyme-linked immunosorbent assay (ELISA), Phytopathology, 72,268, 1982.
123. Zanzinger, D. H. and Tavantzis, S. M., Improved detection of heterologous double-stranded RNA
(dsRNA) antigens by the indirect method of the solid-phase enzyme immunoassay, Phytopathology,
72, 954, 1982.
124. Moffitt, E. M. and Lister, R. M., Application of a serological screening test for detecting double-
stranded RNA mycoviruses, Phytopathology, 65,851,1975.
125. Colby, C. and Duesberg, P. H., Double-stranded RNA in vaccinia virus infected cells, Nature (Lon-
don), 222,940, 1969.
126. Morris, T. J. and Dodds, 1. A., Isolation and analysis of double-stranded RNA from virus-infected
plant and fungal tissue, Phytopathology, 69,854, 1979.
127. Franklin, R. M., Purification and properties of the replicative intermediate of the RNA bacteriophage
R17, Proc. Natl. Acad. Sci. U.S.A., 55,1505,1966.
128. McGinty, R. M., Buck, K. W., and Rawlinson, C. J., Virus particles and double-stranded RNA in
isolates of Phialophora sp. with lobed hyphopodia, Phialophora graminicola and Gaeumannomyces
graminisvar. graminis, Phytopath. Z., 102, 153, 1981.
129. Hayes, R. 1., Buck, K. W., and Brunt, A. A., Double-stranded and single-stranded subgenomic
RNAs from plant tissue infected with tomato bushy stunt virus, J. Gen. Virol. 65, 1239, 1984.
130. Baltimore, D., Purification and properties of poliovirus double-stranded ribonucleic acid, J. Mol.
BioI. 18, 421,1966.
131. Ainsworth, G. C., The number of fungi, in The Fungi: An Advanced Treatise, Vol. III, Ainsworth,
G. C. and Sussman, A. S., Eds., Academic Press, New York, 1968,505.
132. Gibbs, A. J. and Harrison, B. D., Plant Virology: The Principles, Edward Arnold, London, 1976.
133. Buck, K. W., Carlile, M. J., and Border, D. J., unpublished data, 1970.
134. Carlile, M. J., Cell fusion and somatic incompatibility in the myxomycetes, Ber. Dtsch. Bot. Ges.
Bd., 86, S.123, 1973.
135. Goodman, E. M. and Rusch, H. P., Glycogen in Physarum polycephalum, Experientia, 25, 580,
1969.
 69

136. Mori, K., Studies on virus-like particles in Lentinus edodes, in Proc. 1st Intersect. Congr. lAMS,
Vol. 5, Hasegawa, T., Ed., Science Council of Japan, Tokyo, 1975,396.
137. Ainsworth, G. C., IntroductiOn and keys to higher taxa, in The Fungi: an Advanced Treatise, Vol.
IVB, Ainsworth, G. C., Sparrow, F. K., and Sussman, A. S., Academic Press, New York, 1973, 1.
138. Olive, L. S., The Mycetozoa: a revised classification, Bot. Rev., 36,59, 1970.
139. Dieleman-van Zaayen, A., Virus-like particles in a weed mould growing on mushroom trays,. Nature
(London), 216,595, 1967.
140. Dieleman-van Zaayen, A., Igesz, O. and Finch, J. T., Intrecellular appearance and some morpholog-
ical features of virus-like particles in an ascomycete fungus, Virology, 42, 534, 1970.
141. Yarwood, C. E. and Hecht-Poinar, E., Viruses from rusts and mildews, Phytopathology, 63, Illl,
1973.
142. Nienhaus, F., Tobacco mosaic virus strains extracted from conidia of powdery mildews, Virology,
46, 504, 1971.
143. McDonald, J. G. and Heath, M. C., Rod-shaped and sphencal viruslike particles in cowpea rust
fungus, Can. J. Bot., 56, 963, 1978.
144. Faivre-Amoit, A., Keusseoglou, J., Bondoux,P., Lapierre, H., Albouy, J., Molin, G., Kusiak, C.,
and Morand, J.-C., Presence of different types of virus like particles in Ostracoderma sp., a fungus
species often growing in mushroom beds, Ann. Phytopathol., 4,420, 1972.
145. Lapierre, H., Faivre-Amiot, A., and Molin, G., Isolation of virus-like particles associated to Verti-
cillium fungicolaagent of a dry bubble Agaricus bisporus, Ann. Phytopathol., 5,323,1973.
146. Lapierre, H., Lecoq, H., Spire, D., Boissonnet, M., and Chevaugeon, J., Les mycovirus, Ann. Phy-
topathol. 7,236, 1975.
147. Albouy, J. and Lapierre, H., Observations en microscope electronique de souches virosees de Myco-
gone perniciosa (Magn) Cost et Dufour agent d'une mole du champignon de couche, Ann. Phyto-
pathol. 4,353, 1972.
148. Lapierre, H., Faivre-Amiot, A., Kusiak, C., and Molin, G., Particles de type viral associees au My-
cogone perniciosa Magnus, agent d'une des moles du champignon de couche, Compt. Rend. Acad.
Sci. Paris., Ser. D., 274,1867,1972.
149. Huttinga, H., Wichers, H. J., and Dieleman-van Zaayen, A., Filamentous and polyhedral viruslike
particles in Boletus edulis, Nether!. J. Plant. Pathol., 81,102,1975.
150. Inoue, T., Furatani, H., and Nishikado, Y., Virus-like particles in Len tin us edodes, Ann. Phyto-
pathol. Soc. Jpn, 36,356, 1970.
151. Ushiyama, R. and Hashioka,Y., Viruses associated with Hymenomycetes. 1. Filamentous virus-like
particles in the cells of a fruit body of Shiitake, Len tin us edodes (Berk). Sing., Rep. Tottori Mycol.
lnst. (Jpn.), 12, 53, 1973.
152. Yamashita, S., Doi, Y., and Yora, K., Electron microscopic study of several fungal viruses, Proc. 1st
Intersect. Congr. lAMS, 3,340, 1975.
153. Lange, L. and Olson, L. W., Virus-like particles in Synchytrium endobioticum the infectious agent
of potato wort disease, Phytopath. Z., 95,217, 1979.
154. Dieleman-van Zaayen, A. and Temmink, J. H. M., A virus disease of cultivated mushrooms in the
Netherlands. Neth. J. Plant Pathol., 74.48, 1968.
155. Dieleman-van Zaayen, A., Intracellular appearance of mushroom virus in fruiting bodies and basi-
diospores of Agaricus bisporus, Virology, 47,94, 1972.
156. Passmore, E. L. and Frost, R. R., The detection of virus-like particles in mushrooms and mushroom
spawns, Phytopath. Z., 80,85,1974.
157. Passmore, E. L. and Frost, R. R., The detection and occurrence of virus-like particles in extracts of
mushroom sporophores (Agaricus hortensis rCookel Pilat), Phytopath. Z., 95,346, 1979.
158. Passmore, E. L. and Frost, R. R., Virus-like particles in mushroom spawns, Phytopath. Z., 97,68,
1980.
159. Moyer, J. W. and Smith, S. H., Purification and serological detection of mushroom virus-like par-
ticles, Phytopathology, 67, 1207, 1977.
160. Saksena, K. N., Isolation and large-scale purification of mushroom viruses, Dev. Ind. Microbiol. 16,
134, 1975.
161. Molin, G. and Lapierre, H., L'acide nucleique des virus de champignons : Cas de ViruS de I' Agaricus
bisporus, Ann. Phytopathol. 5,233, 1973.
162. Barton, R. J. and Hollings, M., Purification and some properties of two viruses infecting the culti-
vated mushroom Agaricus bisporus, J. Gen. Virol., 42, 231, 1979.
163. Tavantzis, S. M., Romaine, C. P., and Smith, S. H., Purification and partial characterisation of a
bacilliform virus from Agaricus bisporus : a single-stranded RNA mycovirus, Virology, 105, 94, 1980.
164. Tavantzis, M., Romaine, C. P., and Smith, S. H., Mechanisms of genome expression in a single-
stranded RNA virus from the cultivated mushroom (Agaricus bisporus [Lange] Imbach), Phytopath.
Z., 106, 45,1983.
70 Fungal Virology

165. Blattny, C, Kralik, 0., Veselsky, J., Kasala, B., and Herzova, H., Particles resembling virions ac-
companying the proliferation of agaric mushrooms, Ceska Mycol., 27, I, 1973.
166. Corbett, M. K. and Styer, E. L., Intranuclear viruslike partIcles In Phytophthora infestans, Proc.
Am. Phytopath. Soc. 3,332,1977.
167. Styer, E. L. and Corbett, M. K., Intranuclear virus-like particles of Phytophthora infestansand P.
parasitica vaT. parasitica, Mycovirus News., 6, 16, 1978.
168. Dodds, J. A., Association of type 1 virus-like dsRNA with club-shaped particles in hypovirulent
strains of Endothia parasitica, Virology, 107, 1, 1980.
169. Chmel0, R. A. and Kaczmarczyk, W. J., Isolation and partIal characterisation of a virus-like particle
and its genome associated with Endothia parasitica strain 43, in Proc. USDA Forest Service American
Chestnut Cooperators Meeting, Smith H. C. and MacDonald, W. L., Eds., West Virginia University
Books, Morgantown, 1982, 165.
170. Albouy, J., Etude ultramicroscopique du complex viral de la "goutte seche" de carpophores d' Agar-
icus bisporus, Ann. Phytopathol. 4, 39, 1972.
171. Lesemann, D. E. and Koenig, R., Association of clubshaped viruslike particles with a severe disease
of Agaricus bisporus, Phytopathol. Z.89, 161, 1977.
172. Albouy, J., Some morphological changes in fungi induced by fungal viruses in Fungal Viruses, Mol-
itoris, H. P., Hollings, M. and Wood, H. A., Eds., Springer-Verlag, Berlin, 1979,78.
173. Tuveson, R. W. and Peterson, 1. F., Virus-like particles in certain slow-growing strains of Neuro-
spora crassa, Virology, 47,527, 1972.
174. Kuntzel, H., Borath, Z., Ali, L, Kind, J., and Althaus, N.-H., Virus-like particles in an extranuclear
mutant of Neurospora crassa, Proc. Nat. Acad. Sci. U.S.A., 70,574, 1973.
175. Perkins, F. 0., Fine structure of lower marine and estuarine fungi, in Recent Advances in Aquatic
Mycology, Jones, E. B G., Ed., Halsted Press, New York, 1976,279.
176. Kazama, F. Y. and Schornstein, K. L., Herpes-type virus particles associated with a fungus, Science,
177,696,1972.
177. Kazama, F. Y. and Schornstein, K. L., Ultrastructure of a fungus herpes-type virus, Virology, 52,
478, 1973.
178. Volkoff, O. and Walters, T., Virus-like particles in abnormal cells of Saccharomyces carlsbergensis,
Can. J. Genet, Cytol. 12, 621,1970.
179. Kozlova, T. M., Virus-like particles in yeast cells, Microbiologiya, 42, 745, 1973.
180. Weber, H., Ultrastructural evidence for viruses in ascomycetes and fungi imperfecti, in Viruses and
Plasmids in Fungi, Lemke, P. A., Ed., Marcel Dekker, New York, 1979,363.
181. Tikchonenko, T. L, Velikodvorskaya, G. A., Bobkova, A. F., Bartashevich, Y. E., Chaplygina, N.
M., and Maksimova, T. S., New fungal viruses capable of reproducing in bacteria, Nature (London),
249,454, 1974.
182. Tikchonenko, T. L, Viruses of fungi capable of replication in bacteria (PB viruses), in Comprehensive
Virology, Vol. 12" New Characterized Protist and Invertebrate Viruses, Plenum, New York and
London, 1978,235.
183. Tuveson, R. W., Sargent, M. L. and Bozarth, R. F., Purification of a small virus-like particle from
strains of Neurospora crassa, Annu. Meet. Am. Soc. Microbiol. Abst., 1975, 216.
184. Bozarth, R. F., Wood, H. A., and Goenaga, A., Virus-like particles from a culture of Diplocarpon
rosae, Phytopathology, 62,493, 1972.
18S. Spire, D., Virus des champignons, Physiol. Veg., 9,555,1971.
186. Buck, K. W., Lhoas, P., Border, D. J. and Street, B. K., Virus particles in yeast, Biochem. Soc.
Trans., 1,1141,1973.
187. Adler, J., Wood, H. A., and Bozarth, R. F., Virus-like particles in killer, neutral and sensitive strains
of Saccharomyces cerevisiae, J. ViroL, 17,472, 1976.
188. Bruenn, J. A., Virus-like particles of yeast, Annu. Rev. Microbiol. 34,49, 1980.
189. Bruenn, J., The killer systems of Saccharomyces cerevisiae and other yeasts, in Fungal Virology,
Buck, K. W., Ed., CRC Press, Boca Raton, Fla., 1986, Chap. 3.
190. Lindegren, C. C. and Bang, Y. N., The zymophage, Antonie van Leeuwenhoek, J. Microbiol. Serol.
27, I, 1961.
191. Hirano, T., Lindegren, C. C. and Bang, Y. N., J. Bacteriol., 83,1363,1962.
192. Lindegren, C. C., Bang, Y. N., and Hirano, T., Progress report on Zymophage, Trans. N. Y. Acad.
Sci., 24, 540, 1962.
193. Groves, D. P., Clarke, J. 1. and Oliver, S. G., Isolation and characterization of a double-stranded
RNA virus-like particle from the yeast Yarrowia lipolytica, Curro Genet., 7,185, 1983.
194. Rawlinson, C. 1., Hornby, D., Pearson, V. and Carpenter, 1. M., Virus-like particles in the take-all
fungus, Gaeumannomycesgraminis, Ann. Appl. BioI., 74,209,1973.
195. Buck, K. W., A new double-stranded RNA virus from Gaeumannomyces graminis, J. Gen. Virol.
65,978, 1984.
 71

196. Lapierre, H., Lemaire, J.-M. Jouan, B., and Molin, G., Mise en evidence de particules virales asso-
a
ciees une porte de pathogenicitee chez Ie pietin-echaudage des cereales, Ophiobolus grammis Sacc.,
C. R. Acad. Sci. Paris, Ser. D., 271,1833,1970.
197. Lapierre, H. Etude de I'influence des vires sur les champlgnons phytopathogenes du sol, in Perspec-
tives de Lutte Biologique des Champignons Parasites des Plantes Cultivees et les Pourritures des
Tissus Ligneux, Station Federale de Recherches Agronomiques de Lausanne, Switzerland, 1973,62.
198. Buck, K. W., Almond, M. R., McFadden, J. J. P., Romanos, M. A., and Rawlinson, C. J., Prop-
ertIes of thirteen viruses and virus variants obtained from eight isolates of the wheat take-all fungus,
Gaeumannomyces graminia var. tritici, J. Gen. Vlrol., 53,235, 1981.
199. Jamil, N. and Buck, K. W., Apparently identical viruses from Gaeumannomyces graminisvar. tritici
and Phialophora sp. (lobed hyphopodia), Trans. Brit. Mycol. Soc., 83,519, 1984.
200. Jamil, N., Buck, K. W., and Carlile, M. H., Sequence relationships between virus double-stranded
RNA from isolates of Gaeumannomyces graminis in different vegetative compatibility groups, J.
Gen. Virol.65, 1741, 1984.
201. Frick, L. J. and Lister, R. M., Serotype variability in virus-like particles from Gaeumannomyces
graminis, Virology, 85,504, 1978.
202. Wyn-Jones, A. P. and Whalley, J. S., Virus-like particles in Hypoxylon multiforme, Mycopatholo-
gia, 61,63,1977.
203. Tuveson, R. W. and Sargent, M. L., Characterisation of virus-like particles from slow-growing
strains of Neurospora crassa, Genetics, 83,577, 1976.
204. Lechner, J. F., Scott, A., Aaslestad, H. G., Fuscaldo, A. A., and Fuscaldo, K. E., Analysis of a
virus-like particle from N. crassa, Abstr. Annu. Meet. Am. Soc. Micr obiol, 1972, 188.
205. Koons, K., Raid, L. C., and Romaine, C. P., An analysis of the virus-like particles associated with
La France disease, Phytopathology, 73,369, 1983.
206. Shahriari, H., Kirkham, J. B., and Casselton, L. A., Virus-like particles in the fungus Coprinus
lagopus, Heredity, 31, 428, 1973.
207. Blattny, C. and Kralik, 0., A virus disease of Laccaria laccata (Scop. ex. Fr.) Cooke and some other
fungi, Ceska Mykol., 22,161,1968.
208. Ushiyama, R. and Nakai, Y., Viruses associated with hymenomycetes. II. Presence of polyhedral
virus-like particles in shiitake mushrooms, Lentinus edodes (Berk.) Sing., Rep. Tottori Mycol. Inst.,
12,53,1975.
209. Ushiyama, R., Nakai, Y., and Ikegami, M., Evidence for double-stranded RNA from polyhedral
virus-like particles in Lentinus edodes(Berk.) Sing., Virology, 77,880,1977.
210. Ushiyama, R. and Nakai, Y., Ultrastructural features of virus-like particles from Lentinus edodes,
Virology, 123, 93, 1982.
211. Koltin, Y., Berick, R., Stamberg, J., and Ben-Shaul, Y., Virus-like particles and cytoplasmic inher-
itance of plaques in a higher fungus, Nature (London) New BioI., 241, 108, 1973.
212. Zanzinger, D. H., Bandy, B. P. and Tavantzis, S. M., High frequency of finding double-stranded
RNA in naturally occurring isolates of Rhizoctonia solani, J. Gen. Virol., 65, 1601, 1984.
213. Finkler, A., Austen, M., Barash, I., Koltin, Y., and Sneh, B., Characterization of double-stranded
RNA (dsRNA) and virus-like particles in virulent and hypovirulent strains of Rhizoctonia solani,
Abstr. Fourth Int. Congr. Plant Pathology, Melbourne, Australia, Rowprint Services, Victoria, 1983,
114.
214. Rijo, L. and Sargent, J. A.,The fine structure of the coffee leaf rust, Hemileia vastatrix, Can. J. Bot.
52, 1363, 1974.
215. Rawlinson, C. J. and Maclean, D. J., Virus-like particles in axenic cultures of Puccinia graminis
tritici, Trans. Br. Mycol. Soc., 61,590, 1973.
216. Ushiyama, R. and Nakai, Y., Intracellular virus-like particles in Lentinus edodes, J. (Jen. Virol.,46,
507,1980.
217. Coffey, M. D., Palevitz, B. A., and Allen, P. J., The fine structure of two rust fungi, Puccinia
helianthi and Melampsora lini, Can. J. Bot., 50, 231, 1972.
218. Lecoq, H., Spire, D., Rapilly, F., and Bertrandy, J., Mise en evidence de particles de type viral chez
les Puccinia, C. R. Acad. Sci., Paris, Ser. D., 27, 1599, 1974.
219. Bozarth, R. F., Koltin, Y., Weisman, M. B., Parker, R. E., Dalton, R. E., and Steinlauf, R., The
molecular weight and packaging of dsRNA in the mycovirus from Ustilago killer strains, Virology,
113,492, 1981.
220. Mehta, R. J., Nash, C. H., III, and Bozzola, J. J., Virus-like particles and lytic plaque formation in
lawns of Candida albicans, J. Bact., 152, 502, 1982.
221. Nesterova, G. F., Kyarner, Y., and Soom, Y. 0., Virus-like particles in Candida tropicalis, Mikro-
biologiya, 42, 162, 1973.
222. Nesterova, G. F., Karner, J., and Kozlova, T. M., Possible intracellular transfer of virus-like particles
from Candida tropicalis and Saccharomyces cerevisiae, Mikrobiologiya, 43,930, 1974.
72 Fungal Virology

223. Nesterova, G. F., Karner, J., and Kozlova, T. M., On the possibility of transfer of virus-like particles
from the cells of Candida tropicalis IOto the cells of Saccharomyces cerevisiae, Mikrobiologiya, 43,
1095,1974.
224. Rawlinson, C. J., Carpenter, J. M., and Muthyalu, G., Double-stranded RNA virus 10 Colletotri-
chum lindemuthianum, Trans. Brit. Mycol. Soc., 65, 305, 1975.
225. Lecoq. H. and Delhotal, P., Etude des virus de Colletotrichum lindemuthianum. I. Characterisation
des particules virales. Ann. Phytopathol. 8,307, 1976.
226. Isaac, P. K. and Gupta, S. K., A virus-like infection of Alternaria tenUls, 10th Int. Bot. Congr.,
Edinburgh, July 22nd to 28th, 1964,390.
227. Lau, R. H., Reid, J. and Kim, W. K., Virus-like particles in Alternaria tenuis, Mycologia, 73,97,
1981.
228. Mackenzie, D. E. and Adler, J. P., Virus-like particles in toxigenic Aspergilli, Abstr. Annu. Meet.
Am. Soc. Microbiol., American Society for Microbiology, Washington, D. C., 1972,68.
229. Wood, H. A., Bozarth, R. F., Adler, 1. P., and Mackenzie, D. W., Proteinaceous virus-like particles
from an isolate of Aspergillus flavus, J. Virol., 13, 532, 1974.
230. Schmidt, F. R., Davis, N. D., Kelley, V. C., Diener, U. L., and Lemke, P. A., Minor nucleic acid
components in Aspergillus flavus, Abstr. Ann. Meet. Am. Soc. Micro bioi. , 83, 135, 1983.
231. Ratti, G. and Buck, K. W., Virus particles in Aspergillus foetidus:a multicomponent system, J. Gen.
Virol., 14, 165, 1972.
232. Buck, K. W. and Ratti, G., Biophysical and biochemical properties of two viruses isolated from
Aspergillus foetidus, J. Gen. Virol., 27,211, 1975.
233. Buck, K. W. and Ratti, G., Molecular weight of double-stranded RNA: a re-examination of Asper-
gillus foetidusvirus S RNA components, J. Gen. Virol., 37,215,1977.
234. Buck, K. W., Girvan, R. F., and Ratti, G., Two serologically distinct double-stranded ribonucleic
acid viruses isolated from Aspergillus niger, Biochem. Soc. Trans., I, 1138, 1973.
235. Kim, 1. W. and Bozarth, R. F., Intergeneric occurrence of related fungal viruses: the Aspergillus
ochraceous virus complex and its relationship to the Penicillium stoloniferum virus S, J. Gen. Viral.,
66,1991,1985.
236. Day, L. E. and Ellis, L. F., Virus-like particles in Cephalosporium acremonium, Appl. Microbiol.,
22,919, 1971.
237. Cole, G. T., A viral infection of the hyphomycetous fungus, Briosia cubispora, Proc. Microscop.
Soc. Canada, II, 60, 1975.
238. Lapierre, H., Kusiak, C., and Molin, G., Some properties of viral particles of Fusarium culmorum
Smith, Ann. Phytopath, 6, 500, 1974.
239. Chosson, J. F., Lapierre, H., Kusiak, C., and Molin, G., Presence of virus like particles in fungi of
the genus Fusarium, Ann. Phytopath. 5, 324, 1973.
240. Spire, D., Ferault, A. C., Bertrandy, J., Rapily, F., and Skajennikoff, M., Particles de type viral
dans un champignon hyperparasite: Gonatobotrys, Ann. Phytopa th 01. , 4,419, 1972.
241. Dunkle, L. D., The relation of virus-like particles to toxin producing fungi in corn and sorghum,
Proc. 28th Annu. Corn. Sorghum Res. Conf., American Seed Trade Association, Chicago, 1974, 72.
242. Bozarth, R. F., Wood, H. A. and Nelson, R. R., Virus-like particles in virulent strains of Helmin-
thosporium maydis, Phytopathology, 62,748, 1972.
243. Bozarth, R. F., Biophysical and biochemical characterization of virus-like particles containing a high
molecular weight dsRNA from Helminthosporium maydis, Virology, 80, 149, 1977.
244. Spire, D., Ferault, A. C., and Bertrandy, 1., Observation de particles des type viral dans une souche
d' Helminthosporium oryzae, Br. de H., Ann. Phytopa th 01. , 4,359, 1972.
245. Ghabrial, S. A., Sanderlin, R. S., and Calvert, L. A., Morphology and virus-like particle content of
Helminthosporium victoriaecolonies regenerated from protoplasts of normal and diseased isolates,
Phytopathology, 69,312, 1979.
246. Ghabrial, S. A., A Transmissible disease of Helminthosporium victoriae, in Fungal Virology, Buck,
K. W., Ed., CRC Press, Boca Raton, Fla., 1986, chap. 6.
247. Barton, R. J., Mycogone perniciosa virus, Rep. Glasshouse Crops Res. Inst. 1977, 1978, 133.
248. Wood, H. A., Bozarth, R. F. and Mislivec, P. B., Virus-like particles associated with an isolate of
Penicillium brevicompactum, Virology, 44,592, 1971.
249. Volkoff, 0., Walters, T., and Dejardin, R. A., An examination of Penicillium chrysogenum type
virus particles, Can. J. Microbiol. 18, 1352, 1972.
250. Wood, H. A. and Bozarth, R. F., Properties of virus-like particles of Penicillium chrysogenum: one
double-stranded RNA molecule per particle, Virology, 47,604, 1972.
251. Border, D. J., Buck, K. W., Chain, E. B., Kempson-Jones, G. F., Lhoas, P., and Ratti, G., Viruses
of Penicillium and Aspergillus species, Biochem. J., 127, 4P, 1972.
252. Nash, C. H., Douthart, R. J., Ellis, L. F., Van Frank, R. M., Burnett, J. P., and Lemke, P. A., On
the mycophage of Penicillium chrysogenum, Can. J. Microbiol. 19, 97, 1973.
 73

253. Burnett, J. P., Frank, B. H., and Douthart, R. J., Ionic strength effects on the stability and confor-
mation of Pemcillium chrysogenum mycophage double-stranded RNA, Nucleic Acids Res., 2, 759,
1975.
254. Buck, K. W. and Girvan, R. F., Comparison of the biophysical and biochemical properties of Pem-
cillium cyaneo-fulvum virus and Penicillium chrysogenum virus, 1. Vlro/.,34, 145, 1977.
255. Wetmur, J. G., Ruyechan, W. T. and Douthart, R. J., Denaturation and renaturation of Penicillium
chrysogenum mycophage double-stranded ribonucleic acid in tetra-alkylammonium salt solutions,
Biochemistry, 20,2999, 1981.
256. Edmondson, S. P. and Gray, D. M., A circular dichroism study of the structure of Penicillium
chrysogenum mycovirus, Nucleic Acids Res., II, 175, 1983.
257. Edmondson, S. P. Lang, D., and Gray, D. M., Evidence for sequence heterogeneity among the
double-stranded RNA segments of Penicillium chrysogenum mycovirus, 1. Gen. Virol., 65, 1591,
1984.
258. Yamashita, S., Doi, Y., and Yora, K., Intracellular appearance of Penicillium chrysogenum vi-
rus, Virology, 55, 445, 1973.
259. Yazaki, K. and Miura, K., Terminal structure involving a single-stranded stretch in the double-
stranded RNA from Penicillium chrysogenum virus, Virology, 82, 14, 1977.
260. Lemke, P. A., Nash, C. H., and Pieper, S. W., Lytic plaque formation and variation in virus titre
among strains of Penicillium chrysogenum, 1. Gen. Microbiol., 76,285, 1973.
261. Lemke, P. A., Saksena, K. N. and Nash, C. H.,Viruses of industrial fungi, in Genetics of Industrial
Microorganisms, Macdonald, K. D., Ed., Academic Press, New York, 1976,323.
262. Cox, R. A., Kangalingham, K., and Sutherland, E., Thermal denaturation in aCidic solutions of
double-helical ribonucleic acid from virus-like particles found in Penicillium chrysogenum, Biochem.
1.,125,655, 1971.
263. Borre, E., Morgantini, L. E., Ortall, V., and Tonolo, A., Production of lytic plaques of viral origin
in Penicillium, Nature, (London), 229, 568, 1971.
264. Volterra, L., Cassone, A., Tonolo, A. and Bruzzone, M. L., Presence of two Virus-like particles in
Penicillium citrinum, Appl. Microbial., 30, 149, 1975.
265. Benigni, R., Ignazzitto, G., and Volterra, L.,Double-stranded ribonucleic acid in viruses of Penicil-
lium citrinum, Appl. Environ. Microbia/. 34,811,1977.
266. Lai, H. C. and Zachariah, K., Detection of virus-like particles in coremia of Penicillium claviform,
Can. 1. Genet. Cytol., 17, 525,1975.
267. Sodovsky, Y. P., Bobyr, A. D., Boiko, A. L., and Zhovnovataya, V. L., Detection of serological
affinity between viruses affecting fungi of the genus PenicillIUm, Mikrobiol. Zhur., 37, 194, 1975.
268. Buck, K. W. and Kempson-Jones, G. F., Three types of virus particles in Penicillium stoloniferum,
Nature, (London), 225,945,1970.
269. Bozarth, R. F., Wood, H. A., and Mandelbrot, A., The Penicillium stoloniferum virus complex: two
similar double-stranded RNA virus-like particles in a single cell, Virology, 45,516, 1971.
270. Buck, K. W. and Kempson-Jones, G. F., Biophysical properties of Penicillium stoloniferumvirus S,
1. Gen. Virol., 18, 223, 1973.
271. Buck, K. W. and Kempson-Jones, G. F., Capsid polypeptides of two viruses isolated from Penicil-
lium stoloniferum, 1. Gen. Virol. 22, 441, 1974.
272. Dunkle, L. D., Double-stranded RNA mycovirus in Periconia circinata, Physiol. Plant Pathol. 4,
107, 1974.
273. Lecoq, H., personal communication, quoted in Lapierre, H. D., Virus like particles in fungi, in Proc.
1st Intersect. Congr. Assoc. Microbial. Soc., Hasegawa, T., Ed., Science Council of Japan, Int.
Tokyo, 3, 359, 1975.
274. Ferault, A. C., Spire, D., Rapilly, F., Bertrandy, J., Skajennikoff, M., and Bernaux, P., Observation
de particules vir ales dans des souches de Piricularia oryzae, Briosi et Cav., Ann. Phytopathol., 3,
267,1971.
275. Yamashita, S., Doi, Y., and Yora, K., A polyhedral virus found in rice blast fungus, Pyricularia
oryzae, Virology, 55,445, 1973.
276. Yavakhiya, K. G., Characterisation of virus-like particles of lytic and non-lytic Pyricularia oryzae
CAV isolates, Mikol. Fitopatol., 16, 129, 1982.
277. Ushiyama, R. and Nakai, Y., Characterization of fungal viruses, especially a double-stranded RNA
molecule of pyricularia oryzaevirus,Abstr. 6th Int. Congr. Virology, Sendai, Japan, September 1st
to 7th, 1984, 237.
278. Lapierre, H. and Faivre-Amiot, A., Presence de particules virales chez differentes souches de Scler-
otium cepivorum, 7th Congr. Int. Prot. Plantes (Paris), September 2nd to 8th, 1970,542.
279. Lapierre, H., Albouy, J., Faivre-Amiot, A., and Molin G., Mise en evidence de particules virales
dans divers champignons du genre Sclerotium, C. R. Acad. Sci., Paris, Ser. D. 272, 2848, 1971.
280. Bozarth, R. F. and Goenaga, A., Complex of virus-like particles containing double-stranded RNA
from Thielaviopsis basicola, 1. Virol., 24, 846, 1977.
74 Fungal Virology

281. Khandjian, E. W., Roos, U.-P., and Turian, G., Mycovirus d'Allomyces, Pathol. Microbiol., 42,
250, 1975.
282. Khandjian, E. W., Turian, G., and Eisen, H., Characterisation of the RNA mycovirus infecting
Allomyces arbuscula, l. Gen. Virol., 35,415, 1977.
283. Roos, V.-P., Khandjian, E. W., and Turian, G., RNA virus in Allomyces arbuscula:Ultrastructural
localisation during the life cycle, l. Gen. Microbiol., 95,87, 1976.
284. Khandjian, E. W. and Turian, G., Physiomorpholic abnormalities of the virus containing strain (Bali
1) of Allomyces arbuscula, Mycopathologia, 74,57, 1981.
285. Dawe, V. H. and Kuhn, C. W., Virus-like particles in the aquatic fungus, Rhizidiomyces, Virology,
130, 10, 1983.
286. Dawe, V. H. and Kuhn, C. W., Isolation and characterization of a double-stranded DNA mycovirus
infecting the aquatic fungus, Rhizidiomyces, Virology, 130, 21, 1983.
287. Honkura, R., Shirako, Y., Ehara, Y., and Yamanaka, S., Two types of virus-like particles isolated
from downy mildew diseases rice plants, Ann. Phytopath. Soc. lpn., 49,653, 1983.
288. Shirako, Y. and Ehara, Y., Two types of single-stranded RNA viruses isolated from Sclerophthora
macrosora-infected rice plants, Abstr. 6th Int. Congo Virology, Sendai, Japan, September 1st to 7th,
1984,238.
289. Pollack, A., The induction of virus-like particles in the marine phycomycete Thraustochytrium au-
reum, Abstr. Annu. Meet. Amer. Soc. Microbiol., 1975,190.
290. Aist, J. R. and Williams, P. H., The cytology and kinetics of cabbage root hair penetration by
Plasmodiophora brassicae, Can. l. Bot., 49,2023, 1971.
291. Erdos, G. W., Virus-like particles in the acrasid cellular slime mould Guttulinopsis vulgaris, Mycol-
ogia, 73,785, 1981.
292. Perkins, F. 0., Ultrastructure of vegetative stages in Labyrinthomyxa marina (= Dermocystidium
marinum), a commercially significant oyster pathogen, l. Invertebr. Pathol., 13,199, 1969.
293. Marier, J .-F., Vago, C., Devauchelle, G., and Duthoit, J .-L., Infection virale chez les Trichomycetes,
C. R. Acad. Sci., Paris, Ser. D, 273, 1241, 1971.
294. Schnepf, E., Soeder, C. J., and Hegewold, E., Polyhedral virus-like particles lysing the aquatic phy-
comycete Aphelidium sp., a parasite of the green alga, Scenedesmus armatus, Virology, 42,482, 1970.
295. Kuwata, S. and Kubo, S., Rod-shaped particles found in tobacco plants infected with tobacco stunt
agent, Ann. Phytopath. Soc. lpn., 47,264, 1981.
296. Kuwata, S., Kubo, S., Yamashita, S., and Doi, Y., Rod-shaped particles, a probable entity of lettuce
big vein virus, Ann. Phytopath. Soc. Jpn., 49, 246, 1983.
297. Mirkov, T. E. and Dodds, J. E., Detection of dsRNA in lettuce big vein infected lettuce (Lactuca
sativa), Phytopathology, 73,961, 1983.
298. Kuwata, S. and Kubo, S., Properties of two Olpidium-transmitted viruses: tobacco stunt and lettuce
big vein, Abstr. 6th Int. Congr. Virology, Sendai, Japan, September 1st to 7th, 1984, 331.
299. Neinhaus, F. and Still, B., Ubertragung des kartoffel-x-virus durch zoosporen von Synchytrium en-
dobioticum, Phytopathol. Z. 54, 335, 1965.
300. Matsumoto, T., Monoclonal antibody to rice dwarf virus-ribonucleic acid (RDV-RNA), Abstr. 6th
Int. Congr. Virology, Sendai, Japan, September 1st to 7th, 1984, 19.
301. Koltin, Y. personal communication, 1984.
302. Gauze, G. F., Maksimova, T. S., Dudnik, Y. V., Ol'Khovatova, O. L., Velikodvorskaya, G. A.,
Klimenko, S. M., and Tikkonenko, T. I., New strains of viruses isolated from mold fungi of the
genus Penicillium, Mikrobiologiya, 40, 540, 1971.
303. Spire, D., Ferault, A.-C., and Bertrandy, J., Particules de type viral associees a Pyricularia oryzae
Cav. agent de la pyriculariose du Riz, C. R. Acad. Sci., Paris, Ser. D., 277, 1961, 1973.
304. Khandjian, E. W., Roos, U.-P., Timberlake, W. E., Lorenza, E., and Turian, G., RNA virus-like
particles in the chytridiomycete Allomyces arbuscula, Arch. Microbiol., 101, 351, 1974.
305. Khandjian, E. W. and Turian, G., Adenine-rich RNA in the mycovirus of Allomyces arbuscula, l.
Gen. Virol.,36, 195, 1977.
306. Albouy, J. and Lapierre, H., Quelques aspects de I'infection virale chez les champignons (Sclerotium
cepivorum, Ophiobolus graminis, Agaricus bisporus), Ann. Univ. ARERS, 9, 333, 1971.
307. Lapierre, H. D. and Spire, D., Virus-like particles of phytopathogenic fungi, Abstr. 1st Intersect.
Congr. Int. Assoc. Microbiological Societies, Tokyo, Japan, 1974, 72.
308. Buck, K. W., McGinty, R. M., and Rawlinson, C. J., Two serologically unrelated viruses isolated
from a Phialophora sp., l. Gen. Virol., 55, 235, 1981.
309. Rawlinson, C. J. and Buck, K. W., Viruses in Gaeumannomyces and Phialophora spp., in Biology
and Control of Take-All, Asher, M. J. C. and Shipton, P. 1., Eds., Academic Press, London, 1981,
68.
310. Schmidt, F. R., Davis, N. D., Diener, U. L., and Lemke, P. A., Cycloheximide, induction of afla-
toxin synthesis in a nontoxigenic strain of Aspergillus flavus, Biotechnology, I, 794, 1983.
 75

311. Cantino, E. C. and Horenstein, E. A., Gamma and the cytoplasmic control of differentiation in
BlastocIadiella, Mycologia, 48,443, 1956.
312. Myers, R. B. and Cantino, E. C., DNA profile of the spore of BlastocIadiella emersonii: Evidence
for gamma-particle DNA, Arch. Mikrobiol., 78,252, 1971.
313. Myers, R. B. and Cantino, E. c., The Gamma Particle, Monogr. Dev. BioI. No.8, Karger, Basel,
1974, 1.
314. Cantino, E. C. and Myers, R. B., Concurrent effect of visible light on gamma particles chitin synthe-
tase and encystment capacity of BlastocIadiella emersonii, Arch. Mikrobiol., 83,203,1972.
315. Cantino, E. C. and Mills, G. L., The gamma particles in Blastocladiella emersonii : What is it?, in
Viruses and Plasmids in Fungi, Lemke, P. A., Ed., Marcel Dekker, New York, 1979,441.
316. Mills, G. A. and Cantino, E. C., Chitosome-like vesicles from gamma particles of Blastocladiella
emersoniisynthesize chitin, Arch. Microbiol., 130, 72, 1981.
317. Cantino, E. C. and Mills, G. A., The blastoc1adialean y-partic1e: once viral endosymbiont(?), now
"chitosome" progenitor, in Fungal Differentiation: A Contemporary Synthesis, Smith, J. E., Ed.,
Marcel Dekker, New York, 1983, 175.
318. Seale, T. W., Delehanty, J., and Runyan, R. B., Liberation and development of Allomyces arbuscula
mitospores viewed by scanning electron microscopy, f. Bacteriol. 120, 1477, 1974.
319. Pommerville, J. and Fuller, M. S., The cytology of the gametes and fertilization of Allomyces ma-
crogynus, Arch. Microbiol., 109, 21, 1976.
320. Barstow, W. E. and Pommerville, J., The ultrastructure of cell wall information and of gamma
particles during encystment of Allomyces macrogynus zoospores, Arch. Microbiol., 128, 179, 1981.
321. Barstow, W. E., Ultrastructure of the formation of y-particles during zoosporogenesis in Allomyces
and Catenaria anguillulae, Exp. Mycol. 3,28, 1979.
322. Martin, W. W., The ultrastructure of Coelomomyces punctatus, f. Elisha Mitchell Sci. Soc., 87,209,
1971.
323. Loubes, C. and Manier, J. F., Ultrastructural study of Coelomycidium simuliiDebaisieux 1920: its
taxonomic position in the Chytridiomycetegroup, Protistologica, 10, 47, 1974.
324. Lange, L. and Olson, L. W., Zoospore of Olpidium brassicae, Protoplasm a, 90,33, 1976.
325. McNitt, R., Ultrastructure of Phlyctochytrium irregularezoospores, Cytobiologie, 9,307, 1974.
326. Wool, S. H. and Held, A. A., Polyphosphate granules in encysted zoospores of Rozella allomycis,
Arch. Microbiol. 108, 145, 1976.
327. Bartnicki-Garcia, S., Bracker, C. E., Reyers, E. and Ruiz-Herrera, J., Isolation of chitosomes from
taxonomically diverse fungi and synthesis of chitin micro fibrils in vitro, Exp. Mycol. 2, 173, 1978.
328. Farkas, V., Biosynthesis of cell walls of fungi, Microbiol. Rev. 43, 117, 1979.
329. Lovett, J. S., Macromolecular synthesis in Blastocladiella, in Fungal Differentiation: A Contempo-
rary Synthesis, Smith, J. E., Ed., Marcel Dekker, New York, 1983,211.
330. Johnson, S. A. and Lovett, J. S., Gene expression during development of Blastocladiella emerson ii,
Exp. Mycol., 8, 132, 1984.
331. Young, T. W. and Yagiu, M., A comparison of the killer characters in different yeasts and its clas-
sification, Antonie van Leeuwenhoek, f. Microbiol. Serol., 44,59, 1978.
332. Mor, H., Steinlauf, R. and Barash, I., Viruslike particles and double-stranded RNA in Geotrichum
candidum, the casual agent of citrus sour rot, Phytopathology, 74,921, 1984.
333. Hirth, L. and Richards, K. E., Tobacco mosaic virus: model for structure and function of a simple
virus, Adv. Virus Res., 26, 145, 1981.
334. Harrison, B. D. and Robinson, D. J., Tobraviruses in Handbook of Plant Virus Infections and
Comparative Diagnosis, Kurstak, E., Ed., Elsevier/North Holland, Amsterdam, 1981,515.
335. Jackson, A. O. and Lane, L. C., Hordeiviruses in Handbook of Plant Virus Infections and Compar-
ative Diagnosis, Kurstak, E., Ed., Elsevier/North Holland, Amsterdam, 1981,565.
336. Gustafson, G. D., Milner, J. J., McFarland, J. E., Pedersen, K., Larkins, B. A., and Jackson, A.
0., Investigation of the complexity of barley stripe mosaic virus RNAs with recombinant DNA
clones, Virology, 120, 182, 1982.
337. Shirako, Y. and Brakke, M. R., Two purified RNAs of soil-borne wheat mosiac virus are needed for
infection, f. Gen. Virol., 65, 119, 1984.
338. Brakke, M. K., Soil-borne wheat mosiac virus, CMI/AAB Descriptions of Plant Viruses, No. 77,
1971.
339. Putz, C., Composition and structure of beet necrotic yellow vein virus, f. Gen. Virol. 35,397, 1977.
340. Richards, K., Jonard, G., Guilley, H., Ziegler, V., and Putz, C., In vitro translation of beet necrotic
yellow vein virus RNAs and studies of sequence homology among the RNAs using cloned cDNA
probes, f. Gen. Virol.,66, 1553, 1985.
341. Maniloff, J., Das, J. and Christensen, J. R., Viruses of mycoplasmas and spiroplasmas, Adv. Virus
Res., 21,343, 1977.
342. Nienhaus, F., Tobacco mosaic virus strains extracted from conidia of powdery mildews, Virology,
46,504, 1971.
76 Fungal Virology

343. Garcia-Arenal, F., Palukaitis, P., and Zaitlin, M., Strains and mutants of tobacco mosaic virus are
both found in virus derived from single-lesion-passaged inoculum, Virology, 132, 131, 1984.
344. Yarwood, C. E., Virus-like infections from Pseudoperonospora, Sphaerotheca, Thielaviopsis and
Uromyces, Phytopathology, 71,267,1981.
345. Dieleman-van Zaayen, A., Mushroom virus disease in the Netherlands: symptoms, etiology, electron
microscopy, spread and control, Agric. Res. Rep., No. 782, Centre for Agricultural Publishing and
Documentation, Wageningen, 1972, 56.
346. Masri, S. A. and Hiruki, C., A new group of elongated plant viruses having a capsid protein of an
unusually high molecular weight, Can. 1. Plant Pathol. 5,208, 1983.
347. Federici, B. A. and Humber, R. A., A possible baculovirus in the insect parasitic fungus, Strong-
weI/sea magna, 1. Gen. Virol., 35,387, 1977.
348. Purcifull, D. E. and Edwardson, J. R., Potexviruses in Handbook of Plant Virus Infections and
Comparative Diagnosis, Kurstak, E., Ed., Elsevier/North Holland, Amsterdam, 1981, 627.
349. Wetter, C. and Milne, R. G., Carlaviruses, in Handbook of Plant Virus Infections and Comparative
Diagnosis, Kurstak, E., Ed., Elsevier/North Holland, Amsterdam, 1981,695.
350. Hollings, M. and Brunt, A. A., Potyviruses, in Handbook of Plant Virus Infections and Comparative
Diagnosis, Kurstak, E., Ed., Elsevier/North Holland, Amsterdam, 1981,731.
351. Lister, R. M. and Bar-Joseph, M., Closteroviruses, in Handbook of Plant Virus Infections and Com-
parative Diagnosis, Kurstak, E., Ed., Elsevier/North Holland, Amsterdan, 1981,809.
352. Gingery, R. E., Nault, L. R., and Bradfute, O. E., Maize stripe virus: characteristics of a member of
a new virus class, Virology, 112, 99, 1981.
353. Koganezawa, H. Doi, Y., and Yora, K., Purification of rice stripe virus, Ann. Phytopath. Soc. lpn.,
41,148,1975.
354. Toriyama, S., Characterization of rice stripe virus: a heavy component carrying infectivity, 1. Gen.
Virol.61, 187, 1982.
355. Gingery, R. E., Nault, L. R., and Yamashita, S., Relationship between maize stripe virus and rice
stripe virus, 1. Gen. Virol., 64, 1765, 1983.
356. Ackermann, H.-W., Cubic, Filamentous, and pleomorphic bacteriophages in CRC Handbook of
Microbiology, Laskin, A. and Lechevalier, H., Eds., 2nd ed., Vol. II, CRC Press, West Palm Beach,
1978,673.
357. Choppin, P. W. and Compans, R. W., Reproduction of paramyxoviruses in Comprehensive Virol-
ogy, Vol. 4, Fraenkel-Conrat, H., and Wagner, R. R., Eds., Plenum Press, New York, 1975,95.
358. Hull, R., The structure of tubular viruses, Adv. Virus Research, 20, I, 1976.
359. Longworth, J. F., Current problems in insect virus taxonomy, in A Critical Appraisal of Viral Tax-
onomy, Matthews, R. E. F., Ed., CRC Press, Boca Raton, Fla., 1983, 123.
360. Brown, F., Bishop, D. H. L., Crick, 1., Francki, R. 1. B., Holland, J. 1., Hull, R., Johnson, K.,
Martelli, G., Murphy, F. A., Obijeski, J. F., Peters, D., Pringle, C. R., Reichmann, M. E., Schnei-
der, L. G., Shope, R. E., Simpson, D. 1. H., Summers, D. F., and Wagner, R. R., Rhabdoviridae,
Intervirology, 12, I, 1979.
361. Stoltz, D. B., Krell, P., Summers, M. D., and Vinson, S. B., Polydnaviridae - a proposed family
of insect viruses with segmented, double stranded, circular DNA genomes, Intervirology, 21, I, 1984.
362. Bailey, L., Carpenter, J. M. and Woods, R. D., Properties of a filamentous virus of the honey bee
(Apis meJIifera), Virology, 114, I, 1981.
363. Lane, L. C., The nucleic acids of multipartite, defective, and satellite plant viruses, in Nucleic Acids
in Plants, Hall, T. C. and Davies, J. W., Eds., Vol. 2, CRC Press, West Palm Beach, Fla., 1979,65.
364. Van Vloten-Doting, L., Francki, R. 1. B., Fulton, R. W., Kaper, J. M., and Lane, L. C., Tricorna-
viridae - a proposed family of plant viruses with tripartite, single-stranded RNA genomes, Intervi-
rology, 15, 198, 1981.
365. Maniloff, J., Haberer, K., Gourlay, R. N., Das, J., and Cole, R., Mycoplasma viruses, Intervirology,
18,177,1982.
366. Kelly, D. C., Brown, D. A., Ayres, M. D., Allen, C. J., and Walker, 1. 0., Properties of the major
nucleocapsid protein of Heliothis zea singly enveloped nuclear polyhedrosis virus, 1. Gen. Virol., 64,
399, 1983.
367. Lapierre, H., Molin, G., Kusiak, C., and Faivre-Amoit, A., The nucleic acid of mycovirus, Ann.
Phytopathol., 5,322, 1973.
368. Albouy, M. M., Lapierre, H. and Molin, G., Mise en evidence d'un nouveau type de particules dans
des hyphes de carpophores d' Agaricus bisporus, C. R. Acad. Sci. Paris, Ser. D., 276, 2805, 1973.
369. Kornberg, A., DNA replication, W. H. Freeman and Co., San Francisco, 1980,454.
370. Bujarski, J. J., Hardy, S. F., Miller, W. A., and Hall, T. C., Use of dodecyl-{J-D-maltoside in the
purification and stabilisation of RNA polymerase from brome mosaic virus-infected barley, Virology,
119,465, 1982.
371. Girard, M., Baltimore, D., and Darnell, J. E., The poliovirus replication complex: site for synthesis
of poliovirus RNA, 1. Mol. BioI., 24,59, 1967.
 77

372. Grimley, P. M. and Densey, A., Functions and alterations of cell membranes during active virus
infection in Pathobiology of Cell Membranes, Trump, B. F. and Arstila, A. V., Eds., Vol. II, Aca-
demic Press, New York, 1980,93.
373. Gay, 1. L., personal communication, 1984.
374. Bertrand, H. and Pittenger, T. H., Isolation and classification of extranuclear mutants of Neurospora
crassa, Genetics, 71, 521, 1972.
375. Diacumakos, E. G., Garnjobst, L., and Tatum, E. L., A cytoplasmic character in Neurospora crassa.
The role of nuclei and mitochondria, 1. Cell BioI., 26,427, 1965.
376. Mitchell, M. B. and Mitchell, H. K., A nuclear gene suppressor of a cytoplasmically inherited char-
acter in Neurospora crassa, f. Gen. Microbiol., 14, 84, 1956.
377. Lambowitz, A. M. and Siayman, C. W., Cyanide resistant respiration in Neurospora crassa, f. Bact.,
108, 1087, 1971.
378. Rifkin, M. R. and Luck, D. 1. L., Defective production of mitochondrial ribosomes in the rpokyl
mutant of Neurospora crassa, Proc. Natl. Acad. Sci. U.S.A., 68,287, 1971.
379. Sebald, W., Buchner, T., Olbrich, B., and Kaudewitz, F., Electrophoretic patterns of amino acid
incorporation in vitro into insoluble mitochrondrial protein of Neurospora crassa wild type and mi-1
mutant, FEBS Letters, I, 235,1968.
380. Lambowitz, A. M., Lapolla, R. J., and Collins, R. A., Mitochondrial ribosome assembly in Neuro-
spora. Two dimensional gel electrophoretic analysis of mitochondrial ribosomal proteins, f. Cell
BioI., 82, 17, 1979.
381. Pittenger, T. H., Synergism of two cytoplasmically inherited mutants in Neurospora crassa, Proc.
Natl. Acad. Sci. U.S.A., 42,747, 1956.
382. Mannella, C. and Lambowitz, A., Interaction of wild-type and poky mitochondrial DNA in hetero-
karyons of Neurospora, Biochem. Biophys. Res. Comm., 80,673, 1978.
383. Lambowitz, A. M. and Luck, D. J., Studies on the poky mutant of Neurospora crassa. Fingerprint
analysis of mitochondrial ribosomal RNA, f. BioI. Chern., 251, 3081, 1976.
384. Lambowitz, A. M., Chua, N., and Luck, D. J. L., Mitochondrial ribosomal assembly in Neurospora.
Separation of mitochondrial ribosomal precursor particles, site of synthesis of mItochondrial ribo-
somal proteins and studies on the pokymutant, f. Mol. BioI., 107, 223, 1976.
385. Collins, R. A. and Bertrand, H., Nuclear suppresors of the rpokyl cytoplasmic mutant in Neurospora
crassa. III. Effects on other cytoplasmic mutants and a mitochondrial ribosomal assembly in rpokyl,
Mol. Gen. Genet., 161, 267,1978.
386. Grimm, M. F. and Lambowitz, A. M., The 32 S RNA of Neurospora crassa mitochondria is not a
precursor of the mitochrondrial ribosomal RNAs, f. Mol. BioI., 134, 667, 1979.
387. Collins, R. A., Bertrand, H., La Polla, R. J., and Lambowitz, A. M., Mitochondrial ribosome
assembly in Neurospora crassa. Mutants with defects in mitochondrial ribosome assembly, Mol. Gen.
Genet., 177, 73,1979.
388. Collins, R. A., Bertrand, H., La Polla, R. J., and Lambowitz, A. M., A novel extranuclear mutant
of Neurospora with a temperature-sensitive defect in mitochondrial protein synthesis and mitochon-
drial ATPase, Mol. Gen. Genet., 181,13,1981.
389. Bertrand, H., Bridge, P., Collins, R. A., Garriga, G. and Lambowitz, A. M., RNA splicing in Neu-
rospora mitochondria. CharacterIzation of new nuclear mutants with defects in splicing the mito-
chondriallarge rRNA, Cell, 29, 517, 1982.
390. Turna, J. and Grones, J., Origin of virus-like particles in poky mutants of Neurospora crassa, Biol-
ogia (Bratislava), 38,801, 1983.
391. Mannella, C. A., Goewert, R. R., and Lambowitz, A. M., Characterization of variant Neurospora
crassa mitochrondrial DNAs which contain tandem reiterations, Cell, 18, 1187, 1979.
392. Bertrand, H., Collins, R. A., Stohl, L. L., Goewert, R. R., and Lambowitz, A. M., Deletion mutants
of Neurospora crassa mitochondrial DNA and their relationship to the "stop-start" growth pheno-
type, Proc. Nat. Acad. Sci. U.S.A., 77,6032,1980.
393. George, C. X., Gupta, B. M., Khurana, S. M. P., and Nagaich, B. B., Antiviral activity in plants of
a mycoviral double-stranded RNA from Trichothecium roseum, Acta virol., 25, 408, 1981.
394. Kazama, F. Y., Ultrastructural evidence for viruses in lower fungi, in Viruses and Plasmids in Fungi,
Lemke, P. A., Ed., Marcel Dekker, New York, 1979,405.
395. Kazama, F. Y., Zachay, A. L., and Schornstein, K. L., Observations on Thraustochytrium sp. de-
velopment and behaviour in culture, Can. f. Bot., 53,360, 1975.
396. Roizman, B., Carmichael, L. E., Deinhardt, F., de-The, G., Nahmias, A. J., Plowright, W., Rapp,
F., Sheldrick, P., Takahashi, M., and Wolf, K., Herpesviridae: definition, provisional nomenclature
and taxonomy, Intervirology, 16, 201, 1981.
397. Roizman, B., The Herpesviruses, Vol. 1 and Vol. 2, Plenum Press, New York, 1982 and 1983.
398. Surkov, V. V., Chaplygina, N. M., Velikodvorskaya, G. A., and Tikchonenko, T. I., Fungal origin
of PB viruses isolated from Penicillium brevicompactum, Microbiologia, 44,905, 1975.
78 Fungal Virology

399. Amosenko, F. A., Surkov, V. V., and Tikchoneko, T. I., On the presence of PBV1 virus genomes in
mycelium of Peniclllium brevicompactum, Vopr. Virusol., 3,361, 1976.
400. Zambryski, P., Goodman, H. M., Van Montagu, M., and Schell, J., Agrobacterium tumour induc-
tion in Mobile Genetic Elements, Shapiro, J. A., Ed., Academic Press, New York, 1983,506.
401. Depicker, A., Van Montagu, M., and Schell, J., in Genetic Engineering in Plants - an Agricultural
Perspective, Kosuge, T., Meredith, C. P., and Hollaender, A., Eds., Plenum Press, New York, 1983,
143.
402. Okker, R. J. H., Spaink, H., Hille, J., van Brussel, T. A. N., Lugtenberg, B., and Schilperoort, R.
A., Plant-inducible virulence promoter of the Agrobacterium tumefaciens Ti plasmid, Nature (Lon-
don), 312,564, 1984.
403. Preer, J. P., Preer, L. B., and Jurand, A., Kappa and other endosymbionts in Paramecium aurelia,
Bacteriol. Rev., 38, 113, 1974.
404. Bender, W., Spierer, P., and Hogness, D. S., Chromosomal walking and jumping to isolate DNA
from the Aceand rosy loci and the bithorax complex in Drospohila melanogaster, J. Mol. BioI., 168,
17, 1983.
405. Barnes, W. M. and Bevan, M., Kilo-sequencing: an ordered strategy for rapid DNA sequence data
acquisition, Nucleic Acids Res., 11, 349, 1983.
406. Howell, S. H., Physical structure and genetic organisation of the genome of maize streak virus (Ken-
yan isolate), Nucleic Acids Res., 12, 7359, 1984.
407. Mullineaux, P. M., Donson, J., Morris-Krsinich, B. A. M., Boulton, M. I., and Davies, J. W., The
nucleotide sequence of maize streak virus DNA, The EMBO Journal, 3,3063, 1984.
408. Stanley, J. and Gay, M. R., The nucleotide sequence of cassava latent virus DNA, Nature (London),
301,260,1983.
409. Hamilton, W. D.O., Stein, V. E., Coutts, R. H. A., and Buck, K. W, Complete nucleotide sequence
of the infectious cloned DNA components of tomato golden mosaic virus: potential coding regions
and regulatory sequences, The EMBO Journal, 3,2197, 1984.
410. Francki, R. I. B., Hatta, T., Grylls, N. E., and Grivell, C. J., The particle morphology and some
other properties of chloris striate mosiac virus, Ann. Appl. BioI., 91,51,1979.
411. Meints, R. H., Van Etten, J. L., Kuczmarski, D., Lee, K., and Ang. B., Viral infection of the
symbiotic Chlorella-like alga present in Hydra viridis, Virology, 113, 698, 1981.
412. Van Etten, J. L., Meints, R. H., Burbank, D. E., Kuczmarski, D., Cuppels, D. A., and Lane, L. C.,
Isolation and characterization of a virus from the intracellular green alga symbiotic with Hydra viri-
dis, Virology, 113, 704, 1981.
413. Kunkel, W., Proteinkristalle und Tubuli-bundel in Hefezellen. VI. Licht-und elektronenmikrosko-
pische sowie biochemische Untersuchungen zur Alkoholdehydrogenase-activitat isolierter Kristalle,
Z. Allg. Mikrobiol., 17,109,1977.
414. Wojcik, V. H., Isolation and purification of virus-like particles from the aquatic fungus, Rhizidi-
omyces, Abstr. 2nd Int. Mycol. Congr., Tampa, August 27th to September 13, 1977,742.
415. Wilson, J. F., Mitochondrial transplantation studies in Neurospora crassa, Abstr. 11th Bot. Congr.,
Seattle, August 2nd to 9th, 1969,240.
416. Bertrand, H. and Pittenger, T. H., Complementation among cytoplasmic mutants of Neurospora
crassa, Mol. Gen. Genet., 117, 82,1972.
417. Akins, R. A. and Lambowitz, A. M., The [poky] mutant of Neurospora contains a 4-base-pair dele-
tion at the 5' end of the mitochondrial small rRNA, Proc. Natl. Acad. Sci. U.S.A., 81,3791,1984.
417a. Dodds, J. A., New viruses of eukaryotic algae and protozoa, in A Critical Appraisal of Viral Tax-
onomy, Matthews, R. E. F., Ed., CRC Press, Boca Raton, Florida, 1983, 177.
418. Bull, A. T. and Trinci, A. P. J., The physiology and metabolic control of fungal growth, Adv.
Microbial. Physiol., 15,1,1977.
419. Border, D. J., personal communication, 1977.
420. Trinci, A. P. J., Influence of the width of the peripheral growth zone on the radial growth rate of
fungal colonies on solid media, J. Gen. Microbiol., 67,325, 1971.
421. Buck, K. W., unpublished data, 1978.
422. Delhotal, P., Legrand-Pernot, F., and Lecoq, H., Etude des virus de Colletotrichum lindemu-
thianum. II. Transmission des particules virales, Ann. Phytopathol., 8, 437, 1976.
423. Lecoq, H., Boissonnett-Menes, M., and Delhotal, P., Infectivity and transmission of fungal viruses,
in Fungal Viruses, Molitoris, H. P., Hollings, M., and Wood, H. A., Springer Verlag, Heidelberg,
1979, 34.
424. Lemke, P. A., Biochemical and biological aspects of fungal viruses, in Proc. 1st Int. Congo Int.
Assoc. Microbiol. Soc., Vol. 3, Hasegwa, T., Ed., Science Council of Japan, Tokyo, 1975,380.
425. Metitiri, P. O. and Zachariah, K., Virus-like particles and inclusion bodies in penicillus cells of a
mutant of Penicillium, J. Ultrastruct. Res., 40,272, 1972.
426. Boissonnet-Menes, M. and Lecoq, H., Transmission de virus par fusion de protoplastes chez Pyri-
cularia oryzae, Physiol. Veg., 14, 251,1976.
 79

427. Hooper, G. R., Wood, H. A., Myers, R., and Bozarth, R. F., Virus-like particles in Penicillium
brevicompactum and P. stoloniferum hyphae and spores, Phytopathology, 62, 823, 1972.
428. Sensing, G. A., Detroy, R. W., Freer, S. N. and Hesseltine, C. W., Virus-like particles from conidia
of Penicillium species, Appl. Microbial., 26,914, 1973.
429. Crosse, R. and Mason, P. J., Virus-like particles in Penicillium chrysogenum, Trans. Brit. Mycol.
Soc., 62,603, 1974.
430. De Marini, D. M., Kurtzman, C. P., Fennell, D.I., Worden, K. A., and Detroy, R. W., Transmission
of PsV-F and PsV-S mycoviruses during condiogenesis of Penicillium stoloniferum, f. Gen. Micro-
bioI., 100, 59,1977.
431. Hastie, N. D., Brennan, V. and Bruenn, J. A., No homology between double-stranded DNA and
nuclear DNA of yeast, f. Virol.,28, 1002, 1978.
432. Ben-Tzvi, B. and Koltin, Y., RNA polymerase activity in virions from Ustilago maydis, Malec. Cell
BioI., 4, 188, 1984.
433. Schisler, L. C., Sinden, J. W., and Sigel, E. M., Transmission of a virus disease of mushrooms by
infected spores, Phytopathology, 53,888,1963.
434. Dieleman-van Zaayen, A., Means by which virus disease in cultivated mushrooms is spread, and
methods to prevent and control it, Mushroom Growers' Assoc. Bull., 244, 158, 1970.
435. Dieleman-van Zaayen, A., Stored spores of virus-infected mushrooms, Champignoncultur, 18, 221,
1974.
436. Atkey, P. T., Barton, R. J., Hollings, M., and Stone, O. M., Mushroom virus purification and
characterisation, Rep. Glasshouse Crops Res. Inst., /974,122, 1975.
437. Brewer, B. J. and Fangman, W. L., Preferential inclusion of extrachromosomal genetic elements in
yeast meiotic spores, Proc. Natl. Acad. Sci. U.S.A., 77,5380, 1980.
438. Buck, K. W., Viruses of the wheat take-all fungus, Gaumannomyces graminis Var. tritici, in Fungal
Virology, Buck, K. W., Ed., CRC Press, Boca Raton, Fla, 1986, chap. 8.
439. Lhoas, P., Use of heterokaryosis to infect virus-free strains of Aspergillus niger, Aspergillus News-
lett., 11,8,1970.
440. Lhoas, P., Transmission of double-stranded RNA viruses to a strain of Penicillium stoloniferum
through heterokaryosis, Nature (London), 230,248, 1971.
441. Wood, H. A. and Bozarth, R. F., Heterokaryon transfer of virus-like particles associated with a
cytoplasmically inherited determinant in Ustilago maydis, Phytopathology, 63, 1019, 1973.
422. Gregory, P. H. The fungal mycelium: An historical perspective, Trans. Brit. Mycol. Soc., 82, I,
1984.
443. Anderson, N. A.,Variation and heterokaryosis in Rhizoctonia solani, in The Ecology and Physiology
of the Fungal Mycelium, Jennings, D. H. and Rayner, A. D. M., Eds., British Mycological Soc.
Symp., Vol. 8, Cambridge University Press, Cambridge, 1984,367.
444. Caten, C. E., Vegetative incompatibility and cytoplasmic infection in fungi, f. Gen. Microbial., 72,
221,1972.
445. Anagnostakis, S. L., Conversion to curative morphology in Endothia parasitica and restriction by
vegetative incompatibility, Mycologia, 75,777, 1983.
446. Brasier, C. M., Inter-mycelial recognition systems in Ceratocystis ulmi: their physiological properties
and ecological importance, in The Ecology and Physiology of the Fungal Mycelium, Jennings, D. H.
and Raynor, A. D. M., Eds, British Mycological Soc. Symp., Vol. 8, Cambridge University Press,
Cambridge, 1984, 451.
447. Van Zaayen, A., Immunity of strains of Agaricus bitorquisto mushroom virus disease, Neth. f. Plant
Pathol.,82, 121, 1976.
448. Hollings, M., Some aspects of virus disease in mushrooms, Mushrooms Sci., 6,255, 1967.
449. Bozarth, R. F., The problem and importance of transmission of mycoviruses using cell free extracts,
Abstr. 3rd Int. Congr. Virology, Madrid, Spain, 1975, September 10th to 17th, 148.
450. Mitchell, D. J., Herring, R. J., and Bevan, E. A., Virus uptake and interactIOn in yeasts, in Microbial
and Plant Protoplasts, Peberdy, J. F., Rose, A. H., Rogers, H. J., and Cocking, E. c., Eds, Aca-
demic Press, London, 1976, 107.
451. Lhoas, P., Infection of protoplasts from Pencillium stoloniferumwith dsRNA viruses, f. Gen. Virol.,
13,365, 1971.
452. Stanway, C. A. and Buck, K. W., Infection of protoplasts of the wheat take-all fungus, Gaeuman-
nomyces graminisvar. tritici, with double-stranded RNA viruses, f. Gen. Virol., 65,2061, 1984.
453. Detroy, R. W., Freer, S. N., and Fennell, D. I., Relationship between the biosynthesis of virus-like
particles and mycophenolic acid in Penicillium stoloniferum and Penicillium brevicompactum, Can.
f. Microbiol., 19, 1459, 1973.
454. Detroy, R. W. and Still, P_ E., Fungal metabolites and viral replication in Penicillium stoloniferum,
Dev. Ind. Microbiol., 16, 145, 1975.
455. Detroy, R. W. and Still, P. E., Patulin inhibition of mycovirus replication in Penicillium stolon-
ferum, f. Gen. Microbiol., 92, 167, 1976.
80 Fungal Virology

456. Detroy, R. W. and Worden, K. A., Interactions of fungal viruses and secondary metabolites, in
Fungal Viruses, Molitoris, H. P., Hollings, M., and Wood, H. A., Eds, Springer Verlag, Heidelberg,
1979,94.
457. Lhoas, P., Mating pairs of Saccharomyces cerevisiae infected with double-stranded RNA viruses
from Aspergillus niger, Nature (London), New BioI. 236, 86, 1972.
458. Border, D. 1., Electron microscopy of cells of Saccharomyces cerevisiae infected with double-
stranded RNA viruses from Aspergillus niger and Penicillium stolonliferum, Nature (London) New
BioI., 236,87, 1972.
459. Buck, K. W., Biochemical and biological implications of double-stranded RNA mycoviruses, in Bi-
ologically Active Substances: Exploration and Exploitation, Hems, D. A., Ed., John WIley, Chich-
ester, 1977, 121.
460. Buck, K. W., unpublished data, 1973.
461. Ferault, A. C., Les particules de type viral associees a Gaeumannomyces graminis (Sacc.) Arx et
Olivier: Sont-elles responsables de modifications du pouvoir pathogene du champignon?, D.Sc. thesis,
University of Paris, 1983.
462. Lemke, P. A., Coevolution of fungi and their viruses, in Fungal Viruses, Molitoris, H. P., Hollings,
M., and Wood, H. A., Eds, Springer Verlag, Heidelberg, 1979,2.
463. Reilly, 1. F., Bruenn, J. and Held, W., The capsid polypeptides of yeast viruses, Biochem. Biophys
Res. Comm., 121, 619,1984.
464. Romanos, M. A., Buck, K. W., and Rawlinson, C. J., A satellite double-stranded RNA in a virus
from Gaeumannomyces graminis, J. Gen. Virol., 57,375, 1981.
465. Romanos, M. A., The Structure and Function of Some Viruses of Gaeumannomyces graminis var.
tritici, Ph.D. thesis, University of London, 1981.
466. Bostian, K. A., Hopper, J. E. and Bozarth, R. F., unpublished data, 1980.
467. Buck, K. W., Virion-associated RNA polymerases of double-stranded RNA mycovlruses, in Fungal
Viruses, Molitoris, H. P., Hollings, M., and Wood, H. A., Eds., Springer Verlag, Heidelberg, 1979,
62.
468. Hopper, J. E., Bostian, K. A., Kowe, L. B., and Tipper, D. J., Translation of the L-species dsRNA
genome of the killer-associated virus-like partIcle of Saccharomyces cerevisiae, J. BioI. Chern., 252,
9010, 1977.
469. El-Sherbeini, M., Tipper, D. J., Mitchell, D. J., and Bostian, K. A., Virus-like particle capsid pro-
teins encoded by different L double-stranded RNAs of Saccharomyces cerevisiae: their roles in main-
tenance of M double-stranded killer plasmids, Mol. Cell. BioI., 4,2818, 1984.
470. Ikegami, M. and Fraenkel-Comat, H., RNA-dependent RNA polymerase of tobacco plants, Proc.
Natl. Acad. Sci. U.S.A., 75,2122, 1978.
471. Sugiura, M. and Miura, K., Transcription of double-stranded RNA by Escherichia coliDNA depend-
ent RNA polymerases, Eur. J. Biochem., 73,179,1977.
472. Kamen, R. I., Structure and function of the QB RNA replicase, in RNA Phages, Zinder, N. D., Ed.,
Cold Spring Harbor Laboratory, New York, 1975,203.
473. Cooper, P. D., Genetics of picornaviruses, in Comprehensive Virology, Vol. 9, Fraenkel-Conrat, H.
and Wagner, R. R., Eds, New York, Plenum Press, 1977, 133.
474. Banerjee, A. K., Abraham, G. and Colonno, R. J., Vesicular stomatitis virus: mode of transcription,
J. Gen. Virol.,34, I, 1977.
475. Rimon, A. and Haselkorn, R., Temperature-sensitive mutants of bacterophage +6 defective in both
transcription and replication, Virology, 89, 218, 1978.
476. Dawson, W. o. and White, J. L., Characterisation of a temperature-sensitive mutant of tobacco
mosaic virus deficient in the synthesis of all RNA species, Virology, 90, 209, 1978.
477. Dawson, W. O. and White, J. L., A temperature-sensitive mutant of tobacco mosaic virus deficient
in synthesis of single-stranded RNA, Virology, 90, 104, 1979.
478. Cross, R. K. and Fields, B. N., Genetics of reo viruses, in Comprehensive Virology, Vol9, Fraenkel-
Comat, H. and Wagner, R. R., Eds., Plenum Press, New York, 1977,291.
479. Miller, W. A. and Hall, T. C., RNA-dependent RNA polymerase isolated from cowpea chlorotic
mottle virus-infected cowpeas is specific for bromoviral RNA, Virology, 132, 53, 1984.
480. Hollings, M., Taxonomy of fungal viruses, in Fungal Viruses, Molitoris, H. P., Hollings, M., and
Wood, H. A., Eds, Spring-Verlag, Heidelberg, 1979, 165.
481. Buck, K. W., Current problems in fungal virus taxonomy, in A Critical Appraisal of Viral Taxon-
omy, Matthews, R. E. F., Ed., CRC Press, Boca Raton, Florida, 1973, 139.
482. Murant, A. F. and Mayo, M. A., Satellites of plant viruses, Annu. Rev. Phytopathol., 20,49, 1982.
483. Huang, A. S. and Baltimore, D., Defective interfering animal viruses, in Comprehensive Virology,
Vol. 10, Fraenkel-Conrat, H. and Wagner, R. R., Eds., Plenum Press, New York, 1977,73.
484. Koltin, Y., Virus-like particles in Ustilago maydis: mutants with partial genomes, Genetics, 86, 527,
1977.
 81

485. Still, P. E., Detroy, R. W., and Hesseltine, C. W., Penicillium stoloniferum VIruS: altered replication
in ultraviolet derived mutants, J. Gen. virol., 27,275,1975.
486. Buck, K. W., Viruses and killer factors of fungi, in The Eukaryotic Microbial Cell, Gooday, G. W.,
Lloyd, D., and Trinci, A. P. J., Eds., Soc. Gen. Microbiol. Sym. No. 30, Cambridge University
Press, Cambridge, 1980,329.
487. Buck, K. W., Ackermann, H-W., Bozarth, R. F., Bruenn, J. A., Koltin, Y., Rawlinson, C. J.,
Ushiyarna, R., and Wood, H. A., Six groups of double-stranded RNA mycoviruses, Intervirology,
22,17, 1984.
488. Brown, F., Classification and nomenclature of viruses: summary of results of meetings of the Inter-
national Committee on Taxonomy of Viruses in Sendai, September 1984, Interiology, in press.
489. Bozarth, R. F. and Harley, E. H., The electrophoretic mobility of double-stranded RNA in polya-
crylamide gels as a function of molecular weight, Biochem. Biophys. Acta, 432, 329, 1976.
490. Almond, M. R., Biochemical Characterisation of Viruses Isolated from the Phytopathogenic Fungus,
Gaeumannomyces graminisvar. tritici, Ph.D. thesis, University of London, 1978.
491. Ushiyama, R., Studies on a virus associated with shiitake mushroom, Lentinus edodes (Berk.) Sing.,
Rep. Tottori Mycological Institute(Jpn.), 21, I, 1983.
492. Dobos, P. and Roberts, T. E., The molecular biology of infectious pancreatic necrosis virus: a review,
Can. J. Microbiol., 29, 377,1983.
493. Joklik, W. K., Ed., The Reoviridae, Plenum Press, New York, 1983.
494. Boccardo, G., Lisa, V. and Milne, R. G., Cryptic viruses in plants, in Double-stranded RNA Viruses,
Compans, R. W. and Bishop, D. H. L., Eds., Elsevier, Amsterdam, 1983,425.
495. Natsuaki, T., Yamashita, S., Doi, Y., Okuda, S., and Teranaka, M., Radish yellow edge virus, a
seed-borne virus with double-stranded RNA, of a possible new group, Ann. Phytopathol. Soc. (Jpn.)
49,593, 1983.
496. Natsuaki, T., Yamashita, S., Doi, Y., Okuda, S., and Teranaka, M., Two seed-borne double-
stranded RNA viruses, beet temperate virus and spinach temperate virus, Ann. Phytopathol. Soc.
(Jpn.), 49,709, 1983.
497. Grill, L. K., Garger, S. J., Turpen, T. H., Lomeli, S. A., Marsden, M. P. F., and Murry, L. E.,
Involvement of viruses and virus-like agents with the male sterility trait of plants, in Plant Molecular
Biology, Alan R. Liss, New York, 1983, 101.
498. Wickner, R. B., Genetic control of replication of the double-stranded RNA segments of the killer
systems in Saccharomyces cerevisiae, Arch. Biochem. Biophys., 222, 1, 1983.
499. Ball, S. G., Tirtiaux, C .• and Wickner. R. B., Genetic control of L-A and L-(BC) dsRNA copy
number in killer systems of Saaccharomyces cerevisiae. Genetics, 107, 199, 1984.
500. Buck, K. W., Replication of double-stranded RNA mycoviruses, in Viruses and Plasm ids in Fungi,
Lemke, P. A., Ed., Marcell Dekker, New York, 1979,93.
501. Timmis, K .• Cabello, F., and Cohen. S. N .• Utilisation of two distinct modes of replication by a
hybrid plasmid constructed in vivo from separate replicons, Proc. Nat!. Acad. Sci. U.S.A., 71, 4556,
1974.
502. Davison, J., Mechanism of control of DNA replication and incompatibility in Col EI-type plasmids
- a review. Gene,28, I, 1984.
503. Scott. J. R .• Regulation of plasmid replication, Microbiol. Rev.,48, 1, 1984.
504. Lapierre, H., Faivre-Amiot, Kusiak, G., Molin, G., Klusseoglou, P., Albouy, P .• Bondaux, P .• and
Morand, J.-C .• Particules de type viral associees a l'espece, Mycogone perniciosa, Ann. Phyto-
pathol., 3.538, 1971.
505. Weber, H. and Lindner, R., Virus-like particles in yeast protoplasts. I. Demonstration by means of
electron microscopy, Z. Allg. Mikrobiol., IS, 631, 1975.
506. Khadjian, E. W. and Turian. G., In vitro RNA synthesis by double-stranded RNA mycoviruses from
Allomyces arbuscula, FEMS Microbiol. Lett., 2,121,1977.
507. Ratti. G. and Buck, K. W .• Transcription of double-stranded RNA in virions of Aspergillus foetidus
virus S, J. Gen. Virol.,42, 59,1979.
508. Buck, K. W., Romanos. M. A., McFadden. J. J. P., and Rawlinson, C. J., In vitro transcription of
double-stranded RNA by virion-associated RNA polyrnerases of viruses from Gaeeumannomyces
graminis, J. Gen. Virol.,57, 157, 1981.
509. McGinty, R. M., Buck. K. W., and Rawlinson, C. J., Transcriptase activity associated with a type 2
double-stranded RNA mycovirus, Biochem. Biophys. Res. Comm., 98,501,1981.
510. Herring. A. J. and Bevan, E. A., Yeast virus-like particles possess a capsid-associated single-stranded
RNA polymerase, Nature (London), 268.464.1977.
511. Welsh, J. D., Leibowitz, M. J. and Wickner, R. B., Virion DNA-independent RNA polymerase from
Saccharomyces cerevisiae, Nucleic Acids Res., 8, 2349, 1980.
512. Herring, A. J. and Bevan, E. A., Double-stranded RNA containing particles from the yeast Saccha-
romyces cerevisiae and their relationship to the killer character, in Molecular Biology of Nucleocy-
toplasmic Relationships, Puiseux-Dao, S., Ed., Elsevier, Amsterdam, 1975, 149.
82 Fungal Virology

513. Buck, K. W.,Replication of double-stranded RNA in particles of Penicillium stoloniferum virus S,

Nucleic Acids Res., 2, 1889, 1975.
514. Chater, K. F. and Morgan, D. H., Ribonucleic acid synthesis by isolated viruses of Penicillium sto-
loniferum, f. Gen. Virol., 24, 307, 1974.
SIS. Romaine, C. P. and Sriskantha, A., Synthesis of double-stranded RNA by a virus-enriched fraction
from Agaricus bisporus, Phytopathology, 74, 861, 1984.
516. Buck, K. W. and Ratti, G., A model for the replication of double-stranded RNA mycoviruses,
Biochem. Soc. Trans., 3, 542, 1975.
517. Buck, K. W., Semi-conservative replication of double-stranded RNA by a virion-associated RNA
polymerase, Biochem. Biophys. Res. Comm.,. 84, 639, 1978.
518. Ratti, G. and Buck, K. W., Semi-conservative transcription in particles of a double-stranded RNA
mycovirus, Nucleic Acids Res., 5,3843, 1978.
519. Lu, B. C. and Sakaguchi, K., A unique eukaryotic DNA polymerase that transcribes single- and
double-stranded RNA, Genetics, 107, s65, 1984.
520. Ross, I. K., An infectious disorder for meiosis in Corprinus congregatus, in Viruses and Plasmids in
Fungi, Lemke, P. A., Ed., Marcel Dekker, New York, 1979,485.
521. Ross, I. K., The genetic basis for the meiotic disorder of Coprinus congregatus, Curro Genet., 5, 53,
1982.
522. Lemke, P. A., and Nash, C. H., Fungal Viruses, Bacteriol. Rev., 38,29, 1974.
523. Bousquet, J.-F., Le tau x de pyriculol dans les filterats de culture de souches saines ou virosees de
Piricularia oryzaeCavara, Ann. Phytopathol., 5, 112, 1973.
524. Fink, G. R. and Styles, C. A., Curing of a killer factor in Saccharomyces cerevisiae, 69,2846, 1972.
525. Goldblatt, L. A., Aflatoxin, Academic Press, New York, 1969.
526. Woods, D. R., Ross, I. W. and Hendry, D. A., New killer factor produced by killer-sensitive yeast
strain, f. Gen. Microbiol., 81,285, 1974.
527. Hardy, K. G., Colicinogeny and related phenomena, Bact. Rev., 39,464, 1975.
528. Tagg, J. R., Dajani, A. S., and Wannamaker, L. W., Bacteriocins of Gram positive bacteria, Bact.
Rev., 40,722, 1976.
529. Young, T. W. and Philliskirk, G., The production of a yeast killer factor in the chemostat and the
effects of killer yeast in mixed continuous culture with a sensitive strain, f. Appl. Bact., 43, 425,
1977.
530. Young, T. W., The genetic manipulation of killer character into brewing yeast, f. [nst. Brewing, 87,
292, 1981.
531. Puhalla, J. E., Compatibiity reactions on solid medium and inters train inhibition in Ustilago maydis,
Genetics, 60,461,1968.
532. Day, P. R. and Dodds, J. A., Viruses of plant pathogenic fungi, in Viruses and Plasm ids in Fungi,
Lemke, P. A., Ed., Marcel Dekker, New York, 1979,201.
533. Philliskirk, G. and Young, T. W., The occurrence of killer character in yeasts of various genera,
Antonie van Leeuwenhoek f. Microbiol. Serol., 41,147, 1975.
534. Stumm, C., Hermans, J. M., Middelbeck, E. J., Croes, A. F., and de Vries, G. J. M. L., Killer-
sensitive relationships in yeast from natural habitats, Antonie van Leeuwenhoek f. Microbiol. Serol.,
43, 125, 1977.
535. Middelbeek, E. J., Hermans, J. M. H., Stumm, c., and Muytjens, H. L., High incidence of sensitiv-
ity to yeast killer toxins among Candida and Torulopsis isolates of human origin, Antimicrob. Agents
Chemother., 17, 350, 1980.
536. Kandel, J. S. and Stern, T. A., Killer phenomenon in pathogenic yeast, Antimicrob. Agents Chem-
other., 15, 568, 1979.
537. Polonelli, L., Archibusacchi, C., Sestito, M., and Morace, G., Killer system: a simple method of
differentiating Candida albicansstrains, f. Clin. Microbiol.,17, 774,1983.
538. Adler-Moore, J., Subcellular particles in pathogenic fungi, in Fungi Pathogenic for Humans and
Animals, Part B: Pathogenicity and Detection: II, Howard, D. H., Ed., Marcel Dekker, New York,
1985.
539. Ridley, S. P., Sommer S.S. and Wickner, R. B., Superkiller mutations in Saccharomyces cerevisiae
suppress exclusion of M, double-stranded RNA by L-A-HN and confer cold sensitivity in the presence
of M and L-A-HN, Mol. Cell BioI., 4,761, 1984.
540. Last, F. T., Hollings, M., and Stone, O. M., Effects of cultural conditions on the mycelial growth of
healthy and virus-infected cultivated mushroom, Agaricus bisporus, Ann. Appl. BioI., 76,99, 1974.
541. Wach, M. P. and Romaine, C. P., Further characterisation of double-stranded RNA associated with
La France disease, Phytopathology, 73,376, 1983.
542. Barton, R. J. and Atkey, P. T., Detection of viruses in Agaricus bisporus(Lange) Imbach by immu-
nosorbent electron micriscopy, Rep. Glasshouse Crops Res. [nst. 1982,115,1984.
543. Del Vecchio, V. G., Dixon, C., and Lemke, P. A., Immune electron microscopy of virus-like particles
of Agaricus bisporus, Exp. Mycol., 2, 138, 1978.
 83

544. Del Vecchio, V. G., Dixon, C., Fritzinger, J., and Lemke, P. A., Immunochemlcal detection of vlfUS-
like particles and double-stranded RNA from Agaricus bisporus, in Fungal Viruses, Molitoris, H. P.,
Holhngs, M., and Wood, H. A., Eds., Springer Verlag, Heidelberg, 1979, 138.
545. Nair, N. G., Observations on virus disease of the cultivated mushroom Agancus blSPOruS in Aus-
traha, Mushroom Sci., 8, 155, 1972.
546. Atkey, P. T. and Barton, R. J., Virology: Mushroom Agaricus bisporus(Lange) Smg., Rep. Glass-
house Crops Res. Inst. 1975, 124, 1976.
547. Atkey, P. T. and Barton, R. J., Virology: Mushroom Agaricus bisporus(Lange) Sing., Rep. Glass-
house Crops Res. Inst. 1976.125, 1977.
548. Tavantzis, S. M. and Smith, S. H., Virus-like particles transmitted by and detected in spawn of the
cultivated mu,hroom, Agaricus bisporus, Phytopathology, 69, 104, 1979.
549. Barton, R. J. and Atkey, P. T., Mushroom Agaricus bisporus(Lange) Imbach. Rapid detection of
mushroom viruses in spawns, Rep. Glasshouse Crops Res. Inst. 1980,149, 1981.
550. Marino, R., Saksena, K. N., Schaler, M., Mayfield, 1. E., and Lemke, P. A., Double-stranded
ribonucleic acid in Agaricus bisporus, Appl. Environ. Microbiol., 31,433, 1974.
551. Wach, M. P. and Romaine, C. P., Double-stranded RNA in the cultivated mushroom: incidence and
variation, Phytopathology, 73, 376, 1983.
552. Wach, M. P. and Romaine, C. P., Mushroom viruses - incidence, impact and control, Mushroom
News, 31,5, 1983.
553. Deahl, K. L., The occurrence of double-stranded RNA in spawn strains of Agaricus bisporus, Phy-
topathology, 74,843, 1984.
554. Finkler, A., Koltin Y., Barash, I. and Sneh, B., Isolation of a virus from virulent strains of Rhizoc-
tonia solam, f. Gen. Virol.,66, 1221, 1985.
555. Joklik, W. K., Evolution in viruses, in Evolution in the Microbial World, Carlile, M. J. and Skehel,
J. J., Eds., Soc. Gen. Microbiol. Symp., No 24, Cambridge University Press, Cambridge, 1974.
556. Holland, J., Spindler, K., Horodyski, F., Grabau, E., Nichl, S., and VandePol, S., Rapid evolution
of RNA genomes, Science, 215, 1577, 1982.
557. Doolittle, W. F. and Sapienza, C., Selfish genes, the phenotype paradigm and genome evolution,
Nature (London), 284, 601, 1980.
558. Orgel, L. E. and Crick, F. H. C., Selfish DNA: the ultimate parasite, Nature (London), 284, 604,
1980.
559. Shiba, T. and Saigo, K., Retrovirus-like particles containing RNA homologous to the transposable
element copia in Drosophila melanogaster, Nature (London), 302, I 19, 1983.
560. Laporte, D. C., Antisense RNA: a new mechanism for the control of gene expression, Trends
Biochem. Sci., 1984, 463.
561. Gray, C. W., Page, G. A., and Gray, D. M., Complex of fd gene 5 protein and double-stranded
RNA, f. Mol. BioI., 175, 553,1984.
562. Kovacs, E., Bucz, B., and Kolompar, G., PropagatIOn of mammalian viruses in protista. IV. Exper-
imental infection of C. albicans and S. cerevisiae with polyoma virus, Proc. Soc. Exp. BioI. Med.,
132,971,1969.
563. Kovacs, E., Bucz, B., and Kolompar, G., Propagation of mammalian viruses in protista. I. Visual-
isation of fluorochrome labelled EMC virus in yeast and Tetrahymena, Life Sci., 5,2117, 1966.
564. Kovacs, E. and Bucz, B., Propagation of mammalian viruses in protista. II. Isolation of complete
virus from yeast and Tetrahymena experimentally infected with picorna viral particles of their infec-
tious RNA, Life Sci., 5, 211 7, 1966.
565. Kovacs, E., Change in population densities, viability, or multiplication of yeasts and Tetrahymena
infected experimentally with encephalomyocarditis virus, f. Cell BioI., 35, 73A, 1967.
566. Lavrouskin, A. and Treagen, L., Experimental infection of Saccharomyces cere visia e with mammal-
ian viruses, Wasmann f. BioI., 30, 97, 1972.
567. Kovacs, E., Activation of virus production by DMSO in C. albicans experimentally infected with
polyoma-DNA, Experientia, 26, 1296, 1970; Cell BioI., 35, 73A, 1967.
568. Simanis, V., T-Antigen Binding Sites in Cellular DNA Sequences, Ph.D. thesis, University of Lon-
don, 1984.
569. Brants, D. H., Tobacco mosaic virus in Phythium spec., Neth. 1. Pl. Path., 75,296, 1969.
570. Brants, D. H., Infection of pythium sylvaticum in vitro with tobacco mosaic virus, Neth. 1. PI. Path.,
77,175,1971.
571. Nienhaus F. and Mack, C., Infection of Pythium arrhenomanes in vitro with tobacco mosaic virus
and tobacco necrosis virus, Z. Pflanzenkr. Pflanzenschutz, 81,728,1974.
572. Wieringa-Grants, D. H., Infection of Pythium spec. in vitro with C'·-labelled tobacco mosiac virus,
2nd Int. Congr. Plant Pathol. (Minneapolis), Abst. No. 0913, 1973.
573. Coutts, R. H. A., Cocking, E. C., and Kassanis, B., Infection of protoplasts from yeast with tobacco
mosaic virus, Nature (London), 240, 466, 1972.
84 Fungal Virology

574. Koltin, Y. and Day, P. R., SpecIficity of Ustilago maydis killer proteins, Appl. Micro bioI. ,30, 694,
1975.
575. Matthews, R. E. F., Classification and nomenclature of viruses, Fourth Rep. Int. Comm. Taxonomy
of Viruses, S. Karger, Basel, 1981.
576. Pallett, I. H., InteractIOns between fungi and their viruses, in Microbial and Plant Protopiasts, Pe-
berdy, J.F., Rose, A. H., Rogers, H. J., and Cocking, E. C., Eds., Academic Press, London, 1976,
107.
577. Atkey, P. T. and Barton, R. J., Club-shaped virus-like particles. Rep. Glasshouse Crops Res. Inst.
1978,147, 1979.
578. Fields, S. and Winter, G., Nucleotide sequences of influenza virus segments 1 and 3 reveal mosaic
structure of a small viral RNA segment, Cell, 28, 303, 1982.
579. Wright, C. L. and Adler-Moore, J. P., The adjuvant effects of mycoviral dsRNA and polymosinic:
polycytidylic acid on the murine immune response, Biochem. Biophys. Res. Comm., 131, 949,1985.
580. Adler-Moore, J., Subcellular particles in pathogenic fungi, in Fungi Pathogenic for Humans and
Ammals, Part B, Pathogenicity and Detection: II, Howard, D. H., Ed., Marcel Dekker, New York,
1985,103.
581. Reanney, D., The molecular evolution of viruses, in The Microbe 1984 : Viruses, Society for General
Microbiology Symp. 36, Mahy, B. W. 1. and Pattison, J. R., Cambridge University Press, Cam-
bridge, 1984, 175.
582. Roeder, G. S. and Fink, G. R., Transposable elements in yeast, in Mobile Genetic Elements, Shapiro,
1. A., Ed., Academic Press, New York, 1983,299.
583. Boeke, J. D., Garfinkel, D. J., Styles, C. A., and Fink, G. R., Ty elements transpose through an
RNA intermediate, Cell, 40,491, 1985.
584. Garfinkel, D. J., Boeke, 1. D., and Fink, G. R., Ty element transposition: reverse transcription and
virus-like particles, Cell, 42, 507, 1985.
585. Varmus, H. E., Form and function of retroviral proviruses, Science, 216,812, 1982.
586. Summers, J. and Mason, W. S., Replication of the genome of a hepatitis B-like virus by reverse
transcription of an RNA intermediate, Cell, 29,403, 1982.
587. Pfeiffer, P. and Hohn, T., Involvement of reverse transcription in the replication of cauliflower
mosaic virus: a detailed model and test of some aspects, Cell, 33,781,1983.
588. Shiba, T. and Saigo, K., Retrovirus-like particles containing RNA homologous to the transposable
element copia in Drosophila melanogaster, Nature (London), 302, 119, 1983.
589. Mellor, J., Fulton, S. M., Dobson, M. J., Wilson, W., Kingsman, S. M., and Kingsman, A. J., A
retrovirus-like strategy for the expression of a fusion protein encoded by yeast transposon Ty 1,
Nature (London), 313,243,1985.
590. Clare, J. and Farabaugh, P., Nucleotide sequence of a yeast Ty element: evidence for an unusual
mechanism of gene expression, Proc. Nat!. Acad. Sci. U.S.A., 82,2829, 1985.
591. Hauber, J., Nelbock-Hochstetter, P., and Feldman, H., Nucleotide sequence and characteristics of a
Ty element from yeast, NucJ. Acids Res., 13, 2745, 1985.
592. Kuff, E. L., Wivel, N. A., and Lueders, K. K., The extraction of intracisternal A-type particles from
a mouse plasma-cell tumor, Cancer Res., 28,2137, 1968.
593. De Giuli, C., Hanafusa, H., Kawai, S., Dales, S., Chen, 1. J., and Hsu, K. C., Relationship between
A-type and C-type particles in cells infected by Rous sarcoma virus, Proc. Natl. Acad. Sci. U.S.A.,
72,3706, 1975.
594. Newton, A. C., Caten, C. E., and Johnson, R., Variation for isozymes and double-stranded RNA
among isolates of Puccinia striiformis and two other cereal rusts, Plant Pathology, 34, 235, 1985.
 85

Chapter 2

THE KILLER SYSTEMS OF SACCHAROMYCES CEREVISIAE AND

OTHER YEASTS

Jeremy Bruenn

TABLE OF CONTENTS

I. Introduction ................................................................................... 86

II. The Saccharomyces cerevisiaeViruses .................................................. 86

A. Viral RNAs ........................................................................... 87
B. Expression of the Viral Genome ................................................ 91
1. Viral Proteins - Capsid ................................................. 91
2. Viral Proteins - Toxin and Resistance Factor ..................... 93
3. ScV Transcriptase .......................................................... 95
4. In Vivo mRNAs ............................................................ 97
C. Viral Interactions ................................................................... 97
1. Exclusion .................................................................... 97
2. Suppression ................................................................. 99
D. Virus-Host Interactions ........................................................... 99
1. Control of Replication .................................................... 99
2. Nuclear Genes ............................................................ 100

III. Kluyveromyces lactis Plasm ids ......................................................... 101

A. Plasmid DNA ...................................................................... 101
B. Toxin ................................................................................ 101

IV. Cloning Vectors ............................................................................ 103

Acknowledgments .............................................................. , ................... 103

References ............................................................................................ 103

86 Fungal Virology

I. INTRODUCTION

One common strategy of viruses is to establish a carrier state in which the host cell
survives unharmed. Some bacteriophages, for example, may adopt the disguise of plas-
mids. Other strategies vary from the simple nonlethal infection of the filamentous
single-stranded DNA bacteriophages to the complicated integrative pathways pursued
by lambda and the retroviruses. Frequently, among prokaryotes, viruses or plasmids
provide the host with a selective advantage: synthesis of restriction enzymes, colicins,
antibiotic degradative enzymes, or provision of immunity to infection. In one group
of simple eukaryotes, the fungi, two such associations of plasmids or viruses conferring
a selective advantage on the host are known. These are the killer systems of the yeasts
and of Ustilago maydis, a corn smut that adopts a yeast-like mode of growth on labo-
ratory media. Cells with a double-stranded RNA (dsRNA) virus, in several yeasts and
in Ustilago, or with a linear double-stranded DNA (dsDNA) plasmid, in the yeast
Kluyveromyces lactis, synthesize a secreted toxin that kills cells of the same (or in some
cases different) species lacking the virus or plasmid. The easily recognizable phenotype
in the killer systems has made it possible to define their necessary elements. There are
both nuclear and viral genes.
There are killer systems in at least eight yeast genera, although the phenomenon was
first described in Saccharomyces cerevisiae. 1 In the three killer types in Saccharomyces
species (k l , k 2, and k3) and in Yarrowia lipolytica, the toxin is known to be encoded on
a dsRNA.1.2 3 In the other genera, no dsRNAs have been detected! In one of these
genera, Kluyveromyces, the toxin is encoded on a linear dsDNA plasmid.' The char-
acteristics of the Kluyveromyces toxin are also quite different from those of the Sac-
charomyces toxins. These are therefore very different systems that have evolved to a
common purpose: to provide a selective advantage to cells carrying a virus (Saccharo-
myces and Ustilago) or a plasmid (Kluyveromyces). Since the molecular biology of
these two systems is rather different, they will be considered separately.

II. THE SACCHAROMYCES CEREVISIAEVIRUSES

The yeast viruses comprise an excellent model system for studies of the relationship
between a host eukaryotic cell and its persistent dsRNA viruses. There are dsRNA
viruses of plants, bacteria, fungi, invertebrates, and vertebrates. 5 There are dsRNA
viruses that persistently infect animal cells,6 plant cells,'" and insect cells, 10 but in the
fungi persistent infection by dsRNA viruses is the rule rather than the exception. I I ScV,
the Saccharomyces cerevisiae viruses, are similar to other fungal viruses. These are
double-stranded RNA (dsRNA) viruses of apparent icosahedral symmetry, 30 to 40 nm
in diameter, with their genomic dsRNA components separately encapsidated, one mol-
ecule per particle. Fungal virus particles typically have densities in CsCI of 1.3 to 1.45
and S20,w values of 140 to 170. 5 ,1113 The general properties of dsRNA myco-
viruses and their taxonomy, including that of the S. cerevisiae viruses, are discussed in
Chapter 1 of this book.
Most strains of S. cerevisiae have such resident viruses, with about 100 to 3000 par-
ticles per cell." ScV is exclusively cytoplasmic and is not associated with mitochondria
or other known yeast plasmids. 15 17 All the known viruses have one virus species with
a dsRNA (L) of about 4.8 kbp. ScV-L particles are about 35 to 40 nm in diameter,
apparently icosahedral, of S20 w = 160, and with a CsCI density of 1.41. 18 23 Some
strains also have a second, satellite dsRNA of 1.9 kbp (M), encapsidated separately in
particles of the same diameter (ScV -M particles). Many of these strains elaborate an
extracellular toxin (killer toxin) which kills cells not harboring a resident ScV -M virus.
M encodes the toxin and L the major capsid polypeptide, PI (see below). There are a
 87

number of subgroups of M dsRNAs, which encode different toxin and resistance spec-
ificities. 2.24 The viral genomic dsRNAs (L) of these viral subtypes also differ. We have
characterized the dsRNAs of two killer types: we have designated the major RNAs of
killer type 1 L. and M .. the major RNAs of killer type 2 L2 and M 2.2S In some strains,
there are also defective-interfering dsRNAs, designated S, derived from M, by internal
deletion (see suppression).

A. Viral RNAs
The dsRNAs of virus particles from different strains of S. cerevisiae have been in-
vestigated in severallaboratories. 2535 They have been sized by a number of techniques,
none as accurate as those available for dsDNAs.' 4 We estimate that L. is 4.8 kbp, L2
4.9 kbp, M. 1.9 kbp, and M. 1.8 kbp.'5 M. and M2 are separable by agarose gel elec-
trophoresis. 24 The sequences at the ends of the predominant species of these dsRNAs
are given in Table 1.25 .• 8.29.31.34.36.37
The ends of the S dsRNAs derived from M. (see suppression) are identical in the
sequenced region to the M. sequences. 3• Only one of these regions has been sequenced
on both strands, the U-rich end of L •. '8.3' Both L,- and L 2-containing strains have
another L dsRNA 3' U-rich end (GAAAAAAUUCA oH and related sequences) that we
have named L alternate (L.). By cloning cDNAs to L" we have shown that this 3' end
heterogeneity reflects the presence of two dsRNAs in k. strains, L, and L., with little
sequence homology.38 These results have been confirmed by T, fingerprint analysis. 3'
L. and L. have limited internal sequence homology.38 Unlike L" L. terminates with
either AOH or G OH ,38 as does the M, transcript. 40 An L dsRNA with the same 3' termini
as L. exists in k. strains as well.'8
A number of different nomenclatures exist for the ScV dsRNAs. The L, of
Bruenn31.38.4 •. 42 is the LA of Wickner39 .43 and the L2 of Bevan. 44 The L. of
Bruenn31.38.4 •. 42 is the LBC of Wickner39 .43 and the L, of Bevan. 44 The L2 of Bruenn 25.38
is the L A_. of Wickner. 45 Like L. and L 2, LB and Lc share considerable sequence ho-
mology, which has been estimated in this case as about 50070. 3 • Strains may have neither
LA nor L B , either LA or L B , but apparently not both;44.39 similarly, L. and L2 do not
appear to coexist in the same cells. 38 In summary, there are really two families of L
dsRNAs in S. cerevisiae strains: the L. family, consisting so far of L. and L 2; and the
L. family, consisting of LB and Lc and probably similar species in k. strains. There
appears to be unanimity on the nomenclature of M. and M •.
The known sequences at the ends of L, have been extended for about 200 bp with
separated strands.'6.37.46 There is an open reading frame beginning at the AUG at po-
sition 30 to 32 of the L, plus strand, which is probably the gene for PI. 46 The sequences
of four overlapping cDNA clones from a region of about 800 bp, including the coding
region for the putative PI C terminus, have been determined. 46 This region contains a
long inverted repeat of 170 to 340 bp (only une end is defined). The inverted repeat
was mapped on L, by heteroduplex mapping, primer extension, and mapping of stem
and loop structures in denatured L •. From the sequence of this cloned region and the
sequence of the 5' terminus of the L. plus strand, we have placed the P, gene in the 5'
2.3 to 2.6 kb of the L. plus strand, beginning at the first AUG at position 30. 46 There
is no open reading frame through the inverted repeats, and there is room for the PI
gene only in the region of L. 5' to the inverted repeats (Figure 1). There is a subgenomic
mRNA corresponding to this region (see Section B.4).
M, has an internal AU-rich sequence easily cleaved by S. nuclease, which has enabled
Leibowitz to extend the sequences at the ends of M. about 230 bp into the molecule. 34
We have established that the coding strand of L, is the upper strand as shown in Table
1;.8 Leibowitz has identified the plus strand of M. and the probable coding sequence
of the M, toxin precursor beginning at nucleotide 14 of the analogous strand of
M,.34.40.47
 Table 1 00
00
SEQUENCES AT THE ENDS OF THE ScV dsRNAs

U-rich end C-rich end

e
L, pppGAAAAAUUUUUAAAUUCAUAUAACUCCCCAUGC
oHACUUUUUAAAAAUUUAAGUAUAUUGAGGGGUACG
AAAAGAUAAUGGGAAUUACCCAUAUGCA oH
UUUUCUAUUACCCUUAAUGGGUAUACGppp
~......
<:
Z·
L,

L.
pppGAAUAAUUUGAAUAUUCCAUACACUC
oHACUUAUUAAACUUAUAAGGUAUGUGAG
pppGAAUUUUUUC
UAAAUAUAAGAGCUUAUACACAUAUGCA oH
AUUUAUAUUCUCGAAUAUGUGUAUACGppp
GCA OH
g
oHACUUAAAAAAG CGppp

L. pppGAAUUUUUUC GCG OH
OHGCUU AAAAAAG CGppp

M, pppGAAAAAUAAAGAAAUGACGAAGCCAA UGCAACAGCAUAGAAGAAACACACAUCA oH
oHACUUUUUAUUUCUUUACUGCUUCGGUU ACGUUGUCGUAUCUUCUUUGUGGGUAGppp

M, pppGAAAAAAUGAAAGAGACUAC GAAUU ACU ACAGGU ACAUUU ACCU AGCA oH

oHACUUUUUUACUUUCUCUGAUG CUUAAUGAUGUCCAUGUAAAUGGAUCGppp
 89

N p~ C
~1---------------------------~~--------------------L1

N C
,J--"-8 ,;:.Q----'---'--Y----'---'{3----L1-lDtf)____----./ M1
L'

/ /
/ /
/ /
I /
/ /
1 /
/ /
/ 1
I /
I I
/ /
1 /
/ 1
/ I
I I
/ 1
VI /
'------''-------11 S3 , S14

FIGURE I. Structure of L, and M, dsRNAs. N indicates the N terminus and C the C terminus of encoded
polypeptides. PI is the capsid polypeptide and d, a, y, and (J the final proposed cleavage products of the
preprotoxin. The boxed sequences are inverted repeats in L, and the AU-rich sequence in M,. Dotted lines
from M, to S indicate the sequences conserved by internal deletion.

The sequence of the 5' 1 kb of the M, plus strand has been determined from cDNA
clones constructed by oligo-dT priming within the AU-rich region of M,.48so Coding
regions for two toxin polypeptides and an intervening region have been deduced, all
within one preprotoxin polypeptide. The alpha toxin is encoded by bases 146 to 403,
the gamma region by bases 404 to 712, and the beta toxin by bases 713 to 961. Bases
14 to 145 encode the putative signal sequence. 48 A series of processing steps is necessary
for secretion of toxin (see Section B.2). The structure of M, is summarized in Figure 1
and the sequence of the 5' region of M, is shown in Figure 2. Residues 981 to 993 are
present in the clone sequenced by Skipper et al. 50 but not in the clone sequenced by
Bostian et al. 48 The beginning of the AU-rich sequence is evident in the sequence re-
ported by Skipper et al., 50 which extends somewhat beyond the end reported by Bostian
et al. at position 1023. 48 Sequence analysis of a number of such clones, primed at
different positions within the 200 bp AU-rich region, indicates that previous data from
T, and pancreatic fingerprints of M, showing that there were no runs of A longer
than 8 or 9 2330 may be correct: the oligo-dT priming takes place with some mis-
matches. so52
Little is currently known of the functions of the portions of L, 3' to the inverted
repeat or of M, 3' to the AU-rich sequence. Translation of the 3' portion of M, after
cleavage of the dsRNA within the AU-rich region gave a 19 kDa polypeptide, which
was proposed to arise from a coding region on the M, minus strand, in which there
was an apparent open reading frame. 34 ,47 More recent sequence analysis of this region
present in cDNA clones shows that the 3' region of M, has no open reading frames on
either strand,s3 and it has not been possible to repeat the translation results. 54 The
portion of L, 3' to the inverted repeat does have an open reading frame originating
within the second repeat, but it is not known how far it extends. 46
By comparison of the sequences of the yeast viral dsRNAs, we have proposed tran-
scriptase and replicase initiation sites: the transcriptase initiation site at the U-rich end
and the replicase site at the C-rich end.25,31 The rationale for this identification was
that, since ScV-M and ScV-L have the same polypeptide(s) (see Section B.2), they must
have a transcriptase with the same specificity. The transcriptase probably recognizes
the 3' ends of the minus strands of Land M, which should therefore be similar. Only
90 Fungal Virology

10 20 30 40 50 60

G AAA AAT
"
AAA GAA ATG ACG
"
AAGCCA ACC
"
CAA
G'n' TN.
"
GTT
AGA Tee GTC
"
AGT
ATA TTA TTT
"
Met Thr Lys Pro Thr GIn Va] Le'.l Val Arg C: .. ' Val Ser Ue Leu Phe

70 80 90 1UO 110 120

TTC ATC
"
ACA 'l'TA CTA CAC
"
CTA
GTC GTA
"
GCG
CTG AAC
"
GAT
GTG GCC GGT GCA GAA ACA
"
CCT "
Phe lIe Thr Leu Leu His Leu Val Val Ala Leu Asn Asp Val Ala Gly Pro Ala Glu Thr

130 140 160 170 180

* * o~~ 150
* a "" "
GCA CCA GTG TCA TTA CTA CCT CGT GAA GCG CCG TGG TAT GAC AAG ATC TGG GAA GTA AAA
Ala Pro Val Ser Leu Leu Pro Arg Glu Ala Pro Trp Tyr Asp Lys lIe Trp Glu Val Lys

190 200 210 220 230 240

GAT TGG
"
C'I'A
TTA CAG CGT "
GCC
ACA GAT
"
GGC
AAT TGG
"
GGC
AAG TCG ATC TGG GGT TCA "
ACC
"
Asp Trp Leu Leu GIn Arg Ala Thr Asp Gly Asn Trp Gly Lys Ser lIe Thr Trp Gly Ser

250 260 270 280 290 300

" " " "
TTC GTA GCG AGC GAT GCA GGT GTA GTA ATC TTT GGT ATC AAT GTG TGT AAG AAC TGC GTG " "
Phe Val Ala Ser Asp Ala Gly Val Val lIe Phe Gly lIe Asn Val Cys Lys Asn Cys Val

310 320 330 340 350 360

CGT" AAG AAG" "

GGT GAG GAT GAT
"
ATC
AGT ACG GAC TGC GGC CAA ACA CTT TTA CTA GTC
"
GCT
Gly Glu Arg Lys Asp ASp lIe Ser Thr Asp Cys Gly Lys GIn Thr Leu Ala Leu Leu Val

370 380 390 400!: 410 420

" * " * a y " "
AGC ATT TTT GTA GCA GTT ACA TCC GGC CAT CAT CTT ATA TGG GGT GGT AAT AGG CCG GTG
Ser lIe Phe Val Ala Val Thr Ser Gly His His Leu lIe Trp Gly Gly Asn Arg Pro Val

430 440 450 460 470 480

TCG CAG
*
TCA GAT CCT AAT
"
GGC
GCT ACC
"
GTT
GCT CGT "
CGT
GAC ATT TCT GTC GCA GAC "
ACT
"
Ser GIn Ser ASp Pro Asn Gly Ala Thr Val Ala Arg Arg Asp lIe Ser Thr Val Ala Asp

490 500 510 520 530 540

GGG GAT
"
ATT
CCA CTC GAC "
TTT
AGT GCG AAC GAC
"
TTG
"
ATA
TTA AAT GAA GGT ATT AGT
*
CAT
"
Gly Asp lIe Pro Leu Asp phe Ser Ala Leu Asn Asp lIe Leu Asn Glu His Gly lIe Ser

550 560 570 580 590 600

ATA CTC
*
CCA
GCT AAC GCA
"
TCA
CAA TAT
"
GTC
AAA AGA "
TCA
GAC ACA GCC CAC ACG ACA
"
GAA
"
lIe Leu Pro Ala Asn Ala Ser GIn Tyr Val Lys Arg Ser Asp Thr Ala Glu His Thr Thr

610 620 630 640 650 660

AGT TTT
"
GTA
GTG ACC AAC
"
AAC
TAC ACT
"
TCT
TTG CAT
*
ACC
GAC CTG ATT CAT GGT AAT
"
CAT
"
Ser Phe Val Val Thr Asn Asn Tyr Thr Ser Leu His Thr Asp Leu lIe His His Gly Asn

670
"
680
"
690
"
700
*
7!0 7;0 Yh f3
GGA ACA TAT ACC ACG TTT ACC ACA CCT CAC ATT CCA GCA GTG GCC AAG CGT TAT GTT TAT
Gly Thr Tyr Thr Thr Phe Thr Thr Pro His lIe Pro Ala Val Ala Lys Arg Tyr Val Tyr

730 740 750 760 770 780

CCT ATG
"
TGC GAG
CAT GGT
"
ATC
AAG GCC TAC TGT "
TCA
*
ATG
GCC CTT AAT GCC ATG GTG
"
GAT
*
Pro Met Cys Glu His Gly lIe Lys Ala Ser Tyr Cys Met Ala Leu Asn ASp Ala Met Val

790 800 810 820 830 840

TCG GCT
"
AAT
GGT AAC CTG "
TAT
GGA CTA "
GCA
GAA AAG
"
CTG
TTT AGT GAG GAG GGA CAA
*
GAT
*
Ser Ala Asn Gly Asn Leu Tyr Gly Leu Ala Glu Lys Leu Phe Ser Glu Asp Glu Gly GIn

850 860 870 880 890 900

TGG GAG ACG"AAT TAC TAT

"
AAA
TTG TAT
*
TGG
AGT ACT "
GGC
CAG TGG ATA TCG ATG AAG
"
ATG
"
Trp Glu Thr Asn Tyr Tyr Lys Leu Tyr Trp Ser Thr Gly GIn Trp lIe Met Ser Met Lys

FIGURE 2. The sequence of the 5' terminal region of the M, plus strand. The sequence of the eDNA rather
than the RNA is shown. The borders (actual or predicted) between the toxin subunits are indicated. Under-
lined amino acids are possible glycosylation sites.
 91

910 920 930 940 950 960

* * * * *
TTT ATT GAG GAA AGT ATT GAT MC GCC AAT MT GAC TTT GAA GGC TGT GAC ACA GGC CAC
Phe Ile Glu Glu Ser Ile Asp Asn Ala Asn Asn Asp Phe Glu Gly Cys Asp Thr Gly His

970 980 990 1000 1010 1020

* * * * *
TAG GGC ATC GTG TCT GAC CGG GCT ATA TAT MTCTG ATG CGA TAA CTC GAC CCT ACA GAG

1030 1040 1050 1060

* * * *
CAC CTA TAG TAG GCA CMMT AAA ATA AM ATT ATA TT

FIGURE 2 (continued)

the U-rich end has extensive homology, so this should be where the 5' end of the plus
strand originates. We have shown that the transcriptase does initiate at the U-rich end
of L,.28 The homologous 3' end of S (see Section B.3) and M 40 also serves as the
initiation site for the viral transcriptase. The 3' ends of the messenger strands of the
yeast viral dsRNAs are similar to the 3' ends of some plant virus genomic (plus strand)
RNAs, which serve as recognition sequences for coat protein-replicase com-
plexes.z.;,36,ss A comparison of the 3' ends of the ScV and Ilarvirus plus strands is shown
in Figure 3. There is evidence for this secondary structure in L, in the form of Sl
nuclease digestion experiments. J6 The yeast viruses may replicate by a similar mecha-
nism, which probably involves different host factors for M and L (see Section D.2).
Consequently, the putative replicase activity, which must recognize the C-rich termi-
nus, may not require the same sequence at the 3' termini of Land M.
There are at least two other minor dsRNAs present in most strains of Saccharomyces
cere visia e without homology to either M or L, of unknown function, and not packaged
in L, or La PI (see Section B.1). 56 The Yarrowia jipolytica dsRNA virus has a genomic
dsRNA of about 5.2 kbp, which may be analogous to the ScV L.J

B. Expression of the Viral Genome

1. Viral Proteins - Capsid
There appears to be only one major capsid polypeptide in ScV -L, and ScV -M, par-
ticles!,,22,S7 ScV-P1, of about 88 kDa. 22 No other capsid polypeptide has been consist-
ently demonstrated. 22 ,'8 By performing Western blots of ScV viral polypeptides, we
have been unable to find any polypeptides reacting with anti-ScV antibodies that do
not also react with anti-PI antibodies. 59 The 180 kDa polypeptide sometimes present 39
is probably a dimer of PI not dissociated by SDS, since it reacts with anti-P 1 anti-
body.59 There may, however, be small amounts of polypeptides smaller than PI yet
undetected. 59
ScV-PI is also the major capsid polypeptide in ScV-M, particles: SDS-polyacryl-
amide gel electrophoresis (SDS-PAGE) of disrupted ScV-M, and ScV-L, reveals only
ScV -p 1. 57 Tryptic peptide analysis of the ScV -M, major polypeptide confirms that it is
also ScV-PI.60 ScV-M, and ScV-L, also have immunological cross-reactivity.6' The
ScV-La P, apparently does not encapsidate M,.62 L, and L2 ScV-P, are similar in size
and immunologically related;42 ScV -La particles have two large polypeptides" of about
76 and 80 kDa without immunological cross-reactivity to L, P 1 or L2 P ,.37
Although strains having neither ScV-L nor ScV-M and strains having only ScV-L
exist, there are no strains with ScV-M or ScV-S that simultaneously lack ScV-L. This
dependence of ScV -M and ScV -S particles on ScV -L particles and the presence of the
same major capsid polypeptide in ScV-M and ScV-L is due to the fact that L encodes
the major capsid polypeptide of ScV, as shown by Hopper et al. 22 by in vitro transla-
tion of denatured L, dsRNA. The product made in the wheat-germ translation system
 C A U \0
N
U U C C
G A G C
GC C U

-
~
AU U G
CG UA AA ~
AU U AU U C ..,'o
;$
UA
UG
A U
A A
UA
CG
OG
GC g
AU GC CGA UA
U GC U GC AA
CG GC UAGC UA A C
GC UA GC AU AU
AGUUCUGAUCCACUCOGGCAAAAAGAUAAUAUGCAoH AU UA CG
AU AU GC UA
A
GC UA UA C A
UA AU AU OG
AU AU UA CG CG
A U AGGUAUUGCUAAUGCUAAUGCGAOGCoH
AU c
UA
GC
GC
AU
GAAUUACUACAGCAoH
B

FIGURE 3. Secondary structure at the 3' ends of the ScV plus strands: comparison to alfalfa mosaic virus RNAs. A. L.; B, M,; and C. alfalfa mosaic virus RNA4.
 93

precipitates with anti-ScV gamma-globulin, coelectrophoreses with authentic ScV-PI

in SDS-PAGE, and has tryptic pep tides labeled with 35S-methionine of the same chro-
matographic characteristics as those of ScV-Pl. ScV-Pl is also synthesized on dena-
tured L, in the Schreier-Staehe1in in vitro system. 63 L. encodes the ScV-L. capsid poly-
peptide, by similar criteria. 59 In the similar killer dsRNA viral systems in Yarrowia
lipolytica and UstiJago maydis, the major viral capsid polypeptide is in each case about
7S kDa. 3,64
If ScV-L particles consisted solely of one molecule of L and a number of PI mole-
cules, the number of the latter could be determined by measuring the mass of the ScV
particles. The measured mass of ScV -L, by direct measurement using scanning trans-
mission electron microscopy is 8.48 ± 1.58 x 10 6 daltons."9 If the mass of L is taken as
3.2 x 106 daltons (the sodium salt of a dsRNA of 4.7 kb), this implies that each particle
can have only one molecule of L and that there are 59.7 ± 11 molecules of PIper
particle. This corresponds to a simple icosahedron of T= 1 (60 subunits), although the
standard deviation of the mass measurement is unusually large for a virus.

2. Viral Proteins - Toxin and Resistance Factor

There are at least 13 killer toxin and immunity specificities among yeast genera. 2,65,66
Species of Candida, Debaryomyces, Hansenula, Kluyveromyces, Pichia, Saccharomy-
ces, and Torulopsis have killer toxins.' With the exception of a Torulopsis species,
none of the non-Saccharomyces species are killed by the ScV toxins. The widespread
occurrence of killer systems in the yeasts extends to species more recently tested: for
instance, Kluyveromyces lactis" Pichia kluyveri: 7 and Yarrowia (Saccharomycopsis)
lipolytica. 3 Most toxins studied are protease sensitive, heat labile, and stable only be-
low pH 5. ',68,69 There are three toxin and resistance specificities (k" k" and k3) among
Saccharomyces species,2 the best studied of which is the k,.
The k, killer toxin (that encoded by M,) has a pI of 4.5 and is stable and active
within the range of pH 4.2 to 4.6. 70 The toxin is secreted into the medium. As few as
104 molecules of toxin will kill a cell.70,7' The first site of interaction with cells is the
cell wall,71-73 where the receptor is a beta-(1,6)-D-glucan!4 There is a second receptor
on the cell membrane, since spheroplasts of cells resistant by virtue of lack of the cell
wall receptor (krel) are sensitive. 71,72, 74 Effects of treatment of sensitive cells with the
toxin include leakage of potassium ions 72 and acidification of the interior of the cell. 75
Like some bacterial colicins, the ScV and Pichia kluyveri killer toxins appear to me-
diate their effects by providing ion-permeable channels in the membrane, destroying
the proton gradient responsible for amino acid and K+ transport: 67 ,75 The killer toxin is
a protonophore. The resistance (immunity) factor may alter the affinity of the mem-
brane receptor for the toxin. If, as presently assumed, the resistance factor is part of
the preprotoxin, it might bind to the inner portion of the membrane receptor in the
final cleavage reaction which creates the alpha and beta toxin peptides (see following).
The stoichiometric requirement for the immunity peptide is consistent with such a
model!6 The retention of immunity by kexmutants, in which protoxin is not processed
(see following) suggests that the immunity portion of the protoxin may be bound to its
receptor even without the normal proteolytic cleavages, and that in normal processing,
binding may occur to the receptor prior to at least the two final cleavages at the alpha-
gamma and gamma-beta borders (see following).
The presence of M is a necessary condition for the expression of the killer pheno-
type, 27, 77 79 since M codes for the killer toxin and for protein(s) involved in toxin re-
sistance. Denatured M, efficiently directs the synthesis of a polypeptide of 32,000 dal-
tons which is precipitated by anti-toxin gamma-globulin and contains all but one of the
seven tryptic peptides of the purified toxin.s 8,70 The in vitro product also contains four
tryptic peptides not found in the toxin. None of the M, in vitro translation products
94 Fungal Virology

are immunoprecipitated with anti-ScV gamma-globulin. Denatured S dsRNAs (SI, S3,

and S4: see Section C.2) cause the synthesis of an 8000 dalton protein with immuno-
logical cross-reactivity to the toxin. The 32 kDa in vitro product is probably the pre-
protoxin, which can be converted to the glycosylated protoxin (42-43 kDa) in vitro by
dog pancreas membrane vesicle preparations. 80 Unprocessed protoxin accumulates in
yeast sec mutants and kex mutants. 81 There are also M mutants blocked in processing
of the protoxin. 82 From the known sequence of the 5' 1 kb of the plus strand of M I , it
has been deduced that the preprotoxin includes a 44 amino acid leader sequence, fol-
lowed by a toxin alpha peptide of 9.0 kDa, a gamma resistance factor of 12 kDa, and
a toxin beta peptide of 8.5 kDa. Amino terminal sequence analysis of the alpha and
beta toxin peptides has confirmed their assignments to MI sequences. 48 In the current
model for maturation of toxin, the preprotoxin is glycosylated at sites within the re-
sistance factor portion of the preprotoxin, then cleaved into the three peptides. The
secreted toxin is thus a dimer with two subunits of nearly equal molecular weight,48
differing from the previously estimated 11.5 kDa monomer. 70
The preprotoxin is processed by the normal pathway for secretory proteins in yeast.
Protoxin is glycosylated, since it is reduced in size from 43 kDa to about 34 kDa by
endoglycosidase H. 80 This unglycosylated form also accumulates in the presence of
tunicamycin, an inhibitor of the synthesis of the glycosyl donor. 81 There are three asn-
x-thrlser potential sites for glycosylation in the gamma portion of the preprotoxin
(Figure 2), but no such sites in alpha or beta. 48 The mature toxin, consisting of alpha
and beta, is not glycosylated. The size of the protoxin is also consistent with glyco-
sylation at these three gamma sites. Protoxin of normal size accumulates in sed8, sec7
and sed mutants, in which secreted proteins accumulate in the endoplasmic reticulum
and are blocked in transfer to the Golgi (sed8) or blocked in transfer to secretion
vesicles (sec7) or blocked in transfer from secretion vesicles to the exterior (sed).81
Since elongation of carbohydrate chains occurs in the Golgi, only primary glycosyla-
tion appears to take place on the preprotoxin. Preprotoxin is glycosylated by dog pan-
creas membrane vesicle preparations,80 which also cleave off an N-terminalleader pep-
tide of about 1.6 kDa cotranslationally. This cleavage does not appear to occur in
yeast,83 and the predicted cleavage site for cotranslational cleavage (the ala-leu se-
quence at bp 91 of Figure 2) would predict the loss of about 2.9 kDa, compared with
the loss of 1.6 kDa by processing in dog pancreas membrane vesicles. Hence, at least
this cleavage is not shared by yeast and mammals. Elimination of a portion of the
N-terminal leader peptide from the preprotoxin gene results in a loss of glycosylation
but only a 50070 decrease in secretion of (presumably) processed toxin. 76 The signals for
glycosylation and for secretion are at least partially separable in this system. Protoxin
is most stable in sed8 mutants at the nonpermissive temperature, less stable in sec7,
and least stable in sed mutants, suggesting that the cleavages of protoxin normally
occur in secretory vesicles, very late in the normal process of secretion.
TPCK, an inhibitor of chymotrypsin-like proteases, also causes accumulation of
protoxin in some strains. 81 Both KEXI and KEX2 are required for processing of pro-
toxin. 81 Both kex 1 and kex2 mutants are defective in processing of many secreted
proteins; neither secretes active toxin or antigenically reactive material. Protoxin is
stable in kex 1 but not in kex2 mutants, and stable in kex 1 kex2 double mutants,
implying that the KEXI protein (a TPCK sensitive protease) renders protoxin suscep-
tible to the KEX2 protein (a TPCK insensitive protease). The KEX2 gene has now been
cloned and the product identified and characterized. 84 It is an endopeptidase specific
for pairs of basic amino acids, in which one of the pair is usually an arginine. Conse-
quently, KEX2 is probably responsible for the trypsin-like cleavage at the gamma-beta
border (bp 712 in Figure 2). The proposed site of action of the chymotrypsin-like en-
zyme is the trp-gly bond at bp 403 (Figure 2) at the alpha-gamma border. The gene
 95

product of KEXI is the prime candidate for this endopeptidase. The KEXI cleavage
could be required for a conformational change necessary for access of the KEX2 pro-
tease to the gamma-beta cleavage site.
The evidence that the resistance factor is indeed encoded by the 5' 1 kb of the plus
strand of MI is that this eDNA, appropriately inserted into an expression vector DNA,
confers resistance to toxin upon sensitive yeast cell transformants. 7• Preliminary map-
ping of mutations in the preprotoxin gene 48 is also consistent with this assignment. The
sequencing of MI from mutants blocked in processing of protoxin because of altera-
tions in M l 82 and in vitro mutagenesis of MI cDNA 7• should help identify unambig-
uously sites necessary for processing.
The retention of immunity in the kex mutants and the evidence that the final proc-
essing steps take place in vesicles rather than in the Golgi suggest the following model.
Gamma projects out of the secretory vesicle and guides the vesicle to the membrane
receptor, where binding of gamma to the receptor triggers the final two cleavages. The
resulting alpha-beta dimer, held together by several disulfide bonds, would then be
secreted by fusion of the vesicle with the membrane. The binding of gamma to the
inner portion of the receptor could alter the configuration of this trans-membrane
protein so that its external portion, normally recognized by the toxin alpha-beta dimer,
would not now be so recognized. In correspondence with this model, gamma is hydro-
philic and glycosylated, while alpha has a long hydrophobic region. One prediction of
this model is that immunofluorescence binding experiments with antibody to gamma
should fail to demonstrate its presence on the membrane of spheroplasts unless they
are first made permeable to the antibody.
The entry of toxin to its site on the membrane requires interaction with a cell wall
receptor. It has been proposed that interaction of beta with the cell wall receptor causes
the reduction of toxin into alpha and beta subunits, permitting traversal of the cell wall
and access to the cell membrane of the active component of the toxin (alpha).85 It is
equally likely that the only role of the cell wall receptor is to concentrate the (unaltered)
toxin in the periplasmic space. The function of beta could not be, however, the protec-
tion of the producer cell from action of its own toxin,85 either on the inside or the
outside of the membrane, since spheroplasts of mutants lacking the cell wall receptor
(kre mutants) are sensitive to the alpha-beta dimer. 71,72,74 In our model for processing
and secretion, the receptor for alpha/beta action is exclusively on the outside of the
cell membrane and in no danger of interaction with maturing toxin. Since interaction
of the inner portion of the receptor with gamma is prerequisite to secretion, normal
cells are never in danger of being killed by their own toxin. The only case requiring a
more complex explanation is that of the rex mutants, which secrete toxin but are sen-
sitive to it. These may have a defective inner membrane receptor that interacts abnor-
mally with gamma (see Section D.2).

3. ScV Transcriptase
Like all dsRNA viruses, including the fungal viruses,"· 90 ScV-L and ScV-M particles
possess a single-stranded RNA polymerase activity. 91 92 Sc V -S particles also have this
activity. 53 The polymerase is insensitive to DNase, alpha-amanitin, and actinomycin
D, but sensitive to ethidium bromide and pyrophosphate. 92 It requires all four ribo-
nucleoside triphosphates and magnesium ion.92 The major product of the reaction in
ScV -L particles is a single-stranded RNA of the molecular weight of a denaturated
strand of L,.3,91,93 which is entirely complementary to only one of the two strands of
L. .3,9394 The characteristics of ScV _M95 and ScV _SS3 transcripts are similar.
The ScV RNA polymerase activity is not dependent on S-adenosylmethionine
(SAM).·3 The product of the in vitro reaction begins with pppG,.3 but initiation may
not occur in vitro: there is no incorporation of gamma- 32 P-GTP, and the penultimate
96 Fungal Virology

5' T, oligonucleotide of L, (32 nucleotides long) is not labeled in vitro. 63 ,96 The product
serves as mRNA for viral capsid polypeptide synthesis in vitro. 63 The viral polymerase
is thus a transcriptase.
The transcriptase makes a series of 5' pause products.28 The sequences of the smaller
of these have identified the transcript strand as originating, as expected, at the V-rich
end of L. These pause products are similar to the 5' terminal oligonucleotides made by
the reovirus transcriptase. 97 The origin of the longer fragments of transcript also pres-
ent in viral particles in unclear, but they ure not from the 5' end. 46 The M and S
transcripts initiate at the homologous end G." their viral dsRNAs. The strandedness of
the M, transcript has been established by 3' sequencing 95 and the strandedness of the S
transcript by hybridization to M 13 cDNA clones,53 in each case confirming stranded-
ness deduced from the terminal sequences of the viral dsRNAs. 25
Because the Sc V transcripts are extruded from the virus particles,92 we prefer a
model for replication of ScV particles similar to that established for reovirus. The
reovirus transcriptase is conservative and is activated by a partial uncoating of virus
particles in lysosomes; the transcriptase synthesizes ssRNAs that act as mRNAs for the
synthesis of viral proteins; these ssRNAs are also templates for the synthesis of com-
plementary strands within nascent virions; these "sub-viral particles" continue the
cycle of ssRNA synthesis; finally, virions with completed outer capsids cease ssRNA
synthesis. 98 ,99 Because of the similarity between the 3' ends of some of the ScV plus
strands and those of alfalfa mosaic virus (Figure 3), we have postulated that a PI-
replicase complex is assembled at the 3' C-rich ends of the ScV transcripts." Control
of replication would then be coupled to synthesis of PI, a convenient and common
viral strategem. If the formation of stem and loop structures at the 3' termini of the
plus strands were necessary for recognition by a PI-replicase complex, their absence
might require the participation of additional replication factors. M" which does not
have a sequence amenable to the formation of such secondary structure at the 3' end
of the plus strand, does require many more host factors than does L, which has such
3' terminal structures. Many of the mak gene products are unnecessary for L, (see
Section D.2).
There is no definitive evidence on whether ScV transcription in vivo is conservative,
as in reovirus,99 or semiconservative, as in Phi6'oO and AfV -So 90 Our results demon-
strate conservative transcription in vitro.lOo.
Experiments on in vivo replication are inconclusive. '0' More recent experiments lO2
have been interpreted to favor conservative replication in vivo, but again do not rule
out semiconservative replication by a small fraction of the particles. Since transcription
in vitro does not appear to initiate,03,96 it is possible that a minor fraction of particles,
those containing primers, can transcribe in vitro or in vivo. The viral transcriptase
pause products that originate from the 5' end of the transcript'8 and are present in viral
particles prior to transcription 96 in vitro could serve as such primers, explaining the
lack of labeling of the penultimate 5' T, oligonucleotide. There is sufficient molar
excess of the 5' pause products to readily account for twenty times as much synthesis
as is observed. It is unlikely that only elongation of previously initiated transcripts, as
opposed to primed synthesis, is taking place, since the reaction proceeds for at least 3
hours.63
There is a viral replicase. Viral particles synthesize full-sized minus strand in vitro
and all label incorporated into dsRNA is in minus strand. '00. The replicase is thus
conservative as well. This supports the applicability of the reovirus model of replica-
tion.
The only polypeptide so far documented in Sc V is PI, so it may be that a viral
particle with only one polypeptide is capable of transcriptase and replicase activities.
 97

However, a single copy of another polypeptide would not have been detected, and
there is a possible second reading frame for L 1. 46

4. In Vivo mRNAs
Ll and Ml produce both full-length and partial length transcripts. 103 The partial
length transcript of L encodes PI (the major viral capsid polypeptide) and is about 2.3
kb in length. The partial transcript does not bind to poly(U)-sepharose,103 but there is
disagreement about whether the full-length transcript is polyadenylated. 95 ,,03,,04 Both
the full-length and partial length (1.2 kb) M transcripts encode the preprotoxin, and
both are bound to poly(U)-sepharose. Binding of these latter RNAs to poly(U)-sephar-
ose may not indicate polyA 3' termini, since each has an AU tract capable of binding
0Iigo-dT.49,51 In fact, the M in vitro transcript does bind to poly(U)-sepharose and, by
sequence analysis, is not polyadenylated. 9S The preponderance of evidence indicates
that none of the ScV in vivo mRNAs is polyadenylated,95 104 which would not be sur-
prising, since they are made in the cytoplasm, and the ScV virions do not have a
poly(A) synthetase activity. The 5' structure of the in vivo mRNAs has not been inves-
tigated, but, like the in vitro transcripts 63 and the genomic dsRNAs,'9 they are probably
uncapped. ScV virions have no capping activity and, again, their mRNAs are synthe-
sized in the cytoplasm, while the cell's mRNA guanylyl transferase is nuclear. There-
fore, the ScV mRNAs probably differ from yeast nuclear mRNAs by lack of both 5'
and 3' modification.
The partial length transcript of L probably results from cleavage or termination of
transcription between two long inverted repeats in L, and represents the 5' 2.3 to 2.6
kb. 46 A second subgenomic mRNA from the 3' end of the L, plus strand may also
exist,46 The subgenomic M transcript may also be the result of cleavage of the full-
length ScV transcript within the AU-rich region. 95 In fact, cleavage by nucleases occurs
so readily in this region" that the subgenomic Ml mRNA may be an artifact of isola-
tion. It is not yet clear whether the 3' portions of the Land M transcripts are func-
tional.

C. Viral Interactions
1. Exclusion
In crosses of k, killer strains (those with L, and M ,) with k2 killer strains (those with
L, and M 2), M, replaces M2.24 Some nonkiller strains also have a cytoplasmic element
capable of causing exclusion," which is an allele of L I .'",39 ScV-L, replaces SCV-L2 in
these crosses.'" We interpreted exclusion as the result of the preferential replication of
ScV-L, particles. Since M is encapsidated in the polypeptide encoded by L (see capsid
polypeptides), the resultant exclusion of ScV -M, could occur if the L capsid polypep-
tide was incapable of encapsidating M 2. L, and L2 are considerably different in se-
quence!5,4l as are their capsid polypeptides." However, this simple hypothesis does
not account for the fact that L INEX , which prevents exclusion of M2 by L IEXL , is also an
allele of L , ." 4' A more extensive genetic analysis shows that one allele of L. (L.EXd
excludes M2 except when L . NEX is present, but that another allele of L. (L.HOK-NEX)
usually excludes M. only when a nuclear gene, mkt, is present." Since L INEX is indistin-
guishable from L . EXL by hybridization to cDNAs derived from L.HOK-NEX, but quite
distinct from L2,4' the interpretation of exclusion solely as a function of capsid poly-
peptide structure is probably naive.
The k2 strains in which M. is not excludable by L . EXL also have L dsRNAs with the
second phenotype characteristic of most laboratory strains: HOK. HOK was originally
described as the cytoplasmically inherited trait that allows an altered M. (a ski-depend-
ent M.) to replicate in a SKI+ host.",·05 It is now known that these strains did not have
an altered M., but an altered L. (L. EXL ): L. EXL does not replicate M. in a skimutant. 45
98 Fungal Virology

Table 2
GENETIC CROSSES DEMONSTRATING EXCLUSION

Genotypes of Parents and Haploid Progeny

Parent A Parent B Haploid progeny Exclusion of M,

MKT L' HN MKT L' HN M, MKT L' HN M, No

MKT L' HN mktL,M, mktL' HN Yes
MKT L' HN MKTL,M, MKT L' HN M, No
MKT L,E MKT L'HN M, MKT L'HN M, No
MKT L,E MKT I mkt L, M, MKT/mktL'E Yes
MKT L'HN M. MKT L'HN M, MKT L'HN M, Yes
MKT L'HN M. MKT/mktL, M, MKT/mktL'HN M, Yes

Note: L'HN is L ,HoK NEX; L'E is L ,EXL ; mktis the recessive (defective) allele of one of two
genes for maintenance of killer type 2 and MKT its dominant (functional) allele.
In some crosses, only the relevant progeny are shown. L. is usually present in
both parents and has no effect on the crosses.

The EXL-resistant mutants are L2 mutants selected for NEX by mating with an L'EXL
strain under conditions such that the strains with a wild-type L 2, having lost M2 and
become sensitives, would be killed by the toxin produced by the mutant L 2-containing
strains, since these will still have M 2. The mutant L2 alleles also have HOK activity,4S
but it is not clear whether this HOK activity is the result of encapsidation of M, in L2
P I. Since the L2 mutants are derived from an RNA easily distinguishable from L2 by
Northern gel analysis, it should be possible to determine whether both L, and L2 persist
after selection, and hence if the L2 capsid polypeptide is capable of encapsidating M,.
lt has been possible to separate the roles of ScV-L, and ScV-L. in exclusion because
the maintenance of ScV -L, is temperature sensitive and requires MAK3 (see Section
D.3, Nuclear Genes), while the maintenance of ScV-L. is not temperature sensitive and
does not require MAK3. 3 •. 39 Some alleles of L, are also dependent on the product of
MAKIO,39.44 MAK27,45 or PETI8. 44 .45 These are the only nuclear genes necessary for
the maintenance of ScV -L as well as ScV -M, although other nuclear gene products may
modulate the numbers of ScV-L particles (see Section D, Virus-Host Interactions).
The exclusion of SCV-M2 that occurs in crosses in which one parent has SCV-LINEX
and SCV-M2 and the other parent has SCV-LIHOK.NEX and ScV-M, apparently occurs by
a different mechanism: competition between M, and M2 for some limiting component,
in which M, supplants M 2.3. 39 Host gene products must playa critical role in compe-
tition between M, and M 2.45
In summary, there are at least three distinct modes of exclusion of SCV-M2' Exclu-
sion may be caused by SCV-L'EXL if the L species present in the k2 strain does not have
the NEX phenotype (L INEX )' L, replaces L. in these crosses. A second mode of exclu-
sion is mediated by the L species normally present in most laboratory strains of S.
cerevisiae: LIHOKNEX' If either of two known alleles of mkt is present, L, HOK.NEX will
exclude M 2. Finally, if the L species in the k2 strain has the NEX phenotype, exclusion
will still take place if M, is present in the cross. In this case, direct competition appears
to Occur between the M species, a competition that M, invariably wins.
A series of crosses demonstrating exclusion and the proven or predicted genotypes
of the progeny haploids is shown in Table 2. A number of interesting crosses have not
yet been reported. For example, the mkt LIHOK-NEX haploids derived from the second
cross, if their genotype has been accurately deduced, when crossed with MKT
L'HOK NEX M 2, should give 1/2 mkt L, HOK-NEX and 1/2 MKT L IHOK NEX M2 haploid prod-

ucts. Also, the k2 killer progeny of the third cross should now be NEX, while their k.
parent was not.
 99

2. Suppression
Suppressive sensitive strains of S. cerevisiae, '06 when mated to killer strains, produce
sensitive diploids and sensitive spores from those diploids. The resultant sensitive
strains are again suppressive sensitive. Such strains no longer have M dsRNA, but
rather smaller dsRNAs!8,79 also encapsidated in ScV particles,23,6' which are desig-
nated ScV-S.
Since S replaces M in each case, Sc V -S particles appear to be analogous to the defec-
tive interfering (DI) particles of animal viruses. ,o7 A series of experiments has demon-
strated that S dsRNAs, like many DI RNAs, are derived from their parental RNA by
internal deletion; that ScV-S particles are under the same replication control as ScV-M
particles; and that, like DI particles, ScV -S replaces ScV -M by faster replication. The
S dsRNAs consist of two size classes: S3, S14, and S733 of 0.7 to 0.8 kb and Sl, S4,
S2, and S17 of 1.3 to 1.5 kb. '3,30 32,'08 They are all derived from M 2 3,30,3' by internal
deletion.23,31,32 At least one of the larger S dsRNAs (Sl) is derived from a smaller (S3)
by tandem duplication,3o,32 and several other S dsRNAs have been observed to undergo
tandem duplication. 53 Some may have further internal rearrangements or dele-
tions!3,30,32 The "break-points" in M may be similar for many of the S dsRNAs, since
the two examined (S3 and S14) both have about 200 to 250 bp from one end of M and
about 500 bp from the other. 32,53 The 250 bp is from the 5' end of the M, plus
strand. 53 ,54 One of these dsRNAs (S14) has now been completely sequenced: it has 253
bp from the 5' end of the M, plus strand and 540 bp from the 3' end of the M, plus
strand. 53 The structure of these suppressive dsRNAs is diagrammed in Figure 1. It is
likely that all the suppressive dsRNAs are derived by a very similar event, and that the
two size classes of S result from the primary event (generating an S of 700 to 800 bp)
and a tandem duplication (generating an S of 1.4 to 1.6 kb). One possible mechanism
for tandem duplication is for dissociation from one template at the 3' terminus of the
plus strand to be followed immediately by reinitiation at the 3' end of the minus strand
on the same template S dsRNA or on a second S molecule present in the same parti-
cle.23 At least one of the mak genes affects the replication of ScV-S, just as it affects
the replication of ScV -M; '09 this is consistent with the preservation of the terminal
sequences of M in S, and therefore the conservation of transcription and replication
signals.23,30 As expected, ScV -S particles replace ScV -M particles by faster replica-
tion."°
S dsRNAs derived from M" just like their parental dsRNA, exclude M,.45 Clearly,
direct competition between M, and M, in exclusion is very similar to the competition
between M, and its derivative S dsRNAs in suppression, and probably occurs by a
similar mechanism (probably competition for capsid polypeptide during replication).
The structure of the S dsRNAs, their sequence variability, and their competition with
their parental RNA are all properties of DJ particle RNAs.'"115 The lack of correla-
tion between degree of suppressiveness and size of S (for instance, Sl displaces S3) is
also typical of animal virus DI RNAs and may represent the number of copies of a
critical sequence (for instance a capsid polypeptide recognition site) present on the
RNA. No suppressive dsRNAs derived from M, or M3 have yet been described.

D. Virus-Host Interactions
1. Control of Replication
Despite an extensive knowledge of the genetics of the host, very little is known about
the actual mechanisms by which ScV replication is regulated. ScV replication must be
regulated, since the number of virus particles per cell is relatively constant and limited
sufficiently that cellular death does not ensue. Virus replication may be synchronized
with cell division: ScV replication appears to occur in G1 but not in S,'O',116 although
there is some disagreement on this point.'o, There is evidence for nuclear gene modu-
100 Fungal Virology

lation of ScV replication: for instance, the copy number of ScV-L, is affected by the
SKI gene products.' 17

2. Nuclear Genes
The presence of ScV-M wild-type particles confers on their host cells an easily rec-
ognizable phenotype, the ability to secrete killer toxin and immunity to its effects.
ScV-L particles confer no easily recognizable phenotype. Consequently, the search for
chromosomal mutations affecting ScV maintenance or expression has been confined
to those that affect inheritance or expression of M. Several of the mutations presently
characterized do have a noticeable effect on L.
There are at least four general classes of nuclear mutations affecting the k, killer
phenotype: mutations eliminating both killing and resistance (mak);1!· mutations that
eliminate the ability to kill but have no effect on resistance to toxin (kex);"·'"' muta-
tions causing a broad spectrum of resistance to killer toxins (kre); 73 and mutations
eliminating resistance to toxin but not the capacity for killing (rex).'" Note that these
last strains grow normally above pH 4.6. Other nuclear genes also affect ScV -M repli-
cation (e.g., spe2) or expression (e.g., the sec genes). Some of the makmutations are
known to affect M2 as well as M,.24
The mak mutants owe their phenotype to the loss of M in mak haploids or in
mak/mak diploids, ". since M codes for the killer toxin and for toxin resistance. There
are now some 29 mak genes established by complementation and mapping."·,'20 122 It
is unclear what role nuclear genes play in the maintenance of the yeast viruses. Three
genes required for maintenance of M, whose cellular functions are known are MAK8,
which codes for ribosomal protein L3; 123 SPE2, which codes for an enzyme required
for polyamine synthesis; '24 and MAKl, which codes for DNA topoisomerase 1.'25
Nothing is currently known of the nature of ScV-M dependence on these cellular pro-
teins. Polyamines might be necessary for packaging of the viral dsRNAs, as they are
for a number of RNA viruses, but it is not clear why M should have a more stringent
requirement than L. The requirement for DNA topoisomerase I is especially puzzling,
but may be a very indirect effect of these mutations. '25 It has been proposed that L3
plays a role in ScV replication analogous to that of ribosomal protein S1 in Q beta
replication. '23
The rather spectacular multiplicity of nuclear genes affecting the maintenance of
ScV-M remains to be explained. Several of the mak mutations (mak 3, mak 27, and
maklO) are known to affect the maintenance of some alleles of L,.'·,3.,44,45,'2' At least
one mak gene affects the maintenance of S.'o, There are many recessive (Ski) and dom-
inant (KRB) mutations that suppress one or more mak mutations, 126-'2' and recessive
(mkt) and dominant (DET) genes modifying exclusion of ScV-M 2 .45 MKTI is required
for maintenance of M2 but not for maintenance of M,. 24 There are also mutant alleles
of M, (KIL-b) independent of MAK4, 7, 11, and 17.'29 M, is ski dependent in the
presence of L'EXL but not in the presence of L'HOK. 4 5,10S,'2. CLO is required for L. but
not for L,. 56
There are only two kex complementation groups. Wickner and collaborators 130 have
isolated some 30 or more independent mutants of this class and all had alterations of
one of two genes (kexl and kex2). The kexmutants are altered in processing of prepro-
toxin. 5.,., At least one of these genes (KEX2) encodes an endopeptidase involved in
processing of secretory proteins (see Section B.2). The preservation of immunity in kex
mutants suggests a model for binding of the immunity peptide (gamma) to its mem-
brane receptor, for the secretion of toxin (alpha + beta), and for binding to toxin (see
Section B.2).
The rex mutants have not been thoroughly characterized. By the model outlined,
these might be mutants of the membrane receptor for gamma in which the gamma site
 101

was altered, but the outer alpha/beta site was unaltered. The membrane receptor
would have to be altered in such a way that binding of gamma, triggering cleavage of
protoxin, still occurred, but that conformational change of the receptor blocking al-
pha/beta binding did not occur. Since cell wall and cell membrane binding are sepa-
rable, the alpha and beta subunits of toxin may play different roles in these processes
(see toxin and resistance factor). Among another set of nuclear mutations (kre), some
affect cell wall receptors for the toxin. Mutations in krel confer resistance to all killer
toxins tested, but kre 2 and kre 3 mutants are still sensitive to the k2 toxin. 66 ,73
There do not appear to be any copies of the viral dsRNAs in the S. cerevisiaenuclear
genome. 56 ,94,126
III. KLUYVEROMYCES LACTIS PLASM IDS

A. Plasmid DNA
A search of 33 species of yeasts of 17 genera for DNA plasm ids uncovered two linear
dsDNA plasmids in a strain of Kluyveromyces lactis with a killer phenotype. 4 These
DNAs are linear by electron microscopy and by restriction mapping and have sizes of
13.4 kb (pGKI2) and 8.8 kb (pGKII) and CsCI densities of 1.687 g/CC. 4,131 Both plas-
mids are inherited cytoplasmically, and have been transferred to Saccharomyces cere-
visiae by protoplast fusion 132 or by transformation. '33 They have also been fused to
circular yeast vectors for transformation of Kluyveromyces.134 In conjunction with
curing experiments and characterization of mutants, these results have demonstrated
that the pGKII plasmid is responsible for both the killing and resistance func-
tions. '3, ,132,135,136
The coding region for the resistance and toxin genes on pGKll have been roughly
located by virtue of the fact that mutants that retain resistance but lack toxin are
missing 2.6 to 2.9 kb from the central region of the DNA.'31 ,135 The fact that deletion
mutants are lacking internal regions is reminiscent of the DI mutants of ScV and is
probably due to similar requirements for the termini of the dsDNAs for replication.
Linear DNAs must utilize some special strategem to replicate their ends, since all
DNA polymerases require primers. A linear plasmid might use the same strategy as a
chromosome for replication, but this does not seem to the case in K. lactis,'37 pre-
sumably because nuclear replication proteins are not accessible to the cytoplasmic lin-
ear DNAs. Circular DNA plasm ids in S. cerevisiae are replicated in the nucleus, 138 and
artificial linear plasmids with Tetrahymena telomeres and yeast centromeres function
as minichromosomes in S. cerevisiae, presumably replicating in the nucleus. '39 The
terminal regions of pGKII and pGKl2 have been sequenced, and the features charac-
teristic of eukaryotic telomeres (simple tandemly repeated sequences, nicks, and hair-
pins) are not present. Both DNAs have inverted terminal repeats, sequences of about
200 bp in inverted orientation at their ends. The inverted repeats are different in the
two DNAs. The 3' ends of the DNAs have free 3' OH groups, but the 5' ends are
blocked. This is reminiscent of the structure of adenovirus DNA, in which the adeno-
virus 5' terminal protein serves as a primer for replication and becomes covalently
bound to the 5' terminal nucleotide. However, a deletion mutant of pGKl2lacking 8.4
kb from the left end is replicated in the presence of an unaltered pGKl2 plasmid. It
has not been determined how much of the inverted terminal repeat is preserved in this
mutant, nor whether its left 5' end remains blocked. 137
None of the coding regions of pGKl2 have been delineated, but an intact molecule
is necessary for replication of both itself and of pGKll as linear DNAs. 138 Replication
also requires at least one nuclear gene. 131

B. Toxin
In contrast to the Saccharomyces cerevisiae killer toxins, which are effective only
 102 Fungal Virology

- - - - - - - - - - - A In kl-NK '2

-
DRF1 >1
I
ORF t q dRF 3
wo!
-+
IRL IRR

FIGURE 4. Open reading frames in the pGKIl sequence.

against cells of the same genus, the Kluyveromyces lactis toxin is active not only against
nonkiller isolates of the Kluyveromyces genus, but also against many strains of Sac-
charomyces. 4 There is at least one nuclear gene, a recessive allele of which confers
resistance to the toxin. 4
The pH optimum of the Kluyveromyces lactis toxin is quite different from that of
any other killer species characterized. The killer toxins of Saccharomyces, Pichia, De-
baryomyces, Kluyveromyces, Candida, Hansenula, and Torulopsis species character-
ized by Young and co-workers2 all have pH optima of less than 5. The Kluyveromyces
lactis toxin is active at pH 5 to 7, but inactive at pH 4.5. 4 The mode of action of the
Kluyveromyces lactis toxin is also completely different from that of Saccharomyces
(see above). The toxin from the culture medium, purified to electrophoretic homo-
geneity, consists of two polypeptides of 27 kDa and about 80 kDa.'40.141 The toxin
brings about G 1 arrest in Saccharomyces cerevisiae. A number of treatments of yeast
that cause depletion of cyclic AMP cause G 1 arrest: this also appears to be the mode
of action of this toxin. The effects of the toxin on yeast cells are reversed by addition
of cyclic AMP to the medium. The adenyl cyclase activity in a crude membrane prep-
aration from yeast is also dramatically inhibited by the K. lactis toxin. In addition, the
toxin does not exhibit the protonophore characteristics of the S. cerevisiae toxin. 140
Recently, the entire nucleotide sequence of pGK II has been determined. 142 There are
four long open reading frames occupying all but 50 of the 8874 bp, exclusive of the
202 base terminal repeats (Figure 4). Three open reading frames are on one strand,
while the other (ORF3) is on the opposite strand. ORF3 overlaps ORF2 by 10 codons.
Analysis of Northern gels shows that ORF2 encodes a message of 3.5 kb, ORF3 a
message of 1.4 kb, and ORF4 a message of 0.9 kb. These mRNAs were found in a
"polyA+" fraction, which is rather peculiar, since the plasmid is known to be cyto-
plasmic. 137.142 The absence of ORFI mRNA from this fraction, as well as the presence
of contaminating rRNA makes the identification of these mRNAs as poJyA + proble-
matical. It would be interesting to know which RNA polymerase is responsible for
transcription of the Kluyveromyces lactis plasmid DNAs. One possibility is that these
linear DNAs are present in mitochondria,142 whose mRNAs are polyadenylated by a
mitochondrial enzyme.
The internal deletions of pGKIl lacking toxin are missing ORF2, so that this is
probably a structural gene for at least one of the toxin subunits. It is unclear whether
the remaining subunit is encoded by pGKll or pGK12. Both ORF2 and ORF4 have
putative signal peptide sequences, and ORF2 contains seven potential sites for glyco-
sylation situated exclusively in the C-terminal third of the polypeptide. The two toxin
subunits might both be present in a preprotoxin processed from the ORF2 polypeptide
in a manner similar to that for processing of the S. cerevisiae preprotoxin. The in vitro
construction of deletions of pGKll should help resolve these remaining questions.
There are many strains of yeast that synthesize killer toxins but which appear to have
no endogenous dsRNA viruses 2 and no DNA plasmids. 4 Clearly, there are at least two
different types of symbionts that can confer the ability to secrete killer proteins. There
may be others as well.
 103

IV. CLONING VECTORS

Knowledge of the biology of killer toxin synthesis may have practical consequences.
One of the successful strategies for the production of large quantities of eukaryotic
proteins is the construction of cloning vectors in which foreign genes can be expressed
as secreted proteins. The organism of choice for such experiments with eukaryotic
genes is Saccharomyces cerevisiae. Secretory proteins in yeast whose genes have been
cloned include alkaline phosphatase, invertase, alpha factor, and killer toxin. Of these,
only killer toxin and alpha factor are unglycosylated, although their precursors are
glycosylated. Since most of the eukaryotic proteins one might want to produce in quan-
tity are not glycosylated, it may be advantageous to use a cloning vector in which the
secreted product would not be glycosylated. This could be accomplished with a vector
based on the killer toxin cDNA gene, if the eukaryotic gene (probably in the form of a
cDNA) were fused in frame to the toxin gene in such a way as to insure its cleavage
from the preprotoxin-fusion protein precursor during the process of secretion, like the
maturation and secretion of the toxin alpha and beta peptides. Studies now in progress
to understand the maturation of toxin may lead to construction of such vectors from
existing M, cDNA clones.
The K. lactis linear dsDNA plasm ids have quite interesting possibilities as cloning
vectors as well. They should be useful for cloning genes in K. lactis itself, since the
vectors designed for Saccharomyces cerevisiae generally do not replicate in other fungi.
Since these linear plasmids replicate in S. cerevisiae they should also be useful for
cloning genes in that yeast as well, if it becomes useful to have cloned genes on a linear
rather than a circular DNA plasmid.
In addition, the K. lactis pGKll plasmid has the extremely attractive feature that it
encodes a secreted protein. After the elucidation of the structure of the gene(s) respon-
sible, it should be possible to construct cloning vectors in which DNA sequences can
be inserted in frame into this gene(s) in such a way that useful heterologous proteins
can be synthesized and secreted, either from S. cerevisiae or from K. lactis. This may
be less arduous than the construction of useful cDNA cloning vectors from ScV-M,.
The killer systems of K. lactis and S. cerevisiae are thus interesting both for their
biology and for their possible utility in genetic engineering.

ACKNOWLEDGMENTS

I thank D. Doyle and D. Tipper for useful discussions, all my colleagues for com-
munication of data prior to publication, and the National Institutes of Health (grant
GM22200) for support.

REFERENCES

I. Bevan, E. A. and Makower, M., The physiological basis of the killer character in yeast, in Genetics
Today, Vol. 1, Geerts, U. S., Ed., Macmillan, The Hague, 1963,202.
2. Young, T. W. and Yagiu, M., A comparison of the killer character in different yeasts and its classi-
fication, Antonie van Leeuwenhoek 1. Microbiol. Serol., 44,59, 1978.
3. Groves, D. P., Clarte, J. J., and Oliver, S. G., Isolation and characterisation of a double-stranded
RNA virus-like particle from the yeast Yarrowia lipolytica, Curro Genet., 7, 185, 1983.
4. Gunge, N., Tamaru, A., Ozawa, F., and Sakaguchi, K., Isolation and characterization of linear
deoxyribonucleic acid plasmids from Kluyveromyces lactis and the plasmid-associated killer charac-
ter, 1. Bact., 145, 382, 1981.
5. Wood, H. A., Viruses with double-stranded RNA genomes, 1. Gen. Virol., 20,61, 1973.
104 Fungal Virology

6. Taber, R., Alexander, V., and Whitford, W., Persistent reoVlrus infection of CHO cells resulting in
virus resistance, J. Virol., 17, 513, 1976.
7. Grill, L. K. and Garger, 1., Identification and characterization of double-stranded RNA associated
with cytoplasmic male sterility in Vicia faba, Proc. Natl. Acad. Sci. U.S.A., 78,7043, 1981.
8. Lisa, V., Luisoni, E., and Milne, R. G., A possible virus cryptic in carnation, Ann. Appl. BioI., 98,
431, 1981.
9. Lisa, V., Boccardo, G., and Milne, R. G., Double-stranded RNA from carnation cryptic virus, Vi-
rology, 115, 410, 1981.
10. Scott, M. P., Fostel, J. M., and Pardue, M. L., A new type of virus from cultured Drosophila cells:
characterization and use in studies of heat-shock response, Cell, 22, 929, 1980.
11. Saksena, K. N. and Lemke, P. A., Viruses in fungi, in Comprehensive Virology, Vol. 12, Plenum
Press, New York, 1978, 103.
12. Lemke, P. A. and Nash, C. H., Fungal viruses, Bacteriological Reviews, 38,29,1974.
13. Lemke, P. A., Viruses and Plasmids in Fungi, Marcel Dekker, New York, 1979,653 pp.
14. Bruenn, J., Virus-like particles of yeast, Annu. Rev. Microbiol., 34,49, 1980.
IS. Wickner, R., "Killer character" of Saccharomyces cerevisiae: curing by growth at elevated tempera-
ture, J. Bact., 117,1356,1974.
16. Fink, J. R. and Styles, C. A., Curing of killer factor in Saccharomyces cerevisiae, Proc. Nat/. Acad.
Sci. U.S.A., 69,2846, 1972.
17. AI-Aidroos, K., Somers, J., and Bussey, H., Retention of cytoplasmic killer determinants in yeast
cells after removal of mitochondrial RNA by ethidium bromide, Molec. Gen. Genet., 122, 323, 1973.
18. Buck, K., Lhoas, P., and Street, B., Virus particles in yeast, Biochem. Soc. Trans., 1,1141,1973.
19. Herring, A. J. and Bevan, E. A., Virus-like particles associated with the double-stranded RNA species
found in killer and sensitive strains of the yeast Saccharomyces cerevisiae, J. Gen. Virol., 22, 387,
1974.
20. Adler, J., Wood, H. A., and Bozarth, R. F., Virus-like particles from killer, neutral, and sensitive
strains of S. cerevisiae, J. Virol., 17, 472,1976.
21. Oliver, S. E., McCready, S. J., Holm, C., Sutherland, P. A., McLaughlin, C., and Cox, B. S.,
Biochemical and physiological studies of the yeast VLP, J. Bact., 130, 1303, 1977.
22. Hopper, J. E., Bostian, K. A., Rowe, L. B., and Tipper, D. 1., Translation of the L-species dsRNA
genome of the killer-associated virus-like particles of Saccharomyces cerevisiae, J. BioI. Chern., 252,
9010, 1977.
23. Kane, W. P., Pietras, D., and Bruenn, J., Evolution of defective interfering double-stranded RNAs
in yeast, J. Virol., 32, 692, 1979.
24. Wickner, R. B., Plasmids controlling exclusion of the k, killer double-stranded RNA plasmid of
yeast, CelI, 21, 217, 1980.
25. Brennan, V. E., Field, L., Cizdziel, P. and Bruenn, 1. A., Sequences at the 3' ends of yeast viral
dsRNAs: proposed transcriptase and replicase initiation sites, Nucleic Acids Res., 9,4007, 1981.
26. Berry, E. and Bevan, A., A new species of double-stranded RNA from yeast, Nature New BioI., 239,
279, 1972.
27. Bevan, E. A., Herring, A. J. and Mitchell, D. J., Preliminary characterization of two species of
dsRNA in yeast and their relationship to the "killer" character, Nature (London), 245,81, 1973.
28. Brennan, V. E., Bobek, L. A., and Bruenn, J. A., Yeast dsRNA viral transcriptase pause products:
identification of the transcript strand, Nucleic Acids Res., 9,5049, 1981.
29. Bruenn, J. and Keitz, B., The 5' ends of yeast killer factor RNAs are pppGp, Nucleic Acids Res., 3,
2427, 1976.
30. Bruenn, J. and Kane, W., Relatedness of the double-stranded RNAs present in yeast virus-like par-
ticles, J. Virol.,26, 762, 1978.
31. Bruenn, J. and Brennan, V., Yeast viral double-stranded RNAs have heterogeneous 3' termini, Cell,
19,923, 1980.
32. Fried, H. M. and Fink, G. R., Electron microscopic heteroduplex analysis of "killer" double-
stranded RNA species from yeast, Proc. Natl. Acad. Sci. U.S.A., 75,4224, 1978.
33. Holm, C. A., Oliver, S. G., Newman, A. M., Holland, L. E., McLaughlin, C. S., Wagner, E. K.,
and Warner, R. C., The molecular weight of yeast PI double-stranded RNA, J. BioI. Chern., 253,
8332, 1978.
34. Thiele, D. J., Wang, R. W., and Leibowitz, M. J., Separation and sequence of the 3' termini of M
double-stranded RNA from killer yeast, Nucleic Acids Res" 10, 1661, 1982.
35. Vodkin, M. H. and Fink, G. R., A nucleic acid associated with a killer strain of yeast, Proc. Nat/.
Acad. Sci. U.S.A., 70, 1069, 1973.
36. Thiele, D., Hannig, E., and Leibowitz, M., Multiple L double-stranded RNA species of Saccharo-
myces cerevisiae: evidence for separate encapsidation, Molec. Cell. BioI., 4,92, 1984.
37. Reilly, J. D. and Bruenn, J., unpublished results, 1983.
 105

38. Field, L. J., Bobek, L., Brennan, V., Reilly, J. D., and Bruenn, J., There are at least two yeast viral
dsRNAs of the same size: an explanation for viral exclusion, Cell, 31,193,1982.
39. Sommer, S. S. and Wickner, R. B., Yeast L dsRNA consists of at least three distinct RNAs; evidence
that non-Mendelian genes [HOK], [NEXJ, and lEXL] are on one of these dsRNAs, Cell, 31,429,
1982.
40. Thiele, D. J. and Leibowitz, M. J., Structural and functional analysis of separated strands of killer
double-stranded RNA of yeast, Nucleic Acids Res., 10, 6903, 1982.
41. Bobek, L. A., Bruenn, J. A., Field, L. J., and Gross, K. W., Cloning of cDNA to a yeast viral
double-stranded RNA and comparison of three viral RNAs, Gene, 19, 225, 1982.
42. Bobek, L. A., Bruenn, J. A., Field, L. J., Reilly, J. D., and Gross, K. W., Cloning of cDNAs to a
yeast viral dsRNA, in Double-Stranded RNA Viruses, Compans, D. W. and Bishop, D. H. L., Eds.,
Elsevier, New York, 1983,451.
43. Sommer, S. S. and Wickner, R. B., Co-curing of plasmids affecting killer double-stranded RNAs of
Saccharomyces cerevisiae: [HOKj, [NEXJ, and the abundance of L are related and further evidence
that M, requires L, f. Bact., 150, 545, 1982.
44. EI-Sherbeini, M., Bevan, E. A., and Mitchell, D. J., Two biochemically and genetically different
forms of L dsRNA of Saccharomyces cerevisiae exist: one form, L z , is correlated with the [HOK]
plasmid, Current Genet., 7,63,1983.
45. Wickner, R. B., Killer systems in Saccharomyces cerevisiae: three distinct modes of exclusion of M,
double-stranded RNA by three species of double-stranded RNA, M .. L-A-E and L-A-HN, Molec.
Cell. BioI., 3, 654, 1983.
46. Bruenn, J., Madura, K., Siegel, A., Miner, Z., and Lee, M., Long internal inverted repeat in a yeast
viral dsRNA, Nucleic Acids Res., 13, 1575, 1985.
47. Welsh, J. D. and Leibowitz, M. J., Localization of genes for the double-stranded RNA killer virus
of yeast, Proc. Nat!. Acad. Sci. U.S.A., 79, 786, 1982.
48. Bostian, K. A., Elliott, Q., Bussey, H., Burn, V., Smith, A., and Tipper, D. J., Sequence of the
preprotoxin dsRNA gene of type I killer yeast: multiple processing events produce a two-component
toxin, Cell, 36,741,1984.
49. Skipper, N., Synthesis of a double-stranded cDNA transcript of the killer toxin-coding region of the
yeast MI double-stranded RNA, Biochem. Biophys. Res. Comm., 114, 518,1983.
50. Skipper, N., Thomas, D. Y., and Lau, P., Cloning and sequencing of the preprotoxin-coding region
of the yeast MI double-stranded RNA, EMBO f., 3, 107, 1984.
51. Brizzard, B. L. and DeKloet, S. R., Reverse transcription of yeast double-stranded RNA and ribo-
somal RNA using synthetic oligonucleotide primers, Biochim. Biophys. Acta, 739, 122, 1983.
52. Thomas, D., personal communication, 1984.
53. Lee, M., Nemeroff, M., Corstanje, B., Field, L., and Bruenn, J., Conserved regions of yeast-defec-
tive interfering viral double-stranded RNAs, f. Virol., in press.
54. Leibowitz, M., personal communication, 1984.
55. Koper-Zwarthoff, E. C., Th. Brederode, F., Walstra, P., and Bol, J. F., Nucleotide sequence of the
3'-noncoding region of alfalfa mosaic virus RNA 4 and its homology with the genomic RNAs, Nucleic
Acids Res., 7, 1887, 1979.
56. Wesolowski, M. and Wickner, R., Two new double-stranded RNA molecules showing non-Mendelian
inheritance and heat inducibility in Saccharomyces cerevisiae, Molec, Cell. BioI., 4, 181, 1984.
57. Herring, A. J. and Bevan, E. A., Double-stranded RNA containing particles from the yeast Saccha-
romyces cerevisiae and their relationship to the "killer" character, in Molecular Biology of Nucleo-
cytoplasmic Relationships, Vol. 6, Purseox-Dao, S., Ed., Elsevier, Amsterdam, 1975, 149.
58. Bostian, K. A., Hopper, J. E., Rogers, D. T. and Tipper, D, J., Translational analysis of the killer-
associated virus-like particle dsRNA genome of S. cerevisiae: M dsRNA encodes toxin, Cell, 19, 403,
1980.
59. Reilly, J. D., Bruenn, J. and Held, W., The capsid polypeptides of the yeast viruses, Biochem.
Biophys. Res. Comm., 121, 619,1984.
60. Bostian, K. A., Sturgeon, J. A. and Tipper, D. J., Encapsidation of yeast killer double-stranded
ribonucleic acids: dependence of M on L, f. Bact., 143, 463, 1980.
61. Harris, M. 5., Virus-like particles and double-stranded RNA from killer and non-killer strains of
Saccharomyces cerevisiae, Microbios" 21, 161, 1978.
62. Bostian, K., personal communication, 1984.
63, Bruenn, J" Bobek, L., Brennan, V., and Held, W., Yeast viral RNA polymerase is a transcriptase,
Nucleic Acids Res., 8, 2985, 1980.
64. Koltin, y" Mayer, L, and Steinlauf, R., Killer phenomenon in Ustilago maydis: mapping viral func-
tions, Molec. Gen. Genetics, 166, 181, 1978.
65. Philliskirk, G., and Young, T. W., The occurrence of killer character in yeasts of various genera,
Antonie van Leeuwenhoek f. Microbiol. Serol" 41, 147, 1975.
106 Fungal Virology

66. Rogers, D. and Bevan, E. A., Group classification of killer yeasts based on cross-reactions between
strains of different species and origin, J. Gen. Micro., lOS, 199, 1978.
67. Kagan, B., Mode of action of yeast killer toxins: channel formation III lipid bilayer membranes,
Nature (London), 302, 709, 1983.
68. Bussey, H., Effects of yeast killer factor on sensitive cells, Nature New Bioi., 235, 73, 1972.
69. Bussey, H., Physiology of killer factor in yeast, Adv. Microbial. PhySlOl., 22, 93, 1981.
70. Palfree, R. and Bussey, H., Yeast killer toxin: purification and characterization of the protein toxin
from S. cerevisiae, Eur. J. Biochem., 93,487, 1979.
71. Bussey, H., Saville, D., Hutchins, K., and Palfree, R. G. E., Binding of yeast killer toxin to a cell
wall receptor on sensitive Saccharomyces cerevisiae, J. Bact., 140, 888, 1979.
72. Bussey, H. and Sherman, D., Yeast killer factor: ATP leakage and coordinate inhibition of macro-
molecular synthesis in sensitive cells, Biochem. Biophys. Acta, 298,868, 1973.
73. AI-Aidroos, K. and Bussey, H., Chromosomal mutants of Saccharomyces cerevisiaeaffecting the cell
wall binding site for killer factor, Can. J. Micro., 24,228, 1978.
74. Hutchins, K. and Bussey, H., Cell wall receptor for yeast killer toxin: involvement of (1-6)-beta-D-
glucan, J. Bact., 154,161,1983.
75. De la Pena, P., Barros, F., Gascon, S., Lazo, P. S., and Ramos, S., Effect of yeast killer toxin on
sensitive cells of Saccharomyces cerevisiae, J. BioI. Chern., 256, 10420, 1981.
76. LoUe, S., Skipper, N., Bussey, H., and Thomas, D., The expression of eDNA clones of yeast M.
double-stranded RNA in yeast confers both killer and immumty phenotypes, EMBO J., 3,1383,1984.
77. Mitchell, D., Herring, A., and Bevan, E., The genetic control of dsRNA virus-like particles associated
with Saccharomyces cerevisiaekiller yeast, Heredity, 37,129,1976.
78. Tzen, 1. C., Somers, J. M., and Mitchell, D. J., A dsRNA analysis of suppressive sensitive mutants
of "killer" Saccharomyces cerevisiae, Heredity, 33, 132, 1974.
79. Vodkin, M. H., Katterman, F., and Fink, G. R., Yeast killer mutants with altered dsRNA, J. Bact.,
117,681,1974.
80. Bostian, K. A., Jayachandran, S., and Tipper, D. J., A glycosylated protoxin in killer yeast: models
for its structure and maturation, CelI,32, 169, 1983.
81. Bussey, H., Saville, D., Greene, D., Tipper, D. J., and Bostian, K. A., Secretion of Saccharomyces
cerevisiaekiller toxin: processing of the glycosylated precursor, Malec. CelIo BioI., 3,1362,1983.
82. Bussey, H., Sacks, W., Galley, D., and Saville, D., Yeast killer plasmid mutations affecting toxin
secretion and activity and toxin immunity function, Malec. Cell. BioI., 2,346, 1982.
83. Tipper, D., personal communication, 1984.
84. Julius, D., Brake, A., Blaine, L., Kunisawa, R., and Thorner, J., Isolation of the putative structural
gene for the lysine-arginine cleavage endopeptidase required for processing of yeast prepro-alpha-
toxin, Cell, 37, 1075, 1984.
85. Tipper, D. J. and Bostian, K. A., Double-stranded ribonucleic acid killer systems in yeasts, Micro-
bial. Rev., 48, 125, 1984.
86. Alaoui, P., J acquemin, J., and Boccardo, G., Ribonucleic acid polymerase activity associated with
virus-like particles isolated from Penicilium stoloniterumATCC number 14586, Acta Virol., 18,193,
1974.
87. Chater, K. and Morgan, D., Ribonucleic acid synthesis by isolated viruses of Penicillium stoloni-
terum, J. Gen. Virol., 23,307,1974.
88. Buck, K., Replication of double-stranded RNA in particles of Penicillium stoloniterum virus S, Nu-
cleic Acids Res., 2, 1889, 1975.
89. Nash, C., Douthart, R., Ellis, L., VanFranke, R., Burnett, 1., and Lemke, P., On the mycophage of
Penicillium chrysogenum, Can. J. Microbial., 19, 97, 1973.
90. Ratti, G. and Buck, K. W., Semi-conservative transcription in particles of a double-stranded RNA
mycovirus, Nucleic Acids Res., 5,3843,1978.
91. Herring, A. J. and Bevan, E. A., Yeast virus-like particles possess a capsid-associated single-stranded
RNA polymerase, Nature (London), 268,464, 1977.
92. Welsh, J. D., Leibowitz, M. J., and Wickner, R. B., Virion DNA-independent RNA polymerase
from Saccharomyces cerevisiae, Nucleic Acids Res., 8, 2349, 1980.
93. Welsh, J. D. and Leibowitz, M. J., Transcription of killer virion double-stranded RNA in vitro,
Nucleic Acids Res., 8,2365, 1980.
94. Hastie, N., Brennan, V., and Bruenn, J., No homology between double-stranded RNA and nuclear
DNA of yeast, J. Virol., 28, 1002, 1978.
95. Hannig, E., Thiele, D., and Leibowitz, M., Saccharomyces cerevisiae killer virus transcripts contain
template-coded polyadenylate tracts, Malec. Cell. BioI., 4,92, 1984.
96. Brennan, V., Bobek, L., and Bruenn, J., unpublished data, 1982.
97. Yamakawa, M., Furuichi, Y., Nakashima, K., LaFiandra, A. J., and Shatkin, A. J., Excess synthesis
of viral mRNA 5'-terminal oligonucleotides by reovirus transcriptase, J. BioI. Chern., 256, 6507,
1981.
 107

98. Joklik, W., Reproduction of reoviridae, in Comprehensive Virology, Vol. 2, Plenum Press, New
York, 1974, 23!.
99. Silverstein, S., Christman, 1. K., Acs, G., The reovirus replicative cycle, Annu. Rev. BIOchem., 45,
375,1976.
100. Van Etten, 1. L., Burbank, D. E., Cuppels, D. A., Lane, L. C., and Vidaver, A. K., Semiconservative
synthesis of single-stranded RNA by bacteriophage phi6 RNA polymerase, 1. Virol., 33,769, 1980.
100a. Nemeroff, M. and Bruenn, J., Conservative replication and transcriptIOn of yeast viral dsRNAs in
vitro, J. Virol., 57, 754, 1986.
101. Newman, A. M., Elliott, S. G., McLaughlin, C. S., Sutherland, P. A., Warner, R. C., Replication
of double-stranded RNA of the virus-like particles in Saccharomyces cerevisiae, 1. Virol., 38, 263,
198!.
IOZ. Sclafani, R. A. and Fangman, W. L., Conservative replication of yeast double-stranded RNA by
displacement of progeny single strands, Molec. Cell. Biol., 4, 1618, 1984.
103. Bostian, K. A., Burn, V. E., Jayachandran, S., and Tipper, D. J., Yeast killer dsRNA plasmids are
transcnbed in vivo to produce full and partial-length plus-stranded RNAs: models for their synthesis
and for the functional sequence of M,-dsRNA, Nucleic Acids Res., 11,1077,1983.
104. Haylock, R. W. and Bevan, E. A., Characterization of the L dsRNA encoded mRNA of yeast,
Current Genet., 4, 181, 198!.
105. Wickner, R. B., and Toh-e, A., rHOK1, a new yeast non-Mendelian trait, enables a replication-
defective killer plasmid to be maintained, Genetics, 100, 159, 1982.
106. Somers, J. M., Isolation of suppressive sensitive mutants from killer and neutral strains of Saccha-
romyces cerevisiae, Genetics, 74,571, 1973.
107. Huang, A. S., Viral pathogenesis and molecular biology, Bact. Rev., 41, 811, 1977.
108. Sweeny, T. K., Tate, A., and Fink, G., A study of the transmission and structure of dsRNAs asm-
dated with the killer phenomenon in Saccharomyces cerevisiae, Genetics, 84, 27, 1976.
109. Somers, J. M. and Bevan, E. A., The inheritance of the killer character in yeast, Genet. Res., 13, 71,
1969.
110. Ridley, S. P. and Wickner, R. B., Defective interference in the killer system of Saccharomyces cere-
visiae, J. Virol., 45, 800, 1983.
Ill. Guild, G. and Stollar, V., Defective interfering particles of Sindbis virus. V. Sequence relationships
between SV SPD 42S RNA and intracellular defective viral RNAs, Virology, 77, 175, 1977.
112. Holland, 1., Grabau, E., Jones, C., and Semler, B., Evolution of mUltiple genomic mutations during
long-term persistent infection by vesicular stomatitis virus, Cell, 16,495, 1979.
113. Kennedy, S. 1. T., Sequence relationships between the genome and the intracellular RNA species of
standard and defective interfering Semliki forest virus, 1. Molec. Bioi., 108, 491, 1976.
114. Nomoto, A., Jacobson, A., Lee, Y., Dunn, J., and Wimmer, E., Defective interfering particles of
polio virus. Mapping of the deletion and evidence that the deletions in the genomes of DI (1), (II),
and (III) are located in the same region, 1. Molec. Bio!., 128,179,1979.
115. O'Hara, P., Nichol, S., Horodyski, F., and Holland, J., Vesicular stomatitis virus defective interfer-
ing particles can contain extensive genomic sequence rearrangements and base substitutions, Cell,36,
915, 1984.
116. Zakian, V. A., Wagner, D. W., and Fangman, W. L., Yeast L double-stranded RNA is synthesized
during the G I phase but not the S phase of the cell cycle, Molec. Cel!. Bioi., I, 673, 198!.
117. Ball, S., Tirtiaux, C., and Wickner, R., Genetic control of L-A and L-(BC) dsRNA copy number in
killer systems of Saccharomyces cerevisiae, Genetics, 107, 199, 1984.
118. Wickner, R. B., Killer of S. cerevisiae: a dsRNA plasmid, Bact. Rev., 40,757,1976.
119. Wickner, R. B., Chromosomal and non-chromosomal mutations affecting the "killer character" of
S. cerevisiae, Genetics, 76, 423, 1974.
120. Wickner, R. B., Twenty-six chromosomal genes needed to maintain the killer double-stranded RNA
plasmid of S. cerevisiae, Genetics, 88, 419, 1978.
12!. Wickner, R. B., mak mutants of yeast: mapping and characterization, 1. Bact., 140,154, 1979.
122. Wickner, R. B., Killer systems in Saccharomyces cerevisiae, in The Molecular Biology of the Yeast
Saccharomyces, Life Cycle and Inheritance, Strathern, J., Jones, E., and Broach, J., Eds., Cold
Spring Harbor Laboratory, Cold Spring Harbor, New York 1981, 415.
123. Wickner, R. B., Ridley, S. P., Fried, H. M., and Ball, S. G., Ribosomal protein L3 is involved in
replication or maintenance of the killer double-stranded RNA genome of Saccharomyces cerevisiae,
Proc. Nat!. Acad. Sci. U.S.A., 79.4706, 1982.
124. Cohn, M. S., Taber, C. W., Taber, H., and Wickner, R. B., Spermidine or spermine requirement for
killer dsRNA plasmid replication in yeast, 1. Biol. Chem., 253,5225, 1978.
125. Thrash, C., Voelkel, K. A., DiNardo, S., Sternglanz, R., Identification of Saccharomyces cerevisiae
DNA topoisomerase I mutants, J. Bioi. Chem.,259, 1375, 1984.
108 Fungal Virology

126. Wickner, R. B. and Leibowitz, M. J., Dominant chromosomal mutant bypassing chromosomal genes
needed for killer RNA plasmid replication in yeast, Genetics, 87, 453,1977.
127. Toh-e, A., Guerry, P. and Wickner, R. B., Chromosomal superkiller mutants of S. cerevisiae, 1.
Bact., 136, 1002, 1978.
128. Toh-e, A. and Wickner, R. B., A mutant killer plasmid whose replication is dependent upon a chro-
mosomal "superkiller" mutation, Genetics, 91,673,1979.
129. Toh-e, A. and Wickner, R. B., "Superkiller" mutations suppress chromosomal mutations affecting
dsRNA killer plasmid replIcation in S. cerevisiae, Proc. Natl. Acad. Sci. U.S.A., 77,527, 1980.
130. Wickner, R. B. and Leibowitz, M. J., Two chromosomal genes required for killIng expression in
killer strains of Saccharomyces cerevisiae, Genetics, 82, 429, 1976.
131. Wesolowski, M., Algeri, A., Goffrini, P., and Fukuhara, H., Killer DNA plasmids of the yeast
Kluyveromyces lactis. I. Mutations affecting the killer phenotype, Curro Genet., 5,191,1982.
132. Gunge, N. and Sakaguchi, K., Intergeneric transfer of deoxyribonucleic acid killer plasmids, pGK11
and pGKl2, from Kluyveromyces lactis into Saccharomyces cerevisiae by cell fusion, 1. Bact., 147,
ISS, 1981.
133. Gunge, N., Murata, K., and Sakaguchi, K., Transformation of Saccharomyces cerevisiaewith linear
DNA killer plasmids from Kluyveromyces lactis, 1. Bact., lSI, 462, 1982.
134. De Louvencourt, L., Fukuhara, H., Heslot, H., and Weslowski, M., Transformation of Kluyvero-
myces lactisby killer plasmid DNA, 1. Bact., 154, 737,1983.
135. Niwa, 0., Sakaguchi, K., and Gunge, N., Curing of the killer deOXYrIbonucleic acid plasm ids of
Kluyveromyces lactis, 1. Bact., 148, 988, 1981.
136. Wesolowski, M., Dumazert, P., and Fukuhara, H., Killer DNA plasmids of the yeat Kluyveromyces
lactis. II. Restriction endonuclease maps, Curro Genet., 5, 199, 1982.
137. Sor, F., Weslowski, M., and Fukuhara, H., Inverted terminal repetitions of the two linear DNA
associated with the killer character of the yeast Kluyveromyces lactis, Nucleic Acids Res., II, 5037,
1983.
138. Livingston, D., Inheritance of 2 micron DNA plasmid from Saccharomyces, Genetics, 86, 73, 1977.
139. Dani, G. and Zakian, V., Mitotic and meiotic stability of linear plasmids in yeast, Proc. Natl. Acad.
Sci. U.S.A., 80,3406, 1983.
140. Sugisaki, Y., Gunge, N., Sakaguchi, K., Yamasaki, M., and Tamura, G., Kluyveromyces lactiskiller
toxin inhibits adenylate cyclase of sensitive yeast cells, Nature (London), 304,464, 1983.
141. Sugisaki, Y., Gunge, N., Sakaguchi, K., Yamasaki, M., and Tamura, G., Characterization of a novel
toxin encoded by a double-stranded linear DNA plasmid of Kluyveromyces lactis, Eur. 1. Biochem.,
141,241,1984.
142. Stark, M. J. R., Mileham, A. J., Romanos, M. A., and Boyd, A., Nucleotide sequence and transcrip-
tion analysis of a linear DNA plasmid associated with the killer character of the yeast Kluyveromyces
lactis, Nucleic Acids Res., 12, 6011, 1984.
 109

Chapter 3

THE KILLER SYSTEMS OF USTILAGO MA YDIS

Y. Koltin

TABLE OF CONTENTS

I. Phenotypic Expression of Viruses in Ustilago maydis .. .......................... 110

A. Genetics of Interstrain Inhibition ............................................. 110
B. Virus Particles and Interstrain Inhibition ................................... 114
C. Incompatibility .................................................................... 115

II. Characterization of the Viruses of Ustilago maydis ............................... 116

A. Viral Nucleic Acids ............................................................... 116
B. Encapsidation of the Viral dsRNA ........................................... 117
C. Viral Functions and the Segments of dsRNA .............................. 120
D. Relatedness of the Segments of dsRNA ..................................... 126
E. RNA Polymerase Activity ...................................................... 130

III. Killer Proteins .............................................................................. 132

A. Nature of the Killer Toxin ...................................................... 132
B. Specificity of the Toxins ........................................................ 133
C. In Vivo and In Vitro Mode of Action ........................................ 134

IV. Concluding Remarks ...................................................................... 136

References ............................................................................................ 138

110 Fungal Virology

1. PHENOTYPIC EXPRESSION OF VIRUSES IN USTILAGO MA YDIS

A. Genetics of Interstrain Inhibition

The detection of the viruses in Ustilago stemmed from the effort to develop a plate
assay for the identification of the mating alleles that control sexual interaction between
haploid sporidial cultures. 1 In general, the confrontations led to direct contact between
the tested strains with no apparent change in morphology if the mates were incompat-
ible, or the confrontation led to the development of mycelial hyphae typical of the
heterokaryotic state of this organism if the mates were compatible. The formation of
heterokaryotic mycelium is a prerequisite to infection of the host tissue, namely, the
corn plant (Figure 1). A compatible interaction leading to the formation of the heter-
okaryon is known to be governed by multiple alleles at one locus. 2 6 During the course
of the development of the plate assay in some instances, strains known to be compati-
ble, based on their ability to infect jointly corn seedlings, displayed in the plate assay
interstrain inhibition (Figure 2). The strains displaying this effect were of independent
origin and occurred at a low frequency. The inhibition was shown to result from a
secreted substance> thought to be of a proteinaceous nature based on its sensitivity to
proteolytic enzymes and its resistance to nucleases.
Initial indications on the cytoplasmic nature of the inhibitory effect were obtained
by genetic studies using formal crosses and the classical heterokaryon transfer experi-
ments' for the detection of cytoplasmically inherited factors. A cross between the in-
hibitor-producing strain and a sensitive strain was performed. Sexual crosses in Usti-
Jago can be performed only on plants. The inhibitory effect is not effective in plants
and the formation of heterokaryons in seedlings can be accomplished even between a
sensitive strain and one that elicits the inhibitory effect on synthetic medium in plates.
The results of such crosses indicated that the inhibitory effect is determined by a non-
chromosomal factor since no segregation was detected within individual tetrads, each
reflecting the meiotic products of one zygote. In the same tetrads normal segregation
was recorded for other chromosomal markers. In addition, most tetrads yielded prog-
eny that secreted the inhibitory substance, thus further suggesting that the substance is
determined by a nonchromosomal factor. The same crosses suggested the involvement
of another cytoplasmic state in addition to the cytoplasmic state expressed by the se-
cretion of the inhibitory substance. Some tetrads yielded progeny that were uniformly
resistant to the inhibitory substance, yet the same progeny did not express this sub-
stance. In crosses between these progeny and sensitive strains, tetrads displaying uni-
form resistance within the tetrads were obtained, and these results suggested that the
resistance is also determined by a cytoplasmic factor. Thus, 3 cytoplasmic states were
distinguished and designated PI, P2, and P3. PI was assumed to express the cyto-
plasmic factor for interstrain inhibition and a cytoplasmic factor for resistance to the
inhibitory substance. P2 was assumed to be devoid of the cytoplasmic elements for
both cytoplasmic functions, and P3 carries the cytoplasmic element for resistance to
the inhibitory substance but is devoid of the cytoplasmic element that is involved in the
production and expression of inhibitor.
The resistance of strains that did not express an inhibitory factor was not always
transmitted as a nonchromosomal factor. Among the tetrads, some displayed a segre-
gation of the resistance such as would be anticipated from the segregation of a single
gene. Survey of the native strains of Ustilago from the U.S. and from other sources in
the world indicated that among these strains some are sensitive to the inhibitor secreted
by a PI strain and others are insensitive. Crosses between some of these strains yielded
results suggesting the segregation of a single gene. Thus, in addition to the cytoplasmic
resistance (referred to as immunity),. resistance to the inhibitory factor was shown to
be incurred also by a nuclear gene. It is interesting to note that the native PI strain was
 111

FIGURE I . Neoplastic growth on a corn leaf (right) induced by U. maydl5.

Uninfected leaf (left) . The neoplastic tissue contains the sexual stage of U.
maydis.

genotypically carrying the chromosomal gene for resistance to PI and, as will be dis-
cussed below, all the strains expressing the inhibitory function detected thus far carry
a chromosomal gene for resistance. The linkage of the chromosomal gene resistance
and the cytoplasmically determined inhibition suggests an additional role to these genes
beyond the protection against the inhibitory factor determined by the cytoplasmic de-
terminant. Data to be presented below will suggest that these genes play some role in
the maintenance of the cytoplasmic elements or are essential for their initial interaction
with the recipient cell.
The series of crosses between the various cytoplasmic states PI, P2, and P3 did not
reveal any new unexpected interconversions that required a modification of the con-
cept. Crosses between P3 and P2 strains did not yield PI progeny. P2 strains were
found to be stable and never yielded PI or P3 derivatives. Further confirmation of the
cytoplasmic nature of PI and P3 was obtained by cytoduction experiments. Hetero-
karyons formed between compatible strains on solid medium developed dikaryotic hy-
phae. 10 The dikaryotic hypha is unstable in these conditions and dissociates to its mon-
okaryotic components that can be identified if the parental strains were genetically
distinguishable. The descendants of the mating interaction contain mixed cytoplasm.
Thus, cytoplasmic traits are expected to be expressed in association with chromosomal
markers of both parental types if the studied trait is determined by a cytoplasmic ele-
112 Fungal Virology

FIGURE 2. Inters train inhibition. Upper right plate, three strains that
cause the inhibition of growth spotted on a sensitive strain of U. maydis.
Mutual sensitivity of each of the toxm secretors is noticed when each strain
is used as a lawn (upper left and bottom right and left). The mutual sensitivity
among the toxin secretors defines the toxin specificity. The toxin secretors
are the PI, P4, and P6, and the sensitive state is referred to as P2. (From
Koltin, Y. and Day, P. R., Pmc. Natl. Acad. Sci. U.S.A., 73,594, 1976.
With permission.)

ment. The conversion of P2 by PI and by P3 is anticipated if P2 is devoid of the

cytoplasmic element found in PI and in P3. Tests performed by Day and Anagnos-
takis" showed the expected conversion as anticipated according to the three cyto-
plasmic states suggested by Puhalla. '
In the course of the detection of the inters train inhibition, four strains with the
inhibitory capacity were identified. One of these strains was characterized in detail'·6'
but already at that stage of the research it was clear that not all the strains were iden-
tical since mutual sensitivity to the inhibition was noticed. Only two of the four strains
were identical and both were of the PI type (Figure 2). The genetic analysis of two
additional phenotypes (called P4 and P6) was performed by Koltin and Day" The
inheritance of these phenotypes was similar to PI. Both showed cytoplasmic inherit-
ance of the trait and cytoplasmic transmission by the heterokaryon test. Furthermore,
chromosomal genes for resistance to P4 and P6 were detected, and in each case a single
chromosomal gene was shown to confer resistance towards the inhibitory effect of one
of the phenotypes P4 or P6 (Figure 3). Thus, to date 3 independent genes are known
to confer resistance. Each gene confers resistance to only one of the inhibitory sub-
stances. The genes are designated pJr, p4r, and p6r. The major difference in the ge-
netics of the P4 and P6 was noticed in the cytoplasmic transmission experiments. Un-
like the successful transmission of PI to a strain sensitive to PI (P2 pIs),,,,l2 all the
experiments involving cytoplasmic mixing by the formation of heterokaryons have
failed in the case of P4 and P6 if the recipients carried the respective pXs allele. Het-
erokaryons between the donor and pXs recipients were established and cytoplasmic
 113

FIGURE 3. Resistance to the toxms of P4 (left) and P6 (right) determined

by nuclear genes. The toxin secretors plated on a sensItive lawn (top), on a
lawn carrying a nuclear gene for resistance to P4 (rIght), and on a lawn of a
strain carrymg a nuclear gene for resistance to P6 (left).

mixing should have occurred. In spite of the sensitivity of the recipients, heterokaryons
of this type can be formed by a saturation procedure in which the recipients out num-
ber the donors. 12 The development of a hypha typical of the heterokaryon can be seen
and yet, once this heterokaryon is dissociated to its parental components, the inhibi-
tory function is found associated with the genotype of the donor and none of the cells
that were expected to act as recipients display the inhibitory function." The transmis-
sion is effective only if the recipient in the heterokaryon carries the pXr allele. These
conclusions are based on transmission studies in isogenic strains in which the pXrallele
was selected from strains with pXs. In conjunction with these findings, it appears that
the occurrence of the pXr alleles in the natural isolates of P4 and P6 is not a coinci-
dence and may reflect an essential function for the replication or maintenance of the
cytoplasmic factors involved in the inhibitory function.
Furthermore, the precondition for acceptance of the P4 and P6 states and the asso-
ciation of p4r and p6r with the native strains expressing the inhibitory function has
raised difficulties in confirmation of the presence of cytoplasmic immunity to the P4
and P6 substances similar to the one identified in PI. Simple transmission experiments
by heterokaryon transfer are uninformative since transfer will be effective only if both
interacting strains contain the chromosomal gene for resistance. The actual test for
cytoplasmic immunity to the P4 and P6 substances requires the demonstration of the
transmission of such immunity to a sensitive recipient. Two other approaches to verify
the existence of cytoplasmic immunity are (1) analysis of the segregation within tetrads
in a formal cross between a strain displaying resistance or immunity and a sensitive
strain carrying the pXs allele; and (2) by examination of the resistance of diploids
derived from the same cross. The expectations from (1) are similar to those found for
PI in which the tetrads should display uniformity in the resistance of the progeny
114 Fungal Virology

within each tetrad along with the segregation of the chromosomal genes. In an earlier
report,. it was shown that this type of segregation was detected, although in these
crosses the sample was small and problems of viability were noticed. In a more rigorous
and recent analysis, it was shown that only segregational tetrads are obtained by cross-
ing the suspected immune strains with sensitive strains carrying the pXs allele. 66 The
expected segregation of a cytoplasmic factor could not be demonstrated by this type of
analysis. These results do not imply the absence of cytoplasmic immunity but cannot
confirm its existence. The expectations for (2) as a means for the demonstration of
cytoplasmic transmission of an immunity factor could not be properly defined before
the relations between the pXr and pXs alleles were determined. By the isolation of
heterozygote diploids for each of the 3 genes plr, p4r, and p6r, testing their response
to each of the inhibitors determined by PI, P4, and P6, it became clear that the pXs
allele is the dominant allele in each case, since the diploids were P2 pXrlP2 pXs. As
with pls, p4s and p6s also are the dominant alleles. Therefore, if from an interaction
between resistant progeny, derived from anyone of the inhibitory types, and a P2 pXs
a resistant diploid is obtained, some cytoplasmic element must confer the resistance
that is referred to as immunity. Diploids of this type were obtained from P4 and from
P6 and, therefore, it is suggested that in fact a cytoplasmic immunity factor is associ-
ated with these cytoplasmic inhibitory traits but that their transmission is dependent
on the pXr allele. This allele may be pleiotropic in its effect, being recessive with re-
spect to the resistance and yet dominant in the determination of the acceptance, main-
tenance, or replication of the cytoplasmic information. The nature of the immunity
factor and the chromosomal resistance is unknown. The interrelations between the
immunity and resistance and the dominance relations of the chromosomal alleles for
resistance suggest that the latter determine a receptor for the inhibitory substance and
the former acts by inactivation of this inhibitory substance.

B. Virus Particles and the Interstrain Inhibition

The cytoplasmic transmission of the factors that determine interstrain inhibition and
the resistance to the inhibitory factors led to a search for the determinants of these
phenotypes. The search was initiated in strains that were reported as PI, P2, and P3,'4
and among those identified earlier" as recipients of the PI or the P3 determinants.
The profile from sucrose density gradients of material recovered from preparations of
PI strains after cell disintegration and the removal of cell debris indicated the presence
of five reproducible peaks with an absorption at 253.6 nm. Within these peaks, virus
particles 41 nm in diameter were detected (Figure 4). Similar results were obtained with
the P3 strains. However, in six different P2 isolates, particles were not detected and
only in the P2 strains that acted as recipients in the heterokaryon transfer experiment
were virus particles recovered. Thus, the initial findings conformed precisely with the
expectations based on the genetic analysis.
Examination of the P4 and P6 strains led to the detection of virus particles associ-
ated with the P4 and P6 phenotypes. 9 The particles were of the same size and charac-
teristics as the particles of PI. Later experiments 15 - 17 suggested that particles may be
found also in sensitive strains. Examination of sensitive progeny from crosses and of
mutant strains that have lost the inhibitory activity revealed the presence of virus par-
ticles in these strains. These findings were not in contrast with any of the genetic data,
but suggested that the mere presence of particles cannot be directly correlated with the
inhibitory function or the immunity; but perhaps it is the information contained in
these particles that is relevant in relating the interstrain inhibition to the virus particles.
Thus far, mostly the progeny from laboratory crosses in induced mutants were exam-
ined for the presence of viruses. It is unclear whether in nature sensitive strains con-
 115

FIGURE 4. Virus particles 41 nm in diameter recovered from strains o f U. may dis.

taining the viruses are of common occurrence. Of three native P2 strains examined in
repeated tests, one was found to contain virus particles and the others were devoid of
these particles, although both contained the nucleic acid typical to this group of vi-
ruses!

c. Incompatibility
The strains displaying the cytoplasmic factors involved in interstrain inhibition were
from different locations. Each strain displayed one of the inhibitory specificities. The
association of the cytoplasmic elements in single cells was attempted by the perform-
ance of crosses between parental strains displaying different phenotypes!·'··'9 The ini-
tial results suggested that, in a cross between PI and P4, most progeny are of the P4
phenotype and few are PI; in crosses between PI and P6 the inhibitory function of
both strains was lost, and in crosses between P4 and P6 of the two phenotypes intro-
duced by the parental strains only P4 was found in the progeny. In every case in which
the inhibitory effect is still displayed by most progeny , some are devoid of this func-
tion. The results obtained suggested incomplete exclusion of the PI phenotype in the
PI x P4 interaction, mutual exclusion of the inhibitory phenotype in the interaction
between PIx P6, and unilateral exclusion in interactions between P4 and P6 in which
the P6 phenotype is always lost.
As more testers became available and also following some changes in some strains
during their use in the laboratory, modifications of the interactions were noticed. By
use of the proper testers that can detect dual specificities, it became clear that single
cells can harbor the cytoplasmic determinants associated with the PI and the P4 phen-
otypes, although there may be a slight advantage to the P4 phenotype (Figure 5). The
association of PI and P4 in a stable form was generated in the laboratory in crosses,
and strains kept for over a year still displayed the dual inhibitory specificities (P IIP4).
In the interaction between PI and P6, some PI strains that led to the exclusion of P6
were detected, contrary to the mutual exclusion that was noticed earlier. This modifi-
cation suggested some drift in the cytoplasmic factors associated with the inhibition
116 Fungal Virology

FIGURE 5. Dual specificities of PI/P4 displayed by progeny from PI x P4 crosses (center

of plate). The dual specificity is shown by the sensitivity of the testers (plsp4r and plrp4s).
Top row in both plates, P I (left) and P4 (right) spotted on the lawns of the testers. (From
W,dgerson, M. and Koltin, Y., Curro Genet., 5, 127, 1982. With permission.)

factors and, as will be discussed later, the modification in the interaction may be re-
lated to a specific part of the nucleic acids associated with the inters train inhibition.
The incompatibility relations suggest that between PI and P4 there is some similarity,
and both PI and P4 differ from P6.
The exact timing of the exclusion interaction in relation to the life cycle is unknown.
Studies on the interaction between the various cytoplasmic elements in heterokaryons I9
suggested that already at that stage some competition between these elements occurs,
and the outcome is not random and follows specific rules (Figure 6) that are identical
to those seen in the progeny of crosses. P6 is always lost in interactions with both PI
and P4. Although various active mechanisms of exclusion were contemplated, 18 at the
present time it appears that simple intermolecular competition during replication can
account for the outcome of most interactions. I9

II. CHARACTERIZATION OF THE VIRUSES OF USTILAGO MA YDIS

A. Viral Nucleic Acids

The nucleic acids extracted from the partially purified particles of PI were shown to
be double-stranded RNA (dsRNA). The criteria used were its reaction with orcinol and
diphenylamine, the hypochromicity profile and the resistance of these nucleic acids to
ribonuclease at high ionic strength, and the sensitivity to the same enzyme at low ionic
strength. 14 The nucleic acids recovered from purified preparations of virus particles of
P4 and P6 strains as well as from some P2 and P3 strains all displayed the same char-
acteristics. Identical molecules were obtained from total nucleic acid extracts of cells
after treatment of the preparation with RNase and DNase according to the procedure
of Vodkin et al. 20 The use of the special cellulose column designed to separate the
replicative forms of RNA viruses" was later introduced, and is currently used as a
general procedure in the purification of the viral dsRNA either from purified particles
or from whole cells. The recovery from whole cells is feasible, due to the unique prop-
erties of these molecules and their exceptional resistance to nucleolytic digestion. All
the different procedures yielded similar components. The characterization of the
dsRNA extracted from the virions of PI already indicated a characteristic aspect of the
system, namely, that the viruses are multicomponent, and multiple segments can be
recovered. It was unclear whether these segments are associated in one particle or are
 117

108

if.
()
m
r-
r-

•
rn
--t
:I:

...J
...J
w
U
'".,.
"tI

-----
Z
0
'"m
"tI

"tI
:I:
10-1 Z
0
--t
-<
"tI
m
~

5
10
I - ---- --
3 6 8 No Generations

FIGURE 6. Exclusion of P6 by P4 in vegetative cells with dual specificities

isolated from a heterokaryon that was established between a P4 strain and a P6
strain. (From Widgerson, M. and Koltin, Y., Curro Genet., 5, 127, 1982. With
permission.)

separately encapsidated, but a range of segments were consistently recovered in re-

peated extractions (Table 1, Figure 7). Six segments of dsRNA were identified with an
approximate molecular weight of 2.87,2.52,0.93,0.49,0.44, and 0.06 x 10 6 daltons.
Likewise, P4 and P6 were found to be multisegmented with 7 segments in P4 and 5
segments in P6. Five of the segments of PI and P4 were of identical molecular weight,
three of the PI and P6 segments were of identical molecular weight, and three of the
P4 and P6 segments were of identical molecular weight. Thus, each phenotype could
be distinguished qualitatively by its typical segmentation pattern (Table 1 and Figure
7). The sizes of the segments have been reexamined recently in polyacrylamide gel
electrophoresis (PAGE) using known standards that have been characterized by more
than one physical means,22 by agarose gel electrophoresis, and by electron micros-
copy.23 The results indicate that the range of the segments is from 6.2-6.7 kilobase
pairs (kbp) to 360 bp (Table 1), corresponding to molecular weights of 4.3-4.6 x 10"
to 0.25 x 106 , and the total information of the dsRNA, assuming no redundancy among
the segments, as sufficient to encode five to 10 proteins.

B. Encapsidation of the Viral dsRNA

Multiple segments of dsRNA are not unique to the fungal viruses. However, unlike
most animal and plant dsRNA viruses, most fungal viruses seem to encapsidate indi-
vidual molecules in separate capsids. 22 The large number of density components ob-
served in CsCI density gradients in studies of the PI phenotype'4 and later in studies of
the P4 phenotype'4 suggested that the viruses of UstiJago also encapsidate t~ dsRNA
segments individually, and since all the coats are of the same size the difference in the
118 Fungal Virology

-- M

-
FIGURE 7. DsRNA segments in 5070 polyacrylamide
disc gel electrophoresis (PAGE); PI (left), P4 (middle),
and P6 (right). The size of the segments and the desig-
nation of each segment is detailed in Table 1.

content of the nucleic acids accounts for the difference in density. Disruption of virus
particles by an osmotic shock in procedures similar to those used 25 . 26 to examine the
dsRNA content in individual particles provided, however, a confusing picture. These
studies indicated clearly that the amount of dsRNA encapsidated in different particles
is not uniform. In some particles the dsRNA content was within the range of the size
of known dsRNA segments, however, from some particles long molecules were seen
that could be only mUltiples of the known phenol extracted segments. 27 . 66 The com-
plexity of the patterns of segmentation of each of the inhibitory phenotypes rendered
the resolution of the density peaks an impossible task. To resolve the question, simpler
patterns of segmentation were sought. Simplified versions of the wild type complexes
PI, P4, and P6 were found to arise in progeny from crosses between the native strains
and P2 strains" and also at a relatively high frequency after mild mutagenesis. 15.16 The
simpler patterns contained anywhere from I segment to 4 segments (Figure 8), whereas
the original patterns contained 5 segments in P6 to 7 segments in P4. Also, some of
the P2-sensitive strains contained some of the segments found in the native inhibitory
strains. Using a simple pattern, such as one with 2 to 3 segments that offers a chance
for clear separation in CsCl density gradients, the mode of encapsidation of the dsRNA
of the Ustilago viruses was resolved.:l. S
The hydrodynamic properties, the density components, the nucleic acids content,
and the major coat proteins were examined using the virions obtained from a PI strain
and two other virion preparations obtained from strains that do not express the inhi-
bition function and contain only few dsRNA segments. The virions in one strain con-
 119

Table 1
SIZE AND DESIGNATION OF THE dsRNA SEGMENTS
RECOVERED FROM THE VIRIONS OF U. MA YDIS

UmV dsRNA" 2.8OJo PAGE" 1% agarose< 1% agarose d em

HI 6100 6200 6700

H2 4500 5000 4600
H3 3000 3000 3500
H4 2600 3600 3200
MI 1400 1700 1500 1500
M2-6 1000 1200 1100
M2-I,4 960 1100 960 1000
M3 920 1100 960 1000
L 360 350 360

Note: PI is typified by segments HI, H2, MI, M2-I, M3-I, and L; P4 is

typified by segments HI-H4, M2-4, M3-4, and L; P6 is typified by
segments HI, H2, M2-6, M3-6, and L. All the laboratory strains of P6
retain only HI, M2-6, and L.

M2-6 is the M2 segment of P6; M2-I,4 is the M2 segment of PI and P4.

Calculated from the values given by Bozarth et al. 2.
Size standards were the ScV dsRNAs.
d Size standards were dsDNA restriction fragments.

From Field, L. J., Bruenn, J. A., Chang, T. H., Pinchasi, 0., and Koltin,
Y., Nucleic Acids Res., 11, 2765,1983. With permission.

tained one of the heavy segments and one of the medium size molecules, the virions in
the second strain contained two segments of the heavy molecules. In addition, in the
purification of the virions a characteristically sizable sample of empty capsids was
obtained, and this sample provided the basis for the calculation of the molecular
weight of the empty particles.
The results indicated that the size of the full and empty capsids is ca. 43 nm as
determined by their light-scattering characteristics. The capsids from various sources
are similar in size, and this determination is in close agreement with the size obtained
by electron microscopic measurements (41 nm) of negatively stained particles. The
molecular weight of the empty particles is 9.8 x 106 • The molecular weight of the major
coat protein is 73 to 75 K. Other proteins may contribute no more than 1070 of the
mass. 17 Therefore, the capsid is constituted of some 130 subunits. Estimation of the
dsRNA content was obtained from the ratio of the sedimentation velocity of full (175S)
and empty particles (98S). The results suggested that the full particles with the highest
density (1.418 g/cm 1 ) contain some 23.8% dsRNA and their molecular weight is 14 x
106 • This is the upper limit in the content of dsRNA per capsid. These particles should
contain dsRNA segments of molecular weight ca. 4.2 x 106 and, therefore, the particles
contain only one molecule of this size which is equivalent to the heaviest known seg-
ment of the Ustilago viruses and one common to all three phenotypes. Similar calcu-
lations with another molecule of the heavy group lead to the same conclusions that in
fact only one molecule of this size can be contained by a viral capsid. Extraction of the
dsRNA from the particles that banded at different densities indicated that only mole-
cules of one size class are found in virions of a specific density and there is no mixing
between the heavy segment and the medium size segments. Multiple density peaks were
obtained with particles that contained a specific molecule of the medium size class.
These results suggest that the dsRNA segments of the medium class are encapsidated
as individual molecules or as mUltiples. In the studied case up to three copies can be
120 Fungal Virology

FIGURE 8. PAGE of DsRNA from a

P6 phenotype (Jeft) and from a simplified
genome Jacking the inhibitory function
(right).

encapsidated in a single capsid. A clear segregation of the segments is apparent in these

studies although information is not available as yet on the encapsidation of the small
segment. These results suggest that two factors regulate the encapsidation of the
dsRNA since the upper limit is apparently 4.2 x 10 6 daltons, yet a multiple of 3 x
medium size molecules is still below the upper limit that can be encapsidated. There-
fore, it is likely that up to three identical binding sites are available in each particle. If
a heavy molecule is encapsidated, only one molecule is encapsidated and perhaps there
is only one binding site for this class. It is conceivable that the capsid consists of dif-
ferent binding proteins for each class of molecules. The encapsidation pattern of seg-
ments of different sizes in separate capsids, all of which consist of the same major
protein, suggests that the complex operates as a virus helper system or as a complex of
satellite RNAs. Studies on the mapping of viral functions suggest that the system con-
sists of a major virus with satellite dsRNA molecules encapsidated in one-size particles
with an identical major coat protein.

C. Viral Functions and the Segments of the dsRNA

The ability to generate derivatives of the native phenotypes with simpler patterns of
segmentation than those known in PI, P4, and P6 (Figure 8) provided the proper
means for genetic mapping of the functions associated with specific segments. Simpli-
 121

WILD-TYPE VARIANTS
1-- A -1L"'£"'~ E F
2- -
H 3 -
14 -

1-

M
~
2-1 2-4.l:.2 2-1 2-1
...B. I I E li

I
KILLER 4 6 6 6 6 6
SPEC

FIGURE 9. Diagram of the patterns of dsRNA segments obtained in

PAGE. The three inhibitory phenotypes PI, P4, and P6 (left), laboratory-
derived variants expressing the inhibitory function derived (right). Variants
A,B,D,E, and F obtained by crosses between different phenotypes; variant
C originated spontaneously in the laboratory. (From Koltin, Y., Levine, R.,
and Peery, T., Mol. Gen. Genet., 178, 173, 1980. With permission.)

fication of the multicomponent systems of PI, P4, and P6 was obtained by analysis of
progeny from crosses and by mild mutagenesis. In addition, spontaneous alterations
of the segmentation patterns were noticed in two instances. All the laboratory strains
of P6 originally contained five segments (HI, H2, M2-6, M3-6, and L). Those P6
strains used in the laboratory at this time have all lost the H2 and M3-6 segments
(Figure 9). A PI strain that originally contained segments HI, H2, MI, M2-1, M3-1,
and L contains at this time an additional segment similar in size to H4 (Figure 10).
Other PI strains with varying intensities of H4 also arose in the course of the studies.
The dynamics of the genetic drift is not known and the source of the new segment is
unknown. It is clear, however, that an integrated proviral state of these viruses does
not exist. Hybridization of viral transcripts and of cDNA derived from the transcripts
to restricted genomic host DNA showed no trace of hybridization. 2 •
The sequence homology, as will be discussed below, does not suggest the derivation
of smaller segments within the heavy group from the heavier ones. A conceivable al-
ternative is that not all of the segments are maintained in a steady state with respect to
their molar ratios, and some dynamic drift may occur perhaps as a result of simple
genetic alterations such as has been shown for the genomes of various RNA viruses. 3o
The drift may lead even to the detection of dsRNA in cases in which dsRNA was not
detected earlier. McFadden et al. 31 reported recently such a case in Gaeumannomyces
graminis. A hyphal tip culture in which dsRNA could not be detected at the time of
isolation was retested 18 months later and dsRNA was detected with no difficulty.
Similar findings were reported earlier /2 using ascospore-derived cultures. The de novo
origin of strains of Ustilago with the inhibitory potential was reported,33 based on the
122 Fungal Virology

H4

FIGURE 10. DsRNA pattern in 1070 agarose

slab gel from the native strain of PI (left) and
from the same strain after a long period of trans-
fers in the laboratory. Note the difference in the
intensity of segment "4. (From Widgerson, M.
and Koltin, Y . , Curro Genet., 5,127,1982 . With
permission.)

unpublished results of Puhalla. In addition, a strain tested 11.'. and classified by these
authors as a P2 strain, based on its sensitivity to the inhibitory substance and the fact
that it was devoid of viruses, was found later by Koltin and Day 3J· to contain dsRNA
and by Koltin et al. 33b to contain also virions. The same strain seemed to have altered
its properties in the course of the studies since it had been tested for the presence of
dsRNA by Koltin and Day to verify its characteristics as described." ·'· This strain
conformed to the characteristics determined earlier by the various investigators and yet
some 6 months later its properties have altered . Since each strain carries two highly
distinctive genetic markers at two loci, the incompatibility alleles, it is difficult to as-
cribe the change in dsRNA content to some confusion in strains. What is more likely
is, as McFadden et al. 3. suggested, that the titer of the dsRNA is quite low in some
strains beyond the level of detection by the available means. The genetic drift in the
dsRNA molecules or in the host may lead to the replication and expression of some
segments as their selective advantage is altered.
The genetic stability of the dsRNA genomes has not received much attention as yet,
although the study of exclusion phenomena among the Ustilago viruses and the S.
cerevisiae viruses 9 . 1S . 3 • • u was shown to be related to specific dsRNA segments. A phe-
 123

nomenon of this type may be related to the dynamics of the molecular selection as a
function of sequence stability.
The reduced genomes used in the mapping of the viral functions seem to be stable
situations and are with few exceptions cases that are equivalent to deletion mutants if
one considers the native complex of dsRNA in each of the three phenotypes as repre-
senting the full viral complement. The situation is clearly more complex since the seg-
ments are separately encapsidated and, as will be discussed below, the maintenance of
some is not independent. Variants within each class of molecules were detected but the
rules with regard to their interdependence are quite clear. In all the strains that contain
dsRNA at least one of the heavy segments is found. Segments of the medium size and
the light segment are never found in the absence of a heavy segment. These results
suggest some dependence of the medium and the light segments on the information
contained in the heavy segments. Within the heavy group the heaviest segment (HI) is
found most often. Some variants were detected in the laboratory among those derived
from P6 and P4. From P6 the only variants in the heavy segments were those cases in
which segment H2 is missing. Not a single case has been detected, thus far, in which
segment H2 is found in the absence of HI. Among the heavy segments of P4 a number
of variants were detected. The variants were devoid of segment HI but had a number
of different combinations of the other H segments such as H2, H3, and H4; H3 and
H4; and HI, H2, and H4. Segments H2, H3, and H4 were never detected alone in any
of the strains tested. Thus, in the absence of HI, only combinations of other H seg-
ments were detected and it was suggested that some functional overlap exists between
these segments in which those functions found on HI are distributed on the other
segments.
The relationship between the molecular weights of these segments was also sugges-
tive of some interrelations between the H segments. The molecular weights of HI, H2,
H3, and H4 are 4.2, 3.1, 2.7, and 2.1 x 106 , respectively. Therefore, if an essential
function is missing in H2 and is found in H3 or H4, only the combinations can provide
the necessary information for the maintenance of the viral system. 39 With two excep-
tions 34 in all other cases the presence of the heavy segments was associated with the
presence of viral particles. Therefore, from the relations of the heavy segments among
themselves and their relation to the presence of virus particles, it was suggested that
the heavy segments contain the information for maintenance of the other segments and
these segments code for the coat proteins and enzymes such as the replicase and tran-
scriptase. Only for the coat proteins some information is available, since in an in vitro
translation system using the dsRNA derived from a strain lacking the medium and light
segments, the translation products obtained cross-reacted with antibodies derived
against the viral coat protein(s).40 The hybridization data, as will be shown below, do
not confirm the genetic conclusions suggesting functional overlap among the H seg-
ments, since sequence homology between those segments was not found.
Among the medium size segments MI, M2, and M3 of PI; M2 and M3 of P4; and
M2 and M3 of P6, variants are available in which one or more segments are lacking.
It is clear that there are no strains with the inhibitory function and lacking one specific
segment of the medium size class. In PI it is segment Ml-l that seems to determine the
inhibitory function or some signal for its expression, in P4 segment M2-4 plays the
same role, and in P6 segment M2-6 is associated with the inhibitory function. Strong
indications from both genetic studies and in vitro translation experiments suggest that
these segments encode the information for the inhibitory substance. In a cross between
a derivative of P4 devoid of segment M2-4 and a P6 strain, progeny were derived that
express the P6 inhibitory specificity. These progeny contain the dsRNA segments of
the P4 derivative and the M2-6 segment of P6. The phenotype of these progeny was
P6, thus indicating that the inhibitory function and the specificity are associated with
124 Fungal Virology

this specific segment. Therefore, it seems most likely that this segment encodes the
inhibitory substance and M2-4 encodes the inhibitory substance with the P4 specificity,
whereas M2-6 encodes the inhibitory substance with the P6 specificity. An alternative
interpretation suggests that the information encoded by the dsRNA is a signal that
activates the host genes for the inhibitory substance. However, in vitro translation
experiments using the reticulocyte system provided support to the conclusions based
on the genetic analysis that it is the dsRNA that encodes the inhibitory function. Using
the dsRNA obtained from strains containing the M2-4 and from strains containing
segment M2-6, and dsRNA from strains devoid of these segments as a control, a trans-
lation product was obtained that cross-reacts with the antibodies derived against the
purified inhibitory protein. This translation product was obtained only from strains
that contained segments M2-4 and M2-6.
The role of the other segments in the medium size class, M2-1, M3-1, M3-4, and
M3-6 is not known. Judging from the rate of loss of the M3-6 segment in P6, it seems
dispensible and unrelated to the maintenance of the virus, to the expression of the
inhibitory function, and to the expression of the cytoplasmic immunity. Similar re-
duced genotypes have not been detected, thus far, in strains of the PI and P4 pheno-
types among progeny from many crosses. Therefore, it is conceivable that in PI and
P4 other segments of the medium class do play some role in the maintenance and
expression of the inhibitory function and some differences exist between the informa-
tion contained in these segments in the different phenotypes.
The small segment of only 360 bp is quite small for coding anyone of the proteins
known to be encoded by the viral information. The major coat protein is 72 to 75 K
and the preinhibitory substance is a protein of a molecular weight of 19 K. The L
segment was found in every strain that expresses the inhibitory function and in mutants
that secrete the inactive protein. Based on complementation tests between strains af-
fected in the expression of the inhibitory protein, it was suggested 16 that this segment
is involved in some way in the expression of the toxin and is essential for its expression.
As will be discussed, this segment is related to the toxin encoding segment MI-I of PI,
M2-4 of P4, and M2-6 of P6, but it is unclear what its actual function is. In recent
studies,12 efforts were directed towards the mapping of the cytoplasmic immunity on
the viral dsRNA. A PI strain lacking the nuclear resistance allele was obtained by
cytoplasmic transfer of the viral information after fusion of compatible strains of PI
pIrand P2 pIs. The strains complemented each other due to their deficiency in differ-
ent auxotrophic requirements. The heterokaryon was formed in spite of the sensitivity
of the P2 parent by the saturation procedure described earlier. The descendents of the
unstable heterokaryon were selected so as to favor the original P2 with respect to the
auxotrophic requirements. By recurrent selection in the PI inhibitory substance so as
to select the P2 cells that received the cytoplasmic information for the expression of
the inhibitory function and the cytoplasmic immunity, the information for both func-
tions was transmitted in this environment, as was shown earlier. 11 One of the trans-
formed P2 was further treated by elevated temperature and by cycloheximide, in an
effort to cure the strain from cytoplasmic elements to determine whether the inhibitory
function can be cured independently of the immunity function. The efficiency in these
procedures is very low but one such strain was recovered displaying the loss of the
inhibitory function and the retention of the expression of resistance. This strain trans-
mitted the immunity to a strain sensitive to PI, thus confirming the cytoplasmic nature
of the immunity to PI.
The mapping of the function on the small segment was performed by the recovery
of progeny from a cross between the P2 strain transformed to a PI and a strain sensi-
tive to PI. Of 700 tetrads tested, 686 were uniformly PI. Of the tetrads that did not
display the inhibitory function, two showed some segregation of the sensitivity and
 125

FIGURE II. Mapping the immunity function of P 1.

From left to right, typical pattern of PI e'Xpressing the
inhibition and immunity, dsRNA from two strains that
are immune to PI but have lost the inhibitory function,
dsRNA from a strain that has lost both functions.
(From Peery, T., Koltin, Y., and Tamarkin, A., Plas-
mid, 7, 52, 1982. With permission.)

resistance to the PI inhibitory substance, and the others were uniformly sensitive. Since
the transformed P2 strain was genotypically PI pIs and it was crossed with a P2 pls
strain, the segregation of resistance (immunity) within the tetrads was unexpected. Two
strains were examined for the pattern of the M and L segments of dsRNA. One dis-
played an immunity to the toxin and lacked the MI and M2-I segments, the other was
sensitive and lacked MI, M2-I, and L (Figure 11). Another immune strain lacking only
the M2-I segment was obtained by mutagenic treatment of the P2 transformed strain.
Therefore, these results suggested that the immunity is not located on either MI
or M2-1 and it may reside on L. The complete loss of dsRNA in descendents of the
immune strain led again to sensitivity to the PI inhibitory substance.
The proper mutants are not available to resolve which segment is the one that con-
tains the information for cytoplasmic immunity, but since P6 strains devoid of M3 are
very common it is assumed that L is the segment that is associated with the immunity
and therefore strains expressing the inhibitory function always contain the L segment.
However, as was mentioned above, the analogy between the different phenotypes may
not be complete and the final mapping of the immunity function and the role of the
smallest segment is still an open question.
126 Fungal Virology

At the present time, based on the genetic studies and the in vitro translation of
dsRNA extracted from the native inhibitory strains and from strains with simple
dsRNA patterns, it is assumed that the maintenance functions map in the heavy seg-
ments, the toxin is encoded by one of the medium-sized segments, and the small seg-
ment plays a role in the secretion of the toxin and its inactivation in the cell (immunity).
However, no mutants are available that are affected in the replication functions, and
the transcription and some variants in the segmentation patterns are still needed to
make more conclusive statements on the mapping of the known functions and addi-
tional viral functions.

D. Relatedness of the Segments of dsRNA

Among the inhibitory phenotypes, PI, P4, and P6, three distinct patterns of dsRNA
were detected and the variation is within each of the classes of segments. All the pat-
terns conformed to the same basic arrangement of three size classes of molecules with
similarities with respect to the relation of each group to a specific viral function. How-
ever, two basic questions were unresolved, (1) are the segments of a specific phenotype
interrelated and derived perhaps by processing of a master copy?; and (2) are the seg-
ments of a specific size class from the different phenotypes interrelated? Answers to
these questions can provide an insight into functional interrelationships, evolutionary
relations of the dsRNA segments, and the differences and similarities that may also
provide a rational basis for the understanding of the rules of the exclusion phenome-
non. Furthermore, determination of partial sequences, especially the 3' end of the
dsRNA segments, can allow the identification of the site of initiation of the viral tran-
scription.
Sequence homologies were determined by hybridization of isolated dsRNA segments
as probes and with cDNA synthesized from isolated segments of the dsRNA. The
probes were hybridized to Northern blots of dsRNA from all three phenotypes. The
results obtained 23 are shown in Table 2.
The conclusions from the hybridization data are that within the segments of each
phenotype there is no sequence homology between the H segments and the M and L
segments. Only the smallest segment (L) and the M segments bear some sequence ho-
mology. Based on the results with PI and with P6 the homology is between the L
segment and the segments that encode the inhibitory toxin. The results with P6 are
completely clear since the strain used contained only M2-6 of the medium class. In the
case of PI, it was earlier thought that M2-1 is the toxin-encoding segment, in analogy
with the situation in P4. However, since there was not a single case in which Ml was
missing in strains of the PI phenotype, both in the native strain and in a number of
strains derived by complex crosses with P2 strains and with other strains with different
inhibitory phenotypes, it was never resolved whether Ml or M2-1 encode the toxins.
The hybridization data suggested that it is M 1, based on the sequence homology to
M2-4. These data suggested also that if L hybridizes to the segment that encodes the
toxin, then M 1 is the segment in PI that contains the information that encodes the
toxin.
A peculiarity and an apparent contradiction stem from the results of the hybridiza-
tion relations between the segments of the heavy class of P4. The genetic data suggested
a functional overlap between the four heavy segments. The pairing of the segments in
those cases missing the heaviest segment suggested that functions that normally reside
on the heaviest segment are located on different segments among H2, H3, and H4.
The hybridization data show no homology between these segments and HI of P4 and
among themselves. The results are surprising and predict a strong genetic drift within
these segments, at least at the level that prevents the detection of sequence homology.
The occurrence of variations in the number of segments within this group was reported
 127

Table 2
THE INTERRELATEDNESS OF THE dsRNA SEGMENTS AS
DETERMINED BY HYBRIDIZATION OF EACH dsRNA SEGMENT WITH
THE dsRNA SEGMENTS OF EACH OF THE 3 VIRUS COMPLEXES
PI, P4, AND P6

dsRNA on Nitrocellulose

PI P4 P6

Probe HI H2 MI M2+3 L HI H2 H3+4 M2+3 L HI M2 L

P6HI + + +
P6M2 + +
P6L + + +
P4HI + + +
P4H2 + +
P4 H3+4 +
P4M2+3 + + + +
P4L + + + +
PI HI + +
PI H2 + +
PI MI + + + +
PI M2+3 +
PI L + + + +

From Field, L. J., Bruenn, J. A., Chang, T. H., Pinchasi, 0., and Koltin, Y., Nucleic Acids Res., II,
2765, 1983. With permission.

earlier, based on the genetic analysis of progeny from various crosses. However, in a
survey of natural isolates,4' patterns containing only H3 and H4 were found in native
strains and those were not uncommon. Apparently, these segments in native strains are
sufficient to assume the maintenance role ascribed to this group of segments. A com-
parison between those cases in which only H3 and H4 are found versus the occurrence
of all four segments was not performed to determine the interrelatedness of these seg-
ments under different selection pressure. It is conceivable that in the presence of all
four segments the genetic drift is greater and a higher degree of conservation will be
detected in those cases in which only the two segments H3 and H4 occur alone. By
immunological tests using antibodies derived against coats of the virions of P I and
those obtained against P6, both have shown clear cross reaction with virions of all
three phenotypes. A broader survey, including those natural isolates containing only
H3 and H4 from various sources, followed by an immunological analysis of the viral
coat proteins, has not been performed thus far. The peculiarity and apparently discrep-
ancy will be resolved only by a more detailed study of the nucleic acid sequence.
The relatedness of the dsRNA segments among the three phenotypes is shown in
Table 2. The only segments that show cross hybridization are HI of all three virus
complexes. The interrelations between the H2 segments cannot be determined with
respect to P6, since the derivative used lacks H2. The H2 of PI and P4 cross-hybridize.
Whether the H2 originally found in P6 may be also similar to H2 of P4 and PI is
unclear at this time.
The similarity between PI and P4 and the differences between both and P6 is evi-
dent, especially by the lack of hybridization of the toxin-coding segment of P6 (M2-6)
with any of the M segments, particularly the segments that encode the toxin in PI and
P4 (Figure 12). Only the small segment (L) of P6 has some similarity with the L seg-
ment of P4, but even this segment has diverged since the intensity of the hybridization
is lower in the heterologous hybridization than in the hybridization to the homologous
128 Fungal Virology

a
PI) P4 P6 ScV

FIGURE 12. Hybridization of specific dsRNA segments to nitrocellulose containing dsRNA of

all 3 phenotypes and from the S. cerevisiae virus. (a) The dsRNA pattern in a 1.4"70 agarose gel
stained with ethidium bromide. Nitrocellulose filters probed with (b); HI from PI (c); eDNA from
MI; and (d) probed with cDNA from PI. (From Field, L. J., Bruenn, J. A., Chang, T. H.,
Pinchasi, 0., and Koitin, Y., Nucleic Acids Res., II, 2765, 1983. With permission.)

L dsRNA. A rather high degree of similarity is found between the PI and P4 segments.
The M segments of PI and P4 cross-hybridize, and these segments hybridize with the
L segment of both PI and P4. These segments do not hybridize with the M segment
and the L segment of P6. Furthermore, the L segments of P I and P4 cross-hybridize
with each other but do not hybridize with the L segment of P6. Therefore, it seems
that PI and P4 are very similar and are quite different from P6. The hybridization
relations between PI and P4 also suggest that the functions that reside in MI of PI are
located on the M2 or/and M3 of P4 (the genetic data suggest that it is M2). Also, the
results suggest that M I-I, M2-4, and M2-6 all contain a region of homology for the L
segment, yet the divergence of these segments is expressed also in L. This is the reason
that L from P6 does not hybridize with MI and M2-4, and L from PI and from P4
hybridizes with MI and M2-4 but does not hybridize with M2-6.
The homology of the L segment to the M segments and the precise mapping of the
region of homology was determined by heteroduplex analysis (Figure 13). The L seg-
ment hybridizes in all three cases to the end of the M segment. The results strongly
suggest that L is derived from M, especially since the divergence of the M segments is
also expressed in the L segments. These relations are somewhat similar to those re-
ported for the small segments of the S. cerevisiae virus. 42,43 However, that small seg-
ment in S. cerevisiaeinterferes with the replication of the toxin-coding segment, rather
than playa role in its expression as is suggested for L. Some cases of the occurrence of
L without M have been reported. 12 ,3' Thus, L may be derived from M when M is
present, but it appears to have had an origin of replication that permits its maintenance
as an independent unit, providing the coat proteins and replicating functions are avail-
able from an H segment. Nevertheless, the occurrence of L without M is quite rare,
and those cases detected are exceptions rather than the rule. Among the nine patterns
of dsRNA in the survey conducted of natural isolates both from the United States and
from Poland, there is not one case in which an L segment is found in the absence of
an M segment. 41 Most isolates contained a segment equivalent in size to L. Further
support for the interrelatedness of the Land M stems from the stoichiometric relations
of the viral transcripts, as will be discussed below.
The sequence of the 3' ends of the dsRNA segments of all three phenotypes is also
suggestive of some similarities between the M and L segments of P I and P4. Of the
two known 3' ends of MI, only one end is present in L. The same type of relations
holds for L of P4 and the M segments of P4, although M2 and M3 were not separated
in the study. Since both segments have the same 3' sequence it is still possible to resolve
the identity of one 3' as being identical with the 3' end of L. The relations between the
 129

.. . . ...
.' '. -
FIGURE 13 . Heteroduplex of MI and L from P I (the branched molecule). The relaxed circle
is pBR322 and the linear molecule is L. (From Field , L. J ., Bruenn, 1. A., Chang, T. H.,
Pinchasi, 0 . , and Koltin, Y. , Nucleic Acids Res., II, 2765, 1983 . With permission.)

Land M2 of P6 are not as clear. The similarity between the HI segment of all three
phenotypes is evident in the termini of the HI; also, the similarity between the H2
segment of PI and P4 is clearly seen in the terminal sequences.
Heterogeneity within each segment was not detected as earlier reported in the yeast
dsRNA virus. 44 The only heterogeneity detected is in the terminal nucleoside, since
each segment ends with an A or a a, and the duality is seen in every segment and is
thought to be a posttranscriptional addition. The 3' end of all Hand M segments of
the Ustilago viruses are within the consensus sequence A/U A/V A/V A / V A/V A/V
A/U v/a C A/GOH. Some of the segments are very similar to the S. cerevisiaevirus
transcription initiation consensus sequence UVVUUCA/GOH. 44 .45 It is possible that
the 3' end of the transcript of M can be cleaved to yield a message that encodes an
immunity protein. If the small transcript is also replicated, it will form the dsRNA
segment that encodes the immunity factor and is independent of M.
Although some similarities with the yeast dsRNA are quite evident, the hybridization
of the dsRNA from all three phenotypes to dsRNA of the yeast virus yielded no posi-
tive signals. The basic o{ganization and distribution of functions is somewhat similar
as well as the size of the inhibitory protein, yet the sequence encoding those functions
is very dissimilar. Other information relating to the nature of the inhibitory substance
also suggests that the mode of action and physical properties are different, but at the
level of the basic organization of the segment encoding the toxin evident differences
have already been noticed. Heteroduplex analysis of the yeast dsRNA revealed some
unique instability in the middle of the segment that encodes the inhibitory protein. 42
The melting properties of that region implied nearly lOOOJo A+U as the base composi-
tion of the early melting part of the molecule. The denaturation properties of these
molecules were utilized 46 for the localization of the information for the inhibitory pro-
tein, since the molecule could be cleaved to two distinct segments with SI nuclease
under conditions in which the A+U region was unstable and yielded single-stranded
regions. In vitro translation of these segments provided a more precise mapping of the
130 Fungal Virology

15

»
.
~
~

I 10 .., 05
,, ,,
...,, .' \

52 ,, ,,
'\ ,, ;,
04
E ,
1\
.- \ J
Co
u f ,
\

,, \ 03 40

*'en
5 L
\.
'. 02 30
c
'---~ n
\"---
01 20 ~
005 '"
0 2 4 6 8 10 12 14 16 18 20 2224
B T
Tube

FIGURE 14. Distribution of vIrions (. -- .) and RNA polymerase ac-

tivity (0 - 0) in sucrose density gradient used to purify virions of P6.
(From Ben-Zvi, B. S., Koltin, Y., Mevarich, M., and Tamarkin, A., Mol.
CeIl BioI., 4, 188, 1984. With permission.)

encoded information and indicated some basic facts on the organization of this seg-
ment.
In an effort to compare the organization of the S. cerevisiae virus with that of the
Ustilago viruses using more drastic denaturation conditions' 6 such as longer time of
incubation and a concentration of S 1 nuclease some 20 x higher than the concentration
used for the S. cerevisiae virus, no cleavage of the Ustilago dsRNA was detected.
Along with the information concerning the relatedness of the L to the M and the pos-
sible processing and independent expression of the two segments, it seems at this time
that the organization of the segment that encodes the inhibitory protein is basically
different in the two viral complexes.

E. RNA Polymerase Activity

The mode of encapsidation and the relatedness of the segments suggest that the virus
complex is based on a major virus containing the H segments and a number of satellite
dsRNAs. To gain an understanding of the mode of coordinate expression of this com-
plex organization and an insight into the maintenance of the coordinated replication
and expression in a steady state, the RNA polymerase activity related to the dsRNA
viruses was examined. The incorporation of radioactive UTP into TeA precipitable
material was followed in the fractionated sucrose density gradients (Figure 14) used to
purify the virus particles. 29 The activity was found to be related to the virus concentra-
tion and to the growth phase of the cells from which the particles were isolated. Virions
from stationary phase cells display very low incorporation. Examination of these par-
ticles in the electron microscope after negative staining and based on their position in
the sucrose density gradient indicated that the particles obtained from these cells are
constituted mostly of empty coats. The particles obtained from late log phase cells are
very active, and under the appropriate conditions the incorporation can proceed for
more than 24 hours. Using two procedures for the purification of the viral particles,
 131

the results obtained were consistent and suggest that the polymerase activity is associ-
ated with the virions and the reaction is not sensitive to Actinomycin D. The optimal
reaction conditions for the RNA polymerase activity are similar to those defined for
the S. cerevisiaeviruses. 47 . 48 The reaction required 20 mMNa ' + and is sensitive to both
higher and lower concentrations of Na ' +. Also, the reaction requires 5 mMMg2+. The
pH optimum for the polymerase is quite different from that determined for the virus
of S. cerevisiae. Maximum activity of the Ustilago virus was at pH 8.9 and may reflect
the differences in the ecological habitation of the two organisms. RNase activity is
usually also associated with the partially purified virions. Even in a high substrate to
virion concentration, the reaction reaches a plateau within 3 to 4 hours. However, in
the presence of bentonite, a linear reaction can proceed for 5 hr and a plateau is not
reached even after 20 hr. The reaction products obtained are ssRNA molecules and the
transcripts of the viral information are based on their properties.
Over 95070 of the incorporation is in the ssRNA, and only SOlo is found in the dsRNA
and may reflect the early phase of transcription. Within 3 min, RNase-sensitive mate-
rial that had been synthesized is detected. Yet electron microscopic examination per-
formed by Steinlauf48a of the dsRNA extracted from actively transcribing particles
after phenol-SDS extraction and LiCl precipitation, following the procedure of Usala
et al.,.- reveals dsRNA molecules with ssRNA. These results suggest that the transcrip-
tion is by a semiconservative process as in the Aspergillus foetidus viruses. 50 The sen-
sitivity of the newly incorporated material within a very short period, even in reactions
performed at suboptimal temperature (l8°C), is unclear as yet and does not conform
to the expectations based on a semiconservative model. The polymerase activity is
clearly a transcription reaction, and under the conditions used and in particles obtained
from late log phase cells, incorporation of the precursor to dsRNA was not detected.
The mode of the replication of the Ustilago viruses is totally unknown, and the
polymerase activity in log phase cells has not been studied as yet. With the many simi-
larities between the Ustilago viruses and those of S. cerevisiae the replicative activity
should correspond with the logarithmic growth of the cells. 51 Mutants are unavailable,
as yet, to distinguish between the transcriptase and the replicase to determine whether
one or more enzymes are engaged in the two processes. Furthermore, it is unclear
whether the transcriptase is a virus or a host function. The transcriptase reaction is
associated with the particles, but the transcripts are released from the particles. In
reactions containing bentonite, the viruses adhere to the bentonite, yet much of the
transcription products are found in the supernatant. The distribution of the incorpo-
rated UTP between the dsRNA and ssRNA in these experiments is similar to the dis-
tribution found in the total extraction of the nucleic acids after a transcription reaction
and the separation by column chromatography between the ssRNA and the dsRNA.
The ssRNA is mostly in the supernatant, and the dsRNA is in the pellet containing the
virions and the bentonite.
The transcripts are homologous to all three classes of segments H, M, and L. Char-
acterization of the RNA polymerase activity was performed with virions from the P6
phenotype containing segments HI, M2-6, and L. The transcripts obtained hybridize
with all 3 dsRNA segments (Figure 15). Since it is known that segments of Hand M
are separately encapsidated, RNA polymerase activity must be associated with particles
containing each of the segments. It is not clear how the L segments are encapsidated
and, therefore, it is impossible to speculate at this time whether L is transcribed. The
transcription products consist of three size classes of transcripts with molecular weights
of 1.85, 0.41, and 0.1 x 10 6 • These molecular weights are nearly half the molecular
weight of the HI, M2-6, and L segments of dsRNA. Both the hI transcript and the
m2-6 transcripts are very likely polycistronic messages, since more than one function
is assumed to reside on the respective dsRNA segments. The molar ratios of the com-
132 Fungal Virology

FIGURE 15 . HYbridization of in vitro-derived

transcripts from P6 virions to the dsRNA seg-
ments of P6 . The dsRNA segments are marked
H, M , L. (From Ben-Zvi, B. S., Kollin , Y., Mev-
arich , M., and Tamarkin, A., Mol. Cell BioI., 4,
188, 1984. With permission.)

pleted transcripts are 1:5:4 for the hI, m2-6, and 1 transcripts. The similarity of the
molar ratio between m2-6 and I, along with the hybridization data, suggest equal effi-
ciency of transcription of M2 and L or the transcription of m2-6 and I only from
M2-6. The time of synthesis, as shown in pulse-chase experiments, indicates that I is
synthesized subsequent to the synthesis of m2-6. The relations between these two tran-
scripts and the functional role of 1 require further clarification. Sequence analysis
should be very informative in the verification of the relations between M2-6 and Land
in the analysis of its function.

III. KILLER PROTEINS

A. Nature of the Killer Toxin

Studies by Hankin and Puhalla 7 on the nature of the toxic factor of PI have shown
that the substance can be heat inactivated. In addition, the sensitivity of the P I toxin
to some proteolytic enzymes was demonstrated. The activity was precipitated with am-
monium sulfate. These results suggested that the substance is a protein. Tests con-
ducted with the toxin from P4 and P6 cells provided similar results. Treatment of all 3
toxins with RNase and DNase did not affect the activity. Partial purification and char-
acterization of the three toxins indicated that the different toxins can be distinguished 52
in spite of the fact that all three toxins are of a similar molecular weight of ca. 12,000.
 133

The differences between the toxins are reflected by the type of inhibition, the pH
and temperature sensitivity, and by the affinity to concanavalin A. The toxin secreted
by P6 cells forms a clear zone of inhibition with sharp boundaries. The zone of inhi-
bition formed by the toxin of PI and P4 does not have clear boundaries, and creates
the impression that some residual growth occurs within the zone of inhibition. In a
controlled experiment following the survival of cells and growth in the presence of the
different toxins, the results obtained indicated that, using identical concentrations of
toxins and of cell densities, the P6 toxin causes immediate cell death, whereas the PI
and P4 toxins kill 990/0 of the cells and the remaining 1% display delayed replication.
These cells are not resistant to the toxin but reflect a population that always escapes
the treatment. A difference in the sensitivity to pH was also noticed, and whereas the
toxins from P4 cells and from P6 cells are active within a broad range of pH between
pH 4.0 and 9.0, the toxin from PI cells is very sensitive to change in pH, as noticed by
Hankin and Puhalla 7 and is active only if the pH is controlled and maintained close to
pH 7.0. Similar differences were observed in the temperature sensitivity of the three
toxins. The PI toxin is inactivated by short exposure to elevated temperature (2 min at
80°C) whereas the toxins from P4 and P6 cells are only partly inactivated after longer
exposure to 80°C.
Differences in the conformation of the three proteins were noticed in the migration
patterns of the toxins in their native state under nondenaturing conditions in 7.5 %
polyacrylamide gels at different pHs. The rate of migration of all toxins in nondena-
turing conditions is quite distinct. Furthermore, a clear distinction was noticed in the
affinity of the P4 toxin and the P6 toxins to concanavalin A (ConA) although the
interaction was not based on binding affinity by a glycosyl residue. The P4 toxin bound
to ConA in a reversible form. It was eluted from the column with NaCl. The P6 toxin
bound irreversably to this type of column. Alpha-D-mannoside was ineffective in the
elution of both toxins 68 and, therefore, it is assumed that the interaction is an ionic
interaction and the two toxins differ in charge. This difference was confirmed in isoe-
lectric focusing of the two proteins. These results conform with the notion of the di-
vergence between the P6 toxin and that of PI and P4 as reflected at the level of the
nucleic acids.
In spite of the various differences suggesting conformational differences between the
three toxins and some differences in affinity towards specific substrates and sensitivity
to environmental factors, all three toxins have a similar molecular weight and antigenic
determinants. Contrary to earlier reports, 52 all three toxins cross-react with antibodies
derived against the P6 toxin and against the P4 toxin. Furthermore, the pretoxin of
one type derived in in vitro translation cross-reacts with the antibodies derived against
the toxin of another type. 40 The use of specific staining procedures to test whether this
protein is a glycoprotein provided no indications for the association of glucose with
these toxins. However, ribose was found associated in studies with the P4 toxin with
about 10 molecules per toxin molecule. 69 It is unclear as yet whether the occurrence of
ribose is unique to the P4 toxin and the nature of the interaction with the protein.

B. Specificity of the Toxins

To define the specificity of the virus-encoded toxins, tests were conducted to deter-
mine the sensitivity to the toxin of both fungi and bacteria. In the selection of the
organisms to be tested some possible natural competitors of Ustilago maydis53 were
included in the tests. Also, to obtain the broadest spectrum of inhibition, strains of
Ustilago displaying each of the three toxin specificities were used. The tests were con-
ducted as plate assays in which the strain tested for its response to the toxin was used
as a lawn and the strains of Ustilago were spotted on the lawn. The initial sample
included some 51 bacterial specimens representing 25 different species. Inhibition was
134 Fungal Virology

not noticed among any of the bacterial specimens. The tests for sensitivity among the
fungi included primarily members of the Ustilaginales and a few specimens from the
filamentous fungi including some that are known as plant pathogens of corn, such as
He1minthosporium carbonum, H. maydis, H. turcicum, and H. vagans. A number of
isolates from different parts of the world were included in some cases so as to detect
the sensitivity if it was limited to specific geographic locations. The results clearly in-
dicated that among the filamentous forms, including those that are known pathogens
of corn, none are sensitive to the toxin secreted by strains of Ustilago maydis. Among
the Ustilaginales some specimens are very sensitive to the toxins secreted by strains of
U. maydis. Among those displaying the sensitivity were species of Ustilago pathogenic
to wheat, oats, and barley, and those known to grow on various grasses. However,
none of the sensitive species share a common host with U. maydis. Furthermore, sen-
sitivity was detected in one species that is not a pathogen of grasses, such as U. utri-
culosa. Therefore, these results provided no indication that the toxins confer a selective
advantage to the species bearing the necessary viral information for toxin synthesis and
secretion. In addition, the heterothallic life cycle of U. maydis and the incidence of
toxin secretors and strains resistant to the toxin shed some doubt on the selective ad-
vantage of the secretion of the toxin.
These studies have been expanded more recently and tests were conducted with a
collection of the marine and terrestrial yeastlike Basidiomycetes from the collection of
Dr. J. Fell (School of Marine and Atmospheric Sciences, Miami, Florida), including
various species of Rhodosporidium, Rhodotorula, Leucosporidium, Bullera, Candida
albicans, and Torulopsis glabrata were also tested. In none of these tests were there
indications suggesting the sensitivity of fungi other than those classified among the
Ustilaginales. Therefore, it appears that the sensitivity is related to some characteristic
limited taxonomically to the Ustilaginales, such as a specific component of the cell wall
or the cell membranes that acts as receptor for the toxin, and these components are
typical primarily of those Ustilaginales that are pathogenic to the grasses.

C. In Vivo and in Vitro Mode of Action

The expression of the toxin is related to the growth of the cells. The major accumu-
lation of the secreted toxin occurs during the logarithmic phase and early stationary
phase of growth. During the stationary phase a decline in the accumulation of toxin is
noticed. Some differential effect of temperature on growth and toxin secretion can be
detected. In studies with the PI toxin the optimum for toxin secretion was 25°C,
whereas the optimum for growth was 30°C. Recent studies with the P6 toxin suggest
similar relations.
Among the mutants obtained by Koltin and Kandel,16 a few were shown to express
an inactive protein with a molecular weight similar to the native toxin. The in vivo and
in vitro complementation tests performed among the various mutants lacking toxic
activity provided information on the interaction between the molecules and the in-
volvement of the L dsRNA in the expression of the toxin. Complementation can be
detection by the interaction between the secreted molecules when mutants of different
types are plated in close proximity on a sensitive lawn. Mixing of the inactive proteins
from some of the mutants results in the recovery of active molecules. The restoration
of the activity may result from in vitro intracistronic complementation or from an
interaction between the subunits if the secreted toxin of Ustilago is similar to that of
S. cerevisiae!O.71 Two distinct types of mutants were reported based on the comple-
mentation pattern. Recently, a third type was identified that complements the two
mutants reported earlier,72 thus suggesting a higher degree of complexity. Further-
more, some revertants have been detected, suggesting that the inactive toxins of these
mutants result from single base substitutions.
 13S

The involvement of the L dsRNA segment in the expression or the secretion of the
toxin was suggested by a series of complementation tests between nonkiller mutants
that secrete an inactive protein and mutants lacking the toxin coding segment of
dsRNA. The mutants that secrete an inactive protein were complemented in vivo by
two mutants that lacked the toxin coding segment M2-6. These mutants retained the
small segment (L). Following the formation of a heterokaryon between the mutants
that secrete the inactive form of the toxin and the mutant lacking M2-6, an active toxin
was secreted by the heterokaryotic cells and from the haploid descendents of this het-
erokaryon. Thus, these results suggest that the toxin is activated and stabilized during
its secretion by a function encoded by a dsRNA segment other than the one encoding
the toxin. The inactive proteins that were activated by in vitro complementation may
be improperly processed toxins that regained the functional conformation as a result
of intermolecular interaction. This interpretation places the mutation of the nonkillers
that secrete an inactive protein not in the toxin-coding segment but in the L dsRNA
segment. This implies an active role of the L dsRNA in the expression of the toxin.
The kinetics of the interaction between the toxin and the target cell has been ap-
proached very recently73 and the results suggest a single hit kinetics. The site of action
in the intact cells is currently under investigation. A very distinct effect on the cells is
noticed using different ratios of toxins to cells. Using a high toxin/cell ratio leads to
cell lysis within 2 hr. A sublethal dose leads to an initial stimulation of cell division
during 90 to 120 min after exposure to these toxins, followed by an arrest of cell divi-
sion. A time-course study of the incorporation of precursors of nucleic acids and of
proteins indicates that shortly after the exposure of the cells to the toxin protein, syn-
thesis is inhibited, whereas the synthesis of nucleic acids continues.
The studies on the in vitro mode of action were influenced to a large extent by an
effort to study the mode of action using a purified toxin and defined substrates. Fur-
thermore, from the exclusion relations it was postulated that the toxin may playa role
in the exclusion mechanism similar to the DNA restriction enzymes. It was assumed
that the toxins act as dsRNA restriction enzymes. IS
The activity on DNA and RNA (both single stranded and double stranded) was
examined as well as the effect on the synthesis of proteins in an in vitro system. 54 The
DNA substrates used in reactions with toxin purified to homogeneity were the single-
stranded DNA bacteriophage phi-Xl74 and the covalently closed circular DNA of bac-
teriophage PM2 and Simian virus 40 (SV40). Endonucleolytic activity was noticed with
all three substrates. The cleavage appeared as a gradual relaxation of the covalently
closed supercoiled molecules followed by a cleavage of the second strand (Figure 16).
However, no sequence specificity could be noticed in double digests using one of the
restriction enzymes that has only one restriction site in the PM2 and in SV40 DNA.
Efficient cleavage was noticed in ratios of 1: 10 of the substrate to the enzyme in short
reactions of less than 30 min.
The activity on RNA was tested using ribosomal RNA as a ssRNA substrate and
RNA from the U. maydis viruses as a dsRNA substrate. In addition, the degradative
activity of homopolymers of RNA and synthetic dsRNA was examined to test for the
specificity of the nuclease. The purified toxins from the P4 and P6 viruses were used
and in both cases the degradation of ssRNA was clearly noticed. The activity on
dsRNA was inefficient and only slight degradation was noticed in some cases; usually
the molecules seemed insensitive to the nucleolytic activity. In tests performed with the
homopolymers of RNA, the toxin from P4 showed some preference for the poly-rC
and even higher preference for poly-rI, but in both cases the affinity is less clear than
that seen using RNaseA and Tl. The cleavage pattern of the 28S and 16S ribosomal
RNA did not suggest cleavage at specific sites. Further evidence for the cleavage of
ssRNA by the toxins was obtained in studies on the effect of these toxins on the in
136 Fungal Virology

SV40 + + + + +
REACTION
KP4 + +
MIXTURE + +
KP6

TIME 15 30 15 30
INCUBATION

--
MIN. ...,~

--
3:
~ C>
FORMS: 0
;:0
» »

--
-- -
II _L>d C> -l
» 0
;:0
III Z
0
en z
I'T1

FIGURE 16. Cleavage pattern of SV40 DNA obtained by exposure to the toxic protein of
P4 and P6. Forms I, II, and III refer to the supercoil, relaxed, and linear forms of the sub-
strate.

vitro translation in a reticulocyte system. The toxins inhibited the translation and the
polysomes dissociated into monosomes as a function of time of exposure to the toxin
and as a function of toxin concentration, suggesting that the mRNA was affected by
the toxins.
Thus, the purified toxins appear to act as a nuclease that cleaves both ssDNA and
ssRNA and lacks sequence specificity. The catalytic nature of the toxins acting on both
ssDNA and ssRNA is similar to the SI nuclease of Aspergillus oryzae. 55 The gradual
relaxation of the supercoiled molecules followed by single-strand nicking is similar to
the mode of action of colicin E2.56 The mode of action in vivo may still indicate some
sequence specificity or affinity to a specific conformation. The available data provide
no indication that the toxins act on dsRNA, as was suggested originally. In the in vitro
studies, the secreted toxins are not active on dsRNA.
In the in vitro studies, the purified substrates were used and the sensitivity of the
substrates in their natural conformation, such as the ribosomal RNA in the ribosomes,
was not examined. Therefore, the current efforts are directed towards the understand-
ing of the effect of the toxins on the native state of the molecules using intact ribosomes
rather than the purified substrate. Although the toxins are quite small, it is clear that
the protein is divided into a catalytic region and a cell recognition region. The specific-
ity of these molecules seems to be in the recognition site, with only some minimal
requirements for the catalytic activity. The toxin mutants that can be complemented in
vitro retained the catalytic activity and are assumed to have lost the recognition func-
tion. However, information on the size of the toxic molecules that act in vivo is un-
available, and the inactive toxins may be nonfunctional due to an alteration of a site
that must be modified to activate the toxin in vivo such as the case of the diphtheria
toxin. 57 Only by precise mapping of the protein through the accumulation of mutations
in this protein and by gene cloning and sequencing will the complexity of these toxins
be resolved.

IV. CONCLUDING REMARKS

In the study of the virus-host interaction and the genetic organization of the fungal
dsRNA viruses, undoubtedly those systems with a phenotypic expression of the viruses
 137

and a well-defined genetics of the host are the favored systems. Thus far, the clearest
situation is the phenotypic expression displayed by the virus-encoded toxins of U. may-
dis and S. cere visia e. In both organisms the sexual cycle can be regulated and the
genetics is developed, although Ustilago genetics is not nearly as advanced as the ge-
netics of S. cerevisiae. It is these systems that can serve as models for the resolution of
the molecular biology, biochemistry, and genetics of the virus-host interaction. The
biological role of the viruses may require the study of fungal species in which some of
the host functions and the environmental interactions can be clearly defined, such
functions as the production of secondary metabolites or virulence.
In spite of the progress made since the discovery of the "killer phenomenon", many
unresolved aspects remain in the understanding of the biology of the viruses of U.
maydis. The etiology of the virus is unclear. The information on the replication of the
dsRNA viruses in general is only fragmentary and this information as related to the U.
maydis viruses is totally lacking. The transcriptase activity is characterized and its tem-
poral relation to the growth phase is known. However, the interrelations of the tran-
scription and replication are unclear, even to the extent of what serves as the template
for replication, the transcripts, or the dsRNA.
The expression and secretion of the proteinacous toxin is of a broader interest as a
problem in the inactivation of a toxic substance by the producing cell and the process-
ing and secretion of this toxic molecule. The involvement of the host in the secretion
and processing is as yet unknown; and whereas in S. cerevisiae the pretoxin interme-
diates have been identified 5s and an insight gained on the association of the cellular
secretion pathway in the secretion and expression of the toxin,59,60 in U. maydis not a
single nuclear mutation has been identified thus far that affects the processing and
expression of the toxin. Furthermore, the nature of the immunity that is beginning to
unravel in S. cerevisiae 6t is not mapped in all the U. maydis viruses. Some interactions
involving a protein-dsRNA complex are suggested by the available data, but the basic
experimental tests to resolve the nature of the immunity have not been performed. The
factors determining nuclear resistance, conceivably the receptors for the toxins, have
not been identified, although the genetic material is available and mutants resistant to
all three toxins are quite common even among the native population of U. maydis. The
requirements of this host function in the maintenance of the viruses was only recently
noticed.
The relatedness of the dsRNA segments within each virus complex and among the
different complexes has been resolved to a large extent. These studies, along with the
earlier data from S. cerevisiae and more recent data on the mutual occurrence of pop-
ulation of dsRNA molecules within individual cells in which their frequencies are af-
fected by environmental factors,62 indicate the usefulness of these systems in the study
of genetic drift within dsRNA genomes. The dynamics of the dsRNA genomes in a
nonselective environment may offer an important tool in the evaluation of the effi-
ciency of the replication mechanism and the ability of the cellular mechanisms to cope
with replication errors in RNA molecules. These systems can serve as models for the
evaluation of the epidemiological trends of viruses that cause diseases. The genetic drift
in the U. maydis dsRNA viruses is detected with relative ease and has been noticed in
a number of studies thus far.
The information content of the dsRNA segments has been only partially resolved.
The assignment of functions is partly based on the actual translation of the segments
but otherwise it is based on the loss of function in mutants and not on the loss of
specific proteins. The source of the virus-associated transcriptase is as yet unknown.
The functions encoded by segments other than the one encoding the toxin are also
unknown. It is unclear whether all the segments contain the proper regulatory signals
and open reading frames required for expression of the information contained in these
138 Fungal Virology

segments. Some of the segments may function as RNA and not in informational mol-
ecules. These aspects are currently under investigation and will require the translation
of the viral information and additional information on the sequence of the segments
which can be obtained through cloning of the viral genome.
The toxin, as a small protein (83 amino acids), with its recognition function and
catalytic activity, is an attractive biological molecule to be studied in relation to the
specific recognition. The stability of the molecule and its size should permit the fine
mapping of functions in relation to the structure of the protein and can make use of
its combined activity as a model for targeting of molecules. Furthermore, the combined
function and size allow the use of such a molecule to resolve the relation of the size to
the constraints imposed on the specificity of the protein in terms of the identification
of the cell receptors and the substrate specificity.
The benign relations of the dsRNA viruses of the fungi with their hosts consistently
raise the question of the biological role of these viruses, especially as data related to
the organization of these complexes become available. The entire system with the ge-
netic drift as a significant component in the evolution of RNA viruses conveys the
impression that these viruses should be more rare than they are and that their mainte-
nance is quite complex. It is, therefore, quite surprising to learn of the widespread
occurrence of these viruses and perhaps also some as plasmids in so many different
species. The "killer systems" may not be the ideal systems to resolve the biological
role of these viruses and the selective advantages they may confer on their hosts. How-
ever, it is the genetics, molecular biology, and physiology of these systems that provide
the means and methods to address this question in those systems in which the biological
role of these viruses may be better assessed. The fluctuations in morphology, in viru-
lence, and in the synthesis of secondary metabolites as known to occur in fungi at a
rate that cannot be accounted for by simple mutations, have led to the search for
factors such as plasmids and transposable elements. 63 It is conceivable that in part the
source of these fluctuations are the dsRNA viruses and plasmids of the fungi. With the
loss of sexual reproduction by many of the fungi and the complex genetic systems that
regulate the interaction between unlike cells64 limiting the ability even for parasexual
exchange, the rate of drift in RNA viruses may be the solution to flexibility in fungi.
The double strandedness as the preferred molecular structure may be a mere reflection
of the survival of viral infections originating from the RNA plant viruses with a strong
selection against single stranded forms by the action of cellular RNases. Therefore, the
future direction should include the examination of the relatedness of the dsRNA vi-
ruses of plant pathogenic fungi to ssRNA and dsRNA viruses and plasmids in hosts of
specific plant pathogens as a part of the effort to gain an insight into the source of the
fungal dsRNA viruses and to obtain some information that may assist in assessing the
biological role of the fungal viruses.

REFERENCES
1. Puhalla, J. E., Compatibility reactions on solid medium and interstrain inhibition in Ustilago maydis,
Genetics, 60, 461, 1968.
2. Perkins, D. D., Biochemical mutants in the smut fungus Ustilago maydis, Genetics, 34,307, 1949.
3. Holliday, R., The genetics of Ustilago maydis, Genet. Res., 2,204, 1961.
4. Sleumer, H. 0., Uber Sexualitat und Zytologie von Ustilago zeae(Bechm). Ungas., Z. Botan., 25,
209, 1932.
5. Bowman, D. H., Sporidial fusion in Ustilago maydis, J. Agric. Res., 72,233,1946.
6. Rowell, J. B., Functional role of compatibility factors and an in vitro test for sexual compatibility
with haploid lines of Ustilago zeae, Phytopathology, 45,370, 1958.
 139

7. Hankin, L. and Puhalla, J. E., Nature of a factor causing mterstrain lethalIty in Ustilago maydis,
Phytopathology, 61,50,1971.
8. Jinks, J. L., Extrachromosomal Inheritance, Prentice-Hall, Englewood Cliffs, N.J., 1964.
9. Koltin, Y. and Day, P. R., Inheritance of killer phenotypes and double stranded RNA in Ustilago
maydis, Proc. Natl. Acad. Sci. U.S.A., 73,594, 1976.
10. Day, P. R. and Anagnostakis, S. L., Corn smut dikaryon in culture, Nature (London), New Bioi.,
231, 19, 1971.
II. Day, P. R. and Anagnostakis, S. L., The killer system in Ustilago maydis: heterokaryon transfer and
loss of determinants, Phytopathology, 63, 1Ol7, 1973.
12. Peery, T., Koltin, Y., and Tamarkin, A., Mapping the immunity function of the Ustilago maydis P I
virus, Plasmid, 7, 52, 1982.
13. Finkler, A., Peery, T., and Koltin, Y., Mitotic transmission of the Ustilago virus M dsRNA IS de-
pendent on the nuclear resistance genes, Genetics, 107, 232, 1984.
14. Wood, H. A., and Bozarth, R. F., Heterokaryon transfer of virus-like particles associated with a
cytoplasmically inherited determinant in Ustilago maydis, Phytopathology, 63,1019, 1973.
15. Koltin, Y., Virus-like particles in Ustilago maydis: mutants with partial genomes, Genetics, 86,527,
1977.
16. Koltin, Y. and Kandel, J. S., Killer phenomenon in UstiIago maydis. the organization of the viral
genome, Genetics, 88, 267, 1978.
17. Koltin, Y., Mayer, I., and Steinlauf, R., Killer phenomenon in Ustilago maydis: mapping viral func-
tions, Mol. Gen. Genet., 166, 181, 1978.
18. Koltin, Y., Levine, R., and Tamarkin, A., The exclusion among the Ustilago viruses - a dsRNA
restriction modification system? Curro Genet., 2, 185, 1980.
19. Wigderson, M. and Koltin, Y., Dual toxin specificities and the exclusion relations among the Ustilago
dsRNA viruses, Curro Genet., 5, 127, 1982.
20 Vodkin, M., Katterman, F., and Fink, G. R., Yeast killer mutants with altered double-stranded
ribonucleic acid, f. Bacteriol., 117, 681, 1975.
21. Franklin, R. M., Purification and properties of the replicative intermediates of the RNA bacterio-
phage R17, Proc. Nat. Acad. Sci. U.S.A., 5,1504,1966.
22. Bozarth, R. F., The physico-chemical properties of dsRNA mycoviruses, in Viruses and Plasmids in
Fungi, Lemke, P. A., Ed., Marcel Dekker, 1979.
23. Field, L. 1., Bruenn, J. A., Chang, T. H., Pinchasi, 0., and Koltin, Y., Two Ustilago maydis viral
dsRNAs of different size code for the same product, Nucleic Acids Res., II, 2765, 1983.
24. Bozarth, R. F. and Lentz, E. T., Physico-chemical properties and serology of Ustilago maydis PI
and P4, 4th Int. Congr. Virology, The Hague, The Netherlands, p.I72, 1978.
25. Vasquez, C. and Kleinschmidt, A. K., Electron microscopy of RNA strands released from individual
reovirus particles, f. Mol. BioI., 34, 137, 1968.
26. Gomatos, P. J. and Stoeckenius, W., Electron microscopic studies of Reovirus RNA, Proc. Natl.
Acad. Sci. U.S.A., 52, 1449, 1964.
27. Koltin, Y. and Steinlauf, R., The killer phenomenon in UstiIago: electron microscopy of the dsRNA
encapsidated in individual virus particles, Arch. Microbiol., 128, 45, 1980.
28. Bozarth, R. F., Koltin, Y., Weissman, M. B., Parker, R. L., Dalton, R. E., and Steinlauf, R., The
molecular weight and packaging of dsRNA in the mycovirus from Ustilago maydis killer strains,
Virology, 113, 492, 1981.
29. Ben-Zvi, B. S., Koltin, Y., Mevarech, M., and Tamarkin, A., RNA polymerase activity in virions
from Ustilago maydis, Mol. Cell. BioI., 4, 188, 1984.
30. Holland, J., Spindler, K., Horodyski, F., Grabau, Nichol, S., and Var de Pol, S., Rapid evolution
of RNA genomes, Science, 215, 1577, 1982.
31. McFadden, J. J., Buck, K. W., and Rawlinson, C. J., Infrequent transmission of double-stranded
RNA virus particles but absence of DNA provirus in single ascopore cultures of Gaemannomyces
graminis, f. Gen. Virol., 64,927, 1983.
32. Tivoli, B., Ferault, A. C., Lemaire, J. M., and Spire, D., Aggresivite et particules de type viral in
dans haut isolates monoascopores de Gaernannornyces grarninis (Sacc.). Arx et Olivier (Ophiobolus
graminisSacc.), Ann. Phytopathol., 11, 259,1977.
33. Day, P. R. and Dodds, J. A., Viruses of plant pathogenic fungi, in Viruses and Plasmids in Fungi,
Lemke, P. A., Ed., Marcel Dekker, 1979,202.
33a. Koltin, Y. and Day, P. R., unpublished, 1975.
33b. Koltin, Y., Mayer, I., and Steinlauf, R., unpublished, 1978.
34. Koltin, Y. and Day, P. R., Suppression of the killer phenotype in Ustilago maydis, Genetics, 82,629,
1976.
35. Wickner, R. B., Plasm ids controlling exclusion of the K2 killer double stranded RNA plasmids of
yeast, Cell, 21,217,1980.
140 Fungal Virology

36. Sommer, S. S. and Wickner, R. B., Yeast L dsRNA consists of at least three distmct RNAs. evidence
that the non-Mendelian genes IHOK], rNEXI and [EXL) are on one of the dsRNAs, Cell, 31, 429,
1982.
37. Sommer, S. S. and Wickner, R. B., Co-curing of plasm ids affecting killer double stranded RNAs of
Saccharomyces cerevisiae: IHOKI, [NEX] and the abundance of L are related and further evidence
that MI requires L, J. Bacteriol., 150, 545, 1982.
38. Field, L. J., Bobek, L., Brennan, V., Reilly, J. V., and Bruenn, 1., There are at least two yeast viral
double stranded RNAs of the same size: an explanation of the viral exclusion, Cell, 31,193,1982.
39. Koltin, Y., Levine, R., and Peery, T., Assignment of functions to the segments of the dsRNA genome
of the Ustilagovirus, Mol. Gen. Genet., 178,173,1980.
40. Simon, S., Gorecki, M., and Koltin, Y., In vitro translation of the Ustilagovirus dsRNA, Genetics,
97, s76, 1981.
41. Day, P. R., Fungal virus populations in corn smut from Connecticut, Mycologia, 73,379, 1981.
42. Freid, H. M. and Fink, G. R., Electron microscopic heteroduplex analysis of "killer" double-
stranded RNA species from yeast, Proc. Natl. Acad. Sci. U.S.A., 75,4224, 1978.
43. Bruenn, J. A. and Kane, W., Relatedness of the double-stranded RNAs present in yeast virus-like
particles, J. Virol., 26,762, 1978.
44. Bruenn, J. A. and Brennan, V. R., Yeast viral double stranded RNAs have heterogeneous 3' termini,
Cell, 19, 923, 1980.
45. Brennan, V. E., Bobek, L. A., and Bruenn, 1. A., Yeast dsRNA viral transcriptase pause products:
identification of the transcript strand, Nucleic Acids Res., 9, 5049, 1981.
46. Welsh, J. D. and Leibowitz, M. 1., Localization of genes for the double stranded RNA killer virus
of yeast, Proc. Natl. Acad. Sci. U.S.A., 79,786, 1982.
47. Welsh, J. D., Leibowitz, M. J., and Wickner, R. B., Virion DNA independent RNA polymerase
from Saccharomyces cerevisiae, Nucleic Acids Res., 8,2349, 1980.
48. Bruenn, J. A., Bobek, L. A., Brennan, V. E., and Held, W., Yeast viral RNA polymerase is a
transcriptase, Nucleic Acids Res., 13, 2985, 1980.
48a. Steinlauf, R., unpublished, 1982.
49. Usala, S. J., Brownstein, B. H., and Hazelkorn, R., Displacement of parental RNA strands during
in vitro transcription by bacteriophage ,6 nucleocapsids, Cell, 19, 855, 1980.
50. Ratti, G. and Buck, K. W., Semi-conservative transcription in particles of a double-stranded RNA
mycovirus, Nucleic Acids Res., 5,3843, 1978.
51. Bevan, E. A. and Herring, A. J., in Genetics, Biogenesis and Bioenergetics of Mitochondria, Ban-
delow, W., Schweyen, R. J., Thomas, D. Y., Wolf, K., and Kaudewitz, F., Eds., de Gruyter, Berlin,
162, 1976.
52. Kandel, J. and Koltin, Y., Killer phenomenon in Ustilago maydis. Comparison of the killer proteins,
Exp. Mycol., 2,270, 1978.
53. Koltin, Y. and Day, P. R., Specificity of Ustilago maydis killer proteins, Applied Microbiol., 30,
694, 1975.
54. Levine, R., Koltin, Y., and Kandel, J. S., Nuclease activity associated with the Ustilago maydisvirus
induced killer proteins, Nucleic Acids Res., 6,3717,1979.
55. Vogt, V. M., Purification and further properties of single strand specific nuclease from Aspergillus
oryzae, Eur. J. Biochem., 33, 192, 1973.
56. Holland, I. B., Physiology of colicin action, Advan. Microbiol. Physiol., 12, 55, 1975.
57. Drazin, R., Kandel, J., and Collier, R. J., Structure and activity of Diphtheria toxin, J. BioI. Chem.,
246, 1504, 1971.
58. Bussey, H., Saville, D., Greene, D., Tipper, D. J., and Bostian, K. A., Secretion of Saccharomyces
cerevisiaekiller toxin: processing of the glycosylated precursor, Mol. Cell BioI., 3,1362, 1983.
59. Novick, P. and Scheckman, R., Secretion and cell surface growth are blocked in a temperature sen-
sitive mutant of Saccharomyces cerevisiae, Proc. Natl. Acad. Sci. U.S.A., 76, 1858, 1979.
60. Wickner, R. B. and Leibowitz, M. J., Two chromosomal genes required for killing expression in
killer strains of Saccharomyces cerevisiae, Genetics, 82, 429, 1976.
61. Bussey, H., Sacks, W., Galley, D., and Saville, D., Yeast killer plasmid mutations affecting toxin
secretion and activity and toxin immunity function, Mol. Cell BioI., 2, 346, 1982.
62. Weslowski, M. and Wickner, R. B., Two new dsRNA molecules showing non-Mendelian inheritance
and heat inducibility, Mol. Cell BioI., 4, 181, 1984.
63. Shapiro, J. A., Mobile Genetic Elements, Academic Press, New York, 1983.
64. Burnett, J. H., Mycogenetics, John Wiley & Sons, New York, 213,1975.
65. Puhalla, J. E., unpublished data, 1971.
66. Finkler, A., Perry, T., and Koltin, Y., unpublished data, 1984.
67. Bozarth, R. F., unpublished data, 1979.
68. Day, P. R., unpublished data, 1975.
 141

69. Levine, R. and Koltin, Y., unpublished data, 1980.

70. Bostian, K. A., Elliot, Q., Bussey, R., Burn, Y., Smith, A., and Tipper, D. J., Sequence of the
preprotoxin dsRNA gene of type I killer yeast: multiple processing events produce a two-component
toxin, Cell, 36, 741, 1984.
71. Skipper, N., Thomas, D. Y., and Lau, P. C. K., Cloning and sequencing of the preprotoxin-reading
region of the yeast MI double-stranded RNA, EMBO JOUT., 3, 107, 1984.
72. Perry, T. and Koltin, Y., unpublished data, 1984.
73. Perry, T., unpublished data, 1984.
 143

Chapter 4

HYPOVIRULENCE OF ENDOTHIA (CRYPHONECTRIA) PARASITICA

AND RHIZOCTONIA SOLANI

Neal K. Van Alfen

TABLE OF CONTENTS

1. Introduction ................................................................................. 144

II. Hypovirulence of E. parasitica ......................................................... 144

A. Correlative Evidence that dsRNA Is the Cause of
Hypovirulence ..................................................................... 146
B. Transfer of dsRNA Results in Transfer of Hypovirulence .............. 148
C. Loss of dsRNA Results in Loss of Hypovirulence ........................ 148
D. Cell-Free Infection by dsRNA ................................................. 149

III. Hypovirulence of R. solani ........ ...................................................... 149

IV. Nature of dsRNA Associated with Hypovirulence ................................. 150

V. Virulence Expression ...................................................................... 154

A. Host-Pathogen Relations of E. parasitica...... ............................. 155
B. Mechanisms of Virulence Reduction ......................................... 156

VI. Summary ..................................................................................... 158

Acknowledgment ................................................................................... 159

References ............................................................................................ 159

144 Fungal Virology

I. INTRODUCTION

Plant pathogens are organisms that have evolved to occupy specialized niches. The
complexity of extracting nutrients from living hosts is such that we can assume that a
significant portion of the pathogen's genotype is concerned, at least peripherally, with
pathogenicity. Although little is currently known on the subject, it is likely that there
are both structural and regulatory pathogenicity genes. I The invasion of a plant, as in
any developmentally regulated process, probably involves a control of expression of
specific genes at appropriate times. The expression of these pathogenicity genes can be
quantitated, with the relative amount of damage caused to the host being a reflection
of virulence. Low virulence, or hypovirulence, may result from mutations of one or
more pathogenicity genes. Given the potentially large number of pathogenicity genes,
such mutants should be common, yet not be expected to survive in the population of a
pathogen. There are, however, some pathogens that have stable reproducing popula-
tions of hypovirulent variants. The best studied hypovirulent variants are those of the
plant pathogen Endothia parasitica (Murr.) and (Cryphonectria parasitica [Murr.]
Barr), the causal agent of chestnut blight. These hypovirulent forms were first discov-
ered because they were becoming the predominant form in certain localities in Italy,
where damage caused by the pathogen in chestnut plantations was significantly re-
duced. Z These hypovirulent isolates were shown by Grente and Sauret 3 to be dominant
when mixed in a tree with normal virulent isolates. This dominance was found to be
due to transmission of cytoplasmic genes from hypovirulent to virulent isolates.' Hy-
povirulence of E. parasitica thus survives in the natural population of this fungus be-
cause it is transmitted much as a disease is.
As is frequently the case, once a phenomenon is discovered in one organism, other
examples can be found elsewhere. Hypovirulence, or something similar, has been de-
scribed since that time in the plant pathogens Rhizoctonia solani,5 Ceratocystis ulmi, •
and Gaeumannomyces graminis var. tritici. 7 The hypovirulence of E. parasitica and R.
solani have been correlated with the presence of dsRNA 8,' and will thus receive pri-
mary emphasis in this review. For a discussion of dsRNA in C. ulmiand G. graminis
var. tritici, see Chapters 7 and 8 of this book. Hypovirulence in E. parasitica has been
extensively reviewed recently, 10- I. so no attempt will be made to cover the history of its
discovery or to discuss in any detail the attempts to use it in biological control. The
primary emphasis will be the review of current information concerning the biology of
the dsRNA and how it may affect control of virulence expression in fungi.

II. HYPOVIRULENCE OF E. PARASITICA

Sampling the natural population of E. parasitica hypovirulent (H) isolates in Eu-

rope, Grente and Sauret'S identified three major groups of H isolates based upon their
colony morphologies. They designated these three types B, JR, and V* (we have added
an asterisk to their V* type to distinguish it from the commonly used designation of
wild-type virulent isolates as V). B type has much less pigment than virulent (V) forms
of E. parasitica. It also forms very few pycnidia and then only after extended periods.
This is unlike V forms which produce abundant pycnidia in culture within 5 to 10 days.
The JR type of hypovirulence segregates in culture from the B type. Although Grente
did not find this type in nature, it has been isolated by others in Italy. '6 JR type has
pigmented (orange) hyphae and produces conidia from conidiophores on the hyphae
rather than in pycnidia. The third hypovirulence type is not a clearly defined group. It
has been named V* type as well as intermediate,'7 and appears to be a heterogeneous
group of various morphologies. Upon subculture the V* type and B type both are
unstable, segregating to V*, JR, and B types as well as the normal V type in culture.
 145

The V and JR types, on the other hand, are stable upon subculture. IS The predomi-
nance of the three H types in populations of the fungus in Europe was confirmed by
Turchetti. '6
The discovery of morphological markers of hypovirulence has proven to be a great
boon to the study of this phenomenon. Without these genetic markers, progress in
understanding the nature of hypovirulence would have been very slow. With these
markers, it is not necessary to test the virulence of each isolate after every laboratory
manipulation. One of the first discoveries made with the use of these morphological
markers was how H isolates were able to protect chestnut trees from attack by V forms
of the fungus. Using these markers for hypovirulence, Grente demonstrated that
strains of the fungus changed from V to H forms.' He also relied on them to demon-
strate the conversion of V strains to H in culture. When a V and an H strain were
paired on an agar plate and allowed to grow, the V (orange) acquired the phenotype
of the H strain (white). From these studies, Grente and Sauret 15 concluded that hypo-
virulence was an infectious cytoplasmic agent. This hypothesis was confirmed by using
nuclear auxotrophic markers, a lys-H strain and a met-V strain. Pairing the two
strains, both in the host tree and in culture, resulted in transfer of the H phenotype
into the met- strain. The transfer characteristics were as expected for movement of a
cytoplasmic element from the lys- strain to the met- strain. The fact that all recovered
met- isolates had the H phenotype confirmed that hypovirulence was controlled by
transmissible cytoplasmic gene(s).4
Once researchers in North America became aware of the discovery of hypovirulence
in Europe, H isolates of E. parasitica were sought and found on this continent. These
isolates were not as widespread nor as effective in controlling the chestnut blight as
were those in Europe. ,o ,17,2o As with the European H isolates, hypovirulence in North
America was associated with morphological markers that made cultural identification
possible. The morphological markers of the North American strains, however, differed
from those of Europe. The European B, JR, and V* types were not found. The mor-
phologies of North American H strains are too diverse to allow grouping of them into
just a few different types as has been done with the European H strains. The colonies
of North American H types differ from V types generally by the morphology of the
hyphae and growth rates in culture. No pigment difference between H and V types has
been found in North American isolates. The H types are recognizable in culture pri-
marily because the commonly used wild-type virulent colonies are so uniform in their
cultural morphology. 21
As defined by the term hypovirulence, all H isolates have as a common property a
reduction in virulence expression. Virulence, unfortunately, is a property that is diffi-
cult to define, primarily because it is measured relative to standard isolates. Virulence
is considered a unique property of a specific pathogen, and in a pathogen population
there is usually a continuum of virulence expression ranging from highly virulent to
avirulent. Most workers studying E. parasitica have used relative growth rates of the
fungus in chestnut stems as a measure of virulence. This has proven to be a reasonable
assay because of the uniformity of growth rates in host trees of normal isolates of the
fungus. Hypovirulent strains generally grow more slowly in trees. Elliston 22 reported
such a continuum of virulence expression by dsRNA containing isolates of E. parasi-
tica. However, he felt that other parameters of virulence expression should also be
considered in the definition of hypovirulence because some dsRNA-containing, and
presumably hypovirulent, strains grew in stems at rates comparable to normal iso-
lates. 22 Thus, in some virulence tests of E. parasitica, sporulation is also considered in
assessing whether an isolate is virulent or hypo virulent.
Most, but not all, H strains of E. parasitica can transmit their hypovirulence to
normal strains by hyphal anastomosis.23 Other forms of hypovirulence, presumably
146 Fungal Virology

involving mutations of nuclear genes of the fungus, have also been identified. In these
strains hypovirulence is not transmissible by hyphal anastomosis. There are, therefore,
two forms of hypovirulence known, transmissible hypovirulence and nontransmissible
hypovirulence.
The persistence of hypovirulence in natural populations of E. parasitica is certainly
the result of its transmissible nature. In this respect it is a disease of the virulent pop-
ulation. In the study of the etiology of this disease, as with any other, it is important
to ascertain that the disease is caused by a single agent. This is a formidable task, since
fungi tend to accumulate cytoplasmically transmissible agents. Thus, we cannot be
assured that the transmission of hypovirulence from strain to strain is due to the trans-
fer of a single genetic agent. It has not yet been possible to confidently obtain normal
isolates of E. parasitica that have uniform cytoplasmic genes. The task of obtaining
single hypovirulent agents in a background of uniform normal cytoplasmic genes is
even more difficult. These considerations would not be as important if cell-free infec-
tivity assays were available. Unfortunately they are not. In light of these limitations,
we cannot yet conclude that a single agent is responsible for the hypovirulent pheno-
type.

A. Correlative Evidence that dsRNA Is the Cause of Hypovirulence

The best-studied cytoplasmically transmissible agents in fungi are mycoviruses.
Characteristics of some H-type cultures, which include poor growth and frequent sec-
toring, are indicative of the effects of mycoviruses on other fungi. This led researchers
to search for the presence of mycoviruses in H strains. The first indication that dsRNA
was associated with hypovirulence was Moffitt and Lister's24 detection of dsRNA in
two H isolates from France using dsRNA specific antiserum. They did not detect
dsRNA in a V isolate from France nor in one from Michigan. Their attempts to isolate
either the dsRNA or a mycovirus from culture extracts were not successful. 24 Isolation
of dsRNA from H but not V strains was reported by Day et al. 8 These workers discov-
ered that the dsRNA had a segmented genome, and that the number and size of these
dsRNA segments varied from strain to strain. Dodds" later reported that he recognized
three different dsRNA segment patterns, two of which were common to the European
H strains and one of which occurred exclusively in North American strains. He found
a total of 12 different dsRNA segments that could be identified by polyacrylamide gel
electrophoresis, varying in molecular weight from 4.3 to 6.2 MDa. However, no seg-
ment was common to all three groups. This attempt at making sense out of the com-
plexity of segment numbers and sizes was premature since it is now clear that dsRNA
segment diversity is more complex than was recognized at that time. This will be dis-
cussed in greater detail later.
The first report that dsRNA is present in H strains but not V strains was greeted
with the expectation that this finding answered the question concerning the nature of
the cytoplasmically transmissible hypovirulent genome. Although evidence is strong
that the dsRNA is responsible for hypovirulence, it is not yet conclusive. There remain
enough inconsistencies in the correlative data for dsRNA association with hypovirul-
ence that we must remain open to other possibilities. In addition to the few strains that
do not show correlation of dsRNA with hypo virulence, there is a problem reconciling
the large diversity of dsRNA segments with hypovirulence. It appears that there is
either a highly conserved sequence responsible for hypovirulence in an otherwise unsta-
ble genome, or that any dsRNA in E. parasitica can cause hypovirulence. There are
not good precedents in biology for either of these possibilities, so care must be exer-
cised in drawing premature conclusions concerning the cause of hypovirulence in E.
parasitica.
In the initial study of Moffit and Lister 2 • associating dsRNA with hypovirulence,
 147

Table 1
RELATIVE VIRULENCE OF NORMAL AND
dsRNA-CONT AINING STRAINS OF
E. PARASITICA

Relative virulence"
Strain Origin" dsRNA (OJo) Ref.

RCI NA + 4 27
GH2 NA + 14 20
GH5 NA 34 20
GH6 NA + 44 20
GH7 NA 60 20
GH8 NA + 30 20
GHI4 NA + 31 20
GHU2 NA + 36 20
GHU3 NA + 25 20
GHU4 NA + 2 20
GHIB NA 57 20
GHA NA 38 20
CLI NA 109 20
CL2 NA 119 20
CL4 NA 68 20
EP351 NA 182 20
EP47 + 12 22
EP60I NA + 12 22
EP66 + 14 22
EP90 NA + 17 22
EP50 + 17 22
EPI20 + 18 22
EP9 F + 23 22
EP93 NA + 23 22
EP64 I + 30 22
EP51 + 30 22
EP88 NA + 30 22
EP63 + 30 22
EP48 + 30 22
EP90 NA + 38 22
EP92 NA + 41 22
EPI02 NA + 44 22
EP46 94 22
EP49 I + 98 22
EP29 NA 101 22
EPI03 NA + 105 22

Locations where strains were collected. If the strain is a conversion

product, the source of the H strain is given. F, France; I, Italy;
NA, North America.
Relative virulence is expressed as the percentage of average viru-
lence of the normal control strains used in each experiment. Viru-
lence was determined by measuring growth of the fungus in chest-
nut stems.

only two H and two V strains were used. Day et a1. 8 confirmed this association of the
dsRNA with hypovirulence using a total of 28 H and V strains. Since that time, many
more studies have been conducted to confirm the association of dsRNA with hypovi-
rulence. Many of these studies have not been published since it is generally accepted by
all workers in this field that H strains contain dsRNA. Table 1 summarizes the findings
of a couple published studies. This table shows that there is a continuum of virulence
expression in E. parasitica, with dsRNA being associated with most of the less virulent
148 Fungal Virology

strains. As in any continuum, there is a point at which some dsRNA-containing and

dsRNA-free strains are indistinguishable in virulence expression. Such is the case with
EP 49 and EP 103, which contain dsRNA and are virulent. Van Alfen et al. 26 found
that EP 49, like V* strains, is unstable, with some single conidial isolates resulting in
typical European B type colonies. These are much less virulent than the parental EP-
49 strain. This segregation of EP 49 into a less virulent B type suggests that either EP
49 has a mixed infection of different hypovirulence-causing agents or perhaps there
may be a suppressor of hypovirulence in EP 49. 26
Although the correlation between dsRNA content and hypovirulence is very good, it
remains a correlation rather than direct evidence of the role for dsRNA in hypovirul-
ence expression. Fortunately, two other pieces of correlative evidence strengthen the
hypothesis that dsRNA is the cause of hypovirulence in E. parasitica. First, when hy-
povirulence is cytoplasmically transmitted, dsRNA is always transferred into the pre-
viously virulent strain. 23 Second, when dsRNA is cured from a hypovirulent strain, the
strain reverts to the virulent phenotype. 27

B. Transfer of dsRNA Results in Transfer of Hypovirulence

As indicated previously, one of the early clues that hypovirulence was cytoplasmi-
cally transmitted came from the conversion of a V isolate to an H isolate when both
were placed side-by-side in a petri dish. The V isolate under these conditions, starts
growing as a typical V isolate, but after a short time the growing edge assumes the H
phenotype. Anagnostakis and D ay23 used this general method to follow not only con-
version of V to H isolates, but transfer of dsRNA from the H isolate into the V isolate.
In all the colonies that were examined, conversion to H type resulted in transfer of
dsRNA into the previously V colony. They found, however, that the dsRNA from H
strain EP 113 was not always faithfully transferred. This strain contains dsRNA that
is segmented into 5 easily detectable pieces, yet the number of segments that were
detected after transfer into the V strain varied as a function of the strain being tested.
This suggests that the host fungus affects maintenance of particular dsRNA strains.
Regardless of which segments were transferred or whether new ones appeared, all col-
onies that contained any dsRNA had an appearance typical of the B type of hypovi-
rulent colony. In another study, Anagnostakis 28 repeated this basic experiment with
the addition of virulence tests in chestnut trees to confirm that the dsRNA-containing
strains were actually hypovirulent in the tree. She reported that all segments of the
dsRNA from EP 113 were uniformly transferred, even to the V strains that in the
previous study never supported faithful transfer of all dsRNA segments. The reason
for the discrepancy between the two experiments is not clear. It is perhaps the result of
uneven transfer of some other cytoplasmic elements. The existence of other cyto-
plasmic elements was suggested in the second study,>8 since the dsRNA was always
faithfully transferred, yet colony morphologies varied in the recipient strains after
transfer. Even when duplicate transfers of the same strain were conducted, colonies
resulted that differed in morphology. DsRNA from these colonies showing different
morphologies were always identical, as determined by electrophoretic banding pat-
terns. Dodds" also reported variability in detection of all dsRNA segments of this
strain upon subculture.

c. Loss of dsRNA Results in Loss of Hypovirulence

Fulbright" was able to cure dsRNA from H strains by growing these H strains on a
nutrient medium containing cycloheximide. Those strains that were cured of dsRNA
showed an increase in virulence, while those strains not cured of the dsRNA remained
hypovirulent. He also reported that the dsRNA segment pattern was altered in a few
cases, resulting in restoration of colony appearance to normal but with virulence re-
 149

maining low. Fulbright's results are significant because they not only correlate dsRNA
with virulence, but also indicate that dsRNA segment changes can result in phenotypic
changes in the fungus. This indication of phenotypic changes tied to dsRNA changes
should be carefully reproduced to determine if the changes are tied to a specific dsRNA
segment.

D. Cell-Free Infection by dsRNA

All of the experimental strategies to correlate dsRNA with hypovirulence suffer from
the fact that other cytoplasmic elements could possibly be present and be responding
to experimental procedures in the same manner as dsRNA. Thus, even though the
correlative evidence for the role of dsRNA in hypovirulence is very strong, it is not
conclusive. The infection of E. parasitica, with purified dsRNA or viral particles con-
taining dsRNA must be done to eliminate the possible role of other cytoplasmic ele-
ments.
Cell-free infection of fungi with fungal viruses has met with limited success. Chap-
ters 1, 5, and 8 in this book review both the success and failures of these attempts. Our
laboratory has been attempting cell-free infection of E. parasitica with dsRNA for a
number of years, using protoplasts of the fungus. 3o Protoplasts are obtained from a V
strain and then various methods have been used to introduce dsRNA into the proto-
plasts. We found that preparation and regeneration of protoplasts of V and H strains
of the fungus is easily accomplished. Since the dsRNA associated with at least some H
strains is not packaged in viral capsids, but rather in fungal vesicles (see below), we
have been using these naturally occurring membrane vesicles to fuse with fungal pro-
top lasts as a means of introducing the dsRNA into V strains. The only nucleic acid
present in the vesicles is dsRNA. Thus, we feel that these vesicles are ideal delivery
vehicles for introducing dsRNA into a V strain. We also have evidence that the vesicles
contain an RNA polymerase, so enzymes necessary for replication probably are present
within the vesicles. At this stage in our experiments we have evidence that dsRNA can
be transferred into V isolates and that the V isolates are transformed to the H-type
colony morphology (B type of hypovirulence is being used). However, the dsRNA is
not maintained beyond two colony transfers. 31 These preliminary, yet positive, results
suggest that the colony phenotype associated with the European B type of hypovirul-
ence can be successfully transmitted with the dsRNA. We are attempting to improve
maintenance of the dsRNA within the infected strain so that virulence tests can be
conducted. Our results are encouraging since they indicate that cell-free infection of E.
parasitica may be possible.

III. HYPOVIRULENCE OF RHIZOCTONIA SOLANI

The term hypovirulence has been applied to a cultural decline observed in various
isolates of the plant pathogen R. solani.s Some strains of this fungus were found to
degenerate with time when placed into culture. This degenerative disease is character-
ized in the severely diseased strain 189a by: (I) a reduction in the amount of pigment,
(2) an irregular appearance to the culture, (3) reduced growth rate, and (4) production
of few or no sclerotia. 32 The level of disease expression within cultures of this strain
varied with subculturing and location within the culture. Disease-free hyphal tip sub-
cultures could be obtained. These characteristics suggested that a cytoplasmic disease-
causing agent was present within the degenerating cultures. Castanho and Butler"
showed that this disease could be reintroduced by hyphal anastomosis to cured hyphal
tip subculture strains of strain 189a. However, the disease could not be transferred to
other strains within the same anastomosis group.
Virulence of the diseased isolate, 189a, was compared with that of its disease-free
150 Fungal Virology

hyphal tip subculture, 189HT5. The diseased isolate, 189a, killed 12070 of the cabbage
seedlings used in a petri dish virulence test while 189HT5 killed 100070 of the seedlings.
Other types of virulence tests in soil confirmed these results: 189a was essentially avi-
rulent. 5 It was also shown that mixing the two cultures together substantially reduced
the amount of disease compared with that caused by 189HT5 alone. It appeared that
189a was acting much as H strains of E. parasitica, and so the diseased strains of R.
solani were also considered to be hypovirulent. 5 The primary difference between H
strains of E. parasitica and R. solani was that transmission of the disease agent in the
latter was very restricted, being limited to genetically identical individuals. The poten-
tial for practical disease control is also more limited than with E. parasitica since the
diseased strain of R. solani did not survive more than a month in soil. 5
As with hypovirulence of E. parasitica, it was reported that diseased strains of R.
solani contained dsRNA, while healthy ones generally did not. 9 Castanho et al! did
report that dsRNA was occasionally found in the disease-free strain 189HT5. A total
of 14 strains of R. solani were examined, and with the exception mentioned above, it
was found that the other 10 healthy strains did not contain dsRNA while the three
diseased strains did. The dsRNA segments found in the diseased strains were electro-
phoretically unique compared with one another. Viral particles were not detected nor
could they be isolated. On the basis of their correlation of dsRNA with disease, they
postulated that the dsRNA was responsible for the degenerative disease, and thus hy-
povirulence of R. solani. 9
Research in two other laboratories that have further pursued these preliminary stud-
ies have shown that the earlier conclusions may have been premature. Zanzinger et al. 33
reported that 49 of 50 isolates of R. solani examined contained dsRNA. These isolates
ranged from being highly virulent to being essentially avirulent. They concluded that
presence of dsRNA did not correlate with hypovirulence expression by R. solani. This
same conclusion was reached by workers in the laboratories of Y. Koltin and co-work-
ers at Tel-Aviv University.·2.• 3 They have indicated that dsRNA has just the opposite
correlation with virulence expression, i.e., they have evidence of the association of
specific dsRNA segments with augmentation of virulence rather than hypovirulence.
They feel that much of the confusion concerning dsRNA association with R. solani
hypovirulence is due to non-dsRNA artifacts that are isolated along with dsRNA in the
standard method utilizing CF-ll chromatography. These artifacts are seen as high mo-
lecular weight bands on agarose or polyacrylamide gels and are probably DNA. Fur-
thermore, dsRNA has now been detected in virulent isolate 189 by Finkler et al.,63 and
DNA plasmids were detected in two hypovirulent, but not in two virulent, isolates of
R. solani.'· It is clear that further investigations, preferably involving cell-free infectiv-
ity studies, are required to resolve the roles of dsRNA and DNA plasmids in virulence
attenuation or virulence expression in R. solani.

IV . NATURE OF dsRNA ASSOCIATED WITH HYPOVIRULENCE

Superficially, the dsRNA present in H strains of E. parasitica seems typical of the

genome of viruses found in a large number of fungi. The dsRNA in H strains has
multiple segments, is noninfectious in the traditional sense, and is in the approximate
size range of dsRNA found in most fungal viruses that have been characterized. An
important difference is that no viral particles have yet been isolated. Also, the multiple
segments of dsRNA found in H strains, despite considerable effort, have not been
classified into electrophoretic patterns that could represent specific viral types. To
date, we cannot with any certainty identify a specific fungal virus or dsRNA segment
associated with hypovirulence.
The first report of electrophoretic detection of dsRNA in H strains indicated that a
 151

E. parasitica strain
A B c D E F G H J

6.0

5.0

'"0,....

-
x 4.0
.r::
C)
·iii
~
... 3.0
~
~
CJ
CD
'0
:!! 2.0

1.0

o
FIGURE I. Approximate molecular weights of dsRNA segments found in various hypovirulent strains
of E. parasitica. The dsRNA in the columns are from the following strains: A, Type I dsRNA"; B, EP
713 (which has changed in culture from a Type I pattern to that shown); C, Type II pattern"; D, Type
III pattern"; E, strain GH-2; F, strain GHUf; G, strain R-I; H, strain RF-I; I, strain Coli 5; J, strain
Aid T5M (FF-I). DsRNA depicted in columns E through J are from North American H strains collected
in Michigan."

3.3 X 106 Da segment was common to all H strains. No other segments were consistently
associated with hypovirulence.· Dodds 25 later revised the molecular weight estimates of
the dsRNA and reported that the previously reported 3.3 x 10 6 Da segment was not
common to all H strains. Fulbright et al.,'o in a study of H strains in Michigan, found
that dsRNA segments varied in size, number, and concentration. The segment patterns
differed from other North American strains tested by Dodds. Figure 1 summarizes the
dsRNA segment patterns that have been reported to date in H strains of E. parasitica.
The molecular weights listed are only approximations since little physical characteri-
zation of dsRNA from H strains has been done. Most estimates are based upon co-
electrophoresis with standards.
Variability of dsRNA segment patterns has been reported in a number of dsRNA
152 Fungal Virology

viruses. 34 -36 However, in these cases, changes that occur are relatively minor in other-
wise consistent segment patterns. The inability to identify uniform dsRNA segment
patterns in H strains is confusing. It suggests that either the dsRNA is very unstable in
E. parasitica, the particular segment pattern being a function of which fungal strain
the dsRNA infects, or that many different dsRNA viruses are present in H strains. The
evidence for the first possibility is conflicting. As indicated earlier, Anagnostakis and
Day" in one study found that dsRNA from EP-l13 upon transfer by hyphal anasto-
mosis to different E. parasitica strains was unstable, resulting in frequent variations in
segment pattern. In another study!8 using this same dsRNA-containing strain, such
instability was not found. Generally, all segments of dsRNA from H strains are trans-
mitted intact from strain to strain.
There is precedent in other fungi for mixed viral infections resulting in variations in
dsRNA segment patterns!3 This was found to be the case with G. graminis var. tritici,
which contains dsRNA that is present in a variety of segmentation patterns. 37,38 The
evidence in E. parasitica that mixed infections occur is based upon association of phen-
otype with different dsRNA segments, rather than upon the identification of separate
distinct viral particles. Fulbright" found that he could cure E. parasitica of some
dsRNA segments and thereby affect the phenotype of the culture. This would suggest
that there had been a mixed infection. Elliston'8 has genetic evidence of mixed hypo-
virulence factor infections. He demonstrated that one H strain contained at least two
different cytoplasmic agents able to affect hypovirulence expression. He isolated single
conidia and found that these conidia yielded three different colony types, one of which
was wild type, and another the same as the parental. Elliston postulated that the paren-
tal strain (EP 60) contained two agents which could be separated using conidia. It was
assumed that the wild-type segregant no longer contained any hypovirulence agents
and the new colony type was assumed to be a segregant containing one of these agents.
A fourth type containing the other postulated single agent should have been found,
but was not. Perhaps it is indistinguishable in phenotype from the parental type. Al-
though genetic differences have been detected among these segregants, no reports con-
cerning differences in dsRNA segment patterns among them have appeared to date.
The correlation of specific segment patterns to these different phenotypes would be a
major step toward demonstrating both the role of dsRNA in hypovirulence and the
existence of different types of dsRNA agents.
Nucleic acid hybridization techniques are just now being used to sort out the rela-
tionship between different H strains and different segments of dsRNA within a partic-
ular H strain. L 'Hostis et al. 65 5' -end-labeled fragments of total denatured dsRNA
from the French H strain EP 713 and the American strain EP 915 and used these as
probes in a dot hybridization study of homology of the probes with seven other H
strains. They found that the probes hybridized strongly with the homologous RNA,
and that EP 713 showed homology with dsRNA from other European H strains, but
not with those from North America. Likewise, the dsRNA probe from the North
American strain EP 915 hybridized only with itself and other North American strains.
These results suggest that the dsRNA found in H strains collected from North America
is different in sequence from that of H strains collected in Europe. These same conclu-
sions have been supported by recent work in Michigan. 66 The implication of these data
is that there is no single RNA sequence or gene that is responsible for hypovirulence
expression in E. parasitica. Apparently at least two different dsRNA viruses, lacking
genomic homology, can cause hypovirulence in this fungus.
Some of the dsRNA segment diversity within strains could be the result of deletion
mutations or sequence reiterations in other segments. If so, segments within one strain
should show homology in hybridization studies. In an attempt to determine if there is
 153

homology among segments of a single H strain, a small segment of dsRNA from EP

713 was cloned by Drs. R. E. Rhoades and S. A. Ghabrial's groups in Kentucky.67
They prepared cDNA from total dsRNA of this strain, and from the clones of the
cDNAs they identified a short insert (approximately 200 bp) that could be used as a
hybridization probe. They hybridized this probe against Northern blots of the dsRNA
of EP 713 and found that the probe hybridized equally with all of the segments. Since
the probe was limited in size, it is probably premature to speculate concerning the
implications of these findings, but they do indicate that some sequences are common
in all of the segments of EP 713. No homology was detected between this probe and
dsRNA from American H strains, but there was some homology with dsRNA from
other European H strains.
The hybridization studies that have been done to date have suggested that the
dsRNA from European H strains is not homologous with that from North American
H strains. If these preliminary data are confirmed, there must be at least two different
dsRNA types associated with hypovirulence, differing both in RNA sequence and in
phenotypic expression. One of these dsRNA types is of European origin and the other
is North American. The data of Fulbright>o and Elliston" suggest that within these two
general types considerable variation in hypovirulence expression can occur. Fulbright's
data 27 in particular suggest that H phenotype is a function of which dsRNA segment is
present.
One of the implications of the mixed viral infection hypothesis is that either there is
a highly conserved hypovirulence sequence present in all dsRNA entities infecting E.
parasitica., or any dsRNA entity can cause hypovirulence in this fungus. The results to
date suggest that the latter may not be true since, in at least one case, hypovirulence
phenotype is a function of the dsRNA segments present. 27 However, hybridization
studies indicate that more than one hypovirulence sequence has evolved in dsRNA
infecting E. parasitica.
One of the assumptions under which we have been operating is that the dsRNA
associated with hypovirulence is the genome of a virus. No one has yet isolated parti-
cles from H strains that resemble other known fungal viral particles, although there
have been many attempts to do so. Dodds" reported the isolation of a particulate
fraction from EP 113 that he felt could be a viral particle. Negative stains of his puri-
fied preparations showed a club-shaped particle that resembled those that had been
described from Agaricus bisporuS. 19 Dodds" found that most of the dsRNA within EP
113 was contained within the particle and that such club-shaped particles were not
found in V strains. Using Dodd's methods, we have characterized these particles and
found that they do not resemble any known viral particles. 68 They are probably fungal
vesicles that contain the dsRNA. Although some protein can be detected in these par-
ticles, the total protein content in the vesicles is about half the dsRNA content. There
is no evidence for a protein capsid associated with the dsRNA. The major component
of the particle is chloroform-methanol soluble and presumably lipid. Also present in
substantial amounts is carbohydrate of the same neutral sugar composition as the cell
walls of the fungus. The best evidence that these particles are of fungal vesicle origin
is that similar vesicles are found in V strains. The vesicles from V strains contain no
dsRNA and less lipid than those from H strains, but otherwise are quite similar. The
vesicles from H strains can be separated into two fractions on the basis of their density.
The more dense fraction contains most of the dsRNA, although small amounts can be
detected in the less dense fraction. RNA polymerase activity has been found in vesicles
containing dsRNA, but not in those lacking dsRNA.
Electron microscope studies of H and V strains have revealed that H strains of Eu-
ropean origin contain aggregates of particles associated with the endoplasmic reticulum
that are not present in V strains.40 Newhouse et al. 41 using RNA-specific stains showed
154 Fungal Virology

that these particles contain RNA. The particles appeared to be membrane bound and
thus may be the same particles isolated by Dodds. 29 Ellzey et al. 69 also reported the
presence of membrane-bound particles associated with the endoplasmic reticulum in H
but not V strains of the fungus. The particles were 30 to 80 nm in diameter. Both of
these studies were of European H strains. Newhouse and McDonald 70 have been unable
to find comparable aggregates of membrane-bound bodies in thin sections of American
H strains but did observe structures which resemble viral particles scattered throughout
the hyphae. These particles appear to be membrane bound.
The dsRNA associated with H strains probably exists without a capsid within the
fungal cell. The dsRNA in European strains appears to be packaged within fungal
vesicles. All known attempts to isolate virus-like particles from American H strains
have failed. If the dsRNA is the genome of a virus, it is probably a defective virus.
Considering the lack of homology between dsRNA from European and North Ameri-
can H strains, it is interesting that dsRNA from both regions may exist naked within
the fungus. The coincidence is striking since defective fungal viruses are not common.

V. VIRULENCE EXPRESSION

One of the unusual aspects of the presence of dsRNA in E. parasitica is that it

invariably seems to affect virulence expression and sporulation. Other phenotypic
markers associated with the presence of dsRNA such as colony pigmentation, growth,
and morphology are much more variable. Given the apparent lack of homology be-
tween dsRNA of strains collected in Europe and North America, the consistent effect
of dsRNA on virulence expression is unique. If this lack of homology is confirmed, it
indicates a separate evolution toward similar effects on virulence and sporulation of
the fungus. Convergent evolution, while not unusual, does not usually occur toward a
character that would negatively affect the ability of the organism to survive. That it
apparently has occurred in the interaction of dsRNA with E. parasitica may reflect
how virulence and sporulation are regulated in the fungus rather than reflect any ad-
vantage provided by the dsRNA by its ability to affect fungal phenotype. There is
evidence that virulence expression by this fungus is easily perturbed even when dsRNA
is not present. As background to understanding how virulence may be regulated in
fungal plant pathogens, the genetics and biochemistry of virulence expression will be
briefly discussed.
The process of invading and colonizing a host involves a number of distinct stages.
In its attempt to colonize and extract nutrients from a living plant, a pathogen must
first locate or recognize potential entry points into the plant. To gain entry, the path-
ogen must breach constitutive host defenses which may be either physical or chemical.
In addition, it must utilize nutrients that are available in the entry-court. From the
entry-court, the pathogen must penetrate into living tissue. Most pathogens are fairly
specific as to which plant tissue they colonize, so some recognition phenomenon must
be involved to detect tissues. During this stage of colonization the pathogen must cope
with an array of reactive host defense responses and, once successful, be able to dis-
tribute its propagules to other host plants. 42
Pathogen products, generally enzymes or toxins, have been shown to play roles in
this multi-step process of pathogenicity. A pathogen product is considered to be a
virulence factor if loss of that product results in loss of virulence. Most of the products
that have been rigorously demonstrated to be virulence factors are those that are easily
identifiable, such as host-specific toxins. Although most effort, and thus success, in
the study of virulence mechanisms has been directed toward identification of products
that stress the host, other genes without such readily identifiable products must also be
involved in virulence expression. Presumably, mutants of these hypothetical genes
 155

would as likely be avirulent as mutants of genes for host-specific toxin production.

Mutation to avirulence can readily occur in pathogens, but it is normally only a labo-
ratory phenomenon. In the case of E. parasitica, it is not unreasonable to assume that
a significant portion of its genome is involved in virulence expression. Thus, there are
potentially many genes or their products that could be perturbed sufficiently to reduce
virulence of this pathogen.

A. Host-Pathogen Relations of E. parasitica

Relatively little is known about how E. parasitica attacks and kills chestnut trees.
Oriental chesnuts are quite resistant to the pathogen. Cankers can be initiated in these
trees, but spread beyond the colonization point is limited. 43 ,44 In American (Castanea
dentata Borkh.) and European (c. sativa Mill.) chestnut trees the pathogen is able to
rapidly invade healthy periderm, phloem, and cambium tissues. In these two geograph-
ical areas the trees evolved without the pathogen; thus there was little or no selection
for resistance. Live oaks have been reported to be hosts of E. parasitica also, having
cankers limited in size much as on oriental chestnuts. 45 ,46 The destructiveness of E.
parasitica toward chestnut is not a reflection that it is highly evolved as a pathogen,
but rather an indication of just the opposite: that this host and pathogen are primitive
in their relationship. A balanced relationship has developed between E. parasitic a and
the oriental chestnut but not with the chestnuts of Europe and North America.
In understanding virulence expression of E. parasitica, we must be mindful of the
differences between expression of virulence genes of the pathogen and resistance
expression by the host. When V forms of the fungus interact with American chestnut,
virulence expression is unaffected by host defenses. When oriental chestnuts are hosts,
the resistance mechanisms of the host determine the nature of the outcome. Virulence
expression by the pathogen is better studied and understood using nonresistant hosts;
therefore, the description of host-pathogen interactions will be limited to that with the
susceptible host.
The initial step in the interaction between E. parasitica and chestnut trees requires a
preexisting wound in the periderm of the host. Even virulent forms of the fungus are
unable to invade unwounded trees. The wound appears to serve not only as an entry-
court where protective barriers have been breached, but also as a source of sufficient
nutrients to allow the development of the mycelial fans that function to penetrate
healthy host tissue. An important step in virulence expression of E. parasitica is the
organization of the mycelial fan. During colonization of the wound, aggregates of
hyphae form, but the organized extensive aggregates that become mycelial fans do not
develop until about 20 days after infection.47 The initial lesion, which consists of the
wound and nearby cortical tissue, is surrounded by a lignified reaction barrier that may
range up to 15 plant cells thick. 47 Although individual or small aggregates of hyphae
invade the lignified zone, they appear unable to penetrate this barrier. At this stage of
infection, Hebard et al. 47 found no significant difference in the numbers of hyphae in
susceptible hosts produced by V and H isolates of the fungus, nor a consistent differ-
ence in the size of the initial lesions.
Successful penetration of the lignified barrier only occurs by mycelial fans. Hebard
et al. 47 feel that the enlargement of cankers beyond the initial lesion is solely a function
of numbers and rate of formation of mycelial fans. They found that H isolates produce
fewer and thinner mycelial fans than do V isolates of the fungus. The mycelial fans
appear to force their way mechanically through host tissue, killing cells in advance of
the tip of the fan. Hyphae are not observed in front of the growing point of mycelial
fans.47 The death of host cells may be the result of fungal metabolites or perhaps a
futile hypersensitive defensive response by the cells.
The role of fungal metabolites as virulence factors has always held center stage in
156 Fungal Virology

host-pathogen studies, primarily due to the unequivocal role of host-specific toxins.

No such toxins are known to be produced by E. parasitica, although two nonhost-
specific toxins have been reported. 48 50 There is no known role of these toxic fungal
metabolites in the successful penetration and colonization of wound periderm by this
fungus. However, as indicated above, host cells die in advance of the mycelial fans.
This cell death may result in the rapid release of nutrients to the advancing mycelial
fan. Toxic metabolites may be important in the expression of virulence by killing host
cells, thereby increasing the availability of nutrients to the growing mycelial fans.
The two metabolites of the fungus that have been reported to be toxins are the
pigment skyrin and another organic molecule that has been given the name diapor-
thine. 48 50 Skyrin is not formed at the tip of hyphae, but it is generally present in
greatest amounts in the stroma from which fruiting bodies form." Thus skyrin is not
thought to playa role in virulence expression. Its production by H strains is variable,
European white strains producing little, while pigmented American strains probably
produce levels comparable to V strains. Grente and Berthelay-Sauret 17 believe diapor-
thine plays an important role in the host-pathogen relationship of chestnut blight. They
have evidence that diaporthine suppresses the formation of wound periderm by the
host, although no such suppression was observed by Hebard et al. 47 Grente and Ber-
thelay-Sauret have also reported that H strains of the fungus produce very little dia-
porthine. '7
It has recently been reported that oxalate, an organic acid, is produced in large
amounts by hyphal tips of V but not H strains of E. parasitica. 51 Oxalate has been
implicated as a virulence factor in other host-pathogen systems..,,53 It acts both to
lower the pH of the intercellular spaces to a level that is toxic to cells and to chelate
calcium which affects plant cell wall degradation by the pathogen.

B. Mechanisms of Virulence Reduction

If expression of virulence is the sum of many different genes in the pathogen, lack
of virulence can be the result of effects on one or more of these genes. Considering the
variation of hypovirulent phenotypes, there may be many different genes affected with
each different type of H strain representing a different set of virulence genes than have
been perturbed. Correlations between the presence of dsRNA and reduced/increased
production of pathogen metabolites or enzymes beyond those mentioned above may
be discovered. It is possible that many differences will be found between H and V
strains. McCarroll S4 reasoned that hypovirulence is no more than a reflection of re-
duced vigor of the pathogen. He felt that infection with dsRNA reduces growth and
production of many enzymes and metabolites important in virulence expression. The
net effect would be reduced virulence.
McCarroll'sS4 hypothesis is reasonable, but there is evidence that suggests the expla-
nation may not be so simple. As with other host-pathogen systems, that of E. parasitica
with chestnut is susceptible to perturbation at a variety of stages. Reduced effectiveness
as a pathogen at any of these stages could result in reduced virulence expression. Some
H strains of the pathogen appear to be as capable as V strains in colonizing the initial
lesion. 47 One distinction between these H and V strains is expressed at the mycelial fan
formation stage when the lignified wound periderm barrier, erected by the host in
response to infection, is breached. The H strains do not produce mycelial fans in the
number, size, and rate that V strains do.47 The H strains are thus unable to breach the
wound periderm. Whether this reduced mycelial fan formation is a reflection of poorer
nutrient utilization or reduced ability of the H strains to form mycelial fans, is not
known. The latter may be true since the other uniform characteristic of hypovirulence
in E. parasitica is the reduced ability of affected strains to sporulate. 22 Sporulation,
like mycelial fan formation, is a morphogenic event that involves organization of hy-
 157

phae into dense masses. Stroma formation precedes the development of fruiting bod-
ies. 55 Perhaps there are points in both morphogenic events that are common and
equally affected by the presence of dsRNA. The organization of mycelium into a
stroma or into mycelial fans appears to be similar in many respects.
A general reduction of vigor caused by diversion of metabolites and phosphorylated
compounds to viral replication is not frequently observed in any viral disease. 56 In E.
parasitica, growth and vigor of the fungus are not affected by dsRNA in a reproducible
manner. Often the convertant of an H and V strain grows better in culture than the
original V strain. 57 However, some H strains appear quite debilitated in culture. Such
debilitation would be one explanation as to why there is nearly a perfect correlation
between presence of dsRNA and hypovirulence in this fungus. Considering the evi-
dence for mixed infections, debilitation would be the most reasonable explanation as
to why dsRNA of different origins could all cause a reduction in virulence expression
by the pathogen. However, if the dsRNA merely debilitates the fungus by the energy
drain from its replication, the phenotypes would more likely reflect the specific fungal
strain rather than the dsRNA. Evidence to date suggests that the hypovirulence phen-
otype is mostly a reflection of the dsRNA and not the fungal genome. 18,27 If perturba-
tion of morphogenic controls result in hypovirulence expression, it would more likely
be the result of specific effects of the dsRNA or its products on these control processes
rather than general debilitation due to diversion of phosphorylated compounds to
dsRNA replication. The implication of such specific interactions is that there are se-
quences of the dsRNA that cause the hypovirulence phenotype. A replication or trans-
lation product of these sequences could interact with fungal macromolecules to reduce
virulence. For instance, cDNA could be integrated within the fungal genome, or per-
haps RNA or protein derived from the dsRNA binds to fungal DNA, or products of
the dsRNA may affect fungal metabolic pathways at specific points. All of these mech-
anisms would require some specific sequences of the dsRNA to be associated with
hypovirulence.
If specific interactions occur, particularly with the fungal genome, mutations that
mimic the effect of dsRNA at the fungal genome locus would be expected. It is thus
supportive of the hypothesis that specific interactions occur that at least two classes of
mutants have been found which mimic hypovirulent phenotypes. Both of these mutant
types occur at frequencies and under conditions that suggest the mutations may be at
easily perturbable control loci.
In studies of V strain protoplast regeneration, it was discovered that certain V strains
mutate to a white colony color at high frequency when grown on hypertonic media. 58
The morphology of the mutant colonies mimic those of the European B type of hypo-
virulence. The similarity between these white mutants and the B-type colonies includes
other characteristics, such as reduced sporulation and reduced virulence. These mu-
tants are indistinguishable in tests from B-type colonies, with the exception of dsRNA
content. They do not contain dsRNA, nor are they able to transmit their phenotype by
hyphal anastomosis. 58
Genetic analysis of these mutants demonstrated that the locus for pigment (pig) is
located on the nucleus genome and that the 20 mutants tested fall into one complemen-
tation group. It was found, by complementation, that pig- is recessive to pig+ (or-
ange). To date no reversion from pig- to pig+ has been observed. The pigments pro-
duced by E. parasitica are skyrin, oxyskyrin, and rugulosin, all of which are derivatives
of emodin, a methylated anthraquinone. 59 Since the mutants contain much-reduced
amounts of each of these pigments, the mutation must be at a point common to all the
pigments. Although pigment is only one of several detectable phenotypes affected by
the mutation, it is the one most easily analyzed. It is interesting that in the B-type
hypovirulent colonies as well as in these mutants, pigmentation, sporulation, and vir-
158 Fungal Virology

ulence all appear to be tandemly affected. As indicated above, it is possible that a

morphogenic control step could be affected, limiting stroma, mycelial fan, and pig-
ment production. Since pigment is primarily concentrated in the stroma and older
hyphae, it is possible that perturbation of sporulation could also affect pigment pro-
duction. Pigment production, however, is not always tightly tied to sporulation and
virulence. H strains from North America and the JR type H strain from Europe are all
pigmented. In the pig- mutants, pigmentation is reduced much·more than is sporula-
tion. It is not yet known whether the pig locus is the same as the previously described
ere locus of E. parasitica. 60
This description of the pig locus does not explain why its mutation frequency is so
high. We have identified another locus which we have termed pigment control (pc) that
appears to be involved in the high frequency mutation of pig. 58 Strains are tested for
pc by growing on hypertonic media and scoring for white colonies (pig-). In doing
this, we found that only a few strains are mutable at high frequency (up to 40070). The
strains that are mutable have been termed pc-. Results of heterokaryon tests suggest
that pc- is recessive to pc+ and is cytoplasmically transmitted. In some V strains of E.
parasitica it appears that there is an easily perturbable cytoplasmic locus that affects a
nuclear locus which in turn controls pigment and perhaps virulence and sporulation.
Perhaps one of these loci is affected by dsRNA in a reversible manner resulting in
expression of hypovirulence.
This scenario is probably simplistic, since it would require a separate control gene to
control expression of each of the different phenotypes that have been found associated
with hypovirulence. However, the existence of a similar gene was recently described by
Anagnostakis. 6 ' In studying JR type H strains she found that the hypovirulence, spor-
ulation, and deep pigmentation typical of the JR phenotype was controlled by a gene
that she termed flat. She found that flat was expressed whether dsRNA was present or
not, and that when dsRNA was present, the flat morphology was dominant to the B
type of dsRNA expression. Flat is a nuclear gene that appears to be easily mutable
since JR types are derived in high frequency from B type H strains.' 5
These studies show that a phenotype previously thought to be attributable to dsRNA
is in fact the result of a nuclear mutation. Thus this cannot be invoked as an example
of a fungal mutation that mimics hypovirulence. It does show that there is another
easily perturbable site that appears to be a virulence control gene. One of the exciting
results of this study is that it suggests that mutation of the flat locus is affected by the
presence of dsRNA. If this is the case, it is a good example of dsRNA affecting a
specific fungal locus.
While much of the discussion concerning virulence expression by E. parasitica and
the affect of dsRNA on this expression has been speculative, it is hoped that it points
out some questions that are readily addressable. The study of how dsRNA affects
virulence of this pathogen is not just a question for plant pathologists interested in
controlling plant disease, but also one for those interested in knowing how processes
in fungi are regulated.

VI. SUMMARY

It may have occurred to more than one reader of this book that mycoviruses were
not totally fair when they took residence within fungi. While many of the mycoviruses
are latent, causing no apparent ill effects on their hosts, it appears as if the mere
presence of any dsRNA changes the livelihood of the fungus, E. parasitica. Cultural
studies indicate that these unwelcome residents usually do not make E. parasitica sick
in any traditional sense. The change appears to be much more subtle. I have suggested
that perhaps some morphogenic control is affected, but this is just an educated guess.
 159

Whatever the mechanism, any dsRNA within E. parasitica appears equipped to subvert
the normal course of events in the life of this fungus.
The simplest explanation for this common effect of all dsRNA on virulence is that
there is an easily affected regulatory mechanism that controls virulence. The mutants
that mimic the B type of hypovirulence, and the flat mutants that were previously
mistaken for a hypovirulent type, are evidence for the existence of such regulatory
mechanisms. The flaw in this hypothesis is that, if true, the specific manifestation of
hypovirulence would be a function of the fungal strain. From the evidence we now
have, this is not true. The dsRNA, not the fungal strain, determines the phenotype.
The variety of morphological manifestations of hypovirulence mirrors the seemingly
unending variety of dsRNA segment sizes and numbers found in H strains. Should we
eventually be able to sort out all of the specific hypovirulence-causing entities, we will
still be left with the question of how each of these different mycoviruses is able to
express individuality in its phenotypes, and yet commonly reduce virulence.
The mycoviruses that appear to be the cause of hypovirulence, are unusual. In ad-
dition to the common effect on reducing virulence of the fungus, the viruses all appear
to be defective. This conclusion may be premature since only a few strains have been
thoroughly studied, but in those strains that have been studied, there is no evidence yet
for the existence of a capsid. Given the common pathogenic effect of the viruses on
their host and the apparent common lack of a capsid, one could easily speculate that
we are dealing with a single mycovirus that has evolved into different strains. The
preliminary genome hybridization studies contradict this conclusion.
We are obviously at a very early stage in our understanding of hypovirulence. It is
important to continue these studies because not only are hypovirulent strains a biolog-
ical control of an important disease, but they are also interesting biological systems. I
hope the answers to some of the questions I have posed will contribute to our under-
standing of virulence regulation in fungi as well as virus-host interactions.

ACKNOWLEDGMENTS

Studies in the author's laboratory were supported by grants from the National Sci-
ence Foundation (PCM-8402457) and the Utah Agricultural Experiment Station. The
assistance of Dr. Dane R. Hansen and Mr. Lee Barley in the preparation of this chapter
is appreciated.

REFERENCES

1. Tepper, C. S. and Anderson, A. J., The genetic basis of plant-pathogen interaction, Phytopathology,
74, 1143, 1984.
2. Biraghi, A., Possible active resistance to Endothia parasitica in Castanea sativa, Rep. Congr. Int.
Union Forest Res. Org. 11th Rome, 1953,643.
3. Grente, 1. and Sauret, S., L'hypovirulence exclusion phenomene original en pathologie vegetale, C.
R. Acad. Sci. Ser. D, 268,2347, 1969.
4. Van Alfen, N. K., Jaynes, R. A., Anagnostakis, S. L., and Day, P. R., Chestnut blight: biological
control by transmissible hypovirulence in Endothia parasitica, Science, 189, 890, 1975.
5. Castanho, B. and Butler, E. E., Rhizoctonia decline: studies on hypovirulence and potential use in
biological control, Phytopathology, 68, 1511, 1978.
6. Pusey, P. L. and Wilson, C. L., Detection of double-stranded RNA in Ceratocystis ulmi, Phytopath-
ology, 72,423, 1982.
7. Naiki, T. and Cook, R. J., Factors in loss of pathogenicity in Gaeumannomyces graminis var. tritici,
Phytopathology, 73, 1652, 1983.
160 Fungal Virology

8. Day, P. R., Dodds, J. A., Elliston, 1. E., Jaynes, R. A., and Anagnostakis, S. L., Double-stranded
RNA in Endothia parasitica, Phytopathology, 67,1393,1977.
9. Castanho, B., Butler, E. E., and Shepherd, R. J., The association of double-stranded RNA with
Rhizoctonia decline, Phytopathology, 68, 1515, 1978.
10. Anagnostakis, S. L., Biological control of chestnut blight, Science, 215,466, 1982.
11. Elliston, J. E., Hypovirulence and chestnut blight research: fighting disease with disease, f. For., 79,
657, 1981.
12. Elliston, J. E., Hypovirulence, in Advances in Plant Pathology, Vol. I, Ingram, I. D. S. and Wil-
liams, P. H., Eds., Academic Press, London, 1982, I.
13. Van Alfen, N. 1<., Biology and potential for disease control of hypovirulence of Endothia parasitica,
Annu. Rev. Phytopathol., 20, 349, 1982.
14. Van Alfen, N. K. and Hansen, D. R., Hypovirulence, in Plant Microbe Interactions, Vol. I, Kosuge,
T. and Nester, E. W., Eds., Macmillan, New York, 1984,400.
15. Grente, J. and Sauret, S., L' "hypovirulence exclusive" est-elle controlee par des determinants cyto-
plasmiques?, C.R. Acad. Sci. Ser. C,268, 3173,1969.
16. Turchetti, T., Some observations on the "hypovirulence" of chestnut blight in Italy, in Proc. Amer-
ican Chestnut Symp., West Virginia University Books, Morgantown, 1978,92.
17. Grente, J. and Berthelay-Sauret, S., Research carried out in France into diseases of the chestnut tree,
in Proc. American Chestnut Symp., West Virginia University Books, Morgantown, 1978, 88.
18. Elliston, 1. E., Laboratory and field studies of hypovirulent strains: EP-60 from Michigan and EP-
234 from Tennessee, in U.S. Forest Service American Chestnut Cooperators' Meeting, Smith, H. C.,
Ed., USDA For. Servo Gen. Tech. Rep., NE-64, 1981,8.
19. Stambaugh, W. J. and Nash, B. L., Detection and evaluation of hypovirulence in and resistance to
Endothia parasitica in surviving American chestnuts and associated oaks in North Carolina, in Proc.
U .S.D.A. Forest Service American Chestnut Cooperators' Meeting, West Virginia University Books,
Morgantown, 1982, 101.
20. Fulbright, D. W., Weidlich, W. H., Haufler, K. Z., Thomas, C. S., and Paul, C. P., Chestnut blight
and recovering American chestnut trees in Michigan, Can. f. Botany, 61,3164, 1983.
21. Elliston, 1. E., Hypovirulence in Endothia parasitica and suggested procedures for its detection and
analysis, in Proc. U.S.D.A. Forest Service American Chestnut Cooperators' Meeting, West Virginia
University Books, Morgantown, 1982, l.
22. Elliston, J. E., Pathogenicity and sporulation of normal and diseased strains of Endothia parasitica
in American chestnut, Proc. American Chestnut Symp., West Virginia University Books, Morgan-
town, 1978,95.
23. Anagnostakis, S. L. and Day, P. R., Hypovirulence conversion in Endothia parasitica, Phytopath-
ology, 69, 1226, 1979.
24. Moffitt, E. M. and Lister, R. M., Application of a serological screening test for detecting double-
stranded RNA mycoviruses, Phytopathology, 65, 851, 1975.
25. Dodds, J. A., Revised estimates of the molecular weights of dsRNA segments in hypovirulent strains
of Endothia parasitica, Phytopathology, 70, 1217, 1980.
26. Van Alfen, N. K., Bowman, J. T., and Simmons, J. R., The segregation of an Italian virulent isolate
of Endothia parasitica into H and V types, in Proc. American Chestnut Symp., West Virginia Uni-
versity Books, Morgantown, 1978, 106.
27. Fulbright, D. W., Effect of eliminating dsRNA in hypovirulent Endothia parasitica, Phytopathology,
74, 722, 1984.
28. Anagnostakis, S. L., Stability of double-stranded RNA components of Endothia parasitica through
transfer and subculture, Exp. Mycol., 5,236, 1981.
29. Dodds, J. A., Association of type I viral-like dsRNA with club-shaped particles in hypovirulent
strains of Endothia parasitica, Virology, 107, I, 1980.
30. Van Alfen, N. K. and Hansen, D. R., Liposome encapsulated dsRNA for cell-free transmission of
hypovirulence, in Proc. U.S.D.A. For. Servo American Chestnut Cooperators' Meeting, West Vir-
ginia University Books, Morgantown, 1982, 170.
31. Van Alfen, N. K., Hansen, D. R., Miller, S., and Barley, L., Cell-free transmission of hypovirulent
phenotype of Endothia parasitica, Phytopathology, 74 (Abstr.) 833, 1984.
32. Castanho, B. and Butler, E. E., Rhizoctonia decline: A degenerative disease of Rhizoctonia solani,
Phytopathology, 68, 1505, 1978.
33. Zanzinger, D. H., Bandy, B. P., and Tavantzis, S. M., High frequency of finding double-stranded
RNA in naturally occurring isolates of Rhizoctonia solani, f. Gen. Virol., 65,1601, 1984.
34. Rubenstein, R. and Harley, E. H., Reproducible alteration of cytoplasmic polyhedrosis virus double-
stranded RNA genome patterns on laboratory passage, Virology, 84, 195, 1978.
35. Street, 1. E., Croxson, M. C., Chadderton, W. F., and Bellamy, A. R., Sequence diversity of human
rota virus strains investigated by northern blot hybridization analysis, f. Virology, 43,369, 1982.
 161

36. Sabara, M. and Babiuk, L. A., The significance of rota virus electrophoretypic heterogeneity within
individual bovine rota virus isolates, in Double-Stranded RNA Viruses, Compans, R. W. and Bishop,
D. H. L., Eds., Elsevier, New York, 1983,95.
37. Buck, K. W., Almond, M. R., McFadden, J. J. P., Romanos, M. A., and Rawlinson, C. J., Prop-
erties of thirteen viruses and virus variants obtained from eight isolates of the wheat take-all fungus,
Gaeumannomyces graminis var. tritici, 1. Gen. Virol., 53, 235, 1981.
38. Jamil, N., Buck, K. W., and Carlile, M. J., Sequence relationships between virus double-stranded
RNA from isolates of Gaeumannomyces graminis in different vegetative compatibility groups, 1.
Gen. Virol., 65, 1741, 1984.
39. Lesemann, D. E. and Koenig, R., Association of club-shaped virus-like particles with a severe disease
of Agaricus bisporus, Phytopathol. Z., 89,161,1977.
40. Newhouse, J. R., Hock, H. C., and MacDonald, W. L., The ultrastructure of Endothia parasitica,
Comparison of a virulent with a hypo virulent isolate, Can. 1. Bot., 61, 389, 1983.
41. Newhouse, 1. R., Hock, H. C., and MacDonald, W. L., Virus-like particles in Endothia parasitica:
evidence for RNA content and site of replication, presented at Annu. Mtg., Mycological Society of
America, Ames, Iowa, June 26 to 30,1983, Abstr. M27.
42. Hertefuss, R. and Williams, P. H., Eds., Physiological Plant Pathology, Spnnger-Verlag, Berlin,
1976.
43. Jaynes, R. A., Selecting and breeding blight resistant chestnut trees, Proc. American Chestnut Symp.,
West Virginia University Books, Morgantown, 1978,4.
44. Griffin, G. J., Hebard, F. V., Wendt, R. W., and Elkins, J. R., Survival of American chestnut trees:
evaluation of blight resistance and virulence in Endothia parasitica, Phytopathology, 73, 1084, 1983.
45. Baston, W. E., Jr. and Witcher, W., Live oak cankers caused by Endothia parasitica, Phytopathol-
ogy, 58, 1473, 1968.
46. Nash, B. L. and Stambaugh, W. 1., Disease incidence, symptomatology, and vegetative compatibility
type distribution of Endothia parasitica on oak and chestnut hosts in North Carolina, in Proc.
U.S.D.A. For. Servo American Chestnut Cooperators' Meeting, West Virginia University Books,
Morgantown, 1982,74.
47. Hebard, F. V., Griffin, G. J., and Elkins, J. R., Developmental histopathology of cankers incited by
hypovirulent and virulent isolates of Endothia parasitica on susceptible and resistant chestnut trees,
Phytopathology, 74, 140, 1984.
48. Bazzigher, G., Beitrag zur Kenntnis der Endothia parasitica (Murr.) And., dem Erreger des Kastan-
iensterbens, Phytopath. Z., 21, 105, 1953.
49. Boller, A., Gumann, E., Hardegger, E., Kugler, F., Naef-Roth, S., and Rosner, M., Diaporthin, ein
Welke toxin aus Kulturen von Endothia parasitica (Murr.) And., He/v. Chim. Acta, 40,875, 1957.
50. Gaumann, E. and Naef-Roth, S., Uber die Toxine der Endothia parasitica(Murr.) And., Pflanzen-
schutz-Berichte, 19, 9, 1957.
51. Havir, E. A. and Anagnostakis, S. L., Oxalate production by virulent but not hypovirulent strains of
Endothia parasitica, Physiol. Plant Pathol., 23,369, 1983.
52. Bateman, D. F. and Beer, S. V., Simultaneous production and synergistic action of oxalic acid and
polygalacturonase during pathogenesis by Sclerotium rolfsii, Phytopathology, 55,204, 1965.
53. Maxwell, D. P. and Lumsden, R. D., Oxalic acid production by Sclerotinia sclerotiorum in infected
bean and in culture, Phytopathology, 60, 1395, 1970.
54. McCarroll, D. R., Pathogenesis of Endothia parasitica (Murr.) And., Ph.D., dissertation, University
of Tennessee, Knoxville, 1978, 143 pp.
55. Shear, C. L., Stevens, N. E., and Tiller, R. J., Endothia parasitica and related species, U.S. Dep.
Agric. Bull., No. 380, 1917,82 pp.
56. Joklik, W. K., Principles of Animal Virology, Appleton-Century-Crofts, New York, 1980, chap. 6.
57. Hindal, D. F. and Hagen, T. G., Mycelium production in liquid media by several normal, hypoviru-
lent and converted Endothia parasitica isolates, in Proc. U.S.D.A. For. Servo American Chestnut
Cooperators' Meeting, West Virginia University Books, Morgantown, 1982, 153.
58. Van Alfen, N. K., Hansen, D. R., and Miller, S., High frequency mutation of 3. virulence regulatory
site in Endothia parasitica, Phytopathology, 74,833 (Abstr.), 1984.
59. Roane, M. K. and Stipes, R. J., Pigments in the fungal genus Endothia, Virginia 1. Sci., 29, 137,
1978.
60. Anagnostakis, S. L., Genetic analyses of Endothia parasitica: linkage data for four single genes and
three vegetative compatibility types, Genetics, 102, 25, 1982.
61. Anagnostakis, S. L., Nuclear gene mutations in Endothia (Cryphonectria) parasitica that affect mor-
phology and virulence, Phytopathology, 74, 561, 1984.
62. Koltin, Y. and Barash, I., personal communication, 1984.
63. Finkler, A., Koltin, Y., Barash, I., Sneh, B., and Pozniak, D., Isolation of a virus from virulent
strains of Rhizoctonia soIani, 1. Gen. Virol., 66, 1221, 1985.
162 Fungal Virology

64. Hashiba, T., Homma, Y., Hyakumachi, M. and Matsuda, 1., Isolation of a DNA plasmid in the
fungus Rhizoctonia solani, J. Gen. Microbiol., 130, 2067, 1984.
65. L'Hostis, B., Hiremath, S. T., Rhoads, R. E., and Ghabrial, S. A., Lack of sequence homology
between double-stranded RNA from European and Amencan strains of Endothia parasitica, J. Gen.
Virol., 66,351,1985.
66. Paul, C. P. and Fulbright, D. W., Homology relationships with Michigan hypovirulent strains of
Endothia parasitica, Phytopathology, 75, 1325 (Abstr.), 1985.
67. Hiremath, S., L'Hostis, B., Ghabrial, S. A., and Rhoads, R. E., Sequence relationships among
double-stranded RNA species of a hypovirulent stram of the chestnut blight fungus Endothia paras-
itica, Phytopathology, 75, 1325 (Abstr.), 1985.
68. Hansen, D. R., Van Alfen, N. K., Gillies, K., and Powell, W. A., Naked DsRNA associated with
hypovirulence of Endothia parasitica is packaged in fungal vesicles, J. Gen. Virol., 66, 2605, 1985.
69. Ellzey, J. T., Hammons, T. L., and Cooper, M. 0., Ultrastructure of membrane-bounded particles
within hypovirulent strains of Endothia (Cryphonectria) parasitica, Mycologia, in press.
70. Newhouse, 1. R. and MacDonald, W. L., personal communication, 1985.
71. Fulbright, D. W., personal communication, 1984.
 163

Chapter 5

A TRANSMISSIBLE DISEASE OF HELMINTHOSPORIUM VICTORIAE

- EVIDENCE FOR A VIRAL ETIOLOGY

S. A. Ghabrial

TABLE OF CONTENTS

I. Introduction ................................................................................. 164

II. Historical Background .................................................................... 164

III. Viruses of H. victoriae.............. ...................................................... 165

A. General .............................................................................. 165
B. Assay Methods for Virus Detection .......................................... 166
1. Enzyme-Linked Immunosorbent Assay (ELISA) ................ 166
2. Dot Blot Hybridization ................................................. 167

IV. Disease Symptoms and Virus Content. ............................................... 167

V. Pathogenicity of Diseased Isolates of H. victoriae................................. 169

VI. Transmission of H. victoriaeViruses ................................................. 171

VII. Concluding Remarks ...................................................................... 175

References ............................................................................................ 175

164 Fungal Virology

I. INTRODUCTION

There are at least four incentives for investigating the viral etiology of a transmissible
disease of Helminthosporium victoriae Meehan & Murphy, the causal agent of Victo-
ria blight of oats:

1. A virus disease of H. victoriae provides a system for studying the pathological

effects of mycoviruses in a filamentous fungus. This is particularly significant in
view of the fact that the vast majority of fungal viruses are avirulent and appar-
ently have no deleterious effects on their hosts. 1
2. Disease symptoms are valuable markers in screening for infected mycelia in in-
fectivity assays involving virus-free fungal isolates and purified virus. Present
efforts for developing experimental systems for cell-free transmission of myco-
viruses are generally hampered by the lack of recognizable symptoms. 2 To iden-
tify a newly infected mycelium thus requires that individual mycelia produced
from a relatively large number of fungal colonies are screened for virus or viral-
nucleic acids by lengthy purification procedures.
3. H. victoriae presents an excellent model system for a phytopathogenic fungus
whose pathogenicity is dependent on the production of a host-specific toxin "vic-
torin". 35 Since diseased isolates of H. victoriae have been found to be hypovi-
rulent and toxin production has been reported to be controlled by a single nuclear
gene,6 7 investigations of nucleocytoplasmic interactions in the H. victoriae-virus
system will lead to an understanding of the role of mycoviruses in hypovirulence
in this system and possibly in some other phytopathogenic fungi that produce
host-specific toxins. 8 '
4. Diseased H. victoriae isolates, but not normal ones, have been reported to pro-
duce an antibiotic with a broad spectrum of antifungal and antibacterial activ-
ity.lO An investigation of the nature of the disease-induced antibiotic of H. vic-
toriae and whether it is virus-specified will be of considerable interest since the
killer phenomenon has not yet been found in filamentous fungi. However, unlike
this putative antibiotic of H. victoriae, the dsRNA-encoded killer proteins of
virus-infected yeasts and smuts are only lethal or inhibitory to strains of the same
or related species as the producing strains. 11

Although limited research has recently been done on the transmissible disease of H.
victoriae, the attractiveness of the system as outlined above justifies a brief review and
an opportunity to pinpoint priority areas for future research.

II. HISTORICAL BACKGROUND

H. victoriae was first described in 1946 as the causal agent of Victoria blight of
oats.12 It proved to be a highly specialized pathogen since it inflicts its damage on only
those oat varieties with the Victoria-type of resistance to crown rust Puccinia coronata
(Pers.) Cda.4.12.13 The disease caused by H. victoriae rose to epidemic proportions in
1947 and 1948 and caused serious reduction in yields in most oat-growing regions of
the U.S. 4
Although Victoria-derived oat cultivars were subsequently abandoned in the major
oat-cropping areas, they continued to be grown in considerable acreage in some of the
southern states during the 1950s. 6.14 .15 The discovery in 1959 of a transmissible disease
of H. victoriae was based on observations of cultural abnormalities in fungal isolates
obtained from blighted 'Victorgrain' oat plants in Louisiana. 16 Lindberg observed that
some of the H. victoriae colonies isolated from diseased oats were stunted and abnor-
 165

Table 1
PHYSICOCHEMICAL
PROPERTIES OF TWO VIRUSES
FROM HELMINTHOSPORIUM
VICTORIAEa

Virus

Property 1905 145S

Virion
Diameter (nm) 35-40 35-40
eCsCI (g/cm') 1.4325 1.3813-1.4138
RNA
mol wt (X 106) 3.0 2.4
2.2
2.1
2.0
Protein
mol wt (x 10-') 83 92
88 97
106

Data summarized from Sanderlin and Gha-

brial. "

mal. Following an initial period of normal growth, sectors appeared at the margins of
the colonies with concomitant collapse or lysis of existing aerial mycelium and almost
complete inhibition of colony expansion. Lindberg referred to these abnormalities as a
"disease" of the fungus and showed that the disease could be transmitted to normal
colonies via hyphal anastomosis with diseased colonies. 1617 Since the fields of 'Victor-
grain' oats in Louisiana from which diseased and normal H. victoriae were isolated
did not suffer significant yield losses, which was unusual for this serious disease of
oats, it was suggested that a reduction in the pathogenicity level of H. victoriae had
occurred and that this might be attributed to the transmissible disease of the fungus. 6
Twenty-five years ago, Lindberg suggested a viral etiology for the disease of H.
victoriae, but it is only recently that evidence in support of this has been pre-
sented. 18 20 Diseased isolates of H. victoriae have been found to contain two serologi-
cally and electrophoretically distinct viruses designated according to their sedimenta-
tion coefficients as the 1905 and 145S viruses. Normal isolates have been found to be
either virus-free or to contain only the 1905 virus. Evidence to date suggests that either
the 145S virus alone or a mixed infection with the 1905 and 145S viruses is the incitant
of the disease.

II. VIRUSES OF H. VICTORIAE

A. General
The physicochemical properties of the 1905 and 145S viruses are summarized in
Table 1 and an electron micrograph of a purified preparation of the 1905 virus is
shown in Figure 1.
The 1905 is apparently a single component virus since it contains a single species of
dsRNA and bands homogeneously in CsCI density gradients. 18 The reported value of
3 x 10 6 for the molecular weight of 1905-dsRNA, estimated by linear extrapolation
from reovirus dsRNA standards has been confirmed using dsRNAs from HeImintho-
sporium maydis virus, Endothia parasitica (strain EP 113), Penicillium chrysogenum
166 Fungal Virology

FIGURE I. Electron mIcrograph of a purifIed preparation of the 1905 VIrus stamed with 1"70
uranyl acetate. The bar represents 100 nm

virus, and P. stoloniferum viruses Sand F as size markers. 21 22 The occurrence of two
major capsid polypeptides in equimolar amounts in the I90S virions is unusual for
isometric dsRNA mycoviruses. IR However, the relationship between the two polypep-
tides has not been determined, and the possibility that the smaller polypeptide (mol wt
83,000) is derived from the larger one (mol wt 88,000) through partial proteolysis has
not been ruled out. In preliminary experiments, intact virions were incubated with
several proteolytic enzymes prior to analysis by SDS polyacrylamide gel electrophore-
sis. The results from such experiments revealed no change in tl-J.e ratio of the two poly-
peptides and suggested that equal numbers of molecules from each polypeptide were
partially degraded. 24 Purification and characterization, e.g., peptide mapping, of the
individual polypeptides should be made.
The I45S virus is probably a multicomponent virus based on the multiplicity of
density components resolved in CsCI gradients and the isolation of four distinct
dsRNA components (Table 1).'8 The concentration of the I45S virus varies consider-
ably from one preparation to another and is generally much lower than that of the
I90S virus; a ratio of 1:10 of the I45S to the 1905 virus is not uncommon. Attempts to
maintain selections of diseased isolates that consistently produce relatively high yields
of the 145S virus have proved fruitless. This may not be surprising since disease severity
is apparently correlated with the concentration of the I45S virus (see section on disease
symptoms and virus content). ,.
The relatedness of the dsRNA components of the 145S virus needs to be investigated.
It is not known whether the multiple components of dsRNA represent a segmented
genome or whether one or more of these components are satellites or deletion mutants.
The presence of defective interfering particles or suppressive dsRNAs in some of the
diseased cultures would be of considerable interest since it would explain the fluctua-
tion in yield of the 145S virus.

B. Assay Methods for Virus Detection

1. Enzyme-Linked Immunosorbent Assay (ELISA)
An antiserum specific for the 1905 virus was produced and used for virus detection
 167

by the ELISA technique. The lower limits of detection of the 1905 virus by ELISA are
30 to 50 ng/m£ (an A~h~m = 5.0 for the 1905 virus was used). The 1905 virus could be
readily detected by ELISA in mycelial extracts from as little as 10 to 100 mg wet my-
celium depending on the fungal isolate. Antisera specific for the 145S virus are not
available at the present time. The antiserum prepared against a mixture of the 1905
and 145S viruses, and which was satisfactorily used in gel double diffusion tests in a
previous investigation, was found unsuitable for use in ELISA.24 This antiserum ap-
parently contains antibodies to host components since false positives were obtained
with extracts from virus-free fungal isolates. It is doubtful whether this antiserum
would be useful for the detection of the 145S virus by ELISA, even though it should
be possible to cross-absorb the antibodies to healthy components.

2. Dot Blot Hybridization

The dot blot hybridization technique was used for detection of the 1905 virus. Hy-
bridization probes used were either lZP-Iabeled cDNA to dsRNA from the 1905 virus
or fragments of denatured dsRNA labeled with lZp at their 5' ends using polynucleotide
kinase. The dot blot assay detected the 1905-dsRNA in purified preparations at con-
centrations as low as 0.5 to 1.0 ng dsRNA and in total nucleic acid fraction extracted
from as little as 10 mg wet mycelium of strain B-1 of H. victoriae. Z4 Preparation of
labeled probes to the 145S-dsRNAs is currently being undertaken so that the dot blot
hybridization technique can be used in the detection of this virus.

IV. DISEASE SYMPTOMS AND VIRUS CONTENT

The diseased isolates B-1 and A-9, used in previous investigations, are similar in
many respects.'B.I. Each isolate contains both the 1905 and 145S viruses, grows at a
much slower rate than normal isolates, and produces irregular mycelial mats with little
or no sporulation. The two isolates differ, however, in the outstanding disease symp-
toms considered typical of each isolate. Isolate B-1 (Figure 2B) is characterized by
extensive collapse of aerial mycelium in young colonies grown on potato dextrose agar
(PDA) medium. Scattered areas of lysed aerial mycelium are observed throughout the
colony in 4- to 5-day-old cultures with the collapse of aerial mycelium of the entire
culture within 7 to 10 days. Pronounced and rapid collapse of aerial mycelium is also
the main characteristic symptom of isolate B-1 in liquid medium (Figure 3B). Isolate
A-9 is characterized by stunted colonies with sparse aerial mycelium and almost com-
plete inhibition of colony expansion within 5 to 7 days following subculturing (Figure
2C). Sectors of swollen hyphae are observed at the edges of 7- to lO-day-old colonies.
Although collapse of the aerial mycelium of isolate A-9 is limited on PDA it is quite
pronounced within 10 to 14 days in liquid medium. The most characteristic symptoms
in liquid medium, however, consist of poor mycelial growth and production of a large
number of sectors of white mycelium (Figure 3C).
The two isolates also differ in their content of the 1905 virus. ELISA tests with an
antiserum to 1905 virus have consistently shown that isolate B-1 contained a higher
titer of this virus than isolate A-9. 20
The disease symptoms typical of these two isolates have, in general, been faithfully
reproduced upon subculturing for almost a decade. Should one or the other of these
isolates be "cured" of both viruses, disease symptoms characteristic of each isolate
could serve as selectable markers in developing infectivity assays and in understanding
the role of each of the two viruses in disease development.
Three normal isolates have been extensively studied in previous investigations. IB .I•
These are isolates B-2 (Figure 2A), Hv 408, and HW-l. All three isolates produce
uniform mycelial mats and sporulate profusely on PDA. No collapse of aerial myce-
168 Fungal Virology

FIGURE 2. Morphology of three iso- FIGURE 3. Mycelial growth of three

lates of H. victoriae grown on potato isolates of H. victoriae grown on liquid
dextrose agar for 7 days: (A) normal iso- medium of potato dextrose broth for 7
late B-2; (B) diseased isolate B-t; and (C) days: (A) normal isolate B-2; (B) diseased
diseased isolate A-9. isolate B-t; and (C) diseased isolate A-9.

lium is observed in cultures (solid or liquid medium) maintained for as long as 4 weeks
(Figure 3A). Isolates HV 408 and HW-1 were found to be virus-free as determined by
large-scale extraction for virus purification from as much as 0.5 to 1.0 kg wet mycelium
using standard procedures. 18 These two isolates were also shown to be free of the 1905
virus by ELISA using an antiserum specific for this virus. 1o Large scale extractions for
virus purification from isolate B-2, on the other hand, have shown this isolate to con-
tain 1905 virus.'8 The titer of the 1905 virus in mycelial extracts from isolate B-2 was
shown by ELISA tests to be significantly lower than that in the diseased isolates B-1
and A_9. 1o
 169

Previous studies have shown that protoplasts prepared from normal isolates regen-
erated and formed colonies identical to the original cultures; those from diseased iso-
lates, on the other hand, were found to regenerate into three morphologically distinct
types of colonies. t9 In addition to colonies with the morphology of the original diseased
culture (designated type 2 colonies), two other types of colonies were distinguished:
symptomless colonies with vigorous growth (designated type 1) and severely stunted
colonies (designated type 3). Mycelia from type 1 colonies were found to contain only
the 1905 virus with none, or only traces, of the 145S virus. 10 Mycelia from types 2 and
3, on the other hand, contained both 1905 and 145S with the concentration of the 145S
correlating well with disease severity.t9 These results suggested that the 145S virus is
probably the causal agent of disease. However, a role for the 1905 virus in disease
development cannot be ruled out since all diseased isolates contain both 1905 and 145S
viruses.
Type 1 colonies produced abundant aerial mycelium which usually did not collapse.
The hyphae at the periphery of these colonies were long, with few branches, and the
cells had uniform size and normal cytoplasmic contents (Figure 4A).'0 In contrast, the
hyphae at the edges of type 3 colonies showed profuse branching and most of the cells
were swollen with granular or disorganized cytoplasmic content. Some of the cells lysed
and the exuded cytoplasmic contents could be observed adhering to the sides of hyphae
(arrows, Figure 4B and Figure 5A). Thin sections of such cells revealed the presence of
a large number of aggregates of virus-like particles (VLPs), both in the exuded cyto-
plasmic contents and inside the cell (Figure 5B, C, and D).25 Some of the VLPs seen in
these aggregates have electron-dense cores (arrows, Figure 5C).
Ultrastructural studies were made of severely diseased (type 3) and apparently
healthy (type 1) hyphae derived from the same diseased isolate (B-1) and comparable
to hyphae shown in Figure 4. Thin sections of apical cells, as well as of cells near the
hyphal tip of severely diseased hyphae, revealed that the cytoplasm and cellular organ-
elles had degenerated and the cells were almost entirely filled with large masses of
closely aggregated VLPs. 2 0 Large crystalline arrays or loosely aggregated VLPs that
were surrounded by a membrane and membraneous structures (Figure 6) were fre-
quently detected in such cells. 25 Thin sections of comparable cells from apparently
healthy hyphae appeared normal and no aggregates of VLPs could be detected. 20

v . PATHOGENICITY OF DISEASED ISOLATES OF H. VICTORIAE

Lindberg compared the virulence of diseased and normal isolates on oat seedlings
using two methods of inoculation. 6 Inocula consisting of mycelial plugs were placed
either in the axil of the cotyledonary leaves or added to the soil at the time of seeding.
The results showed that diseased isolates had a markedly reduced virulence. Whereas
93 to 100o,to of the plants exposed to normal mycelium were killed, only 13 to 20% of
those treated with diseased mycelium were destroyed. In treatments where mixed in-
ocula of normal and diseased mycelia were placed in soil at seeding time, 50% of the
plants survived. When seeding was delayed 30 days following soil infestation, the per-
centage of survival for plants exposed to mixed inocula was as high as for those treated
with diseased mycelium alone. These results suggested that biological control of the
normal virulent isolates had occurred in soil.
The pathogenicity of H. victoriae depends on the production of the host-specific
toxin "victorin" and this, in turn, appears to be controlled by a nuclear gene. S • 7 The
virulence of different fungal isolates may, thus, be evaluated by assaying for toxin
production in culture. s Using the root growth inhibition bioassay, Lindberg reported
that toxin production by normal isolates was two- to tenfold higher than that by dis-
eased isolates. 6 He attributed this to differences in growth rates between the two types
170 Fungal Virology

FlGURE 4. Morphology of hyphae at the edges of colonies ?foduced from

protoplasts of a diseased isolate (B-1) of H . yictoriae grown on potato dex-
trose broth for 5 days . Hyphae from : (A) type 1 colonies; and IB) type 3
colonies examined in the stereomicroscope (magnification x 43). Arrows in
(B) point to exuded cytoplasmic contents from lysed cells. (From Ghabrial,
S. A., Sanderlin, R. S., and Calvert, L. A. , Phytopath%gy,69, 312, 1979.
With permission.)

of isolates. Sanderlin tested several normal and diseased isolates of known virus con-
tent for toxin production (Table 2).13 The results indicated that presence or absence of
virus is not directly related to toxin production. This is not surprising in view of the
fact that toxin production is conditioned by a nuclear gene. The finding that all dis-
eased isolates (including those tested by Lindberg) produced negligible or low levels of
victorin suggests that virus infection may indirectly affect toxin production. Thus,
toxin production may be affected not only by a mutation in a nuclear gene, but also
by nucleocytoplasmic interactions in virus-infected fungal isolates. This hypothesis
could be experimentally tested by: (1) monitoring toxin production in isolate HW-l (a
 171

potent producer of toxin, Table 2) following infection with one or both viruses; or (2)
determining whether "curing" of the diseased isolates A-9 and B-1 would result in
their regaining the capacity to produce high levels of victorino

VI. TRANSMISSION OF H. VICTORIAE VIRUSES

Ghabrial and Mernaugh inoculated protoplasts from a virus-free isolate, Hv 408,

with a puified virus preparation (containing both the 1905 and 145S) in the presence
of polyethylene glycol. 20 Dilutions of fusing protoplasts were plated on a protoplast
regeneration medium (PDA containing 20070 sucrose w/v). Colonies which formed
from regenerated pro top lasts were individually transferred to PDA (at least 200 colo-
nies from each treatment) and observed for disease symptoms. Stunted and abnormal
colonies were detected at a low frequency (about 1%) among colonies derived from
virus-inoculated pro top lasts (Figure 7 B, C, and D). All control colonies, on the other
hand, were similar to the original normal isolate; they grew vigorously, produced uni-
form mycelial mats, and sporulated profusely (Figure 7A). The newly diseased colonies
showed reduced or no sporulation, and hyphae at the edges of the colony were initially
swollen and profusely branched. However, normal, vigorous, and profusely sporulat-
ing sectors arose frequently at the colony periphery (Figure 7C and D).
Evidence that the disease was transmitted to the virus-free isolate, Hv 408, through
the use of fungal protoplasts and purified virus preparations has been based on:

1. Detection of stunted and abnormal colonies showing some of the characteristics

of the disease of H. victoriae among colonies derived from virus-treated, but not
control, protoplasts. These results were reproduced in three experiments. How-
ever, we have failed to detect such colonies in three other experiments (Ghabrial
and Mernaugh);20.25
2. Thin sections of hyphae from newly diseased, but not from control colonies,
revealed the presence of aggregates of VLPS;20 and
3. Virus particles in small numbers (1 to 3 particles per field at a magnification of
46,000) with similar size and morphology to the viruses used in the inoculum were
detected in mycelial extracts (lyophilized mycelium equivalent to 2 g wet weight
suspended in 1 ml buffer) from newly diseased colonies by immune electron mi-
croscopy using an antiserum to a mixture of the 1905 and 145S viruses diluted
1: 100 with buffer (Ghabrial and Mernaugh)!0.25 Unequivocal evidence for dis-
ease transmission using purified virus preparations, however, must await the re-
isolation and characterization of the virus(es) in the newly diseased colonies.

Maintenance of colony morphology of the newly diseased colonies through serial

subcultures required that mycelial fragments from the grey non sporulating center of
the colony be used for transfer (Figure 7B, C, and D). Using this procedure, some of
the stunted and abnormal colonies have maintained their morphology through re-
peated transfers for more than 4 years. Subculturing of mycelial fragments from dark,
vigorous and sporulating sectors of these colonies, however, gave rise to apparently
normal vigorous colonies similar in morphology to the original Hv 408 isolate. 24 The
production of these apparently normal sectors at high frequency has been a major
problem in our attempts to increase mycelium from newly diseased colonies in liquid
medium for the purpose of virus purification. The diseased mycelium is apparently
diluted out in such cultures since vigorous, profusely sporulating mycelium becomes
predominant. Concentrated extracts from as much as 100 g wet mycelium from such
cultures, analyzed by sucrose density gradient centrifugation, failed to reveal the pres-
ence of either the 1905 or 145S viruses!4 ELISA tests for the 1905 virus in mycelial
172 Fungal Virology

FIGURE 5.
 173

FIGURE 6. Thin sec tion o f a hypha from severely diseased type 3 colonies showing aggre-
gates of virus particles encl osed by a membrane and membra neou s structures. Bars = 200 nm ,
(unpublished micrograph , co urtesy of S. A. Ghabrial and R. L. Mernaugh.)

Table 2
VIRUS CONTENT AND VICTORIN
PRODUCTION BY SEVERAL NORMAL AND
DISEASED ISOLATES OF
HELMINTHOSPORIUM VICTORIAE
Victorin
Fungal Colony production
isolate morphology· Virus content' (units),

Hv 408 N None 0
HW-l N None 1000
B-2 N 1905
B-1 0 1905 + 145S
A-9 0 1905 + 145S 10

N = normal ; 0 = diseased
Virus content of the various fungal isolates was determined
by sucrose density gradient centrifugation as described pre-
viously."
One milliunit of victorin will inhibit root elongation by 50"70
(data from Sanderlin").

FIGURE 5. Light and electron microscopic examination of severely diseased hyphae from type 3 colonies
of a diseased isolate (B-1) of H. victoriae showing cells that have exuded some of their cytoplasmic contents .
(A) Hyphae examined with phase contrast microscopy (magnification x 1,000); (B) thin section of a hypha
Showing a large number of aggregates of virus-like particles inside the cell and in the exuded material.
Framed areas inside and outside the cell are, respectively, shown in (C) and (0) at a higher magnification.
Arrows in (C) point to particles with electron dense cores . Vs = Vesicles; L = Lipid inclusion . Purified virus
negatively stained with phosphotungstic acid is shown in (E). Bar = I lIm in (B); bar represents 100 nm in
(C), (D), and (E) . (Unpublished micrograph, courtesy of S. A . Ghabrial and R. L. Mernaugh .)
174 Fungal Virology

FIGURE 7. Morphology of colonies produced in

transmission experiments using protoplasts from a vi-
rus-free fungal isolate (Hv 408) and a purified virus
preparation (containing both 1905 and 145S viruses).
Five-day-old colonies produced from control (A) and
virus-treated protoplasts (B, C, and D). (From Gha-
brial, S. A. and Mernaugh, R. L., in Double-Stranded
RNA Viruses, Compans, R. W. and Bishop, D. H. L.,
Eds., Elsevier Biomedical, New York, 1983,441. With
permission.)

extracts (1 g wet mycelium per 2 ml buffer) from newly diseased colonies were also
negative_ This suggests that the concentration of the 1905 virus, if present in these
colonies, is lower than 100 ng/g wet mycelium since the lower limit of detection by
ELISA is 50 ng/ml and a sample volume of 0_2 ml is used. Future plans for detection
of the 1905 and 145S viruses in newly diseased colonies of isolate Hv 408 will make use
of nucleic acid hybridization techniques (dot blot hybridization).
Since isolate Hv 408 apparently supports only limited virus replication, it is not
suitable for use in optimizing the conditions required for developing efficient infectiv-
ity assays for H. vietoriae viruses. We have been considering the following H. vieto-
riae isolates for use in transmission experiments:

1. Isolate B-2. This normal isolate is of special interest since it is presumably derived
from the same fungal culture as the diseased isolate B-l. 5 The latter is known to
be susceptible to both 1905 and 145S viruses. Cultures of isolate B-2 used in
previous investigations contained only the 1905 virus. 18 However, we have not
been able to detect the 1905 by ELISA in mycelial extracts from some recently
obtained cultures of this isolate. Interestingly these cultures were derived from
single conidial isolates obtained from a culture that has been stored on silica gel
for 2 to 3 years. The present cultures of B-2 will be tested for virus by immune
electron microscopy, dot blot hybridization, and large-scale extraction for virus
purification prior to its use in transmission experiments.
2. The diseased isolates A-9 and B-l. If freed of both the 1905 and the 145S viruses,
these two isolates will be ideal for use in transmission experiments since the dis-
ease symptoms typical of each isolate have been extensively studied. Several ap-
proaches will be attempted, including hyphal-tip subculturing, use of single con-
idial isolates, and/or treatment with cycloheximide.
 175

VII. CONCLUDING REMARKS

Several lines of evidence have been presented in support of a viral etiology for the
disease of H. vic toria e. Diseased isolates of the fungus contain two serologically dis-
tinct, isometric viruses designated, according to their sedimentation coefficients, as the
1905 and 145S viruses. The 1905 virus does not appear to have deleterious effects on
its host since normal fungal isolates have been found which contained this virus alone.
The 145S virus, on the other hand, is presumed to be the causal agent of the disease.
The concentration of this virus in diseased mycelium has been shown to be correlated
with symptom severity. It cannot be ruled out, however, that mixed infection with the
two viruses is required for disease development since all known diseased isolates con-
tain both viruses.
In ultrastructural studies, disease severity has been correlated with degeneration of
the cytoplasm and accumulation of large masses of closely aggregated, virus-like par-
ticles in apical cells as well as in cells near the hyphal tip.
Evidence that the disease has been transmitted to a normal, virus-free isolate (Hv
408) through the use of fungal protoplasts and purified virus preparations was based
on the detection of stunted and abnormal colonies among those derived from virus-
treated, but not control, protoplasts. Furthermore, virus particles were detected by
immune electron microscopy in mycelial extracts and by electron microscopy in thin
sections of hyphae from colonies produced by inoculated, but not by control, proto-
plasts. This evidence, however, is incomplete since isolation and characterization of
the nucleic acid and protein components of the virus(es) in the newly diseased colonies
have not yet been accomplished. This information has been difficult to obtain because
isolate Hv 408 apparently supports only limited virus replication. Nucleic acid hybrid-
ization techniques will be used in the future for virus detection in the newly diseased
colonies of isolate Hv 408. Attempts will also be made to obtain virus-free cultures of
fungal isolates known to be susceptible to virus infection for use in transmission exper-
iments.

REFERENCES

I. Ghabrial, S. A., Effects of fungal viruses on their hosts, Annu. Rev. Phytopathol., 18, 441, 1980.
2. Lecoq, H., Boissonet-Menes, M., and Delhotal, P., Infectivity and transmission of fungal viruses, in
Fungal Viruses, Molitoris, H. P., Hollings, M., and Wood, H. A., Eds., Springer, Berlin, 1979, 34.
3. Meehan, F. and Murphy, H. C., Differential phytotoxicity of metabolic by-products of Helminthos-
porium victoriae, Science, 106, 270, 1947.
4. Litzenberger, S. C., Nature of susceptibility to Helminthosporium victoriae and resistance to Pucci-
nia coronata in Victoria oats, Phytopathology, 39, 300, 1949.
5. Luke, H. H. and Wheeler, H. E., Toxin production by Helminthosporium victoriae, Phytopathology,
45,453, 1955.
6. Lindberg, G. D., Reduction in pathogenicity and toxin production in diseased Helminthosporium
victoriae, Phytopathology, 50,457, 1960.
7. Scheffer, R. P., Nelson, R. R., and Ullstrup, A. J., Inheritance of toxin production and pathogenicity
in Cochliobolus carbonum and Cochliobolus victoriae, Phytopathology, 57, 1145, 1967.
8. Pringle, R. B. and Scheffer. R. P., Host-specific plant toxins, Annu. Rev. Phytopathol., 2, 133,
1964.
9. Hollings, M., Mycoviruses: viruses that infect fungi, Adv. Virus Res., 22, I, 1978.
10. Lindberg, G. D., Disease-induced antibiotic production in Helminthosporium victoriae, Phytopath-
ology, 54 (Abstr.), 898, 1964.
I I. Buck, K. W., Viruses and killer factors of fungi, in The Eukaryotic Microbial Cell, Gooday, G. W.,
Lloyd, D .• and Trinci, A. P. J., Eds .• Cambridge University Press, London, 1980.329.
12. Meehan, F. and Murphy, H. C., A new Helminthosporium blight of oats, Science, 104,413, 1946.
176 Fungal Virology

13. Litzenberger, S. C. and Murphy, H. C., Methods for determining resistance of oats to Helminthos-
porium victoriae, Phytopathology, 37,790, 1947.
14. Ivanoff, S. S., Bowman, D. H., and Rothman, P. G., Oat diseases in Mississippi, Plant Dis. Rep.,
42, 521, 1958.
15. Scheffer, R. P. and Nelson, R. R., Geographical distribution and prevalence of Helminthosporium
victoriae, Plant Dis. Rep., 51,110,1967.
16. Lindberg, G. D., A transmissible disease of Helminthosporium victoriae, Phytopathology, 49, 29,
1959.
17. Lindberg, G. D., Transmission of a disease of Helminthosporium victoriae, Phytopathology 50
(Abstr.), 644, 1960.
18. Sanderlin, R. S. and Ghabrial, S. A., Physicochemical properties of two distinct types of virus-like
particles from Helminthosporium victoriae, Virology, 87, 142, 1978.
19. Ghabrial, S. A., Sanderlin, R. S., and Calvert, L. A., Morphology and virus-like particle content of
Helminthosporium victoriae colonies regenerated from protoplasts of normal and diseased isolates,
Phytopathology, 69,312, 1979.
20. Ghabrial, S. A. and Mernaugh, R. L., Biology and transmission of Helminthosporium victoriae
mycoviruses, in Double-Stranded RNA Viruses, Compans, R. W. and Bishop, D. H. L., Eds., Elsev-
ier Biomedical, New York, 1983,441.
21. Bozarth, R. F. and Harley, E. H., The electrophoretic mobility of double-stranded RNA in polya-
crylamide gels as a function of molecular weight, Biochem. Biophys. Acta, 432,329, 1976.
22. Dodds, J. A., Revised estimates of the molecular weights of dsRNA segments in hypovirulent strains
of Endothia parasitica, Phytopathology, 70, 1217, 1980.
23. Sanderlin, R. S., Biophysical properties of virus-like particles occurring in healthy and diseased iso-
lates of Helminthosporium victoriae Meehan and Murphy, Ph.D. thesis, University of Kentucky,
Lexington, 1977.
24. Ghabrial, S. A., unpublished data, 1984.
25. Ghabrial, S. A. and Mernaugh, R. L., unpublished data, 1984.
 177

Chapter 6

THE D-FACTOR IN CERATOCYSTIS ULMI - ITS BIOLOGICAL

CHARACTERISTICS AND IMPLICATIONS FOR
DUTCH ELM DISEASE

C. M. Brasier

TABLE OF CONTENTS

I. Biological Characteristics of d-Factors ............................................... 178

A. Distribution in the C. ulmi Subgroups ...................................... 178
B. The "d-Reaction" ................................................................ 179
C. The Cytoplasmic Location and Spread of d-Factors ..................... 179
D. Effect on Growth and Reproduction ......................................... 186
1. Mycelial Growth ......................................................... 186
2. Growth Stability ......................................................... 190
3. Comparative Effects of d 1 _ and d 1 -Factors ........................ 191
E. Effect on Conidial Viability .................................................... 192
F. Effect on Perithecial Formation ............................................... 193
G. Regulation of d-Factor Transmission ........................................ 193
1. Regulation by Vegetative Incompatibility .......................... 193
2. Reduced Transmission via Conidia .................................. 194
3. Restricted Transmission to Ascospores ............................. 196
H. Pathogenic Behavior of d-Infected Isolates ................................. 197

II. Impact of d-Factors on Dutch Elm Disease ......................................... 199

A. Spread of d-Factors during the Disease Cycle .............................. 200
1. Bark Phase ................................................................ 201
2. Beetle Phase ............................................................... 201
3. Feeding Groove Phase .................................................. 202
4. Xylem or Pathogenic Phase ........................................... 202
B. Frequency of d-Infection in Nature .......................................... 203
C. Potential of d-Factors in Disease Control. .................................. 204

Acknowledgments .................................................................................. 207

References ............................................................................................ 207

178 Fungal Virology

Dutch elm disease, currently by far the most destructive tree disease in the northern
hemisphere, is caused by a combination of the vascular wilt pathogen Ceratocystis
(Ophiostoma) ulmi and vector elm bark beetles of the family Scolytidae. The fungus
exists as at least three genetically isolated subgroups, the highly pathogenic Eurasian
(EAN) and North American (NAN) races of the aggressive strain, and the more weakly
pathogenic nonaggressive strain. 1 The non aggressive strain is now believed to have
been responsible for the first rather milder epidemics of Dutch elm disease which oc-
curred during the 1920s and 1930s. The EAN and NAN races of the aggressive strain
are responsible for the present massive second epidemics of the disease, as a result of
which most mature elms will probably be killed across an area extending from the
Rocky Mountains to the east coast of North America and throughout Europe to at
least as far east as Tashkent in Central Asia. 2 - 4 As a further consequence of this second
wave of disease, it is likely that the "old" nonaggressive strain will be replaced by the
aggressive, so that following generations of young elms will continue to be attacked by
one or another form of the aggressive strain. Unless the pathogenicity of the aggressive
strain attenuates in some way, this may result in the elm, at least in the medium term,
being reduced to a scrub or marginal population throughout most of the current epi-
demic areas. 3 5
Into this already complex picture can now be introduced the d-factor, a recently
discovered cytoplasmically transmissible genetic determinant of C. ulmi capable of
exerting a deleterious effect on its growth and reproductive fitness. 6 By analogy with
other cytoplasmically transmitted diseases of fungal pathogens, the properties of the
d-factor raise a number of questions about its potential impact on Dutch elm disease,
particularly during the forthcoming postepidemic period when other selection pres-
sures on C. ulmi are likely to intensify.
This paper describes in detail the known biological characteristics of the d-factor,
including for the first time an account of its effects on the pathogenicity of C. ulmi
isolates, drawing on much previously unpublished experimental work by the author. It
also presents a theoretical assessment based on combined laboratory and ecological
information of the potential for d-factors in natural and man-managed control of
Dutch elm disease.

I. BIOLOGICAL CHARACTERISTICS OF d-FACTORS

A. Distribution in the C. ulmi Subgroups

The d-factor was first described in a French isolate of the NAN aggressive subgroup
of C. ulmi (H321, see Reference 6), and it is with the NAN form of the fungus that
most experimental work with d-factors has been carried out. Since then, on the basis
of the occurrence of d-reactions (see below), d-factors have been identified among
populations of NAN isolates from most countries in which the NAN is known to be
present, and have also been shown to be a common component of populations of the
EAN aggressive and nonaggressive subgroups of the fungus (for the distribution and
properties of the C. ulmi subgroups see References 1 to 5). Estimates of the frequency
of d-factor infection, or d-infection, in EAN and NAN populations in particular will
be given later.
Since the three C. ulmi subgroups (NAN, EAN, and nonaggressive) are genetically
and reproductively isolated sub populations of the fungus, each having its own char-
acteristic properties and range of variation, 15 it is likely that the d-factors of each
subgroup will be qualitatively different as a consequence of co-evolution with their
respective host population. Moreover, there is also some evidence for qualitative dif-
ferences between d-factors within a single subgroup such as within the NAN (see later).
 179

To allow for qualitative differences between d-factors and to discriminate between d-

factors originating from different isolates, they have been assigned code numbers on
the basis d l , d" . . . d". Thus, d l is the original d-factor discovered in French NAN
isolate H321, now designated H321d l . On transmission of the dl-factor from H321d l
to another isolate, 13F zP ll , the latter is denoted 13F zP ll d l , and so on. An isolate be-
lieved to be carrying more than one d-factor as a result of laboratory-controlled trans-
missions is denoted as such, e.g., 13F zP ll d 1 d z •

B. The "d-Reaction"
The "d-reaction" is a characteristic though variable pattern of mycelial growth and
sometimes of pigmentation that occurs when a diseased or d-infected "donor" isolate
is paired with a "healthy" recipient isolate on elm sapwood agar (ESAY (shown dia-
grammatically in Figures la to c). As a result of the transmission of the d-factor from
the donor, sectors of newly d-infected weak or altered growth habit develop at the
periphery of the young recipient colony along the junction line between the two colo-
nies (Figure la). These usually continue to develop as expanding sectors of altered
mycelium (Figure 1b), often giving a "butterfly" appearance when the colonies are
fully grown (Figure lc).
The appearance of a "d-reaction" varies according to environmental condi-
tions such as medium, temperature, and the spacing apart and relative growth rates of
donor and recipient isolates, such that there is in reality no such thing as a "typical"
d-reaction. Reactions involving a slower growing donor are often of a rather different
configuration and more striking in appearance (Figures 2a, d to f) than reactions be-
tween isolates of similar growth rates (Figures 2b, c). This may be because slower
growing donors really do tend to cause stronger d-reactions. Indeed, in d-reactions
involving very sick and slow-growing donors, the recipient colony is often so badly
affected by the d-factor that its growth becomes ragged and then stops either along its
leading edge (Figure 2e) or even around its entire periphery (Figure 20.
A subtle secondary structure is often seen in many d-reactions in which more deeply
shadowed triangular patches form along the leading edge and along the boundary of
infected and healthy mycelium of the newly infected sectors (Figures 2b, c, arrowed).
Many d-reactions are characterized by the production of red-brown pigment, either
within or along the perimeters of the newly infected sectors of the recipient (Figures 3a
to d) or both. Pigment lines corresponding to the shadowed triangular patches de-
scribed above may also be present (Figure 3a). Occasionally the pigment line at the
junction of infected and healthy mycelium may be double. The agar around the edge
of very slow-growing donor colonies and the edge of recipient colonies arrested by
d-factor transmission is very often strongly pigmented (Figures 2e, 0.
Since the subgroups of C. ulmi have somewhat different colony morphologies and
growth rates under given environmental conditions, d-reactions tend to take on a
slightly different appearance according to which subgroup is being studied. In addi-
tion, d-reactions between EAN aggressive, or between nonaggressive isolates are more
frequently of an arrested growth type (Figures 2d, f). This is thought to be due to the
interaction of the d-factor with the "up-mut" factor, which is present in EAN and
nonaggressive isolates but not in NAN isolates,1.5 and is discussed below.

C. The Cytoplasmic Location and Spread of d-Factors

As with most fungal viruses and plasmids, primary transmission of the d-factor is
believed to occur as a result of the formation of viable fusions between either leading
or more probably lateral hyphae of the donor and recipient isolates, and subsequent
secondary transmission between d-infected and healthy hyphae of the recipient (see
below). That the d-factor is a cytoplasmically located rather than a nuclear factor was
180 Fungal Virology
FIGURE 1. a through c: Diagram of the development of a d-reaction between a healthy recipient isolate, R, and a d-infected donor isolate, D, on ESA
medium. (a) Initial reaction; (c) developed d-reaction. Small arrows indicate direction of d-factor spread in the recipient via hyphal fusion; ds = newly d-
infected sector of recipient colony; vi = vegetative incompatibility barrage (if present); 0 = inoculation point. d through f: Experimental investigation of the
cytoplasmic location and spread of d-factors. (d) If the donor d-infected isolate (D) carries a nuclear gene marker for MBC (fungicide) tolerance, and
subcultures are taken from the recipient at the points shown, the following phenotypes are commonly obtained: b., MBC tolerant, d-infected; +, MBC
sensitive, d-infected; . , MBC sensitive, healthy. (e) If the donor d-infected isolate (D) carries a cytoplasmic (mitochondrial) marker for chloramphenicol
tolerance and the recipient (R) is healthy but carries the nuclear marker for MBC tolerance, and subcultures are taken from the points shown, growth on
single or double-drug medium usually occurs as follows: ., growth on MEA + chloramphenicol, or ~ sometimes growth on MEA + chloramphenicol and
on MEA + MBC, but not on MEA + MBC + chloramphenicol; +, growth on MEA + MBC, or (very occasionally) Ell growth on MEA + MBC and on MEA
+ chloramphenicol, but not on MEA + MBC + chloramphenicol. (f) When the donor isolate H321d' (D) was d-infected and the recipient 13F,P" (R) was
healthy, subcultures from points I and 8 were consistently d-infected, those at 2 and 7 often d-infected, and those at points 3 to 6 healthy. See text for further
details of experiments.

.....
00
182 Fungal Virology

FIG U R E 2
 183

demonstrated by effecting d-reactions between an MBC-(fungicide) tolerant d-infected

donor isolate (H321 toll d') and an MBC-sensitive recipient on ESA (see Figure la).
MBC tolerance is a stable single locus nuclear marker. 8 When subcultures taken from
the newly infected sectors of the recipient were tested for MBC tolerance and for ability
to transform a healthy culture, they were shown to be d-infected but still MBC sensitive
while subcultures from the morphologically normal or healthy region of the recipient
remained non-d-infected and MBC sensitive, and those from the donor, d-infected and
MBC tolerant (Figure 1a).
In another experiment, an NAN isolate carrying a genetic marker for chloramphen-
icol tolerance (CR ), a cytoplasmic marker of presumed mitochondrial location shown
in reciprocal crosses to be classically maternally inherited,24 was infected with the
d 2 -factor via ad-reaction (logl/3-8d 2 donor + H106 CRl recipient ... H106 CRld 2 ) on
ESA agar. Thus, the derivative H106CRld 2 carried both the presumed mitochondrial
marker and the d 2 -factor. It was then paired with its wild-type self (i.e., healthy chlor-
amphenicol sensitive) H106 to induce a further d-reaction (H106CRld 2 + H106) and
15 subcultures were taken from the resulting newly d-infected sectors of H106. All of
these were confirmed in further tests to be d 2 -infected, but chloramphenicol sensitive.
Five other subcultures taken from the H106 CRld 2 side of the junction line between the
two colonies were still found to be chloramphenicol tolerant. The same result was
obtained with four other isolate combinations: (W4 + H106CRld 2 ; H363 + H106
CRld 2 ; RDT38 + RDT38CRld 2 , and H363 + RDT38CRld 2 ). These observations sug-
gested that the d 2-factor was unlikely to be located within the mitochondria - certainly
not on the putative chloramphenicol-tolerant mitochondria - and that unless the latter
were at a selective disadvantage in the recipient's cytoplasm compared with wild-type
mitochondria, mitochondria were probably not transmitted to the recipient along with
the d 2 -factor.
To investigate this aspect further, a similar experiment to the above was carried out
but using MBC-tolerant recipient colonies in the combinations W2toll + H106CRld 2 ,
RDT38 toll + H106CRld 2 , W2toll + RDT38CRld 2 , and RDT38toll + RDT38CRld 2 •
Fifteen subcultures were taken from the newly d 2 -infected sectors of each recipient,
and five subcultures from the donor, along the junction line between the two colonies
as shown in Figure Ie. On removal, each subculture was split and tested for its ability
to grow on MEA + MBC, MEA + chloramphenicol, and MEA + MBC and chloram-
phenicol, and for its ability to give a d-reaction against a healthy recipient. Subcultures
from appropriate control isolates were used to confirm the efficacy of the test media.
The results broadly confirmed those of the first experiment. All 60 subcultures from
the newly d-infected segments in the recipients (Figure Ie) were MBC tolerantld 2 -in-

FIGURE 2. Appearance of d-reactions. (a) Developing d-reaction between a slower growing d-infected
donor isolate H321d l (below) and healthy recipient isolate 13F,P II (above), showing darker red-brown pig-
mented segments of newly d-infected growth developing in the recipient (arrowed). The white barrage is a
vegetative incompatibility reaction (see below). (b) Developed d-reaction between recipient FIP" (above) and
donor H321d l (below) in which the newly d-infected segments of the recipient show as altered, rather "shad-
owed" areas of growth (arrowed). (c) Similar d-reaction between recipient FIP,. (above) and donor H321d l
(below) showing a different structure to the "shadowed" newly d-infected segments (arrowed). (d) Growth
of recipient EAN isolate H332 (above) partially arrested by transfer of a d-factor from slow-growing EAN
donor P123d' (below, arrowed). Note the abnormal and irregular growth of the newly d-infected segments
of the recipient. (e) Growth of recipient NAN isolate W4 partically arrested by the transfer of the d'-factor
from slow-growing NAN donor logl/3-1Sd' (arrowed). Note the pigmented edge to the d-infected area of
the recipient colonies. (f) Growth of recipient EAN isolate H277 completely arrested around its colony
margin by the transfer of the d'-factor from slow-growing NAN donor Logl/3-8d' (arrowed). Note dark
pigment around entire margin of H277 colony. (c) also shows a fully vegetatively incompatible "wide" or
w-reaction barrage at the junction line between the two colonies; (a) shows a partially incompatible "nar-
row" or n-reaction, and (b) a slightly incompatible "line-gap" or Ig-reaction. All cultures were grown on
ESA medium at 20°C.
184 Fungal Virology

FIGURE 3
 185

fected. All but two of the 20 control subcultures taken from the donor colonies (Figure
Ie) were, as expected, chloramphenicol tolerant/d 2 -infected (although two were chlor-
amphenicol sensitive, presumably as a result of a local preponderance of wild-type
mitochondria in the donor mycelium concerned). Of additional interest, however, was
the fact that some subcultures from the donor colonies which grew on MEA + chlor-
amphenicol also grew on MEA + MBC (9 out of 18; Figure Ie shown at position @ )
and that some subcultures from one recipient grew not only on MEA + MBC but also
on MEA + chloramphenicol (7 out of 15; Figure Ie summarized at $). However, none
of these subcultures able to grow on both single drug media separately could grow on
the double drug medium MEA + MBC + chloramphenicol. Hence, they were shown
to be mixtures of donor and recipient mycelium and not to be nuclear/cytoplasmic
hybrids between recipient and donor - which would have been expected to grow on
the double drug medium!5 Thus they must have originated from intermittent intrusion
of hyphae or spores from the donor into the recipient's mycelium or vice versa. This
result therefore indicated that donor hyphae carrying chloramphenicol-tolerant mito-
chondria were present among portions of a recipient's mycelium to which d 2 -factor
transmission had occurred, yet no equivalent movement of donor mitochondria into
the recipient's mycelium could be detected. This seems to imply either preferential,
possibly even active, movement of the d 2 -factor into the recipient's hyphae or selective
exclusion of donor mitochondria from the recipient. Clearly more work of this or a
more molecular nature is required to resolve such possibilities. Nevertheless, the evi-
dence of both the above experiments strongly suggests that the d-factor is more likely
to be either a dsRNA mycovirus or a DNA plasmid than mitochondrial DNA.
The possibility that a d-factor spreads to parts of a recipient mycelium beyond the
visibly altered sectors of the d-reaction was investigated in an experiment involving the
d'-factor. Fourteen replicates of the pairing donor H321d' toll + recipient 13F 2 Pll were
set up on ESA medium. Following the development of d-reactions, samples were taken
from the recipients (13F 2P ,,) in each replicate at the eight positions illustrated in Figure
If and tested for d-infection by pairing against healthy 13F 2 P ll and H321d'. The sam-
ples were found to be consistently d-infected at positions 1 and 8 in all 14 replicates,
often infected (54070) at points 2 and 7, but never infected at points 3 through 6. Thus
little d-infection was detected beyond the visibly d-infected segments of the recipient.
That which was detected (points 2 and 7) was usually within 5 to 7 mm of the visibly
infected segments and was probably due to the intermingling of lateral hyphae and
spores from the infected segment rather than to d-factor transmission.
This result tends to support the view that transmission occurs mainly via fusions
between lateral hyphae and that resulting spread within a newly infected hypha tends
to be proximal (towards its growing tip) rather than distal. Thus at position 5 (Figure
1f) where no contact of donor and recipient lateral hyphae is likely because donor and
recipient leading hyphae would be in direct opposition, no transmission was visibly
initiated or detected. Only at position x (Figure 1f) is primary transmission initiated,
and it seems reasonable to suggest that this was facilitated by the fact that leading
hyphae of donor and recipient were growing tangentially towards each other, allowing
mixing and fusion of laterals which, combined with proximal internal spread and sec-

FIGURE 3. a through d: Pigment development in d-reactions on ESA medium. Photographs taken with
back-lighting. (a) Single pigment line at the boundary between healthy mycelium and newly d-infected seg-
ments in a recipient (same culture plate as Figure 2c). (b) Pigment lines formed both at the junction between
donor and recipient colonies and at the boundary of healthy and d-infected segments within the recipient;
(c, d) Pigment zones corresponding to newly d-infected segments of recipient colonies. Pigment lines are also
present in each case. e through f: Severely d-infected colonies on MEA medium. (e) d-Infected EAN isolate
P124d" showing extremely stunted growth after 10 days at 20°C. (f) Unstable "amoeboid" d-infected col-
ony-type after 14 days at 20°C (EAN isolate H277d 2 ).
186 Fungal Virology

(/)
2
Q)

ell
o
(/)

i
~
o
Z

50 55 60
Mean colony dia (mm) at 5 days

FIGURE 4. Comparative growth of subcultures taken from the healthy 0 and newly d'-
mfected segments D of recipient colonies. Above, subcultures from recipient I3F,P •• paired
with donor H32Id'. (See also Figure 2c). Below, subcultures from recipient 13F ,P" also paired
with H321d'. (See also Figure 2b). Arrows indicate group means. Note the much slower growth
rate of the newly d'-infected mycelia. Cultures grown on ESA medium at 20 c C.

ondary transmission between laterals of the recipient, led to the characteristically ex-
panding visibly infected segment. Where the effect of infection is so severe that hyphal
tip growth stops (Figures 1e; 2d to f) lateral fusion may ensure continued transmission
although a "segment" is not produced. However, it should be noted that in the pre-
vious experiment (Figure 1e) in contrast to the present experiment, d-infection was
detected in the recipient at the equivalent of position 5. This suggests that the pattern
of d-factor transmission may be different where the donor colony is very small and
severely d-infected, possibly because with such colonies growth of tips of leading hy-
phae is abnormal and disrupted (and hence more equivalent to growth of lateral hy-
phae), so that transmission occurs at an earlier stage of the encounter between donor
and recipient.

D. Effect on Growth and Reproduction

1. Mycelial Growth
An important and general feature of d-infection is that it has a detrimental effect on
linear growth. Indeed, a striking aspect of d-reactions already described is the reduc-
tion in growth of the recipient that may follow d-factor transmission. Subcultures
taken from such newly infected areas of a recipient in a d-reaction often grow much
more slowly than their previously healthy counterparts, as illustrated in Figure 4 which
shows the colony diameters on ESA of subcultures from newly d'-infected segments of
a recipient in comparison with those taken from noninfected areas of the same colony.
However, it would be wrong to assume either that the impact of d-factors on growth
is universally negative or that it is a simple interaction, as the results of the following
experiments will show.
 187

In an early experiment investigating the effect of acquisition of the d 2 -factor on the

radial growth of nine NAN isolates and one EAN isolate using malt extract agar
(MEA)', shown in Figure 5, most d 2 -infected isolates grew more slowly than their
healthy counterparts during the 48 to 168 hr measurement period,7 and were also much
less vigorous in growth habit, being generally rather thin and "moth eaten" in appear-
ance (see Figure 6). The single EAN isolate H277d 2 grew extremely feebly (Figure 6f)
and later developed into a colony with highly degenerate and unstable "amoeboid"
morphology (Figure 3f and see later). Interesting exceptions to the above were isolates
W2d" H301d" and H363d 2 • These three isolates not only grew faster than their healthy
counterparts, but were also more vigorous and striate-petaloid in appearance (Figures
5, 6d). In a repeat of the original d 2 transfer to W2, the derivative W2d 2 again grew
more vigorously, indicating that this was more than a chance phenomenon. However,
after a period in stock culture, the same d 2 -derivatives of all three isolates were found
to grow more slowly than their healthy counterparts on MEA. Some other NAN recip-
ients of the d 2 -factor have behaved similarly, showing an initial increase in growth
vigor but a decrease after a period in culture.
Another interesting aspect of the same growth-rate test, illustrated in Figure 5b, was
that the d 2 -infected isolates grew much more slowly than the healthy isolates during the
first 0 to 48 hr of growth, with a mean initial colony diameter (excluding H277d 2 ) at
48 hr of 7.8 mm, compared with 11.3 mm for the healthy isolates. The same effect
occurred in two similar growth-rate comparisons on MEA. It appears that d-infected
isolates often make a much slower start (0 to 48 hr) but tend to accelerate in growth
rate such that over the following 120 hr of the measurement period they show only a
slightly slower average growth rate than do healthy isolates.
The detrimental effect of d-infection on growth is generally found to be much more
marked on ESA than on MEA agar. This is illustrated by the experiment in Figure 7
which involved the same isolates as those shown in Figure 5. Much larger differences
were obtained between the growth rates of the d 2 -infected and healthy groups. More-
over, the three d 2 -infected isolates which grew faster than their healthy counterparts in
the experiment on MEA (Figure 5) certainly did not do so on ESA. In appearance,
while the healthy colonies were vigorous and striate-petaloid, the d 2 -infected colonies
were wispy and thin, most with intense red-brown pigmentation near the centre of the
colony. Several repeat experiments on ESA involving isolates carrying the d 2 -factor
and their healthy counterparts have given very similar results.
The more severe effect of d-infection on growth on ESA as compared with MEA has
proved to be a consistent phenomenon, of particular interest because ESA is a medium
derived directly from host material. It is also one of the few media on which mature
perithecia and synnemeta are formed by all three C. ulmi subgroups and on which
clear well-structured vegetative incompatibility reactions occur. 9 On the whole, the
growth of the fungus on ESA is more structurally differentiated and developmentally
advanced than on MEA. Therefore, while the stronger interaction between d-infection
and mycelial growth and differentiation on ESA could be an artifact, it might also be
a result of co-adaptation on elm and hence a response by the d-factor, fungus, or both
to particular elm-related chemical constituents in the medium such as bark polyp hen-
ols. Since the red-brown pigment produced by severely d-infected isolates on ESA
might provide some clues as to the physiology of this interaction, its chemical identity
and role in d-infection would be worthy of investigation. Since it could itself conceiv-
ably be a polyphenol, it is worth bearing in mind that some polyphenols are thought
to have antiviral properties in fungi. ' ° The possibility that a toxin is involved in
d-infection should also be considered.
188 Fungal Virology

a
8

Vl
III 6
3
til
o
Vl

4
o
o
Z
2

11 10 9
0
14 b
0
0
0
c9
.r: • 0
co 00
'<t 10
0
•
til
.0 0

• •
E
E
til
•
•
"'0
>.
c:
•
•
0
0 5

•
0

c:
til
III
~

•
0
0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0
Mean radia I growth rate mm/day-1 (over 48-168h)

FIGURE 5. Growth rates of healthy NAN and EAN isolates 0 compared with their newly d'-infected
counterparts II on MEA medium. (a) Mean radial growth rates measured over the 48 to 168 hr growth
period at 20°C. Key: 1 = isolate W2IW2d'; 2 = RDT38; 3 = H363; 4 = H623; 5 = H351; 6 = H106; 7 = W4;
8 = H301; 9 = H625; 10 = H277 (EAN); II = W2toll. Arrows indicate group means with isolate no. 10
omitted. Isolates are NAN agressive unless otherwise stated. Note the transient stimulation of the growth of
W2(= I), H363 (= 3), and H30l (= 8) by d' infection; the very slow growth rate of EAN isolate H277d' (=
10) compared with H277; and the poor growth of W2tolld' (= 11), the MBC-tolerant form of W2. See also
Figure 6 for photographs. (b) Colony diameter after only 48 h plotted against the above radial growth rates.
Note the apparent improvement in growth of many d'-infected isolates following the initial 48 h period, and
the consistently poor growth of EAN isolate H277 d' (= 10) and W2 wild' (= 11) throughout the measure-
ment period.
 189

FIGURE 6 Effect of mfection with the d'-factor on mycelIal growth. Photographs of colonies after 7-
days growth on MEA at 20°C. Healthy colony above, d'-infected counterpart below. a through c: NAN
isolates H351, H625, and HI06, showing poor growth of d'-infected counterparts. (d) NAN Isolate W2
showmg unusual transient stimulation of growth of W2d'. (e) Isolate W2toll; note the effect of MBC tol-
erance in reducing the growth of W2 toll compared with W2, and the particularly severe effect of the d'-
factor on the growth of W2 toll. (f) EAN isolate H277, showing the partIcularly severe effect of the d'-factor
on its growth.
190 Fungal Virology

10 a

'"OJ 42
8Fl 52
ro
0

'"
"0 5
0
z

10 20

Mean colony d,a (mm) al 5 days

FIGURE 7. (a) Comparative growth rates of healthy 0 and d'-infected II NAN and EAN isolates on ESA
medium. Note the more severe effect of d'-infection on ESA compared with MEA medium shown in Figure
S. Key: isolate nos. I-II as in Figure S. Arrows indicate group means with isolates 10 and II omitted. (b)
(Inset) Growth rates of d' versus d'-infected forms of H321ss1S. Key: 12 = H321ss1S; 13 = H32lsslSd'; 14
= H32IssISd'.

2. Growth Stability
In the growth-rate test illustrated in Figure 7 the EAN isolate H277d 2 was again
found to be particularly severely affected by d-infection. Indeed d-infected isolates of
both the EAN and nonaggressive subgroups often exhibit severe and sudden degener-
ation after a period in culture and because of this can be difficult to maintain. Many
eventually produce unrecognizably altered and clearly very sick and scarcely growing
colony types (Figures 3e, f) and may have to be discarded when derivatives of wild-
type morphology can no longer be recovered. Others may recover following sin-
gle sporing (see later). NAN isolates seem in general to be better able to cope with
d-infection, but a proportion degenerate to the unstable "amoeboid" colony type (Fig-
ure 3f) in which sectors of faster, near wild-type growth appear, especially following
exposure to light, but then stop growing to be replaced by others. When tested, such
"near wild-type" growth sectors are usually also found to be d-infected. Most "amoe-
boid" NAN isolates obtained directly from elm bark or twigs in nature by the author
have been shown to be d-infected.
The greater sensitivity of EAN and nonaggressive isolates to d-infection in culture
may be due to the combined effect on the recipient of the d-factor and the "up-mut"
factor. The latter is a normal component of EAN and nonaggressive genomes but is
apparently absent in the NAN aggressive. The "up-mut" factor normally causes some
colony irregularity and unevenness in EAN and in nonaggressive isolates and, when
fully expressed, induces a reversible and rather unpredictable switch from wild-type
colony morphology to a distinct "up-mut" form of slower growth and extremely reg-
ular appearance.'·s It seems likely that the additional influence of the up-mut factor is
at least partly responsible for the more severe growth disturbance which occurs when
the EAN and nonaggressive subgroups are d-infected.
A further clue to the causes of sudden degeneration of d-infected isolates comes
from certain NAN isolates. Although many NAN isolates seem better able to cope with
d-infection in culture, common exceptions are NAN isolates carrying the genetic
marker for MBC (fungicide) tolerance such as the MBC-tolerant form of isolate W2,
W2 toll, shown in Figure 6e. MBC-tolerant isolates are typically slower growing than
their normal MBC-sensitive counterparts, although they carry only a single allelic dif-
 191

ference l l (cf. the growth of W2 toll with W2 in Figure 5). When d-infected, they too
are inclined to sudden and severe colony degeneration (see the growth of W2 tolld 2 in
Figures 5, 6, and 7), whereas their d-infected but otherwise normal counterparts are
not. Similarly, while NAN isolates RDT38 and H106 grow normally on MEA, and
their chloramphenicol-tolerant healthy counterparts RDT38CRl and HI06CRl grow
only a little slower, the newly d 2 -infected forms of the latter, RDT38 CRld 2 and
H 106 CRld>, are degenerate, grow extremely slowly, and are very difficult to maintain.
Thus a common factor in the tendency to severe degeneration when d-infected ap-
pears to be a slower intrinsic growth rate, whether resulting from the influence of the
"up-mut" factor, pleiotropic effects of loci conferring drug resistance, or some other
possibly external cause. This relationship could conceivably be due to a lower level of
some form of active resistance to d-infection in slower growing mycelia, such that the
resistance mechanism is overcome by d-infection and normal cellular integration
breaks down. Alternatively, an explanation might be found in growth dynamics. Pos-
session of an intrinsically faster growth rate may favor growth stability in spite of
d-infection, simply because the rate of hyphal tip extension either somehow exceeds
d-factor replication rate or leads to the hyphal tip continually escaping the immediate
effects of d-infection. Slower growth would favor degeneration because the effects of
the d-factor would catch up with the growing tips. Somewhere along the scale from
fast to slow growth rate, a point is reached at which the impact of d-infection on
growth becomes critical, the point at which this occurs varying according to other
environmental stresses on the mycelium. The latter hypothesis is supported by evidence
that on MEA, growth of some d-infected mycelia begins slowly and then accelerates
while that of others remains slow (see Figure 5b). It is also of interest to recall at this
point that slow-growing and unstable d-infected mycelia, as donors, often cause much
more intense d-reactions in recipients (Figs. 2d-f). Possibly they carry a higher initial
d-factor load as a result of a higher d-factor replication rate/hyphal extension rate
ratio, with the result that more d-factors are transmitted to the recipient, with yet more
deleterious results. Those mycelia which begin slow and remain slow growing may have
higher initial d-factor loads such that no "escape" from the effects of d-infection by
the leading hyphae is possible.

3. Comparative Effects of d'- and d 2 -factors

On the whole, isolates carrying the d'-factor (H321d' and related isolates) appear to
grow better than isolates carrying the d 2 -factor. Unfortunately, critical comparisons of
the effects of different d-factors against a single genetic background are difficult to
obtain because of the limitations imposed on d-factor transfer by the vegetative incom-
patibility system (see later). However, as a result of a chance observation of a d-reac-
tion against a d 2 -infected donor it was possible to select a d 2 -infected form of isolate
H321ss15 (a derivative of H321d' which had been freed from d' factor by single spor-
ing). A d'-infected derivative of H321ss15 was therefore reconstructed and the relative
growth rates of H321ss15' and H321ss15d 2 were compared (see Figure 7b). The latter
proved much slower growing than the former, and both were slower growing than their
non-d-infected counterpart, H321ss15. This result suggests that there may be consid-
erable differences in the quantitative effects on growth caused by different d-factors.

E. Effect on Conidial Viability

C. ulmi produces abundant conidia (Sporothrix stage) on agar media and, more
importantly, in beetle-breeding galleries in diseased elm bark. '2 Preliminary observations
suggested that conidial viability of d-infected mycelia differed from that of healthy
mycelia. A series of experiments was therefore initiated in which spores from similar
aged d-infected or healthy colonies on MEA were spread on MEA slide cultures, main-
tained at 100070 humidity for 16 hr, and examined for viability and germ-tube length.
192 Fungal Virology

Table 1
EFFECT OF d-INFECTION ON CONIDIAL GERMINATION

Mean germ-tube length

0/0 Germination· 0/0 Reduction, (/Am)" % Reduction,
d-infected d-infected
Isolate No. Healthy d-Infected vs. healthy Healthy d-Infected vs. healthy

Experiment I
(age II days")
H363/H363d' NT NT 13.5 18.2 -39.3
CI13/CII3d' NT NT 10.5 8.7 -27.7
Experiment 2
(age 6 days")
H363/H363d' 80 24 -70.0 3.6 3.1 -14
H277 /H277d' 80 30 -62.5 5.1 4.1 -19.6
C1l3/C113d' 80 46 -39.5 4.9 5.0 +2.0
Experiment 3
(age 11 days")
H363/H363d' 49 32 -34.5 7.2 3.8 -52.6
W4/W4d' 67 41 -38.8 10.0 8.5 -15.7
H106/H106d' 56 37 -33.9 9.5 11.1 +17.4
Experiment 4
(age 7 days")
W4/W4d' 95 41 -56.8 22.3 14.0 -37.2
H106/H106d' 92 80 -13.0 21.4 18.3 -14.6
CI13/CI13d' 94 64 -31.9 22.6 18.1 -19.9
% Reduction -42.3 -20.5
overall

Visual assessment of 100 spores per isolate.

Measurement of 25 or more germ-tubes per isolate.
Age of colony on MEA. NT = Not tested.

The results are shown in Table 1. Even with healthy colonies, the highest germina-
tion levels achieved were only ca. 95070. Higher germination levels overall occurred in
6 to 7 day-old as compared with 11 day-old colonies. Nonetheless, there were usually
much lower levels of germination in d-infected than in "healthy" isolates, the per-cent
reduction in germination in d-infected isolates ranging from 13 to 70070 with a mean
over all tests of 42.3070. The level of reduction for a single isolate varied from one
experiment to another, though this is perhaps not surprising for a cytoplasmic phenom-
enon which exhibits such complex environmental interactions as d-infection.
The mean length of germ tubes of d-infected isolates was also generally lower than
that of healthy isolates, with length reductions ranging from 14 to 52070 (overall mean
reduction, 20.5070). It should also be noted that these germ-tube measurements apply
only to surviving viable spores, and that with the d-infected material a proportion of
the remaining germinating spores for which germ-tube lengths were measured may
have been free from d-infection (see later). This may partly account for the fact that
with one isolate in one test (HI06, Experiment 3) the mean germ-tube length of
d-infected material was greater than that of the healthy material.

F. Effect on Perithecial Formation

Ascospores are important for the spread of C. ulmi both during the saprophytic
phase in elm bark and on the vector beetles. s. [2. [3 The fungus is heterothallic, with two
mating types A and B controlled by the A and B alleles of the mating-type or mt locus. 9
Both mating types are required for perithecial formation which involves the fertiliza-
 193

tion of ascogonia or protoperithecia (9) by spores of opposite mating-type acting as

spermatia. 9
The effect of d-infection on perithecial formation on ESA seems to depend on the
severity of its effect on growth, but experiments on this aspect have been difficult to
assess quantitatively. In isolates in which growth rate on ESA is only moderately af-
fected by d-infection such that the mycelium reaches the edge of the plate, the fre-
quency of protoperithecial formation usually differs little from that of a faster-growing
healthy counterpart. When the protoperithecia are fertilized by conidia of opposite
mating-type, resulting perithecial frequencies in the d-infected and healthy isolates are
also similar. However, when growth on ESA is actually disrupted by d-infection, as in
"amoeboid" or severely degenerate isolates, protoperithecial formation is often very
patchy and of much lower frequency overall than in a healthy counterpart. When such
protoperithecia are fertilized by conidia, resulting perithecia are often not only com-
mensurately patchy but tend to develop much more slowly, have shorter, thicker more
irregular necks, and the production of a sticky drop of ascospores at the tip of the neck
is delayed and sometimes prevented.

G. Regulation of d-Factor Transmission

The transmission of d-factors from infected thalli to other individuals, whether be-
tween mycelia or via spores, is far from unrestricted but is regulated or controlled by
at least three difference mechanisms. These will now be considered.

1. Regulation by Vegetative Incompatibility

Like a number of other outcrossing ascomycetes,14 16 C. ulmi has a "heterogenic"
vegetative or somatic incompatibility system which restricts the occurrence of func-
tional hyphal fusions between adjacent mycelia to certain, usually very similar geno-
types. In C. ulmi the system is polygenic, with multiple alleles possibly at at least one
of the loci concerned. 9 In fully vegetatively compatible reactions (or c-reactions) the
vegetative compatibility (v-c) loci of both mycelia are qualitatively identical, and fully
functional fusions occur between them. In fully incompatible reactions (or w-reactions)
the alleles at all of the v-c loci are different, and functional fusions allowing intermin-
gling of nuclei and cytoplasm are largely prevented. Partially compatible (n, 1, and 19)
reactions also occur in which the mycelia have some but not all v-c alleles in common. 9
Since it restricts formation of confluent hyphal fusions, it is hardly surprising yet
extremely significant to find that vegetative incompatibility acts as a major regulator
of d-factor transmission between mycelia. In a series of experiments assessing the
spread of the d 2 -factor between mycelia of wild NAN isolates on the basis of the oc-
currence of d-reactions, transmission occurred in only ca. 4070 of the fully incompatible
or w-reactions, ca. 50% of partially compatible n-reactions, and 100% of the compat-
ible or c-reactions (see Table 2). Fl backcross isolates of varying relatedness to isolate
H321d 1 gave a slightly higher rate of transfer across w-reactions (Table 2). The latter
could reflect the cytoplasmic relatedness of the mycelia concerned, the influence of the
d-factor itself, or some qualitative aspect of H321 v-c loci. Recent evidence 25 suggests
that the latter is probably responsible, and that certain w-reactions, while appearing
morphologically typical, are more likely to allow transient cytoplasmic continuity, pos-
sibly because of a delay in the expression of vegetative incompatibility.
Nevertheless, in broad terms, these results indicate strongly restricted transmission
of d-factors between fully vegetatively incompatible mycelia, less restricted transmis-
sion between partially compatible mycelia (n-reactions), and virtually unrestricted
transmission between compatible mycelia. Since fully incompatible reactions are, with
certain exceptions the commonest form of interaction between thalli in nature in both
the aggressive and nonaggressive subgroups of C. ulmi,5.9.26 vegetative incompatibility
194 Fungal Virology

Table 2
TRANSMISSION OF d-FACTORS ACROSS VEGETATIVE
INCOMPATIBILITY REACTION TYPES IN C. ulmi

Wild NAN isolates· Backcross isolates'

Reaction type against
d-infected donor w n 1I1g c w n l/2g c

No. of isolates tested 79 8 3 21 5 6 IO 9

No. giving d-reaction as 3 4 2 21 2 10 9
recipient
"10 d-Factor transmission <4 50 66 100 20 33 100 100

111 Wild NAN recipient isolates paired against d-infected donor W2tolld' (belonging
to the NAN v-c "super-group"). Pairings carried out on ESA medium.
Selected F,-F. backcross progeny of varying relatedness to H321 (see Reference 9)
paired as recipients against d-infected H321d' as donor. Pairings on ESA medium.

From Brasier, C. M., in The Ecology and Physiology of the Fungal Mycelium, Jennings,
D. H. and Rayner, A. D. M., Eds., Cambridge University Press, 1984,451. With per-
mission.

must be of considerable importance in restricting the spread of d-factors in the field.

The "leakiness" of some "fully incompatible" reactions, however, will probably allow
some degree of spread of d-factors between different v-c groups. In addition, partially
compatible reactions (n-, 1-, and Ig-reactions) might act as a bridge between different
v-c groups: a particular d-factor might spread from one v-c group to another via an
intermediary with some genetic similarity to both!!.22
Unless otherwise stated, descriptions of tests for d-reactions between donors and
recipients in other parts of this paper usually involve vegetatively compatible mycelia
(i.e., c-reactions).

2. Reduced Transmission via Conidia

The effect of d-infection in reducing the viability of conidia has already been de-
scribed. To investigate the level of d-infection among surviving viable spores, conidia
from a lO-day-old colony of H321d 1 on MEA were spread on MEA, germinated, and
germlings picked off after 24 hr and again at 48 hr (in an attempt to favor slow grow-
ers) and the radial growth rates of 34 of the resulting colonies examined on MEA at
20°C.
The resulting single-spore derivatives showed a considerable range of growth rates
(Figure 8a), and four isolates in particular grew much more slowly over the 48- to 168-
hr growth measurement period. A comparison of the growth rates of these isolates
during their first 48 hr with their growth between 48 to 168 hr (Figure 8b) also showed
that these isolates had started growing only moderately and continued to grow poorly,
while other isolates which began slowly became faster, and yet others started faster and
sustained faster growth (cf. also Figure 5b). When the five slowest-growing and five of
the faster-growing isolates were tested for their ability to give a d-reaction as a donor
when paired with a healthy vegetatively compatible recipient, all five slow growers
caused d-reactions but only one of the faster growers did so (Figure 8). Moreover,
when all ten isolates were paired with H321d 1 , the latter caused d-reactions only against
the four isolates shown to be negative for d-infection in the first test. Thus this result
confirmed that four of the five faster-growing isolates were probably free from
d-infection.
In a similar experiment with EAN isolate Pll1d s , the four slowest and five fastest
 195

15 a -

10 -

o
Z

-
• '--

•
•
r-;-
0 0

• 0 0
•
~

...
<0 36 b
;;; ,
0

E 28
E
0, ,x
, , •
•• • •
, x

'" 'x'
•
"0 20 "'0'
>- , 0
co , ,
"0
0
,,
12
"co
'"
'"
:::ii

3.0 3 5 4 0 4 5 5 0

Mean radial grow1h rate mm/day-1 (48-168h)

FIGURE 8. Growth rates of single conidial derivatives of H321d' on MEA. (a) Distribution of mean radial
growth rates (measured over 48 to 168 hr). (b) Plot of colony diameters at 48 hr against the above radial
growth rates. 0. subsequently tested and shown to be healthy; e.
subsequently shown to be d'-infected.

growing of 30 single-spore derivatives of PIll d S were tested for their ability to cause
d-reactions as donors against a healthy EAN recipient (H224). One single-spore deriv-
ative, which was by far the slowest isolate and also highly unstable, caused a strong
d-reaction; another slow-growing derivative, a weak d-reaction; and the remaining
seven derivatives, no d-reaction. In a further test the latter all acted as recipients when
paired with PIlI d S , and were thus confirmed apparently non-d-infected.
In a third experiment, a comparison was made between the frequency of d-infection
in conidial germlings and in the yeast-stage of C. ulmi, the latter being of prime im-
portance during the pathogenic phase of the fungus in elm xylem. 17 Conidia from the
surface of a 14-day-old MEA colony of H321d l were suspended in 2 ml sterile water,
a proportion spread on MEA, and 50 resulting germlings picked off over 24 to 48 hr
and stored at 5°C when 1 cm in diameter. The remainder of the conidial suspension
was used to seed 200 ml of Tchernoff's liquid culture medium to induce the yeast-
stage, and shaken at room temperature for 6 days, after which 0.1 m£ aliquots were
196 Fungal Virology

spread on MEA and a further 50 germlings picked off. From each of the two sets of
50 germlings, 35 colonies, one derived directly from the germinating conidia and the
other from the same conidia but via the yeast-stage, were then paired as donors with a
healthy recipient (13F 2 P ll ) to examine their level of d-infection. Only twelve (34070) of
the 35 conidial derivatives and 13 (380/0) of the yeast-stage derivatives gave d-reactions
and were thus detectably d-infected, the remaining 62 to 66070 being apparently healthy.
All three of the above experiments indicate that only a proportion of conidial germ 1-
ings from a d-infected mycelium are likely to be carrying an immediately effective
d-factor load, and that slow-growing derivatives are more likely to be infected than
fast-growing ones. It appears that the d-factor can be either lost or suppressed during
conidiogenesis. However, it is also of interest to note from the last experiment that
although in the yeast-stage many generations and hence many cell divisions are likely
to have occurred, the frequency of d-infection in the spore population at the end of
the experiment (ca. 38070) was similar to that at the beginning (ca. 34070). Thus a mech-
anism may exist which, in the absence of intense selection against d-infection such as
may occur in host xylem, enables the frequency of d-infected spores to remain roughly
constant in each generation. The explanation for the above observations may lie in the
phenomenon of latency, discussed in Chapter 7. It should also be remembered that
during the yeast-stage, as with conidial germination, the actual level of d-infection in
each spore generation may be much higher than indicated, because many more heavily
d-infected spores may simply fail to germinate.

3. Restricted Transmission to Ascospores

The transmission of d-factors to ascospores has also been investigated in a number
of experiments. In the first, sexually compatible isolates H301d 2 and H351d 2 , each of
a different mating type but the same vegetative compatibility type, were paired on ESA
medium, and 15 F, single ascospore progeny sampled from each of 3 resulting perithe-
cia. They were then tested for their ability to give d-reactions against healthy forms of
H301 and H351. None of the 45 progeny gave d-reactions against either parent, and
were thus apparently free from d-infection. In a very similar test using the mating
combination H321d' x 13F.P,d', a further set of 45 F, progeny originating from three
perithecia also caused no visible d-reactions against healthy forms of their parents
(H321ss15 and 13F.P,).
In another experiment, protoperithecia of isolate H351d 2 were fertilized with conidia
of the sexually compatible and vegetatively compatible healthy isolate H106. Fifteen
of the resulting perithecia were surface sterilized and ascospores allowed to ooze en
masse from the ostiole. Each ascospore mass was then allowed to form a single colony,
and the 15 resulting mass ascospore co10i1ies were tested for d-infection against H106.
Although the female parent H351d 2 was severely d-infected and degenerate in mor-
phology on ESA, all 15 mass ascospore colonies grew normally on ESA, and none gave
d-reactions against H106.
Thus ascospore progeny are apparently substantially free from d-infection. More-
over, when compared with their parent isolates, ascospore-derived colonies are often
more stable and regular in appearance in culture. The latter phenomenon is particularly
noticeable with ascospore progeny of the nonaggressive subgroup. Wild-type nonag-
gressive isolates are particularly prone to degeneration in culture, but their ascospore
progeny tend to be much easier to maintain. As already discussed (Section I.D.2),
d-infection may be responsible for much instability in wild isolates of C. ulmi, partic-
ularly isolates of the EAN and nonaggressive subgroups.
It is important to note again here the experimental distinction between heritability
of d-infection and of chloramphenicol resistance. The latter, as already mentioned, is
a cytoplasmically and probably mitochondrially located trait which in C. ulmi has
 197

I
6

rn 2
(I)

'orn"
o
-
I
o
o
Z
4

o
o 20 40 60 80

Pa thogenicity (mean % defoliation)

FIGURE 9. Comparative pathogenicity (mean "10 defolia-

tion) on English elm of healthy (above) and d'-infected (be-
low) NAN isolates. Arrows indicate group means. (r, isolate DVI
3/8d'; 6" isolate H321d'.

isolates at 58.6070. What was unexpected, however, was the much greater range of
defoliation caused by the infected isolates (29.5 to 81.8%) compared with the nonin-
fected isolates (46.8 to 74.3%) and the fact that four of the d 2 -infected isolates
(RDT38d>, H623d', W2d', H363d') actually caused considerably more mean defolia-
tion than their healthy counterparts (Figure 10).
Eighteen weeks after inoculation, the originally d'-infected isolates were reisolated
from the lowest diseased branch of each replicate (the furthest point of disease below
the inoculation point) and tested for d-infection against a healthy recipient. Of the
eleven originally d 2 -infected isolates, only one was still detectably d-infected in any of
its replicates. This isolate, H363d 2 , was not only still d-infected in three of its four
replicate reisolations, but, interestingly, was also by far the most pathogenic isolate in
this test (Figure 10, no. 3).
Overall, therefore, the results of this experiment confirmed that a d-factor is likely
to be lost during dissemination in the xylem. This may be a result of two interacting
mechanisms. It may be due, firstly, to the elimination of d-factors during sporogenesis
in the yeast-phase comparable to that already described for conidia and for liquid
cultures (Section 1.0.2), or; secondly, to selection favoring the spread of the resulting
fitter and presumably faster-growing non-d-infected components of the xylem popu-
lation. Noninfected genotypes might be especially favored during the penetration of
pit membranes where hyphal growth is probably involved. It is also conceivable that
non d-infected hyphae or spores would be better able to cope with active host resistance
and be more effective in producing toxins and enzymes.
In the light of these observations, the sustained d-infection of isolate H363d 2 within
the tree seems all the more remarkable, since it is necessary to assume that in this isolate
198 Fungal Virology

proved to be faithfully maternally inherited in reciprocal crosses. 24 In contrast, the

d-factor, as shown above, is not usually inherited via the ascospores even when the
female parent is d-infected. This provides further evidence, together with that dis-
cussed in Section I.C, that the d-factor is not directly associated with mitochondrial
DNA.

H. Pathogenic Behavior of d-Infected Isolates

In nature C. ulmi usually enters the vascular system of healthy elms via wounds
made by the vector elm bark beetles. The comparative pathogenic ability of different
C. ulmi isolates is therefore usually tested by inoculating spore suspensions of the
fungus directly into the xylem (for methods, see Reference 7). Once in the xylem, the
fungus exists mainly as the yeast-stage. 17 . 1 • Disease is assessed on the basis of the degree
of wilting of the crown that results.
In a preliminary experiment on the influence of d-infection on pathogenicity, 13
healthy isolates related to H321d ' (mainly either single ascospore lines from back-
crosses involving H321 9 , or its single conidial derivatives; see Figure 8) were compared
with H321 d I and 12 similarly related d I-infected derivatives (newly infected recipient
isolates or d'-infected single conidial lines; see Figure 8). Spore suspensions of each
isolate derived from 16-day-old colonies on MEA were inoculated into 4-year-old
clonal Ulmus procera near the top of the crown, with four replicates per isolate, in
June 1982. Disease was assessed as per-cent defoliation at 4-week intervals thereafter.
Unfortunately drought stress in a high proportion of the inoculated trees prevented
individual healthy vs. d-infected isolate comparisons from being made. However, by
12 weeks the mean defoliation caused by the pooled d'-infected isolates, 36.91110 (range
7.6 to 58.81110), was a little less than that caused by the healthy isolates, at 44.71110 (range
23.8 to 58.31110), as shown in Figure 9.
The possibility of either loss or persistence of d-infection as the fungus spread down-
wards through the xylem of a tree was examined by destructively sampling at the end
of the summer (16 weeks) three replicate trees of the following two isolates: DVI3!8d ' ,
the least pathogenic isolate (7.61110 mean defoliation), and H321d I (47.21110 mean defol-
iation) (Figure 9). For each replicate, the fungus was reisolated (1) from the xylem at
the original inoculation point, and (2) at regular positions on the main stem above and
below it, including the lowest point of downward disease spread. The latter point av-
eraged only 23 cm for the replicates of DVI3!8d' and 57 cm for those of H321d'.
Although DVI3!8d ' was still found to be d-infected at the inoculation points and at
several other reisolation points close by, it remained d-infected at only 9 out of 19
reisolation points in the 3 trees overall (470'f0). In contrast, H321d' was found to be
entirely free from d-infection at all seventeen reisolation points including the 3 original
inoculation points. This somewhat unexpected result showed that d-infection could be
quickly lost as the fungus spread through the tree. It also suggested a relationship
between level of disease caused and the sustained level of d-infection of the fungus.
In June 1983 the pathogenic ability of 11 NAN isolates with and without the
d 2 -factor was compared. Prior to the experiment the growth rates of the isolates were
tested on ESA (see Figure 7). Each isolate was then inoculated to four replicate trees
of 5-year-old U. procera. Conidial inoculum from MEA plates was applied to scalpel
cuts in the xylem (at a point near the top of the crown) using paint brushes in an
attempt to simulate beetle-mediated infection, and the inoculation point sealed with
PVC tape. This inoculation technique did, however, result in a rather high overall
incidence of failed inoculations (151110) which was independent of whether an isolate
was d-infected or healthy. The pathogenicity distributions of the d 2 -infected and non-
infected isolates at 12 weeks is shown in Figure 10. The overall mean defoliation caused
by the d 2 -infected isolates was 56.41110, very close to that of the equivalent noninfected
 199

5
j
4
11
3
8

- 2
Ul
Ql
9 7 10
<1l
0
4 2 6 3 5
-
~
0

j
0

0
z
3

2
9 8 4

7 2 3
o ~~r-~--~--~--+---~~---+--~--~--+-~~~--~
20 30 40 50 60 70 80 90
Pathogenicity (mean % defoliation)

FIGURE 10. Comparative pathogenicity (mean "70 defoliation) on English elm of

healthy isolates (above) and d'-infected isolates (below). Key: I = isolate W2IW2d'; 2
= RDT38; 3 = H363; 4 = H623; 5 = H35I; 6 = H106; 7 = W4; 8 = H30I; 9 = H625; 10
= H277 (EAN); 11 = W2 toll. Arrows indicate group means. Note the higher level of
pathogenicity of the d'-infected forms of H363 (= 3), RDT38 (= 2) and H623 (= 4), and
W2 (= 1) compared with their healthy counterparts. All isolates NAN aggressive unless
otherwise stated.

d-infection may actually have been favored by selection. If this is so, then the d-factor
itself must be responsible for the very high level of pathogenicity of this isolate. This
conclusion is strengthened by earlier evidence that H363d 2 grew faster and more vig-
orously on MEA than did its noninfected counterpart (Figure 5a) (although it was not
faster on ESA) and by the fact that two other d 2 -infected isolates showing higher path-
ogenic ability than their healthy counterparts, W2d 2 and H301d 2 , also grew more vig-
orously on MEA (Figure 5a) - but again not on ESA. Thus occasionally d-infection
may actually promote both pathogenic ability and growth, adding a further twist to an
already complex behavioral picture. The contrast between the apparently consistently
severe impact of d-infection on the fungus on ESA medium and its occasional stimu-
lation of the fungus on MEA and in the xylem may, as already postulated, result from
specific nutrients in ESA inducing a more differentiated state in the mycelium. Cer-
tainly in the xylem the fungus probably remains in a highly undifferentiated or juvenile
state. 17 . IS

II. IMPACT OF d-FACTORS ON DUTCH ELM DISEASE

Up to this point the d-factor has been considered mainly from the point of view of
its biological effects on its host under experimental conditions. An intriguing aspect of
the d-factor, however, is that its host is itself the cause of one of the most destructive
plant diseases in the northern hemisphere. Of considerable additional interest, there-
fore, is the wider impact that d-factors may have on the disease causing ability of C.
ulmi, including not only any control that they may exert via the fungus on natural
levels of Dutch elm disease, but also their potential as an agent in man-managed disease
200 Fungal Virology

o
A
~

~
, ~~

r:J~~

~ ~y oL· '\

({\:'I~: ;
2

~
FIGURE 11. The main growth and dispersal phases of C. ulmi in relation to the annual cycle of Dutch
elm disease. A, Bark (saprophytic) phase. In late summer and autumn trees weakened by the disease become
breeding sites for the vector beetles (1). Beetle larvae cut galleries in the bark (2) and C. ulmifruits in the
breeding galleries (3). B, Flight phase. In spring and summer, adult beetles emerge carrying spores of C.
ulmi (4). C, Feeding-groove phase. The newly emerged beetles fly to feed in twig crotches of healthy elms
(5). As a result of beetle feeding, the pathogen may be introduced into the xylem. D, Xylem (pathogenic)
phase. Infected twigs and branches wilt, showing characteristic streaks or spots in the infected annual ring
(6). (After Reference 5; diagram adapted and redrawn from Peace, T. R., Pathology of Trees and Shrubs,
Oxford University Press, London, 1962,419. With permission.) (From Brasier, C. M., in Advances in Plant
Pathology, Vol. 5, Ingram, D. S. and Williams, P. H., Eds., Academic Press, London, in press. With
permission. )

control programs. In order to properly assess this aspect of d-factor behavior it is

necessary to consider the impact of d-factors not upon the response of individual C.
ulmiisolates, as in a Petri dish or inoculated tree, but upon the sum of their effects on
a local C. ulmipopulation or upon the C. ulmipopulation at large. It is also necessary
to consider the potential for spread and expression of d-factors during the contrast-
ingly different stages of the Dutch elm disease cycle.

A. Spread of d-Factors during the Disease Cycle

The annual cycle of Dutch elm disease is summarized in Figure 11. So far as C. ulmi
is concerned, four ecologically distinct phases of the fungal cycle can also be identified
within it: s (1) the bark or long saprophytic phase; (2) the beetle phase; (3) the feeding
groove phase; and (4) the pathogenic or xylem phase (D). The spread of d-factors will
be discussed in relation to these phases.
 201

1. Bark Phase
The fungus spends some 8 to 10 months of the year as a saprophyte associated with
beetle breeding galleries in diseased elm bark (Figure IIA), the bark population usually
comprising a complex mosaic of different C. ulmi genotypes and hence of different
vegetative incompatibility types. 9 The fungus also appears to undergo periods of colo-
nization and recolonization of the bark in association with a distinct sequence of
fruiting involving conidial, synnematal, and perithecial formation '2 ,13 The bark phase
thus provides considerable opportunity for mycelial interactions of different types over
a long period, and is therefore probably of great importance for the spread and expres-
sion of d-factors.
Most primary bark colonization is probably initiated by spores introduced by breed-
ing beetles. On present evidence, many d-infected spores will probably fail to germinate
(Table I) and many surviving d-infected germ lings will be severely disadvantaged, and
hence be liable to be selected against, particularly during the first 24 to 48 hr of ger-
mination (cf. Figure 5b). Those able to initiate mycelia, although still presumably at a
growth disadvantage, will be liable to transmit d-factors via hyphal fusions to adjacent
germlings or mycelia of the same v-c group, and very occasionally to mycelia of differ-
ent v-c groups (Table 2). The frequency of v-c groups in a given bark population is
therefore of much significance.
Good evidence for such spread of d-factors during the bark phase has come from
the work of J. F. Webber and C. M. Brasier6,l3,27 who killed a tree with a genetically
marked (MBC tolerant) NAN aggressive isolate W2 toll by injection into the xylem and
allowed the bark to be colonized naturally by beetles carrying "wild-type" (or MBC
sensitive) C. ulmi (see also Reference 5). As it grew out of the xylem into the bark
throughout the following winter and spring, the progress of W2 toll was monitored by
sampling. It became established within the genetic mosaic in the bark, and in one
isolation grid, five of the ten samples of W2toll recovered from a 125 cm 2 area of bark
were found to be very severely d-infected and unstable. This could only have been due
to W2 toll acquiring a d-factor from an adjacent C. ulmimycelium. The five d-infected
samples (designated isolates "log 1/3 - 7 to 8, 9, 13, and I5"!) carried what was later
to be designated the d 2-factor, a d-factor which has subsequently been transferred to
many other isolates in laboratory studies (see earlier pages).
The secondary dispersal or recolonization of the bark by C. ulmivia asexual conidia
or synnematospores, probably aided by mites and other microfauna,'2,13 will almost
certainly lead to further spread of d-factors, subject to the constraints on transmission
already discussed above. Acting as a safety net for the fungus, however, in addition to
vegetative incompatibility, will be the ascospores. C. ulmi often produces a flush of
perithecia in the bark during midwinter. 12 The resulting ascospores will tend to initiate
mycelia not only of diverse v-c groups5,9 but apparently largely free from d-factors (see
Section I.G). In addition, perithecial formation by severely d-infected mycelia is also
likely to be reduced and delayed. Thus ascospore production is likely to limit the extent
of d-factor spread and enlarge the reservoir of "healthy" inoculum in the spore pop-
ulation.
Beetles of the new generation, emerging from the bark in the spring (Figure II B),
are thought to acquire their spore inoculum largely from fruiting structures of C. ulmi
in the pupal chambers. '3 The extent of d-infection of the inoculum will depend upon
the success of d-infected mycelia in colonizing the pupal chambers, and on the relative
proportions of asexual spores (conidia and synnematospores) versus ascospores pro-
duced therein. Present evidence suggests that asexual spores tend to predominate.

2. Beetle Phase
During the flight of newly emerging beetles to healthy trees for crotch feeding (Fig-
202 Fungal Virology

ure lIB, C), considerable spore mortality occurs as a result of environmental stresses
such as desiccation and exposure to UV light. Thus, while some 60 to 100070 of beetles
may leave the bark contaminated with C. ulmi, only 10 to 55% of beetles may arrive
at the feeding groove still contaminated. '3 In view of the much lower viability of d-
infected spores even in ideal environments (e.g., Table 1), the mortality of d-infected
spores during flight is likely to be extremely heavy compared with healthy conidia and
ascospores, and the flight period is hence likely to lead to intensive selection against
d-infection.

3. Feeding Groove Phase

Infection of healthy elms occurs as a result of spores carried on beetles being depos-
ited in the feeding grooves during crotch feeding (Figure llC). Before xylem infection
takes place, a short mycelial phase is thought to occur. 13 The impact of d-infection is
likely to be similar to that discussed for the primary colonization of bark. Slow grow-
ing d-infected germlings are likely to be at a competitive disadvantage compared with
healthy ones, because xylem infection is more likely to be achieved by germ lings with
a faster linear growth rate. Moreover, a high frequency of d-infection among a popu-
lation of spores on a beetle arriving in a feeding groove could, as a result of low spore
viability, conceivably prevent infection altogether. Thus Webber and Brasier 13 esti-
mated that on average a minimum inoculum of ca. 10,000 spores per beetle might often
be required for xylem infection to result. If, on a beetle carrying 10,000 spores, 40 to
60% of the spores were inviable as a result of d-infection (cL Table 1), xylem infection
might not take place. The chances of infection being caused by beetles with even
smaller spore loads of, say, 250 to 5000 spores (common spore loads on the largest
vector, Scolytus scolytus) 13 might be severely reduced by d-infection. These conclusions
are supported by a recent experiment on infection of artificial feeding grooves with
d'-infected and healthy isolates which showed that, with an inoculum of 50,000 spores,
the d'-factor reduced the chances of xylem infection by around 10- to 50-fold.'8
Feeding grooves also provide another opportunity for transmission of d-factors via
hyphal fusions, particularly where adjacent spores and resulting mycelia are of the
same v-c group. This is likely to be a common occurrence since the inoculum will
frequently comprise sticky spore masses derived from a single synnema or group of
synnemata in the pupal chamber.13 Where mixed genotypes occur, as is more likely to
be the case with ascospore inoculum, transmission is likely to be opposed by vegetative
incompatibility reactions between the germlings. Multiple use of feeding grooves by
beetles, recently demonstrated by Webber,'8 must enhance the chances of d-factor
spread.

4. Xylem or Pathogenic Phase

Once entry to the xylem has been gained, the fungus spreads within the tree mainly
in its the yeast-like state. 17 Wilting of affected branches results from toxin action and
from restriction of sap flow (Figure IlD).'8.19 In some cases, d-infection may lead to
restricted disease development, particularly where the resulting low growth fitness of
the pathogen interacts with other limiting factors such as high temperature, drought,
or a high level of host resistance (cL isolate DVI 3/8, Section I.H). However, it appears
that in most cases d-infection will quickly be lost as a result of the combined effects of
sporogenesis in the yeast phase and of selection favoring faster growing d-factor-free
components of the fungal population in the xylem. Thus, present evidence suggests
that in general, though with some important exceptions (isolate H363d'), the more
extensive the disease development in a tree the more likely it is that a d-factor has been
left behind.
Webber and Brasier have shown that there are in fact two cycles of the fungus in
 203

Table 3
FREQUENCY OF d-INFECTION AMONG ISOLATES OF
THE EAN AND NAN v-c "SUPER-GROUPS" FROM
POLAND, ROMANIA, AND BRITAIN

No. super-group "'0

Sample site Site type isolates tested" d-Infected

Poland 1980 EAN Epidemic front area 26 31

Romania EAN old epidemic area 14 7
Britain 1983 NAN old epidemic area 52 6

Tested for ability to give d-reactions against EAN recipient isolate H224 or NAN
recipient isolate W2 on EAS medium.

Dutch elm disease, a "bark-to-bark" or saprophytic cycle in which inoculum is trans-

ferred from breeding material to breeding material, often with an intervening flight
phase, and a "bark-to-xylem-to-bark" cycle involving a separate loop for the patho-
genic phase in the xylem, the latter eventually feeding back into the saprophytic gene-
pool through being released into the bark. 13 Present experimental evidence suggest that
C. ulmi feeding back from the pathogenic phase into the bark is likely to be substan-
tially free from d-infection and hence cytoplasmically fitter. Since considerable areas
of bark are sometimes colonized by fungus originating from the xylem,13,26 the cyto-
plasmic "clean-up" of C. ulmiin the pathogenic phase is likely to be a major factor in
limiting overall levels of d-infection in C. ulmi populations.
Thus d-infection is probably promoted by the saprophytic phase and largely opposed
by the pathogenic. It is also opposed by vegetative incompatibility, ascospore forma-
tion, and, to an extent by conidial formation. During the bark phase, beetle flight
phase, feeding groove phase, and in the xylem its effects may be minimized by selec-
tion. In view of all these apparent constraints, therefore, it is of interest to consider
the frequency of d-infection recorded among wild isolates.

B. Frequency of d-Infection in Nature

Little information is available so far on d-infection levels in bark populations of C.
ulmi apart from that already discussed. Fortunately this is not the case with xylem or
pathogenic phase populations, owing to the fact that extensive sample surveys of the
fungus have been carried out in recent years (see References 1 through 5).
An interesting feature of the European EAN and NAN aggressive populations re-
vealed by the surveys is that a large over-represented and widely distributed vegetative
compatibility group, called a "super-group" has been identified in each (a genetically
different v-c group in each case)'. The remaining portion of each population tends to
be highly polymorphic for vegetative-compatibility groups. Such "super-groups" pres-
ent an ideal opportunity to assess d-infection levels in the laboratory via d-reactions
because of the absence of vegetative incompatibility. In nature, their existence seems
likely to increase the frequency of d-factor transmission, and present data support this
view (see Table 2). Thus, among the small surviving sapling elms in the old epidemic
area of northeastern Romania, the EAN super-group represents about 19070 of the
EAN population and of these some 7% were found to be detectably d-infected, often
becoming severely debilitated in culture. At the current epidemic front area (among
mainly large mature elms) in Poland, the EAN super-group represents ca. 47% of the
population and of these a much higher frequency, ca. 31 %, are d-infected. Similar
data are available for the NAN super-group, though so far only from the old epidemic
areas in southern Britain (see Table 3).
These data show that natural d-infection levels can sometimes be surprisingly high
204 Fungal Virology

even in a sample of the C. ulmi population that has been subject to selection in the
xylem, although it should be pointed out that since only young shoots of I to 2 cm
diameter were collected during the above sampling, the opportunity for loss of d-infec-
tion may have been limited. Nevertheless it would seem reasonable to suppose that
d-infection levels recorded at the above sample sites would probably have been even
higher in the bark phase.

C. Potential of d-Factors in Disease Control

Disease control is of two types: that exerted on a pathogen by existing biotic and
abiotic factors in the pathogen's environment, which could be termed natural biologi-
calor natural environment control; and that in which either biotic or abiotic factors
are deliberately brought to bear on the pathogen by man in an attempt to suppress
disease, which could be called artificial control. The distinction between natural and
artificial biotic control is inevitably blurred because, where the biotic factor concerned
is already a part of the natural environment of the pathogen, at least initially artificial
control must be attempted in the context of natural control. Moreover it is important
first to understand the degree of natural control of a disease (if any) by a given biotic
factor in order to see how, when, and where its use as an artificial control could be
most successfully applied.
This is particularly true of the d-factor. As a disease of C. ulmi it must be exerting
some degree of natural biological control on the fungus, and it may offer some poten-
tial as a means of artificial control through being deliberately disseminated in the C.
ulmi population. The question is, how great is its natural impact on C. ulmi during the
present Dutch elm disease epidemics, and how, when, and where (if anywhere) could
this impact be usefully exploited.
To summarize from evidence already presented, since the d-factor appears to exert
its most deleterious effects on the vigor of mycelia and asexual spores, its greatest
degree of natural control is likely to be exerted in the bark, beetle, and feeding groove
phases of the fungus (Figure IIA through C) in which growth and sporulation fitness
are vitally important, rather than in the vascular wilt or pathogenic phase in the xylem.
Thus d-infection probably reduces the ability of the fungus to colonize elm bark, to
disperse and survive the long bark phase, and to produce inoculum for the next gen-
eration of emerging beetles. It must severely reduce the viability of spore inoculum on
flying beetles and on beetles in feeding grooves. It undoubtedly limits the ability of the
fungus to initiate xylem infection via the feeding groove. The carry-over of the fungus
to new beetle-breeding material is also likely to be reduced by d-infection of beetle-
borne inoculum. Within the xylem d-infection may sometimes delay the onset of dis-
ease but only occasionally restrict the spread of the pathogen.
In spite of such concise, rather confident statements of likely d-factor effects, there
remain too many unanswered questions and too little quantified information to allow
more than a speculative assessment of its control potential. For example, at what
threshold does the d-infection level in a C. ulmipopulation have to be in order to exert
a significant depressant effect on the overall incidence of Dutch elm disease in a given
elm population? Or indeed, is it actually possible for d-infection either alone or jointly
to force the annual incidence of disease into a downward spiral, given the various
factors that are likely to oppose it?
An important aspect underlying these questions, apart from the qualitative effective-
ness of d-factor themselves, is the size and genetic quality of the C. ulmi population
on which they are acting. This is particularly pertinent in the context of the present
epidemics of Dutch elm disease since, for a number of reasons, the C. ulmipopulation
is likely to be both qualitatively and quantitatively very different in the postepidemic
as opposed to the epidemic period. First, recent evidence suggests that at the onset of
 205

present outbreaks it is the nonaggressive strain of C. ulmi that initially shows a strong
population surge. This is followed by a surge in the aggressive strain, following which
the nonaggressive strain rapidly declines to be replaced and possibly even eliminated
by the aggressive (whether EAN or NAN race).3 .• As a result, once most mature elms
have been killed across North America, Europe, and southwest Asia, the next genera-
tions of seedling and sucker elms may be largely confined to being a scrub or marginal
population under constant attack by one or another form of the aggressive strain (see
References 3 to 5). It can also be argued that, during the epidemics, when the C. ulmi
population would be very large, the d-factor would probably exert little impact since
the fungus could afford the losses involved, whereas during the postepidemic period,
when the fungal population will be small, the impact of d-factors on the disease could
be much greater. 3-6 Indeed, the very future of the field elm as a major forest and
landscape feature in the epidemic outbreak areas may depend upon some form. of
attenuation of the aggressive strain in the postepidemic period. The d-factor could
conceivably exert a significant impact on the aggressive strain, not perhaps through
causing attenuation directly via the genes controlling its pathogenicity or virulence
which are in any case probably largely under nuclear gene control,20 but through re-
ducing its overall effectiveness throughout the disease cycle, including its infection
potential in the feeding groove. The latter will effectively break the cycle of disease.
Here the data for d-infection levels among samples of the EAN and NAN super-
groups in nature (Table 3) may be significant. All three samples are thought to be from
freely outcrossing C. ulmi populations, one (Poland) of epidemic and two of postepi-
demic status and, as previously mentioned, the d-infection level in the former appears
to be much higher. Following the lines of the above argument, and assuming that these
data are not merely sampling artifacts, then the high d-infection levels in the Polish
sample could reflect the fact that d-infection is less damaging to the fungus during the
epidemics than in the postepidemic period. In addition the small size of the EAN and
NAN v-c "super-groups" at the postepidemic sites in Romania and the U.K. (Table 3)
may reflect a need for increased reliance on ascospores for dispersal and survival, a
situation which could well have the effect of suppressing d-infection in the population
not only through the process of ascosporogenesis itself but through constantly gener-
ating new v-c groups via outcrossing.
During the postepidemic period, therefore, the potential for spread of d-factors may
be reduced as a result of changes in C. ulmi population structure, and, in particular,
increased reliance on sexual as opposed to asexual propagules. s The latter could result
from beetles having to fly greater distances in order to find suitable feeding or breeding
materia!.s Indeed, other features of the postepidemic period which could conceivably
lead to more intensive selection against d-infection are the small size and number of
the surviving seedling elms and the greatly reduced vector population. Thus it seems
likely that the threshold of d-infection level required for maximum effect of the
d-factor on the disease would be rather higher during the epidemics than in the post-
epidemic period. To put this another way, a 7!1{o d-infection level in the postepidemic
period might exert as much or more impact on the disease as would a 31!1{o level during
the epidemics.
Much needs to be learned about d-factor variability and the comparative effects of
different d-factors, but it is already clear that there are qualitative differences be-
tween factors (Figure 7). In this regard it is possible that d-factors had co-evolved
separately with the three subgroups of the fungus prior to the overlapping of their
ranges in the present epidemics, such that d-factors from one subgroup might tend to
be more deleterious when transmitted to another. 3 Thus d-factors of the nonaggressive
sort might be more deleterious in the cytoplasm of the aggressive sort and vice versa, a
possibility yet to be critically investigated, although the NAN d 2-factor has been found
206 Fungal Virology

to have a particularly severe effect on an EAN isolate (H277, Figure 6f). Such inter-
subgroup d-factor transmission could conceivably exert a significant impact on the
course of the current epidemics. The rapid decline of the nonaggressive strain in the
presence of the aggressive strain' 5 might be due in part to its having acquired delete-
rious d-factors from the latter. NAN or EAN aggressive isolates may well have ac-
quired d-factors from the nonaggressive isolates and from each other. While the v-c
system will tend to restrict such transmission, the occasional transmission of d-factors
across w-reactions (Table 2) would be enough to ensure some inter-subgroup transmis-
sion, and in any case it has already been shown that some v-c genotypes are common
to both EAN and NAN isolates, e.g., the identity of EAN isolate H277 with the NAN
supergroup, cited above.
An obvious and attractive way of using the d-factor in artificial control would be to
breed and release vector scolytid beetles carrying inocula with high levels of d-infection
in order to spread d-factors in feeding grooves and in the bark phase. 6 A major obsta-
cle to the success to such an approach, apart from the effectiveness of the d-factors
themselves, is again likely to be the vegetative incompatibility system of the fungus,
which would tend to restrict transmission between different genotypes. Where v-c
"super-groups" exist in the C. ulmi population they would provide ideal targets for
increasing d-infection levels. Since it is likely that the remainder of the C. ulmi popu-
lation would also have to be "targeted", it might well be necessary to spread d-infected
spores comprising a large number of v-c groups, as is the case with dissemination of
the h-factor in Endothia parasitica 21.22 (and see Chapter 4). However, in a regularly
outcrossing fungus with a comparatively short generation time such as C. ulmi, apart
from the v-c super-groups, it is not the identification of particular v-c groups in the
population that may be of overriding significance (as is sometimes stated to be the case
with Endothia), but the potential numberof different v-c groups that could arise in the
population by genetic recombination per disease cycle (cf. References 5, 9). This will
depend on the extent of polymorphism at the v-c loci in the local C. ulmi population
concerned.
The artificial spread of d-factors could be used for different purposes in different
contexts: for example, either during or after the present epidemics as part of short-
term local disease-control programs, or in the postepidemic period as part of a longer-
term effort to accelerate attenuation or debilitation in the aggressive strain population
as a whole. Locally, the d-factor might be a useful component of an integrated control
program in which other disease management practices such as sanitation, beetle pher-
omone lures, and disease resistant elms were incorporated. Such considerations are at
present entirely theoretical, but it is worth noting that d-factors could well be more
effective in combination with resistant elms, not only by restricting initial infection
levels via feeding grooves but also by acting in concert with internal host resistance to
restrict pathogen development in the xylem. For this same reason, d-factors might
cause more of a problem to the fungus on the more resistant native elms of Asia than
among the more susceptible elms of Europe and North America. Similarly, hotter drier
climates which would probably take a higher initial toll of beetle inoculum and tend to
restrict pathogen development in the xylem might enhance both natural and artificial
disease control by d-factors.
Finally, returning to the question of d-factor effectiveness, it may be that a particu-
larly deleterious d-factor suitable for artificial spread by man could either be selected
from the wider d-factor gene pool or developed in the laboratory by genetic manipu-
lation. This is one of several reasons why a substantial collection of isolates of the now
fast disappearing nonaggressive strain of C. ulmi should be conserved in a "gene
bank". Also pertinent here is the fact that the only isolate of C. ulmi obtained from
the Himalayas - one which cannot be readily assigned to any of the NAN, EAN, or
 207

nonaggresive sub-groups" - has recently been shown to give a d-reaction when paired
with its own "healthy" single spore derivatives 2S although transmission to other C.
ulmi isolates has yet to be attempted. The ability to manipulate rather than to select d-
factors will depend upon progress in understanding their molecular structure and intra-
cellular behavior, which are the subjects of the next Chapter.

ACKNOWLEDGMENTS

I am indebted to Susan Kirk and Susan Kent for excellent technical assistance; to
John Gibbs and Joan Webber for critical reading of the manuscript; to Joyce Ander-
son and George Gate for help with word processing and photography; and to the Eu-
ropean Economic Community for financial support of the sample surveys in Romania
and Poland.

REFERENCES

1. Brasier, C. M., Occurrence of three sub-groups within Ceratocystis ulmi, in Proc. Dutch Elm Disease
Symposium and Workshop Winnipeg, Manitoba, October 5-9,1981, Kondo, E. S., Hiratsuka, Y.,
and Denyer, W. B. G., Manitoba Department of Natural Resources, Manitoba, Canada, 1982,298.
2. Brasier, C. M., Dual origin of recent Dutch elm disease outbreaks in Europe, Nature (London), 281,
78, 1983.
3. Brasier, C. M., The future of Dutch elm disease in Europe, in Research on Dutch Elm Disease in
Europe, Burdekin, D. A., Ed., For. Comm. Bull., 60, Her Majesty's Staionery Office, London, 1983,
96.
4. Brasier, C. M., Recent genetic changes in the Ceratocystis ulmipopulation, in Populations of Plant
Pathogens: Their Dynamics and Genetics, Wolfe, M. S. and Caten, C. E., Eds., Blackwell, Oxford,
in press.
5. Brasier, C. M., The population biology of Dutch elm disease, in Advances in Plant Pathology, Vol.
5, Ingram, D. S. and Williams, P. H., Eds., Academic Press, London, in press.
6. Brasier, C. M., A cytoplasmically transmitted disease of Ceratocystis ulmi, Nature (London), 305,
220, 1983.
7. Brasier, C. M., Laboratory investigation of Ceratocystis ulmi, in Compendium of Elm Diseases,
Stipes, R. J. and Campana, R. J., Eds., American Phytopathological Society, St. Paul, Minnesota,
1982,76.
8. Brasier, C. M. and Gibbs, J. N., MBC tolerance in Ceratocystis ulmi, Ann. Appl. BioI., 80, 231,
1975.
9. Brasier, C. M., Inter-mycelial recognition systems in Ceratocystis ulmi: their physiological properties
and ecological importance, in The Ecology and Physiology of the Fungal Mycelium, Jennings, D. H.
and Rayner, A. D. M., Eds., Cambridge University Press, London, 1984,451.
10. Tavantizis, S. M. and Smith, S. H., Isolation and evaluation of a plant-virus-inhibiting quinone from
sporophores of Agaricus bisporus, Phytopathology, 72,619, 1982.
II. Webber, J. F., Implications of fungicide tolerance, in Rep. Forest. Res. Edin., 33, Her Majesty's
Stationery Office, London, 1983.
12. Lea, J. and Brasier, C. M., A fruiting succession in Ceratocystis ulmi and its role in Dutch elm
disease, Trans. Br. Mycoi. Soc., 80,381,1983.
13. Webber, J. F. and Brasier, C. M., The transmission of Dutch elm disease: a study of the processes
involved, in Invertebrate-Microbial Interactions, Anderson, J. M., Rayner, A. D. M. and Walton,
D., Eds., Cambridge University Press, London, 1984,271.
14. Esser, K. and Blaich, R., Heterogenic incompatibility in plants and animals, Adv. Genet., 17, 107,
1973.
15. Anagnostakis, S. L., Vegetative incompatibility in Endothia parasitica, Exp. Mycol., I, 306, 1977.
16. Caten, C. E. and Jinks, J. L., Heterokaryosis: its significance in wild homothallic Ascomycetes and
Fungi Imperfecti, Trans. Br. Mycoi. Soc., 49, 81,1977.
17. Banfield, N. M., Distribution by the sap stream of spores of three fungi that induce vascular wilt
disease of elm, f. Agric. Res., 62,637, 1941.
208 Fungal Virology

18. Elgersma, D. M., Host-parasite interactions in Dutch elm disease, in Research on Dutch elm Disease
in Europe, Burdekin, D. A., Ed., For. Comm. Bull., 60, Her Majesty's Stationery Office, London,
1983,78.
19. Scheffer, R. J., Toxins in Dutch elm disease, in Research on Dutch Elm Disease in Europe, Burdekin,
D. A., Ed., For. Comm. Bull., 60, Her Majesty's Stationery Office, London, 1983,82.
20. Brasier, C. M., Genetics of pathogenicity in Ceratocystis ulmi and its significance for elm breeding,
in Resistance to Diseases and Pests in Forest Trees, Heybroek, H. M., Stephan, B. R., and von
Weissenberg, K., Eds., Pudoc, Wageningen, Netherlands, 224, 1982.
21. Anagnostakis, S. L., The mycelial biology of Endothia parasitica. II. Vegetative incompatibility, in
The Ecology of Physiology of the Fungal Mycelium, Jennings, D. H. and Rayner, A. D. M., Eds.,
Cambridge University Press, London, 1984,499.
22. Kuhlman, E. G., Bhattacharyya, B. L., Nash, B. L., Double, M. L., and MacDonald, W. L., Iden-
tifying hypovirulent isolates of Cryphonectria parasitica with broad conversion capacity, Phytopath-
ology, 74,676, 1984.
23. Brasier, C. M., The origin of Dutch elm disease, in Report on Forest Research 1983,32, Her Majes-
ty's Stationery Office, London.
24. Brasier, C. M. and Kirk, S. A., Maternal inheritance of chloramphenicol tolerance in Ophiostoma
ulmi, Trans. Br. Mycol. Soc., in press.
25. Brasier, C. M., unpublished, 1985.
26. Mitchell, A. G., unpublished, 1985.
27. Webber, J. F. and Brasier, C. M., unpublished, 1981.
28. Webber, J. F., unpublished, 1985.
 209

Chapter 7

THE MOLECULAR NATURE OF THE D-FACTOR IN CERATOCYSTIS

ULMI

H. J. Rogers, K. W. Buck, and C. M. Brasier

TABLE OF CONTENTS

I. Introduction ................................................................................. 210

II. The d 2 Factor ............................................................................... 210

A. Parallel Acquisition of the d 2 -Factor and dsRNA Segments ............ 210
B. Loss of the d 2 -phenotype and dsRNA Segments in Ascospore
Progeny ............................................................................. 211
C. Transfer of the d 2 -Factor and dsRNA Segments by Hyphal
Anastomosis ....................................................................... 212
D. Loss of the d 2 -phenotype and dsRNA Segments after Tree
Inoculation ......................................................................... 212
E. Effect of Conidiogenesis on Transmission of the d 2 -Factor and dsRNA
Segments ............................................................................ 212
1. Type I Single Conidial Isolates ....................................... 212
2. Type II Single Conidial Isolates ...................................... 213
a. General ............................................................ 213
b. Implications of Latency of the d 2 -Factor. ................. 213
3. Type III Single Conidial Isolates ..................................... 215
F. Conclusions ........................................................................ 216

III. The dl-Factor ............................................................................... 216

IV. Comparison of d 1- and d 2 -Factors in C. ulmi with Hypovirulence

Determinants in Endothia parasitica .................................................. 217

V. Other Reports of DsRNA and Plasmids in Ceratocystis ulmi ................... 218

References ............................................................................................ 219

210 Fungal Virology

I. INTRODUCTION

D-factors are cytoplasmically transmitted genetic determinants that have the poten-
tial to cause degenerative disease in Ceratocystis ulmi. I Their biological characteristics
and implications for Dutch elm disease have been reviewed. 2 The molecular nature and
mode of action of d-factors are of considerable interest since, once identified, it may
be possible to manipulate them to provide effective biological control agents for Dutch
elm disease.
The failure to transmit d-factors into ascospore progeny of genetic crosses between
isolates when the female parent is d-infected argues that they are probably not associ-
ated with the mitochondria. 2 Genetic defects in fungi associated with mitochondria,
such as senescence in Podospora anserina, show classical maternal inheritance. 3 ,4 How-
ever, obligate anaerobes probably need to retain a proportion of normal mitochondria
to survive. This was shown for P. anserina when treatment of senescent cultures with
ethidium bromide led to their rejuvenation with loss of senescence-specific DNA and
recovery of normal mitochondrial DNA.5 Consistent with this observation is the report
that when female parental senescent isolates of P. anserina were crossed with male
parental healthy isolates, 10070 of the ascospore progeny were healthy. 3 Thus it is pos-
sible that isolates of C. ulmi carrying the d-factor could have a mixture of normal and
defective mitochondria, and that protoperithecia could be formed preferentially from
cells with normal mitochondria. Hence a mitochondrial location for d-factor cannot
be completely eliminated, although its complete exclusion in genetic crosses makes this
unlikely.
The other possibilities for a cytoplasmic factor are a plasmid and a virus. Several
DNA plasmids have been described in fungi. The best characterized of these reside
either in the nucleus or in the mitochondrion, but there is a group of plasmids whose
subcellular location is unassigned,. some of which may reside in the cytoplasm. A
number of different types of virus or virus-like particles have also been described in
fungi, of which the most common are the isometric particles with genomes of double-
stranded RNA (dsRNA).6 DsRNA, either free in the cytoplasm or associated with
membranous vesicles, has also been reported in Endothia parasitica in which it is as-
sociated with cytoplasmically transmitted hypovirulence. 7
Brasier 2 has commented that the phenotypes of C. ulmi isolates infected with dif-
ferent d-factors can differ significantly, suggesting that the molecular nature of differ-
ent d-factors could also be different. This is not surprising since it is known that pro-
tein toxins secreted by killer strains of some yeasts are encoded by dsRNA segments,
whereas in other yeasts (different) protein toxins are encoded by DNA plasmids. '4
Investigations on the molecular nature of d-factors have only recently commenced.
Here we summarize the results of some recent experiments· on the nature of the
d 2 -factor, which show a close, but not absolute, association between this factor and
specific segments of dsRNA. The possibility that the d I-factor could be associated with
dsRNA is also discussed.

II. THE d 2 -FACTOR

A. Parallel Acquisition of the d 2 -Factor and dsRNA Segments

The discovery of the d 2 -factor has been described by Brasier. 2 Briefly, a young elm
tree was killed by inoculation with an NAN aggressive isolate of C. ulmi, W2 toll,
which carried a single nuclear gene for tolerance to MBC fungicide. Subsequently the
bark became colonized by beetles carrying wild-type C. ulmi genotypes, and during the
following winter and spring MBC fungicide-tolerant isolates were recovered to monitor
the progress of W2 toll. Some of these were found to be infected with ad-factor,
 211

Table 1
DISTRIBUTION OF dsRNA SEGMENTS IN ISOLATES OF
C. ulmi

dsRNA segment

Number" A B C D E F G H J K L

+ + + + + +
2 + + + + + + + + +
3 + + + + + +
4 +
5 + + + + + + +
6 + + + + + + + + +
7 +
8 + + + + + + +
9 + + + +
10 +

Note: A, isolates Log1l3-8 and LogI/3-15; B, isolates W2, W2toll, Log1l3-63

and LogIl3-83; C, ascospore isolate derived from a cross between isolates
Log1/3-8 and H351; D, non-d-infected isolate obtained from a tree inocu-
lated with isolate LogI/3-8; E through L, single conidial isolates derived
from isolate LogIl3-8. (+) indicates the presence of dsRNA segment.

" Segment numbers refer to those in isolate LogIl3-8.

subsequently designated d 2 • Analysis of the dsRNA composition of two of these

d 2 -infected isolates, designated Log 1/3-8, and Log 1/3-15, revealed in both cases ten
segments of dsRNA with molecular weights ranging from 2.40 x 10 6 to 0.30 X 10 6 ,
which have been numbered 1 to 10 in order of decreasing size (Table 1). In contrast,
the original isolate, W2 toll, and two other MBC fungicide-tolerant, non-d-infected
isolates, Log 1/3-63, and Log 1/3-83 which had been reisolated from the tree in the
same experiment as Log 1/3-8, contained only two segments of dsRNA with the same
molecular weights as dsRNA segments 1 and 8. It is clear that when isolate W2 toll
acquired the d 2 -factor in the tree, presumably by transmission via hyphal anastomosis
from a wild C. ulmiisolate, eight new segments of dsRNA (2,3,4,5,6,7,9 and 10)
were also acquired. However, the W2 toll reisolates which had not acquired the
d 2 -factor, i.e., Log 1/3-63, and Log 1/3-83, had not acquired any new dsRNA seg-
ments either. On elm sapwood agar (ESA) medium isolate Log 1/3-8 grew significantly
more slowly than either of isolates W2toll or Log 1/3-63, suggesting a possible corre-
lation between slow growth, the d 2 -factor, and multiple segments of dsRNA.

B. Loss of the d 2 -phenotype and dsRNA Segments in Ascospore Progeny

Brasier 1.2 has reported that sexual reproduction in C. ulmi generally leads to loss of
the d-phenotype in the ascospore progeny. To determine if loss of the d-phenotype is
paralleled by loss of dsRNA, a genetic cross was carried out between isolate Log
1/3-8 (d 2 -infected) as the female parent with another non-d-infected, MBC-fungicide-
sensitive isolate of opposite mating type (H351) which contained no dsRNA. This was
done by fertilizing protoperithecia formed by isolate Log 1/3-8 with conidia from iso-
late H351. Of 20 single ascospore isolates analyzed, 5 were tolerant, and 15 were sen-
sitive, to MBC-fungicide. None of these were d-infected, i.e., none gave ad-reaction
when tested against isolate H351, and all had growth rates within the normal range.
Furthermore, 19 of the isolates had lost all 10 dsRNA segments and the remaining
isolate had lost 9 of the 10 dsRNA segments, retaining only segment 2. Hence loss of
the d 2 -factor correlated well with loss of multiple dsRNA segments and recovery of
212 Fungal Virology

normal growth rates. The complete absence of transmission of the d 2 -factor from the
maternal parent substantiates similar experiments reported by Brasier 2 which suggest
that the d-factors are not associated with mitochondria.

C. Transfer of the d 2 -Factor and dsRNA Segments by Hyphal Anastomosis

To determine if transfer of the d 2 -factor from one isolate to another by hyphal an-
astomosis was also accompanied by transfer of all ten dsRNA segments, d-reactions
were set up between isolate Log 113-8 and the following isolates, which were all in the
same v-c group as Log 113-8 and which were all sensitive to MBC-fungicide: 2 H351,
PI, P3, and P4. The latter three isolates were dsRNA-free, single ascospore progeny
derived from the genetic cross between isolates Log 113-8 and H351. Converted iso-
lates, i.e., isolates to which the d 2 -factor had been transmitted, were selected from the
recipient areas of the d-reactions by their sensitivity to MBC-fungicide and their ability
to give a strong d-reaction when tested against isolate H351. Converted isolates also
grew slowly on ESA medium. From all four transmissions, converted isolates were
found to contain all ten dsRNA segments, i.e., transmission of the d 2 -factor by hyphal
anastomosis is accompanied by faithful transmission of the multiple dsRNA segments.

D. Loss of the d 2 -Phenotype and dsRNA Segments after Tree Inoculation

Brasier2 has reported that when elms were inoculated with d-infected C. ulmi iso-
lates and then reisolations were made from the xylem, both at the inoculation point
and at other points to which the disease has spread, the d-phenotype was found to have
been lost in a large proportion of the reisolates. To test if such loss was also accom-
panied by loss of dsRNA segments, twelve young elms were inoculated with conidial
suspensions of isolate Log 113-8 and five other d 2-infected isolates (two trees per iso-
late); reisolations were made (one or two per tree) from the lowest twig in each tree
showing signs of infection. All of a total of eight reisolations had lost the d-phenotype,
as judged by their failure to give a d-reaction when tested against isolate H351. So far
only one of these has been examined for its dsRNA content; and this one, from the
tree inoculated with isolate Log 113-8, was found to have lost seven of the original ten
dsRNA segments, retaining only segments 2, 5, and 6. Thus, again, loss of the
d-phenotype correlates well with loss of dsRNA segments. However, it would be inter-
esting to examine the dsRNA content of a much greater range of the reisolates which
had apparently lost the d-phenotype, particularly in view of the possibility that, in
some of these isolates, the d-infection might have become latent (see Section II.E.2.).

E. Effect of Conidiogenesis on Transmission of the d 2 -Factor and dsRNA Segments

Single conidial isolates derived from isolate Log 113-8 were of three types: (I) slow-
growing isolates which retained all ten dsRNA segments, designated type I isolates; (2)
fast-growing isolates which retained all ten dsRNA segments, designated type II iso-
lates; (3) fast-growing isolates which had lost some of the dsRNA segments, designated
type III isolates. The fast-growing isolates tend to predominate over the slow-growing
ones, and within the fast-growing isolates type II is the most abundant (in one experi-
ment the ratio of type II to type III was about 3:1).

1. Type I Single Conidial Isolates

Twelve isolates of this type from one experiment were examined. All had retained
dsRNA segments with the same molecular weights as those of isolate Log 113-8, and
all had growth rates on ESA medium significantly slower than those of non-d-infected
isolates and mostly similar to that of isolate Log 1/3-8. When tested against a non-
d-infected isolate of the same v-c group, all gave strong d-reactions. Type I isolates
therefore represent conidia into which the d-factor has been faithfully transmitted and
 213

in which the d-phenotype is fully expressed. Here again d-factor transmission corre-
lates with the transmission of all the dsRNA segments.

2. Type II Single Conidial Isolates

a. General
Twenty-eight type II isolates from one experiment were examined. On ESA medium
most had growth rates similar to those of non-d-infected isolates; a few grew more
slowly but still significantly faster than isolate Logll3-B. Surprisingly, although all
these isolates were found to contain ten dsRNA segments with the same molecular
weights as those of isolate Log 1I3-B, none gave a d-reaction when first tested against
a non-dinfected isolate of the same v-c group. At first sight this result might be taken
to indicate that despite all the correlations described above, the d'-factor was not, after
all, associated with dsRNA. However, further investigations have shown that these
isolates carried the d'-factor in a latent form, i.e., a form in which its phenotypic
expression could not be detected. Most isolates of this type could be induced to give a
d-reaction under one or more of the following conditions: (a) growth of the isolates on
cellophane/malt extract agar medium or ESA medium prior to testing for ad-reaction;
(b) use of different non-d-infected recipients in d-reaction tests; (c) making isolations
from putative d-reaction zones and testing these isolates in further d-reactions. In gen-
eral, after d-reactions had occurred, the converted isolates, i.e., the recipients, were
slow growing and themselves able to give d-reactions. In a few cases merely storing an
isolate at 4°C caused it to revert from a type II to a type I isolate.
Out of 23 type II isolates tested, five still failed to give a d-reaction despite numerous
attempts. It was decided to investigate whether these more refractory isolates could
revert to type I via a second cycle of conidiogenesis. It was found that a proportion,
ranging from 10OJo to 53OJo, of the single conidial progeny from each of these five type
II isolates were slow growing and "amoeboid'" in morphology. Most of these gave
strong d-reactions. Hence it appears that all five of these more refractory type II iso-
lates can revert to type I on conidiation. Several faster-growing, second-cycle conidial
isolates were also capable of giving d-reactions, whereas others which did not give
d-reactions were found to have lost dsRNA segments. Although these isolates have not
been exhaustively investigated, it seems likely that conidiogenesis of type II isolates
produces a mixture of types I, II, and III conidia.
Five d'-infected slow-growing, second-cycle single conidial isolates from one of the
five "refractory" type II isolates were examined for their dsRNA content. All con-
tained the ten segments characteristic of isolate Log 1I3-B. It is concluded that all
conidial isolates which retain the ten dsRNA segments of isolate Log 1I3-B also retain
the d'-factor. However, whereas in some isolates (type I) the d'-factor is fully ex-
pressed, in others (type II) the d'-factor is latent and only expressed after induction by
a variety of methods.

b. Implications of Latency of the d 2 -Factor

If a population of single conidial isolates from a d'-infected isolate were examined
for growth characteristics and ability to give a d-reaction, it might be concluded that
the d'-factor had been transmitted to only a minority of conidia. In fact it appears that
transmission occurs to a majority of conidia. Hence estimates based only on one or a
few tests probably significantly underestimate the frequency of d'-factor transmission.
It appears that a second cycle of conidiogenesis is the only sure way of determining
whether or not an isolate is infected with the d'-factor. This would obviously be very
laborious to carry out with large number of isolates. Once the molecular nature of the
d'-factor has been identified unequivocally, a molecular marker would probably be the
best way of monitoring transmission.
214 Fungal Virology

From a biological point of view, a latent phase could help to ensure the survival of
ad-factor, since presumably a fast-growing, latently d-infected isolate would not be
selected against in the same way as a slow-growing, overtly d-infected isolate. Latency
could also explain the variation in growth rates observed following d-infection of var-
ious isolates. 2 However, it is not yet certain whether latency occurs with d-factors other
than d 2 •
Latency also has implications regarding the molecular nature of the d 2 -factor. Pos-
sible ways in which latency could develop include the following:

1. A DNA copy of the dZ-factor (or the dZ-factor itself if it is DNA) could become
reversibly integrated into nuclear DNA. In the integrated state, the functions
responsible for the d-phenotype could be repressed. Analogies for such integra-
tion are phage J... in bacteria,. RNA and DNA tumor viruses in animals,lo,l1 and
Agrobacterium Ti and Ri plasmids in plants. 1z This mechanism would appear
unlikely if the d-factor is dsRNA because, firstly, there is no evidence for eDNA
synthesis or integrated proviral DNA from any dsRNA mycovirus 6 and, second,
the ten dsRNA segments are present in normal amounts in the latently dZ-infected
cells. However, the possibility that a second cytoplasmic factor may be involved,
such as an episomal DNA which could interact with dsRNA, should be consid-
ered.
2. A host protein required for expression of the d 2 -factor may not be produced
during the latent phase. This could be due to an epigenetic event, such as repres-
sion of synthesis as a result of changes in the concentration of effector metabo-
lites, or a genetic event, such as DNA transposition, e.g., movement of a gene
from an expressing to a nonexpressing site, or transpositional inactivation of a
gene. The high frequency of conversion to, and reversion from, the latent state
would require a highly efficient transposition event, such as occurs in mating-
type switching in homothallic yeast strains. 13 These mechanisms would be com-
patible with a dsRNA model for d 2 factor.
3. A host protein could be produced in the latent phase which suppresses the
dZ-phenotype either by producing a product to neutralize the effect of the
dZ-factor or by preventing the expression of the dZ-factor, e.g., at the level of
transcription, translation, or protein processing. Suppression seems to be an un-
likely explanation for latency since it has been observed that an overtly dZ-in-
fected isolate will give a strong d-reaction when paired with a latently dZ-infected
isolate; i.e., latency is a recessive characteristic whereas it would be expected that
suppression would be dominant.
4. In the latent phase the level of the dZ-factor could be so reduced that its expres-
sion is insufficient to give an overt dZ-phenotype. Reversion to overt dZ-infection
would be the result of multiplication of the dZ-factor back to its former high
levels. This explanation of latency is attractive for two reasons: first, there is
evidence that the dZ-factor is completely eliminated in some conidia (see Section
II.E.3.); therefore, partial elimination, perhaps in a majority of conidia, appears
feasible; second, methods used to induce reversion to the dZ-phenotype, such as
slow growth at 4°C or growth on ESA medium, could lead to the replication of
the dZ-factor, "catching up"z on that of the cell and hence to greater cellular
concentrations of dZ-factor. However, it does not explain why reversion by a
second round of conidiogenesis is the most effective way of demonstrating la-
tency. If the level of dZ-factor were reduced by one cycle of conidiogenesis it
would be expected to be reduced even further in a second cycle. Nevertheless it
remains possible that during the process of conidiogenesis, about which little is
known at the molecular level, increased dZ-factor replication could occur during
 215

the early stages to give a level sufficient for overt expression; subsequent events
could then be similar to those occurring in the first cycle leading to formation of
types I, II, and III conidia. The "reduced d 2 factor level" hypothesis would ap-
pear to be inconsistent with a dsRNA d 2 -factor, because normal levels of all ten
dsRNA segments are found in the latent phase. Even if the dsRNA segments were
heterogeneous, as found in L dsRNAs in yeast,14 a reduction in intensity of one
or more dsRNA bands in gel electrophoresis would be expected to be consistently
observed in the latent phase, but it is not. Another possibility is that the level of
a second cytoplasmic factor, which could itself be the d 2 -factor or could be re-
quired to interact with dsRNA to produce the d-phenotype, is reduced in latency.
5. The final hypothesis to be considered is that latency results from mutation of the
d 2 -factor to an inactive form. This would require an unusually high rate of mu-
tation both for conversion to, and reversion from, the latent state. However, the
possibility that the mutation rate of the d 2 -factor could be influenced by a second
cytoplasmic factor should be considered. There is evidence that the high frequen-
cies of mutation observed for the flat and pig loci of Endothia parasitica are
controlled by cytoplasmic elements. 7

3. Type III Single Conidial Isolates

The type III single conidial isolates retained from two to seven dsRNA segments in
various combinations (Table 1). All these isolates had lost segments 4,7, and 10, and
out of twelve isolates examined, four had lost only these three segments. This suggests
that segments 4, 7, and 10 could be satellites dependent on one or more of the other
segments for their replication. If dsRNA replication also requires host proteins' 14 and
the supply of such proteins becomes limiting during conidiogenesis, satellite dsRNAs
would tend to be eliminated before their helpers, as observed for satellite dsRNAs in
killer strains of Saccharomyces cerevisiad 4 and Ustilago maydis. 15 It is noteworthy that
dsRNA segments 7 and 10 have molecular weights in the same range as the M and L
satellite dsRNAs of U. maydis. 15 Loss of the other dsRNA segments appeared to be
fairly random with little evidence for satellitism. Segment 1 was lost whenever segment
8 was lost (not vice versa), but the number of isolates in which this happened (three) is
too small to establish a correlation. It is interesting, however, that segments 1 and 8
were found together in isolate W2 toll before it acquired the d 2 factor (see Section
II.A.). The most stable dsRNA segments appeared to be 6 and 2 which were lost only
once and twice, respectively, out of twelve isolates, possibly reflecting less dependence
on host proteins than the other dsRNAs.
None of the type III single conidial isolates could be induced to give ad-reaction
despite numerous attempts using the methods outlined in Section II.E.2. In order to
determine whether the d 2 -factor could be latent in type III isolates, seven isolates were
selected for a second cycle of conidiogenesis to see if reversion to overt d-infection
occurred as it did with the type II isolates. The seven isolates included the four which
had lost only dsRNA segments 4, 7, and 10, and three others which had lost six, seven,
and eight segments of dsRNA, respectively (Table 1, columns J, K, and F). For each
of these isolates, 56 conidia were isolated (a total of 292) and, after selection for slow
growth rate, tested for ability to give a d-reaction. Only two overtly d-infected single
conidial isolates were identified (from one first-cycle conidial isolate which had lost
dsRNa segments 4, 7, and 10). The d 2 -factor in only one of these second-cycle isolates,
together with the remaining seven dsRNA segments, has been transmitted by hyphal
anastomosis to a recipient isolate which then also showed overt d-infection.
From these results it appears that, out of those tested, all of the first-cycle single
conidial isolates which have lost four or more dsRNA segments (including segments 4,
7, and 10) and more than 99070 of the first-cycle isolates which have lost only segments
216 Fungal Virology

4, 7, and 10 have also lost the d 2 -factor. This suggests that segments 4, 7, and 10 could
be involved in some way in the d 2 -phenotype. However, the one second-cycle conidial
isolate which apparently retains the d 2 -factor while having lost dsRNA segments 4, 7,
and 10 remains to be explained. Possible explanations are as follows:

(1) dsRNA segments 4,7, and 10 could be present in this isolate but in amounts too
small to be detected. This could easily be checked using gel transfer hybridization
techniques with radioactive probes capable of detecting one dsRNA molecule in
500 cells.'6 However, the same technique would have to be applied to all the
isolates to establish a difference between d 2 _ and non-d'-infected isolates.
(2) The d'-factor could be a cytoplasmic element distinct from dsRNA, but factors
resulting in loss of dsRNA segments could also result in loss of the d'-factor. This
would be consistent with the explanation put forward for latency, based on re-
duction of the level of the d 2 -factor in conidia (Section II.E.2.b., paragraph 4).
If levels of d'-factor are reduced when all dsRNA segments are retained (as in
type II isolates), d'-factor could be eliminated altogether when some dsRNA seg-
ments are lost. The second-cycle isolate which showed the d'-phenotype arose
from a first-cycle (type III) isolate in which the d'-factor was apparently latent;
the latter isolates could have retained low levels of d 2 -factor.
(3) The second-cycle isolate which showed the d'-phenotype could be a mitochon-
drial mutant which arose during conidiogenesis, assuming that any isolate with a
cytoplasmic mutation resulting in slow growth might give a reaction, similar to
d-reactions, when paired with other isolates. Evidence from pairings of donor
slower-growing chloramphenicol-tolerant (presumed mitochondrially mutant)
isolates with recipient chloramphenicol-sensitive isolates is against this, since the
mutant mitochondria were apparently not transmitted to the recipient.' However,
this could more readily be resolved by checking for maternal inheritance through
the sexual stage.

F. Conclusions
The d 2 -factor shows a close, but not absolute, correlation with dsRNA segments 4,
7, and 10. It remains possible that the d'-factor is a cytoplasmic element other than
dsRNA, e.g., a DNA plasmid, which was fortuitously associated with dsRNA in the
original isolate, and which is eliminated in a manner similar to dsRNA in conidia and
ascospores. Alternatively, the d 2 -factor could be a cytoplasmic element which requires
dsRNA for its stability or expression. Unequivocal identification of the d'-factor will
require its isolation (and characterization), followed by transformation of non-
d-infected isolates with the isolated d'-factor.

III. THE d'-FACTOR

The d'-factor, in isolate H321 of the NAN subgroup of the aggressive strain of C.
ulmi, was the first d-factor to be discovered. ' , This isolate contains seven segments of
dsRNA with molecular weights ranging from 1.20 x 106 to 0.39 X 106, numbered 1 to 7
in order of decreasing size. All segments, along with the d '-factor, are faithfully trans-
mitted by hyphal anastomosis to non-d-infected isolates. Fifty single conidia produced
by isolate H351 were investigated. Four single conidial isolates which had lost the abil-
ity to give a d-reaction contained no detectable dsRNA. Another six isolates which
gave strong d-reactions were found to contain three to five of the original dsRNA
segments. Loss of segments appeared to be random except that segments 2 and 5 were
retained in all of these four isolates. Most of the single conidial isolates gave weak,
irreproducible, or no d-reactions, suggesting that the d'-factor had either been lost or
 217

much reduced in quantity; some of these could have been latently d '-infected in an
analogous way to the latent d 2 -infections.
In summary, little can be said as yet about the molecular nature of the d '-factor
beyond the fact that d'-infected isolates contain dsRNA, and loss of dsRNA is corre-
lated with loss of the d'-factor. Whether the dsRNA segments in d'- and d 2 -infected
isolates are related is unknown; but the molecular weights of segments 2 and 5, which
were the only segments to be retained in all the d '-infected single conidial isolates, did
not correspond to those of any segment in d 2 -infected isolates.

IV. COMPARISON OF d 1 _ AND d 2 -FACTORS IN C. ULMIWITH

HYPOVIRULENCE DETERMINANTS IN ENDOTHIA PARASITICA

There appear to be both similarities and differences between d'- and d 2 -factors in C.
ulmi and hypovirulence determinants in Endothia parasitica. 7 Isolates infected with d ,-
and d 2 -factors contain dsRNA, and all isolates of E. parasitica carrying a cytoplasmi-
cally transmissible factor for hypovirulence also contain dsRNA. However, most hy-
povirulent isolates of E. parasitica have dsRNA segments in the molecular weight range
5 to 7 X 106 , whereas the largest dsRNA segments in d'- and d 2 -infected isolates are 1.2
x 106 and 2.4 x 106 , respectively. For both fungi, evidence for association between
dsRNA and hypovirulence or d-phenotypes is entirely correlative, and unequivocal
proof that dsRNA segments are the genetic determinants of these effects is lacking.
For both fungi, ascospore isolates are free from hypovirulence determinants or
d-factors, and are generally free from dsRNA. Hypovirulent strains of E. parastica are
apparently unable to form protoperithecia and hence cannot function as the female
parent in a genetic cross; d-infected strains of C. ulmi can form protoperithecia, al-
though with reduced efficiency. 2 For both fungi loss of hypovirulence or d-phenotype
on conidiogenesis is only partial. Less is known about latency in E. parasitica than it
is in C. ulmi. However, some dsRNA-containing isolates of E. parasitica are as virulent
and fast growing as dsRNA-free isolates and there is evidence that, in at least some of
these, hypovirulence may be latent. 7
In Italy hypovirulent isolates of E. parasitica have become predominant and a host-
parasite balance appears to have been achieved, leading to the reestablishment of the
chestnut (reviewed by Van Alfen 7 ). Although this has not yet occurred with C. ulmi,
neither has it happened with E. parasitica in America. The key to these differences may
lie in the relative roles of conidia and ascospores as agents of transmission, the role of
vectors in transmission, and the distribution of vegetative-compatibility groups in the
fungal population. Wind-borne as cos pores may be the most important method of
spread of E. parasitica in America, whereas in Italy spread of hypovirulent isolates of
this fungus probably involves a vector (as yet unidentified) carrying conidia or mycelial
fragments. With C. ulmi both conidia and ascospores are important at different stages
of the life cycle, and vector transmission is well established. 2 The apparent ease with
which the d-factors are lost (or become latent) when isolates spread in the xylem phase
may limit their efficacy in suppressing virulence or disease development. If d-infected
isolates of C. ulmi and hypovirulent isolates of E. parasitica are to be exploited for
the biological control of Dutch elm disease and chestnut blight generally, we must learn
more about the molecular nature and mode of action of the genetic determinants of d-
phenotype and hypovirulence and the ways in which they interact with both nuclear
and other cytoplasmic genes. The example of E. parasitica in Italy shows that a cyto-
plasmic factor can lead to effective biological control of a disease.
218 Fungal Virology

v. OTHER REPORTS OF dsRNA AND PLASMIDS IN CERATOCYSTIS

ULMI

Hollings and co-workers I7 ,18 could find no dsRNA in one isolate each of the aggres-
sive and nonaggressive strains of C. ulmi and concluded that dsRNA was not involved
in the lower level of pathogenity in the nonaggressive strain. Although this conclusion
is probably correct, the method of detection of dsRNA, using an antiserum to poly
rI:poly rC, was evidently not sensitive enough since one of the isolates examined, W2,
is the one in which we subsequently demonstrated two segments of dsRNA (see Section
II.A.).
In isolates of C. ulmi, dsRNA was first detected by Pusey and Wilson. 19 Subse-
quently in a study of C. ulmi isolates from different parts of the U.S. ,>0,21 these au-
thors found that "less aggressive" isolates from states with a long history of Dutch
elm disease usually contained multiple segments of dsRNA (molecular weights 2.0 x
10 6 to 0.4 X 10 6 ), whereas "more aggressive" isolates from states with a short history
of the disease were generally dsRNA-free or contained fewer dsRNA segments. The
results were interpreted as supporting the notion that dsRNA might attenuate patho-
genicity. However, comparing the locations of the isolates with the published distri-
bution of the aggressive and nonaggressive strains of C. ulmi in the U.S. 22 suggests
that the multiple dsRNA segments actually correlate with the distribution of the non-
aggressive strain, the lower pathogenicity of which can be ascribed to multiple genetic
differences at the nuclear, rather than the cytoplasmic, level. 23 The results of Pusey
and Wilson 20 21 could indicate that multiple dsRNA segments are more common in the
nonaggressive than the aggressive strain of C. ulmi, but far more isolates than the
relatively few (14) studied would be needed to validate this suggestion, as well as fur-
ther tests to confirm the subgroups into which the various isolates should be placed. 24
Pusey and Wilson 21 also noted that single conidial isolates from one "less aggres-
sive" isolate were of three types. One type contained all seven dsRNA segments present
in the parent strain and was less pathogenic than a second type with only one segment,
or a third type having no dsRNA. Parallels with the behavior of d l - and d 2-infected
isolates on conidiogenesis (Sections Il.E. and III.) are obvious, and it is possible that
the parent isolate, or indeed some of the other isolates with multiple dsRNA segments,
could have been d-infected. Unfortunately no transmission experiments were carried
out to determine if the "less aggressive" trait was due to a cytoplasmic factor.
Attempts to isolate virus particles from dsRNA-containing isolates of C. ulmi have
so far been unsuccessful. 21 This could be because the dsRNA is not encapsidated, or is
associated with membranous vesicles as in the case of E. parasitica, 7 or that virus
particles were present but were not purified by the methods used.
Recently "plasmids" of chain lengths 22, 18, 3 (two species), 2 and 0.54 kb were
reported in C. ulmi isolates. 25 The five smaller "plasmids" were found in two isolates
of the aggressive strain and one nonsporulating isolate, all of which were distinct in
cerato-ulmin (CU) production, but not in two nonaggressive isolates, negligible in CU
production. CU is a polypeptide toxin which may contribute to the pathogenicity of C.
ulmi. 26 Transmission experiments will be presumably required to determine if CU pro-
duction is associated with one or more of these "plasmids". The nature of the "plas-
mids", i.e., RNA or DNA, circular or linear, was not reported. The size of the two
larger molecules (22 and 18 kb) is within, or close to, the size range for fungal mito-
chondrial DNAs,4 whereas the four smaller ones are in the size range of the dsRNA
segments found in C. ulmi. Care should be taken in the proper identification and
characterization of plasmids. Recently a circular DNA from Fusarium oxysporum,
previously believed to be a plasmid,27 was shown to be mitochondrial DNA.28 Further-
more, Filetici et al. 28 have reported that M and L dsRNAs and the 2-/.Im DNA plasmid
 219

from Saccharomyces cerevisiae copurify when an alkaline lysate method, commonly

used for DNA plasmid extraction from bacteria, is employed. If authentic cytoplasmic
DNA plasmids can be demonstrated in C. ulmi clearly, they would be additional fac-
tors to be considered along with dsRNA in investigating the molecular nature of
d-factors.

REFERENCES

1. Brasier, C. M., A cytoplasmIcally transmitted disease of Ceratocystis ulmi, Nature, (London), 305,
220, 1983.
2. Brasier, C. M., The d-factor in Ceratocystis ulmi- Its biological characteristics and implications for
Dutch elm disease in Fungal Virology, Buck, K. W., Ed., CRC Press, Boca Raton, Fla., 1986,
chap. 6.
3. Marcou, D., Notion de longevitee et nature cytoplasmique du determinant de la senescence chez
quelques champignons, Ann. Sci. Nat. BioI. Bot. Veg., 12, 653, 1961.
4. Bockelmann, B., Osiewacz, H. D., Schmidt, F. R., and Schulte, E., Extrachromosal DNA in fungi
- organization and function in Fungal Virology, Buck, K. W., Ed., CRC Press, Boca Raton, Fla.,
1986, chap. 9.
5. Koll, F., Begel, 0., Keller, A., Vierny, C., and Belcour, L., Ethidium bromide rejuvenation of se-
nescent cultures of Podospora anserina: loss of senescence-specific DNA and recovery of normal
mitochondrial DNA, Curro Genet., 8, 127, 1984.
6. Buck, K. W., Ed., Fungal Virology - an overview in Fungal Virology, CRC Press, Boca Raton,
7. Van AIfen, N. K., Hypovirulence of Endothia (Cryphonectria) parasitica and Rhizoctonia solani in
Fungal Virology, Buck, K. W., Ed., CRC Press, Boca Raton, Fla., 1986, chap. 4.
8. Rogers, H. J., Buck, K. W., and Brasier, C. M., unpublished data, 1985.
9. Campbell, A., Bacteriophage A, in Mobile Genetic Elements, Shapiro, J. A., Ed., Academic Press,
New York, 1983,66.
10. Weiss, R. A., Teich, N., Varmus, H. E., and Coffin, J. M., Eds., Molecular Biology of Tumor
Viruses. RNA Tumor Viruses, Cold Spring Harbor Laboratory, Cold Spring Harbor, New York,
1982.
11. Tooze, J., Ed., Molecular Biology of Tumor Viruses. DNA Tumor Viruses, Cold Spring Harbor
Laboratory, Cold Spring Harbor, New York, 1980.
12. Zambryski, P., Goodman, H. M., Van Montagu, M., and Schell, J., Agrobacterium tumor induc-
tion, in Mobile Genetic Elements, Shapiro, J. A., Ed., Academic Press, New York, 1983,505.
13. Haber, J. E., Mating-type genes of Saccharomyces cerevisiae, in Mobile Genetic Elements, Shapiro,
J. A., Ed., Academic Press, New York, 1983,537.
14. Bruenn, J., The killer systems of Saccharomyces cerevisiae and other yeasts, in Fungal Virology,
CRC Press, Boca Raton, Fla., 1986, chap. 2.
15. Koltin, Y., The killer systems of Ustilago maydis, in Fungal Virology, CRC Press, Boca Raton, Fla.,
1986, chap. 3.
16. McFadden, J. J. P., Buck, K. W., and Rawlinson, C. J., Infrequent transmission of double-stranded
RNA virus particles but absence of DNA proviruses in single ascospore cultures of Gaeumannomyces
graminis, J. Gen. Virol., 64,927, 1983.
17. Hollings, M., Mycoviruses: viruses that infect fungi, Adv. Virus Res., 22, 3, 1978.
18. Hollings, M., Stone, O. M., Barton, R. J., and Atkey, P. T., Dutch elm disease, Rep. Glasshouse
Crops Res. lnst. 1973,124, 1974.
19. Pusey, P. L. and Wilson, C. L., Double-stranded RNA in Ceratocystis ulmi, Phytopathology, 69,
542, 1977.
20. Pusey, P. L. and Wilson, C. L., Possible role of fungal viruses in the distribution and spread of the
Dutch elm disease fungus, J. Arboric., 69, 1215, 1981.
21. Pusey, P. L. and Wilson, C. L., Detection of double-stranded RNA in Ceratocystis ulmi, Phytopath-
ology, 72,423, 1982.
22. Gibbs, J. N., Houston, D. R., and Smalley, E. B., Aggressive and non-aggressive strains of C. ulmi
in North America, Phytopathology, 69, 1215, 1979.
23. Brasier, C. M., Inheritance of pathogenicity and cultural characters in Ceratocystis ulmi: hybridiza-
tion of protoperithecial and non-aggressive strains, Trans. Br. Mycol. Soc., 68,45, 1977.
220 Fungal Virology

24. Gibbs, J. N. and Brasier, C. M., Correlation between cultural characters and pathogenicity In Cera-
tocystis ulmifrom Britain, Europe and America, Nature (London), 241,381,1973.
25. Takai, S., lizuka, T., and Richards, W. C., Discovery of plasm ids in CeratocystJS ulmi, Phytopath-
ology, 74, 833, 1984.
26. Stevenson, K. 1., Slater, J. A., and Takai, S., Cerato-ulmin - a wilting toxin of Dutch elm disease
fungus, Phytochemistry, 18, 235,1979.
27. Marriott, A. C., Archer, S. A., and Buck, K. W., Mitochondrial DNA in Fusarium oxysporum is a
46.5 kilobase pair circular molecule, J. Gen. Microbiol. 130, 3001,1984.
28. Filetici, P., Junakovic, N., and Ballario, P., Rapid alkaline preparation for yeast circular covalently
closed DNA molecules, Curro Genet., 9, 123, 1985.
 221

Chapter 8

VIRUSES OF THE WHEAT TAKE-ALL FUNGUS,

GAEUMANNOMYCES GRAMINIS V AR. TRITIeI

K. W. Buck

TABLE OF CONTENTS

I. Introduction ................................................................................. 222

A. The Gaeumannomyces/Phialophora Complex ............................ 222
B. Take-All Disease of Wheat and Barley ...................................... 222
C. The Discovery of Viruses in Gaeumannomyces and
Phialophora spp ................................................................... 223

II. Virus Groups and Virus Variants ...................................................... 223

A. Viruses of Ggt ..................................................................... 223
B. Viruses of Gggand Gga ................................................... ...... 227
C. Viruses of Phialophora sp. (1.h.) and P. graminicola .................... 227
D. Conclusions ........................................................................ 227

III. Virus Transmission and Epidemiology ............................................... 228

A. Transmission into Conidia ...................................................... 228
B. Transmission into Ascospores ................................................. 228
C. Transmission by Hyphal Anastomosis ....................................... 229
D. Epidemiology of Host and Virus .............................................. 230
E. Infection of Pro top lasts ...................................................... '" 231

IV. Virus Infection and Pathogenicity of Ggt............................................ 231

A. Self-Inhibition ..................................................................... 231
B. Pathogenicity ...................................................................... 231
C. Summary and Future Prospects ............................................... 233

References ............................................................................................ 234

222 Fungal Virology

I. INTRODUCTION

A. The Gaeumannomyces/Phialophora Complex

Gaeumannomyces graminis is a homothallic, ascomycete fungus which exists in
three varieties. (1) G. graminisvar. tritici (Ggt) is the cause of take-all disease of wheat
and barley. (2) G. graminis var. avenae (Gga) is the cause of take-all disease of oats
and take-all patch disease of turf grasses; it also infects wheat and barley. (3) G. gra-
minis var. graminis (Ggg) is the cause of crown (black, brown) sheath rot of rice; it is
usually not pathogenic for wheat or barley. All three varieties have a wide grass host
range.
Two related imperfect fungi are parasites of species of the Gramineae, but are only
moderately or weakly invasive. A Phialophora sp. with lobed hyphopodia (l.h.), for-
merly called Phialophora radicicola var. radicicola, which is weakly pathogenic, has
been isolated from roots of wheat, maize, barley, and rye and is probably the ana-
morph of Ggg; there is no record of its host range apart from agricultural crops, but a
similar lobed hyphopodiate fungus occurs on grasses in Australia. Phialophora gra-
minicola, formerly called Phialophora radicicola var. graminicola, is common in nat-
ural, amenity and agricultural grasslands, but is usually only found in cereal crops such
as wheat when these follow grass leys or are infested with grass weeds; it is essentially
nonpathogenic. P. graminicola is probably the anamorph of Gaeumannomyces cylin-
drosporus. The taxonomy of Gaeumannomyces and Phialophora spp. and their eco-
logical relationships have been reviewed comprehensively by Walkeri and Deacon,>
respectively.

B. Take-All Disease of Wheat and Barley

Take-all disease of wheat and barley is essentially a root rot disease. 2 6 The causative
fungus, Ggt, grows towards the roots of susceptible plants as a chemotactic response
to root exudates. Runner hyphae grow on the surface of the root and form a mycelium
from which infection hyphae are produced. The infection hyphae may penetrate the
epidermis of the plant at or below the soil surface, e.g., through root hairs, straw
bases, and lower leaf sheaths. The plant attempts to stop the infection by producing
lignitubers, sheaths of cell wall materials impregnated with lignin, around the infection
hyphae. However, the infection hyphae overcome this barrier by producing extracel-
lular enzymes which degrade the lignitubers from inside, and they often grow out
through the apices of the lignitubers into the next cell layer .
All six cell layers of the cortex of seminal root axes are penetrated, followed by entry
into the endodermis and invasion of the stele. The phloem is rapidly colonized and
destroyed near the point of invasion and the root below this point senesces. The xylem
is colonized more slowly than the phloem but soon becomes blocked by gum-like de-
posits. The disruption of transport processes, consequent on invasion of the stele, is
the main cause of damage to the plant. In the field, infected seedlings and young plants
are often killed, leaving bare patches (hence, "take-all"); infection of older plants
leads to stunting, reduced tillering, and dead, bleached inflorescences or "white-
heads".
After a crop has been harvested, the fungus may survive either as a saprophyte in
dead infected cereal residues or as a root parasite in weed grasses such as Agropyron
repens. Such sources may act as inoculum when the next crop is grown. Ascospores,
discharged from perithecia which are often formed around the straw bases of heavily
infected plants, represent another possible source of inoculum. The role of ascospores
in the biology of Ggt has remained uncertain. Although the seminal roots of seeds
germinating on the surface of moist soil are readily infected by ascospores, it has not
been possible to infect roots growing within natural soils, probably because of com-
 223

petition with soil and root surface microflora. Infection by wind-borne ascospores,
however, remains an attractive explanation for the sudden outbreaks of take-all in
fields when other sources of inoculum would not be expected to be present. Further-
more, the fact that the pathogen has retained the ability to produce ascospores suggests
that this property may be selected for in nature.
Ggt produces two types of phialospores, (1) germinating phialidic conidia, known
only in culture, and (2) smaller, nongerminating phialospores, occasionally produced
from germinating ascospores. Only the latter are produced in vivo and hence their role
as possible sources of inoculum for spread of take-all in nature is unknown. Some
isolates also produce moniliform cells, chlamydospores, and micro sclerotia in culture,
but it is not known whether these resting structures occur in nature.
When wheat or barley are grown in monoculture, take-all is rarely damaging in the
first year. However, after a severe outbreak of the disease, which may occur in the
second year or later, there is generally a depression of the disease, called take-all de-
cline, in subsequent crops. Take-all decline appears to be the result of suppression of
the disease by a heat-sensitive, transferable factor which develops in the soil. Many
hypotheses have been put forward with regard to the nature of the take-all decline
factor and the mechanism of suppression, but no single explanation appears to be
completely adequate. 7 The various hypotheses may be grouped into three categories:
(1) reduction of pathogenicity C'f isolates of Ggt (e.g., by viruses or genetic changes);
(2) nutrient changes in the rhizosphere of wheat; (3) inhibition of hyphal growth, either
before or during colonization of the root (e.g., by competing or antagonistic micro-
flora, hyperparasitism, reduction in the trophic response of hyphae, or after root col-
onization, that is, by antibiotics produced by heat-sensitive organisms or fluorescent
pseudo monads) .

C. The Discovery of Viruses in Gaeumannomyces and Phialophora spp.

Isometric virus particles were first discovered in a field isolate of Ggt in France. '.10
Since then, similar particles have been found in isolates of Ggt, Gga, Phialophora sp.
(l.h.), and P. graminicola from many other parts of the world (e.g., Africa, Australia,
Japan, North America, and the U.K. 11 16. The principal motivation for such investi-
gations was the possibility that virus infection might lead to a reduction in pathogen-
icity and that virus-infected hypovirulent isolates might be used to protect plants (e.g.,
by prior seed inoculation) from subsequent infection by virulent isolates." However,
virus-oriented studies have also led to discoveries with implications for other aspects
of the biology of Ggt and related fungi, for example, the discovery of a diffusible,
broad-spectrum antifungal inhibitor'· (Section IV.A), recognition of the diversity of
individuals within Ggt populations (Section II.A, III.C) and re-evaluation of asco-
spores as possible propagules for the spread of Ggt in nature (Section I1I.B,D).

II. VIRUS GROUPS AND VIRUS VARIANTS

A. Viruses of Ggt
Virus particles from Ggt, like those from many fungi, are isometric and have gen-
omes of dsRNA.,··,s.20.21 Considerable variation in the properties of viruses from dif-
ferent Ggt isolates has been observed, and some Ggt isolates have been found to be
infected with as many as four different viruses. Twenty-two viruses from thirteen iso-
lates of Ggt have been classified into four groups, based on the physical and serolog-
ical properties of the particles and the numbers and sizes of their dsRNA and capsid
polypeptide species."·"·22.23 Viruses within a group were found to be serologically re-
lated, whereas no relationships were detected between viruses from different groups.
For completeness, a fifth group is added here to accommodate viruses with dsRNA
224 Fungal Virology

Table 1
PROPERTIES OF GGTVIRUSES IN GROUPS I TO V·

Mol wt
Size of of capsid Buoyant
Particle SO 20 dsRNA seg- polypeptides density CsCI
Group diam(nm) (S units) ments (kbp) (x 10-') A 260 / Azso (g/ml)

34-36 109-128 1.5-1.9 54-60 1.4-1.5 1.36-1.37

II 34-36 33-148 2.0-2.3 68-73 1.4-1.5 1.35-1.37
III 39-41 159-163 4.7-6.3 84-87 1.6 1.40-1.41
IV 27-29 110-127 1.8 66 1.7 1.40
V 39-41 N.D. 6.5-10.5 94-125 N.D. N.D.

Modified from Buck. 22

Note: N.D., Not determined

segments in the range 6.5 to 10.5 kbp, although this may need to be subdivided when
more is known about individual members. The ranges of properties for the viruses in
the different groups are given in Table 1.
It is likely that there five groups will accommodate all, or most, of the Ggt viruses
so far described from different laboratories, although often the particles have not been
characterized sufficiently to make unequivocal assignments. The first Ggt virus to be
described9.10 from French isolate 911 had a diameter of 35 nm (revised value 24 ) and a
sedimentation coefficient of 116S, placing it in group I. Subsequently particles of 35
nm and 26 to 27 nm diameters were detected in a range of English 11 and French 25
isolates (including 911), and probably included groups I, II, and IV viruses. Rawlinson
et al." reported particles of diameter 35 nm, sedimenting at 148S, with a capsid poly-
peptide mol wt of 70,000, placing them in group II, whereas particles of diameter 27
nm sedimented at 11 OS and probably also had a capsid polypeptide of 70,000 daltons,
corresponding to group IV. Almond 15 reported a 27 nm virus with a capsid polypeptide
of 66,000 daltons, i.e., group IV. Frick and Lister" reported that virus particles from
American isolates of Ggt had modal diameters of 35 nm, 39 nm, and 41 to 42 nm
(possible groups I, II, III, and V); one specific virus preparation had polypeptide spe-
cies of ca. 70,000 daltons, suggesting that it contained a group II virus.
The distributions of size ranges of dsRNA segments from 152 isolates of Ggt, ana-
lyzed by Almond,15 J amil et al. 23 and Stanway, 26 are shown in Table 2. It is clear that
dsRNAs corresponding to virus groups I (or IV), II, and V are common, whereas group
III occurs relatively infrequently. In this analysis, virus groups I and IV, which cannot
be distinguished on the basis of dsRNA sizes alone, are represented as a composite
sum. Numbers of dsRNAs in different isolates varied from one to ten, and some iso-
lates contained dsRNAs of sizes corresponding to all the virus groups. The diversity of
patterns of dsRNA segments is illustrated by the occurrence of 132 different patterns
of segments within the 152 isolates. Only a minor proportion of these isolates con-
tained dsRNA segments with sizes outside of the ranges of the five virus groups. Five
isolates contained dsRNA segments of 3.3 to 3.8 kbp, i.e., between the group II and
III ranges, but attempts to purify virus particles containing dsRNA of this size have
not been successful. A few isolates contained dsRNA segments of sizes below the group
I range, i.e., in the range 1.4 to 1.0 kb, but in these isolates dsRNA segments in the
group I range were also present; it is likely, therefore, that these dsRNAs are satellites,
as shown for one Ggt isolate. 27
The biochemistry and genome organization of dsRNA mycoviruses, including Ggt
viruses,27.28 have been reviewed. 24 Groups I and II viruses have apparently similar gen-
 225

Table 2
DISTRIBUTION OF dsRNA
SEGMENTS IN 152 ISOLATES OF
GGT

DsRNA segments
with sizes in the
range of virus Number of OJo of total
group(s) isolates isolates

31 20.4
II 8 5.3
V 18 11.8
I + II 27 17.8
I+ V 25 16.4
I1+V 4 2.6
I+I1+11I 5 3.3
I+I1+V 24 15.8
II + I1I+ V 2 1.3
I + II + III + V 8 5.3
Total I 120 78.9
Total II 78 51.3
TotalllI 15 9.9
Total V 81 52.9
Range 3.3 to 3.8 kbp 5 3.3
Range 1.0 to 1.4 kbp 10 6.6

Data from Stanway."

orne organizations with two molecules of dsRNA, and both have been placed in the
Partitiviridae family of dsRNA mycoviruses. 29 .3o Group III viruses have a genome of
one molecule of dsRNA and have been placed in the Totiviridae family of dsRNA
mycoviruses. 29 .3o For all three groups, individual virus isolates may have more than the
minimum number of dsRNA segments. These additional dsRNA segments are proba-
bly satellite or defective dsRNAs!1.31.32 The only group IV virus to have been thor-
oughly characterized,22 and at least one member of group V, 15.16 have only one dsRNA
segment; although their properties are distinct from the group III viruses, which also
have one dsRNA segment, little is known as yet about their genome organization.
Eventually it should be possible to group all the Ggt viruses, along with other dsRNA
mycoviruses, in families, genera, and species.
The Ggt virus groups were established on the basis of differing physicochemical
properties and absence of serological relationships between viruses in different groups.
However, even with these groups there are differences between individual viruses, for
example, there are small differences in sizes of dsRNA and capsid polypeptide species
(Table 3), serological differences, and differences in the amount of sequence homol-
ogy. In group I, viruses 019/6-A, 38-4-A and 3bla-C (all from English isolates of Ggt)
were moderately closely related serologically to each other, but the other viruses in this
group (from English, French, and Japanese isolates) were only distantly related to
these viruses and to each other. The pattern of relationships suggests the presence of
several epitopes on the surface of the virus particles. This would also explain why pairs
of viruses, e.g., 019/6-A and F6-C, which were not serologically related to each other,
were both related (albeit not equally) to other viruses, e. g., 38-4-A, OgA-B, and 3bla-
C. Virus 01-1-4-A (from a Japanese isolate) appeared to be closely related to virus
019/6-A using a rabbit antiserum to the latter virus, but no relationship could be de-
tected with a mouse antiserum to the same virus. This suggests that the two viruses
have one closely related epitope which is more immunogenic than their other (unre-
226 Fungal Virology

Table 3
VARIATION OF PROPERTIES OF INDIVIDUAL VIRUSES IN
GROUPS I TO V

S,o,w Size of dsRNA segments Mol wt of capsid

Group Virus' (S units) (kbp) polypeptides (X 10-3)

019/6-A 126 1.85, 1.73 60

87-I-L N,D. 1.85,1.76,1.48 60
38-4-A 115 1.85, 1.73, 1.59 55
3bla-C 115 1.85, 1.73, 1.62 55
45/9-A 117 1.89, 1.78, 1.66, 1.62 55
OgA-B 125 1.89, 1.78 55
45/101-D 113 1.66, 1.53 55
01-1-4-A 109 1. 78, 1.66 55
F6-C 128 1.85, 1.73 54
II TI-A 133 2.17,2.14 73
F6-B 133 2.33, 2.27, 2.11 73
3bla-BI) 2.33,2.24 73
140
3bla-B2) 2.11,2.08 73
74-A N.D. 2.33,2.24 73
45/101-A 140 2.33,2.24 73
45/101-B N.D. 2.08,2.02 68
OgA-A 135 2.17,2.02 68
III 87-I-H N.D. 6.1 84
3bla-A 163 6.0,5.1 87
F6-A 159 6.3,4.7 87
IV 45/101-C 127 1.80 66
V F3-A N.D. 9.2,7.1 125
FIO-A N.D. 9.2 94

Note: N.D. Not determined.

Origins of Ggt isolates from which viruses were obtained. Australia: TI (Carnarvah,
1974, wheat); England: 019/6 (Rothamsted Highfield, 1972, barley after wheat); 38-4
(Rothamsted Hoosfield, 1973, continuous barley); 45/9 (Rothamsted Little Knott,
1972, twelfth wheat); 45/101, spontaneous variant of 45/10 (Rothamsted Little Knott,
1972, twelfth wheat); 3bla (Rothamsted Little Knott, 1970, third wheat); OgA (Selby.
1972, second wheat); 87-1 (Rothamsted Highfield. 1981. second wheat after 3 years'
lucerne); 74 (Rothamsted Highfield. 1981. wheat. after barley. barley. wheat); France:
F3. F6. FlO (Le Rheu. 1974, second wheat); Japan:019/6 (Gunma, Tokyo. 1950.
wheat).

Data from Buck et al..·· Almond." McFadden et al.,'· Buck." Jamil et al..23 and Jamil
and Buck."

lated) epitopes in the rabbit, but less immunogenic than the others in the mouse. Anal-
yses of sequence homology using solid phase hybridization 33 were in general agreement
with the serological results. RNA from virus 3bla-C hybridized strongly with RNA
from viruses 019/6-A and 38-4, weakly with RNA from virus 01-1-4, and not at all
with RNA from virus 45/101-D. Even for closely related viruses the degree of sequence
homology between the different RNA components may differ. For example oligonu-
cleotide fingerprinting and solution hybridization analysis showed that whereas the
RNAs 1 of viruses 019/6-A and 38-4-A were 95070 homologous, the RNAs 2 of these
viruses (which encode their coat proteins) were only 50% homologous. 27 Similar vari-
ations were found in the viruses of groups II and III.
Differences between viruses from isolates from the same country can be as great as
those from different countries. This was emphasized both by the studies of Frick and
Lister" who demonstrated serotype variability between viruses isolated from a single
 227

field at Vincennes, Indiana, U.S. and those of Jamil et al. 23 who showed that sequence
homology between dsRNAs extracted from isolates from a single field at Rothamsted,
England was comparatively rare. In the latter studies, sequence homology with dsRNA
from virus 87-1-L (group I, Table 3) was detected in only two out of twelve isolates
having dsRNA in the group I size range. Furthermore, no homology could be detected
between dsRNA from virus 74-A (group II, Table 3) and dsRNA in the group II range
from six other isolates.

B. Viruses of Ggg and Gga

Rawlinson and Muthyalu 34 reported a serological relationship between virus particles
from an Australian isolate (Gl) of Ggg and particles from an unspecified isolate of
Ggt. Subsequently the virus from isolate Gl was purified and found to have the follow-
ing properties;35 diameter, 35 nm; sedimentation coefficient, 143S; buoyant density in
CsCI, 1.38 g m- I ; mol wt of major capsid polypeptide, 65,000; size of dsRNA seg-
ments, major, 2.37, 2.24, minor, 1.88, 1.70, 1.69 kbp. The particle properties and the
sizes of the major capsid polypeptide and dsRNA species correspond to a group II Ggt
virus (Tables 1 and 3). The minor dsRNAs are either satellites or belong to an unde-
tected group I virus. Interestingly, satellite dsRNAs appear to be rare in Ggtviruses in
Group II, although they are relatively common in group I viruses (Table 3). Gl virus
did not react with antisera containing antibodies to the following Ggt viruses:
019/6-A, 38-4-A, 3bla-(A, B, and C), OgA-(A and B), and Tl-A. The fact that Raw-
linson and Muthyalu 34 found a relationship to particles from another Ggt isolate does
not conflict with these findings, but further emphasizes the diversity of serotypes in
Ggt viruses.
Very little information is available on Gga viruses beyond reports of the occurrence
of particles of diameters 35 and 27 nm and serological relationships between particles
of a Welsh Gga isolate and particles from English and American isolates of Ggt.12.13.34

C. Viruses of Phialophora sp. (l.h.) and P. graminicola

Analysis of six isolates of Phialophora sp. (l.h.) from one location revealed five
different patterns of dsRNA segments and particles of two diameters, suggesting the
presence of viruses corresponding to groups I, II, and V. 36 Particles, 26 to 27 nm in
diameter, have also been reported. 12,37 Two serologically unrelated viruses, both of 35
nm diameter, purified from isolate 2-2, had the following properties. 38 Virus 2-2-A,
sedimentation coefficient, 116S, capsid polypeptide mol wt 60,000 sizes of dsRNA
segments, 1.92, 1.78, 1.50 kbp; virus 2-2-B, sedimentation coefficient, 122S, capsid
polypeptide mol wt 66,000, sizes of dsRNA segments, 1.92, 1.82, 1.50 kbp. Both vi-
ruses are similar to the group I Ggtviruses, although the capsid polypeptide mol wt of
2-2-B is closer to that of the group II viruses. Neither virus reacted with antisera to the
following group I and group II viruses: 019/6-A, 38-4-A, 3bla-(B and C), OgA-(A and
B), F6-(B and C), Tl-A. However, recently39 virus 2-2-A was found to be indistinguish-
able serologically and in other properties from another group I Ggt virus (87-1-L).
Serological relationships between other unspecified Ggt and Phialophora sp. (l.h) vi-
ruses have also been reported. 12
A virus from P. graminicolahad diameter, dsRNA segments (Table 4), and sedimen-
tation coefficient (115S)34 similar to Ggt group IV viruses, although no serological
relationship to (unspecified) Ggt viruses was detected. 34 More recently Ferault37 found
particles of 35 and 26 nm diameter in French isolates of P. graminicola and showed
serological identity between the smaller particles and particles of similar size in isolate
911 of Ggt.

D. Conclusions
Particles of similar types occur in Ggt, Gga, Ggg, Phialophora sp. (l.h.), and P.
228 Fungal Virology

Table 4
VIRUS PARTICLES AND dsRNA IN ISOLATES OF
PHIALOPHORA SP (L.H.) AND P. GRAMINICOLA

Diam of virus Size of dsRNA segments

Fungal isolate" particles (nm) (kbp)

Phialophora sp. (I.h.)

2-2 35,40 8.9,1.91,1.86,1.79,1.51
12-2 35, 40 8.9,2.29
17-3 and 55-4 35 I. 73, 1.66
24-4 35 2.46,2.36
74/1007-2 35 1.97, 1.86, 1.79, 1.51
Phialophora graminicoJa
1348-2 30 1.92, 1.79, I. 70, 1.66

Origin of isolates. Rothamsted, England: continuous barley, Hoos·

field. 1973: 2·2,12·2,17·3,24·4,55·4;1974: 74/1007·2; first wheat
after ley, Summerdells, 1974: 1348·2.

graminicola and can be classified into (at least) five groups. There is considerable di-
versity between viruses in different groups and smaller, but significant, variation be-
tween viruses within groups. Fungi in the Gaeumannomyces/Phialophora complex ex-
ist as a diverse population of individuals which can be recognized by their differing
viruses and dsRNA profiles.

III. VIRUS TRANSMISSION AND EPIDEMIOLOGY

A. Transmission into Conidia

Ggt viruses are generally transmitted with high efficiency into germinable phialidic
conidia (Section I.B). Three viruses, F6-(A, B, and C), were transmitted to all of ten
single conidial isolates, together with their associated seven dsRNA segments, except
that in one isolate the largest of the three virus F6-B dsRNA segments was absent. ,.
Similarly, four viruses, 3bla-(A, B 1, B2, and C) and their associated nine dsRNA seg-
ments were faithfully transmitted into two single conidial isolates (3bla conD and 3bla
conE). ,. Virus exclusion can apparently occur during conidiogenesis, however. Stan-
way'· could not detect dsRNA in seven out of fifteen single conidial isolates derived
from five virus-infected Ggt isolates.

B. Transmission into Ascopores

Lapierre et a1." reported that an ascospore isolate, 911-1, derived from virus-infected
Ggtisolate 911 (Section II.A), was virus-free. However, after several years of storage,
viruses reappeared in isolate 911-1. 25 This suggests that low levels of particles, origi-
nally undetected, could have been transmitted to 911-1 and could have increased in
concentration over a period of time. Rawlinson et al. 11 did not detect virus particles in
56 single ascopsore cultures arising from four virus-infected Ggt isolates, but again
the method of detection (electron microscopy of subcellular fractions) may not have
been sensitive enough to detect low levels of particles. McFadden et al." could not
detect virus particles in hyphal tip cultures of Ggt conidial isolate F6SM8; however,
using a more sensitive method (immunosorbent electron microscopy, ISEM) it was
shown that virus particles were, in fact, present, although only at the level of one or
two particles per cell. After a further period of 18 months, involving several subcul-
tures, the level of virus had risen to about 450 particles/cell, similar to the level in the
original conidial isolate. 33 Such an increase, from levels only detectable by a sensitive
 229

method, could be analogous to the "re-appearance" of particles in the ascospore iso-

late 911-1.
More definitive experiments on ascospore transmission were carried out by Mc-
Fadden et al. 19 using gel transfer hybridization methods capable of detecting one
dsRNA molecule in 500 cells or one DNA provirus molecule in 60 cells, and a combi-
nation of large-scale virus extraction and ISEM, capable of detecting one virus particle
in 1000 cells. Of eight single ascospore isolates derived separately from conidial isolates
3bla conD and 3bla conE (Section III.A), six were completely free from dsRNA and
virus particles. Two, however, were found to be infected with virus B 1, one of the four
viruses in the original conidial isolates. The levels of virus (20 and 50 particles per cell,
respectively) could easily have been detected by standard electron microscopy. Inter-
estingly the two dsRNA segments of virus BI in the ascospore isolates were slightly
smaller than those of this virus in the original conidial isolates. No DNA provirus
molecules could be detected in two of the virus-free ascospore isolates or in one of the
virus BI-infected ascospore isolates.
A virus exclusion mechanism exists during sexual reproduction in Ggt, and this can
lead to complete elimination of virus in some ascospores. However, it is now clear that
transmission of virus particles into ascospores can occur. The frequency of transmis-
sion may possibly be isolate dependent. Although it was infrequent in 3bla conD and
3bla conE,'9 Tivoli et al. 49 reported that seven out of eight single ascospore isolates
derived from isolate 911-1 (after its virus content had increased on storage) contained
virus particles.

C. Transmission by Hypha! Anastomosis

Rawlinson et al. 11 reported transmission of virus particles between compatible iso-
lates of Ggt as a result of hyphal anastomoses, although they noted, as have others
(Davis,41 Chambers and Flentje,42 Nielson,43 Cunningham 44 ), that most pairwise com-
binations of isolates appeared to be incompatible, giving reactions similar to the "bar-
rages" formed between colonies of incompatible isolates of other fungi (Caten 45 , Esser
and Blaich46 ), i.e., clear zones containing few living hyphae and often bounded on each
side by narrow zones of intense black pigmentation. Jamil et al. 23 argued that if such
incompatibility barriers are effective in Ggt, close relationships might be limited to
viruses within a vegetative compatibility (v-c) group. To determine if this were so, they
first investigated incompatibility reactions between 31 isolates of Ggt from a single
field at Rothamsted and classified them into 18 v-c groups. This number of v-c groups,
from a fairly small number of isolates from one location, again emphasizes the diver-
sity of individuals in Ggt populations. Fourteen isolates, from nine of these v-c
groups, were then selected which contained easily detectable dsRNA segments, and
labeled cDNA probes were prepared from three of them, representative of the virus
groups I to IV, for use in gel-transfer hybridization experiments. In most isolates, no
dsRNA homology could be detected. However, close relationships were detected be-
tween dsRNAs in three isolates, each in a different v-c group. These results suggested that
v-c barriers may be operative in Ggt, but that transmission of viruses between some
isolates in different v-c groups could have occurred, particularly as all the isolates came
from the same locality.
There are, however, other explanations for the occurrence of closely related and
possibly identical dsRNAs in Ggt isolates of different v-c groups. For example, virus
infection could predate the evolution of the v-c groups. If infection persisted, subse-
quent evolution would be expected to result in virus diversification, as has been ob-
served. Nevertheless, if the number of v-c groups is large (which seems likely) and the
number of possible virus mutations is limited by structural and replication constraints,
closely related viruses could still persist in some isolates which have diverged into dif-
230 Fungal Virology

ferent v-c groups. The finding of related viruses in isolates from different countries,
such as America, England, or Japan, suggests a long history of association between
these viruses and their hosts (assuming that transmission of extracellular virus particles
is not possible47 ). On the hand, if ascospores are important in the epidemiology of Ggt
(see Section III.D), infections which persist over long periods of time seem unlikely.
A possible solution, which could accommodate these apparently contrasting obser-
vations, is suggested by the finding of apparently identical viruses in isolates of Ggt
and Phialophora sp. (l.h.) (Section II.C). It is possible that this Phialophora sp., which
does not generally produce ascospores, might act as a reservoir for subsequent infec-
tion of virus-free (ascospore?) isolates. It has been observed that Phialophora sp. (l.h.),
when grown on solid media alongside isolates of Ggt, often gives rise to "barrage"
reactions along the interface of the two colonies, similar to those formed between
incompatible Ggt isolates. Transmission of viruses between Phialophora sp. (l.h.) and
Ggtmay, therefore, also occur across vegetative incompatibility barriers. Verification
of this hypothesis will require transmission experiments with genetically marked iso-
lates.

D. Epidemiology of Host and Virus

Rawlinson et al." observed that Ggt isolates from first cereals after fallow or non-
suseptible break crops were usually virus-free or occasionally infected with particles of
27 nm diameter (group IV). However, more than half of 145 isolates from cereals,
after two to twelve consecutive crops, contained particles of diameter 27 and/or 35 nm
(group I or II). In a further study 34 the introduction of Ggt and the appearance of its
virus was studied in Barnfield, Rothamsted, which had carried no cereal crops from
1856 to 1967.34 From 1968 to 1974 the station carried spring barley and spring wheat
in alternate years, and in 1975 an area was sown to winter wheat. No virus was found
in take-all isolates for the period 1972 until May 1975, but in July, 17 out of 38 isolates
from widely separated plants contained particles of diameter 27 nm.
It was considered 11 that the predominance of virus-free isolates in first-year cereals
could indicate that ascospores were a more important source of Ggt inoculum than
had been hitherto appreciated (Section I.B). This would also be consistent with the
observed diversity of individuals in Ggt populations in a single field (Sections II.A,
III.C; see also Rayner et al. 48 on the consequences of the individualistic mycelium).
Appearance of virus could then result from transmission and mUltiplication of viruses
which were present in a proportion of the ascospores or transmission of viruses from
a Phialophora reservoir. Neither of these hypotheses would explain the prior appear-
ance of 27 nm diameter particles.
Another possibility is that colonization of fields previously free from Ggt could be
by wind-borne fragments of Ggt-infected debris from crops in nearby plots. Such in-
ocula could be virus-infected, but conditions in first crops could be suppressive to virus
replication. Stanway26 has noted that virus-infected Ggt isolates sometimes appear to
lose virus after being passaged on wheat seedlings grown in Levington's compost.
Changing soil conditions as a result of successive cereal crops could relieve the condi-
tions suppressive to virus replication. This could explain the relatively synchronous
appearance of virus in Ggt growing on widely separated plants in a field. Prior appear-
ance of particles of 27 nm diameter could be explained if such particles are the least
sensitive to suppression.
A third possibility is that Ggtcould acquire virus particles from its host plant. Cryp-
tic dsRNA viruses have been found in many plants, including members of the Grami-
neae, and resemble dsRNA mycoviruses of the Partitiviridae family, 26,64 which includes
Ggt groups I and II viruses (Section II.A). However, there is no evidence to support
this suggestion as yet. No serological relationships between several plant cryptic viruses
 231

and Ggt viruses have been found so far 6' and Ggt isolates did not gain particles after
infection and reisolation from wheat seedlings grown in sand."

E. Infection of Protoplasts
Recently infection of protoplasts of Ggt with purified viruses has been achieved. 50
The virus inoculum consisted of a mixture of viruses 3bla-A, 3bla-BI, 3bla-B2, and
3bla-C (Table 3). An ascospore isolate derived from 3bla conD was chosen as host for
virus infection because (1) it had been shown to be completely virus-free, and (2) it
should be susceptible to the 3bla viruses. Infection was achieved only in the presence
of polyethylene glycol and only with virus particles, not with isolatea dsRNA. Out of
30 clones regenerated from single protoplasts in two independent experiments, three
were found to be infected, two with virus BI and one with virus B2. The levels of virus
in the newly infected colonies were similar to those of the respective viruses in isolate
3bla and infections were completely stable through three subcultures, after which the
virus had undergone at least 10" cycles of replication. The absence of colonies infected
with viruses 3bla-A and 3bla-C could be due to lower infectivity of these viruses or
insufficient colonies having been examined.
Although fungal protoplasts do not generally occur in nature, the possibility that
anastomosing hyphae might be susceptible to infection with extracellular virus has been
considered.' 7 If this occurs in nature it would clearly be an important consideration in
the epidemiology of Ggt viruses.

IV. VIRUS INFECTION AND PATHOGENICITY OF GGT

A. Self-Inhibition
Many, possibly most, isolates of Ggtsecrete low molecular weight substances, termed
Q-factors, when incubated under appropriate conditions, which include low pH (pH
3.5 to 5.0).'8.51 Q-factors inhibit the growth, not only of a wide range of related and
unrelated fungi, but also of the Ggt isolates which secrete them. It has been
suggested 5'.S2 that Q-factor could have evolved as a mechanism to protect Ggt from
competing organisms in conditions unfavorable for growth, for example, when it sur-
vives saprophytically in host residues between susceptible crops. Self-inhibition under
these conditions would be unimportant. Q-factor secretion could result from, or be a
contributing factor to, suppressive conditions such as take-all decline. '6 This notion
would be consistent with recent observations that suppressive soil acts primarily by
affecting the emergence of hyphae from propagules 8 and that the tendency of Ggt
cultures to produce less disease (after a period in culture) correlated with an increased
potential to produce inhibitor suppressive to their own growth. 52
Q-factors were discovered as a result of investigations to determine whether virus-
infected Ggt isolates containing satellite dsRNAs could secrete polypeptide toxins
analogous to those secreted by Saccharomyces cerevisiae 53 and Ustilago maydis. 54
However, it has been shown unequivocally that Q-factors are not polypeptides'8 and
their synthesis is not affected by the presence or absence of virus particles or dsRNA. 5'

B. Pathogenicity
Pathogenicity in Ggt is difficult to quantify because (I) there are several different
methods of measurement which may not give equivalent results, 55 (2) the various meth-
ods may be influenced differently by environmental factors,55 and (3) isolates readily
lose their pathogenicity on storage. 56 .57 Nevertheless, Ggt isolates from nature do ap-
pear to vary in pathogenicity from strongly invasive to hypovirulent,58 even though
isolations are usually made from lesioned roots which may preferentially select the
232 Fungal Virology

more virulent individuals of a population. 59 Asher and co-workers·o have shown, by

analysis of progeny from genetic crosses between several virulent isolates and one hy-
povirulent isolate, that pathogenicity is inherited as a polygenic trait with no obvious
single major pathogenicity gene. Since there are no wheat varieties resistant to Ggt,
single gene differences analogous to the so-called "virulence" genes, associated with
race-specificity in other plant pathogens,·' would not be expected in the take-all fun-
gus.
The finding that hypovirulence can result from nuclear mutations in no way implies
that all cases of hypovirulence result from this cause. In Endothia parasitica hypovi-
rulence can result from either cytoplasmic or nuclear factors.· 2 Although Blanch·3
could find no evidence for extranuclear factors in determining pathogenicity, she em-
ployed homokaryotic hyphal tip or ascospore cultures which may have lost cytoplasmic
factors present in the original field isolates. In particular, it is well documented that
the concentrations of virus particles in hyphal tip cultures in many fungi, including
Ggt, are generally much reduced","," and they may be completely excluded in some
ascospores (Section III.B). Nevertheless evidence for cytoplasmic factors associated
with differences in growth rate of hyphal tip cultures of Ggt has been obtained. 55
There are several reports that the pathogenicities of some ascospore progeny differ
significantly from those of their parent field isolates.·0 .".57 In some cases these have
been correlated with reduction of virus particles to undetectable levels. However,
whereas Lemaire et al.' o found that an ascospore isolate 911-1 (which contained levels
of virus particles too low to be detected by standard methods) was much more patho-
genic than its weakly pathogenic, virus-infected parent isolate 911 (Sections II.A,
III.B), Rawlinson et al." found that some apparently virus-free ascospore isolates were
much less pathogenic than their strongly pathogenic, virus-infected parents. The 1: 1
and 1:3 segregations of pathogenicities in ascospores from single asci" suggest the
involvement of nuclear rather than cytoplasmic determinants and it is possible that the
original field isolates could have been heterokaryotic or intimate mixtures of individ-
uals. Contrary to a suggestion by Blanch et al.,·o this simple pattern of segregation
does not necessarily conflict with their report of polygenic control of pathogenicity.·o
Rather, the results suggest that whereas the strongly and weakly pathogenic isolates
crossed by Blanch et al.·o differed in many of their pathogenicity genes, the two com-
ponents in the field isolates of Rawlinson et al." probably differed in only one of their
pathogenicity genes. Indeed it would be expected that for stable association in a field
isolate, whether as an intimate mixture of individuals or as a heterokaryon, the two
component nuclei may have rather few differences.
In a study of 145 field isolates of Ggt, Rawlinson et al." found no consistent asso-
ciation between the presence or absence of virus particles (27 and 35 nm diameter) and
either pathogenicity or several other characteristics, such as unusual growth, morphol-
ogy, pigmentation, lysis, or readiness to form perithecia. Isolates with one kind of
particle were mostly more pathogenic and those with both kinds less pathogenic than
those in which virus was not detected but the differences were slight and probably not
significant. These results showed clearly that virus particles (35 and 27 nm diameter;
groups I, II, and IV) did not generally cause hypovirulence in Ggt. However, they did
not take into account viruses in groups III and V (40 nm diameter), which were later
shown to be present in some of the isolates studied, or the variation which was subse-
quently discovered in particles of particular diameters.
To investigate whether particles in a particular group might affect pathogenicity in
a proportion of virus-infected Ggt isolates Stanway2. made a large number of Ggt
isolation from wheat plants growing in several fields at Rothamsted and measured their
pathogenicities and the sizes of their dsRNA segments. Data for 100 of these isolates,
analyzed according to their pathogenicities and size classes of dsRNA segments, are
 233

Table 5
PATHOGENICITIES AND dsRNA SIZE
CATEGORIES OF FIELD ISOLATES OF
GGT

Pathogenicity

Low Medium High

Total number of isolates 38 37 25

Number of isolates with:
No dsRNA 2 8 8
DsRNA category'
A 4 6 2
B 0 0 2
C 6 6 3
A+C 10 2 6
B+C 3 7 2
A+B+C I3 8 2
Total with dsRNA 36 29 17

dsRNA categories: A, 10.5 to 4.7 kbp (virus groups III

and V); B, 2.3 to 2.0 kbp (virus group II); C, 1.9 to 1.5
kbp (virus groups I and IV) and below.

given in Table 5. The results indicated that no size class of dsRNA either alone or in
combination absolutely reduced pathogenicity. However, isolates with dsRNA, partic-
ularly those with dsRNA in all three categories, apparently have a greater chance of
being weakly pathogenic than those with no dsRNA. 571170 of the isolates containing all
three categories of dsRNA and 44% of those with dsRNA in one or more categories
were of low pathogenicity compared to 11 % of those with no dsRNA.
Most of the patterns of dsRNA segments in the 100 isolates were different, and it
remained possible that reduction in pathogenicity required specific segments or com-
binations of segments of dsRNA. In this context it is noteworthy that whereas a large
proportion of natural isolates of Ustilago maydis contain dsRNA with many different
patterns of segments, only a small proportion are killer strains. The ability of such
strains to secrete polypeptide toxins depends on the presence of an H segment, a spe-
cific M segment, and an L segment. 54 To investigate this possibility single conidial
isolates were obtained from one of the Ggt isolates (06S14) which contained nine
dsRNA segments with sizes ranging from 9.8 to 1.0 kbp (categories A, B, and C) which
was consistently weakly virulent in a range of different pathogenic tests. 26 Two conidial
isolates which retained all nine dsRNA segments were weakly pathogenic, but one con-
idial isolate, in which dsRNA could not be detected, was highly pathogenic. The pos-
sibility remains that the original field isolate 06S14 was a heterokaryon or an intimate
mixture of individuals and that the different conidial isolates represented the two pa-
rental types. This seems unlikely because this isolate was weakly pathogenic, even in
full season pathogenicity tests when it would be expected that the more invasive com-
ponent of a mixture would be selected. If 06S14 was heterogeneous the weakly patho-
genic component must have been extremely effective in cross-protection.

C. Summary and Future Prospects

Consideration of all the evidence indicates that viruses do not generally cause a re-
duction of pathogenicity in Ggt. However, it remains possible, even likely in view of
the results with isolate 06S14 (Section IV.B), that a minority of viruses or combinations
of viruses with specific dsRNA segments do result in hypovirulence. In this context it
234 Fungal Virology

is noteworthy that out of 152 isolates of Ggt lS ,23,26 06S14 is the only one to contain
dsRNA segments in the size range of the U. maydis M segments which encode polypep-
tide toxin. Another rare phenotype could be caused by a cytoplasmic factor. During a
study of vegetative incompatibility in Ggt, Jamil and Carlile 65 discovered an isolate
which, when paired with other isolates, caused a lytic effect that appeared to be trans-
missible. Unfortunately, on subculturing, the isolate became spontaneously cured of
this property. In searching for explanations of such phenomena we must consider not
only viruses, but also DNA plasmids, which have recently been reported in Ggt.66
In view of the likelihood that reduction of pathogenicity in Ggt by viruses or other
cytoplasmic agents occurs only infrequently, viruses can be excluded as general expla-
nations for take-all decline (Section I. B), cross-protection 17,67 or loss of virulence dur-
ing storage. 25 ,57 However, if reduction of pathogenicity, e.g., by 06S14 dsRNAs, can
be proved by transmission and infection experiments, they would have potential as
biological control agents. Now that protoplast infection has been achieved, 50 such
dsRNAs could be introduced into Ggt isolates in a range of v-c groups, and combi-
nations of such isolates could be used in field trials to assess their efficacy in protecting
plants from damage by take-all. The advantage over other possible biological control
agents 67 ,68 would be the possibility of the hypovirulence-inducing dsRNAs spreading
through a pathogen population.

REFERENCES

I. Walker, J., Taxonomy of take-all fungi and related genera and species, in Biology and Control of
Take-all, Asher, M. J. C. and Shipton, P. J., Eds., Academic Press, London, 1981, 15.
2. Deacon, 1. W., Ecological relationships with other fungi: competitors and hyperparasites, in Biology
and Control of Take-all, Asher, M. J. C. and Shipton, P. J., Eds., Academic Press, London, 1981,
75.
3. Skou, J. P., Morphology and cytology of the infection process, in Biology and Control of Take-all,
Asher, M. J. C. and Shipton, P. J., Eds., Academic Press, London, 1981, 175.
4. Manners, J. G. and Myers, A., Effects of host growth and physiology, in Biology and Control of
Take-all, Asher, M. J. C. and Shipton, P. J., Eds., Academic Press, London, 1981,237.
5. Hornby, D., Inoculum, in Biology and Control of Take-all, Asher, M. J. C. and Shipton, P. 1., Eds.,
Academic Press, London, 1981, 271.
6. Shipton, P. J., Saprophytic survival between susceptible crops, in Biology and Control of Take-all,
Asher, M. J. C. and Shipton, P. J., Eds., Academic Press, London, 1981,271.
7. Hornby, D., Take-all decline: a theorist's paradise, in Soil-Borne Plant Pathogens, Schippers, B. and
Gams, W., Eds., Academic Press, London, 1979, 133.
8. Wildermuth, G. B., Rovira, A. D., and Warcup, 1. H., Mechanism and site of suppression of Gaeu-
mannomyces graminisvar. triticiin soil, Trans Br. Mycol. Soc., 84,3,1985.
9. Lapierre, H., Lemaire, 1. M., Jouan, B., and Molin, G., Mise en evidence de particules virales
associees a une parte de pathogenicite chez Ie Pietin-echaudage des cereales, Ophiobolus graminis
Sacc., C. R. Hebd. Seanc. Acad. Sci. Paris, Ser. D, 271, 1833, 1970.
10. Lemaire, J. M., Lapierre, H., Jouan, B., and Bertrand, G., Decouverte de paticules vir ales chez
certaines souches d' Ophiobolus graminis, agent du Pietin-echaudage des cereales: consequences
agronomique previsibles, C. R. Hebd. Seanc. Acad. Agric. Fr., 56, 1134, 1970.
11. Rawlinson, C. J., Hornby, D., Pearson, Y., and Carpenter, J. M., Virus-like particles in the take-all
fungus, Gaeumannomyces graminis, Ann. Appl. BioI., 74, 197, 1973.
12. Rawlinson, C. 1. and Muthyalu, G., Take-all fungi and associated viruses from other countries.
Similar viruses in Gaeumannomycesspp. and Phialophoraspp., Rep. Rothamsted Exp. Stn. for 1974,
Part 1, 228, 1975.
13. Frick, L. J. and Lister, R. M., Serotype variability in virus-like particles from Gaeumannomyces
graminis, Virology, 85,504, 1978.
14. Buck, K. W., Almond, M. R., McFadden, J. J. P., Romanos, M. A., and Rawlinson, C. J., Prop-
erties of thirteen viruses and virus variants obtained from eight isolates of the wheat take-all fungus
Gaeumannomyces graminisvar. tritici, 1. Gen. Virol., 53,235, 1981.
 235

15. Almond, M. R., Biochemical Characterisation of Viruses Isolated from the Phytopathogenic Fungus,
Gaeumannomyces graminis var. tntici, Ph.D. thesis, University of London, 1978.
16. Rawlinson, C. J. and Buck, K. W., Viruses in Gaeumannomyces and Phialophora spp., in Biology
and Contral of Take-all, Asher, M. J. C. and Shipton, P. J., Eds., Academic Press, London, lSI,
1981.
17. Lemaire, J. M., Doussinault, G., Lucas, P., Perraton, B., and Messager, A., Possibilites de selection
pour l'aptltude a la premunition dans Ie cas du Pietin-echaudage des cereales (Gaeumannomyces
graminis), Cryptogam. Mycol., 3,347, 1982.
18. Romanos, M. A., Rawlinson, C. 1., Almond, M. R., and Buck, K. W., Production of fungal growth
inhibitors by isolates of Gaeumannomyces graminis var. tritid, Trans. Br. Mycol. Soc., 74, 79, 1980.
19. McFadden, J. J. P., Buck, K. W., and Rawlinson, C. J., Infrequent transmission of double-stranded
RNA virus particles but absence of DNA proviruses in single ascospore cultures of Gaeumannomyces
graminis, J. Gen. Viral., 64, 927, 1983.
20. Lapierre, H., Molin, G., Kusiak, C., and Faivre-Amiot, A., L'acid nUcleique des virus de cham pig-
non, Ann. Phytopathol., 5, 322, 1974.
21. Carpenter, J. M. and Rawlinson, C. J., Properties of nucleic acid from VLP in G. graminis, Rep.
Rothamsted Exp. Stn. for 1973, Part 1, 119, 1974.
22. Buck, K. W., A new double-stranded RNA virus from Gaeumannomyces graminis, J. Gen. Viral.,
65,987, 1984.
23. Jamil, N., Buck, K. W., and Carlile, M. J., Sequence relationships between virus double-stranded
RNA from isolates of Gaeumannomyces graminis in different vegetative compatibility groups, J.
Gen. Viral. 65,1741,1984.
24. Lapierre, H., Etude de l'influence des virus sur les champignons phytopathogenes du sol, in Perspec-
tives de Lutte Biologique des Champignons Parasites des Plantes Cultivees et les Pourritures des
Tissus Ligneux, Station Federale de Recherches Agronomiques de Lausanne, Switzerland, 62, 1973.
25. Ferault, A. C., Tivoli, B., Lemaire, J. M., and Spire, D., Etude de l'evolution comparee du niveau
d'agressivite et du contenu en particuies de type viral d'une souche de Gaeumannomyces graminis
(Sacc.) Arx. et Olivier (Ophiobolus graminis Sacc.), Ann. Phytopathol., 11, 185, 1979.
26. Stanway, C. A., Double-stranded RNA Viruses and Pathogenicity of the Wheat Take-all Fungus
Gaeumannomyces graminisvar. tritid, Ph.D. thesis, University of London, 1985.
27. Romanos, M. A., Buck, K. W., and Rawlinson, C. J., A satellite double-standed RNA in a virus
from Gaeumannomycesgraminis, J. Gen. Viral.,57, 375,1981.
28. Buck, K. W., Romanos, M. A., McFadden, 1. J. P., and Rawlinson, C. J., In vitro transcription of
double-stranded RNA by virion-associated RNA polymerases of viruses from Gaeumannomyces gra-
minis, J. Gen. Viral.,57, 157, 1981.
29. Buck, K. W., Fungal Virology - an overview in Fungal Viralogy, Buck, K. W., Ed., CRC Press,
Boca Raton, Fla., 1986, chap. I.
30. Brown, F., Classification and nomenclature of viruses: Fifth Report of the International Committee
on the Taxonomy of Viruses, Karger, S., Basel, in press, 1985.
31. Buck, K. W., Current problems in fungal virus taxonomy, in A Critical Appraisal of Viral Taxon-
omy, Matthews, R. E. F., Ed., CRC Press, Boca Raton, 1983,139.
32. Jamil, N. and Buck, K. W., Apparently identical viruses from Gaeumannomyces graminisvar. trWci
and Phialophorasp. (lobed hyphopodia), Trans. Br. Mycol. Soc., 83,519,1984.
33. McFadden, J. J. P., Detection and Characterization of Viruses in Conidial and Ascospore Isolates of
Gaeumannomyces graminis var. tritid, Ph.D. thesis, University of London, 1982.
34. Rawlinson, C. J. and Muthyalu, G., Virus-infected isolates of G. graminis var. tritici in Barnfield
soil. Relationships of viruses in Gaeumannomyces spp. and Phialophora spp., Rep. Rothamsted
Exp. Stn. for 1975, Pt. 1,256, 1976.
35. McGinty, R. M., Mycoviruses in Isolates of Gaeumannomyces and Phialophora spp., Ph.D. thesis,
University of London, 1981.
36. McGinty, R. M., Buck, K. W., and Rawlinson, C. J., Virus particles and double-stranded RNA in
isolates of Phialophora sp. with lobed hyphopodia, Phialophora graminicola and Gaeumannomyces
graminisvar. graminis, Phytopathol. Z., 102, 153, 1981.
37. Ferault, A. C., Les particules de type viral associees a Gaeumannomyces graminis (Sacc.) Arx et
Olivier: Sont-elles responsables de modifications du pouvoir pathogene du champignon?, D.Sc. The-
sis, University of Paris, 1983.
38. Buck, K. W., McGinty, R. M., and Rawlinson, C. J., Two serologically unrelated viruses isolated
from a Phialophorasp., J. Gen. Viral., 55,235,1981.
39. J amil, N. and Buck, K. W., Apparently identical viruses from Gaeumannomyces graminis var. tritici
and Phialophora sp. (lobed hyphopodia), Trans. Br. Mycol. Soc., 83,519, 1984.
40. Rawlinson, C. J., Muthyalu, G., and Deacon, Natural transmission of viruses in Gaeumannomyces
and Phialophora spp., Abstr. 2nd Int. Mycol. Congr., Tampa, Florida, August 27 to September 13,
1977,558.
236 Fungal Virology

41. Davis, R. J., Studies on Ophiobolus graminis Sacco and the take-all disease of wheat, J. Agri. Chem.,
31, 801, 1925.
42. Chambers, S. C. and Flentje, N. T., Studies on oat-attacking and wheat-attacking isolates of Ophiob-
olus graminisin Australia, Aust. J. BioI. Sci., 20, 927,1967.
43. Nilsson, H. E., Studies of root and foot rot diseases of cereals and grasses. I. On resistance to
Ophiobolus graminis Sacc., Ann. Agri. ColI. Sweden, 35,275, 1969.
44. Cunningham, P. C., Some consequences of cereal monoculture on Gaeummanomyces graminis
(Sacc.) Arx and Olivier and the take-all disease, EPPO Bull., 5,297, 1975.
45. Caten, C. E., Vegetative incompatibility and cytoplasmic infection in fungi, J. Gen. Mlcrobiol., 72,
221,1972.
46. Esser, K. and Blaich, R., Heterogenic incompatibility in plants and animals, Adv. Genet., 17,107,
1973.
47. Buck, K. W., Replication of double-stranded RNA mycoviruses, in Viruses and Plasmids in Fungi,
Lemke, P. A., Ed., Marcel Dekker, New York, 1979,93.
48. Rayner, A. D. M., Coates, D., Ainsworth, A. M., Adams, T. 1. W., Williams, E. N. D., and Todd,
N. K., The biological consequences of the individualistic mycelium in The Ecology and Physiology
of the Fungal Mycelium, Jennings, D. H. and Rayner, A. D. M., Eds., Br. Mycol. Soc. Symp. No.
8,509, 1984.
49. Tivoli, B., Ferault, A. C., Lemaire, 1.-M., and Spire, D., Agressivite et particules de type viral dans
huit isolats monoascopores de Gaeumannomyces graminis, Ann. Phytopathol., 11, 259, 1979.
50. Stanway, C. A. and Buck, K. W., Infection of protoplasts of the wheat take-all fungus, Gaeuman-
nomyces graminis var. tritici, with double-stranded RNA viruses, J. Gen. Virol., 65, 2061, 1984.
51. McGinty, R. M., McFadden, J. J. P., Rawlinson, C. J., and Buck, K. W., Widespread inhibitor
production in culture by isolates of Gaeumannomyces gramims var. tritici, Trans. Br. Mycol. Soc.,
82,429, 1984.
52. Naiki, T. and Cook, R. 1., Relationship between production of a self-inhibitor and inability of Gaeu-
mannomyces graminis var. tritici to cause take-all, Phytopathology, 73, 1657, 1983.
53. Bruenn, J. A., The killer systems of Saccharomyces cerevisiae and other yeasts, in Fungal Virology,
Buck, K. W., Ed., CRC Press, Boca Raton, Fla., 1986, chap. 2.
54. Koltin, Y., The killer systems of Ustilago maydis, in Fungal Virology, Buck, K. W., Ed., CRC Press,
Boca Raton, Fla., 1986, chap. 3.
55. Asher, M. J. C., Pathogenic variation, in Biology and Control of Take-all, Asher, M. J. C. and
Shipton, P. 1., Eds., Academic Press, London, 1981, 199.
56. Chambers, S. C., Pathogenic variation in Ophiobolus graminis, Aust. J. BioI. Sci., 23, 1099, 1970.
57. Naiki, T. and Cook, R. J., Factors in loss of pathogenicity in Gaeumannomyces graminis var. tritici,
Phytopathology, 73, 1652, 1983.
58. Asher, M. 1. C., Variation in pathogenicity and cultural characters in Gaeumannomyces graminis
var. tritici, Trans. Br. Mycol. Soc., 75, 213, 1980.
59. Asher, M. J. C., Isolation of Gaeumannomyces graminisvar. triticifrom roots, Trans. Br. Mycol.
Soc., 71, 322, 1978.
60. Blanch, P. A., Asher, M. J. C., and Burnett, J. H., Inheritance of pathogenicity and cultural char-
acters in Gaeumannomyces graminis var. tritici, Trans. Br. Mycol. Soc., 77,391, 1981.
61. Van der Plank, J. E., Priniciples of Plant Infection, Academic Press, New York, 1975.
62. Van Alfen, N. K., Hypovirulence of Endothia (Cryphonectria) parasitica and Rhizoctonia solani, in
Fungal Virology, Buck, K. W., Ed., CRC Press, Boca Raton, Fla., 1986, chap. 4.
63. Blanch, P. A., Pathogenic Variation in Gaeumannomyces graminis, D. Phil. thesis, University of
Oxford, Oxford, England, 1977.
64. Milne, R., personal communication, 1985.
65. Jamil, N. and Carlile, M. J., personal communication, 1983.
66. Honeyman, A. L. and Currier, T. C., The isolation and characterization of two linear DNA elements
from Gaeumannomyces graminis var. tritici, the causative agent of "take-all disease" of wheat,
Abstr. 83rd Annu. Meet. Am. Soc. Microbiol., New Orleans, American Society for Microbiology,
Washington, D. C., 1983, 135.
67. Wong, P. T. W., Biological control by cross-protection, in Biology and Control of Take-all, Asher,
M. J. C. and Shipton, P. 1., Eds., Academic Press, London, 1981,417.
68. Cook, R. J. and Reis, E., Cultural control of soil-borne pathogens of wheat in the Pacific North
West of the U.S., in Strategies for the Control of Cereal Disease, Jenkyn, J. F. and Plumb, R. T.,
Eds., Blackwell, Oxford, 1981, 167.
 237

Chapter 9

EXTRACHROMOSOMAL DNA IN FUNGI - ORGANIZATION AND

FUNCTION

Bernhard Bockelmann, Heinz Dieter Osiewacz, Frank Rainer Schmidt,

and Erika Schulte

TABLE OF CONTENTS

I. General Introduction ...................................................................... 238

II. Mitochondrial DNA ....................................................................... 238

A. Organization and Expression .................................................. 238
1. Organization of the Mitochondrial Genome ...................... 238
2. Expression of the Mitochondrial Genome ......................... 243
a. Transcription .................................................... 243
b. RNA Processing ................................................. 243
c. Translation ....................................................... 245
B. Transposition of Mitochondrial DNA ....................................... 247
1. Neurospora crassa ....................................................... 247
2. Saccharomyces cerevisiae .............................................. 247
3. Podospora anserina ..................................................... 248
C. Altered Mitochondrial DNA and Mitochondrial Plasmids ............. 249
1. Altered mtDNA and Plasm ids Derived from mtDNA .......... 249
a. rho- Mutants in yeast.. ........................................ 249
b. Poky Mutants of Neurospora crassa ....................... 252
c. Stopper Mutants of Neurospora crassa ...... .............. 252
d. Ragged Mutants of Aspergillus amstelodami ............ 255
e. Senescence in Podospora anserina ...... .................... 256
f. Senescence in Podospora curvicolla ........................ 262
g. Senescence in Cochliobolus heterostrophus? ............. 263
h. Senescence ("Kalilo Cytoplasms") in Neurospora
intermedia ........................................................ 263
2. Other Mitochondrial Plasmids ........................................ 263
a. Mitochondrial Plasmids in the Genus Neurospora ...... 263
b. Linear Plasmids of Claviceps purpurea..... ............... 265

III. Nonmitochondrial Plasmids ............................................................. 266

A. Nuclear Associated Plasmids ................................................... 266
1. 2 /Am DNA of Saccharomyces cerevisiae............................ 266
a. Biophysical Properties and Structure ...................... 266
b. Replication ....................................................... 266
c. Interconversion .................................................. 268
d. Function .......................................................... 268
2. The Ddp 1 Plasmid in Dictyostelium discoideum ................ 268
B. Plasmids of Unknown Association ........................................... 268

IV. Genetic Engineering with Fungal Extrachromosomal DNA ..................... 269

A. yeasts ................................................................................ 269
B. Filamentous Fungi. ............................................................... 270

References ............................................................................................ 272

238 Fungal Virology

I. GENERAL INTRODUCTION

The first evidence for the occurrence of genetic information outside the nucleus was
provided by the observations of Correns l ,2 and Baur 3 in 1909 that the inheritance of
specific markers in plants does not conform to Mendelian rules.
This extranuclear or extrachromosomal inheritance was first analyzed extensively in
Saccharomyces cerevisiae by Ephrussi and collaborators.' 6 Ephrussi studied a specific
yeast mutation which gives rise to slow-growing so-called "petite colonies". Crosses
between such mutants and normal growing wild cells led to normal growing diploid
colonies ("grande colonies"). The analysis of asci formed by these diploids revealed
that all four ascospores led to "grande colonies". These results are in contrast to the
expected 2:2 segregation of nuclear factors, and indicate that the "petite" phenotype
is extranuclearly inherited. It may be worth mentioning suppressive petites here, in
which, after crossing with wild cells, 4:0 petite:grande ratios occur in some cases. Fur-
ther analysis by Ephrussi and collaborators resulted in the postulation of a cytoplasmic
factor controlling the "petite" phenotype which is localized in mitochondria. 7
The genetic trait in mitochondria was first identified as DNA in 1963. 8 10 The petite
mutation and other extranuclearly inherited phenotypes in fungi have been demon-
strated to result from alterations of the standard mitochondrial (mt) DNA. These al-
terations lead in several cases to the formation of mitochondrial plasmids, which are
able to replicate autonomously. The presence of a plasmid in several systems is corre-
lated with a specific phenotype, as for instance in Podospora anserina or Aspergillus
amstelodami (Table 1).
Apart from these plasmids, which are homologous to parts of the mitochondrial
genome, in the last few years several other plasm ids have been found in fungi. These
are either associated with mitochondria and are without any homologies to the mt
"chromosome" or they are associated with the nucleus, e.g., the 2 /Am plasmid in
Saccharomyces cerevisiae. In yet another group of fungal plasmids, the location of the
plasmid is unknown at present.
This contribution will deal with all extranuclear DNAs in fungi, with the exception
of killer plasm ids which are dealt with in a separate part of this book. The first portion
of this review presents a brief summary of our current knowledge of the mitochondrial
genome, including plasm ids derived from the standard mtDNA. This is followed by a
discussion of the structure and function of those fungal plasmids with no homology to
the mitochondrial DNA. Finally, the practical implications of fungal extrachromo-
somal DNA in molecular cloning and biotechnology will be evaluated. For further
details on the organization and expression of the mitochondrial genome, the reader is
referred to recent reviews and monographs. lo ,33.'2

II. MITOCHONDRIAL DNA

A. Organization and Expression

1. Organization of the Mitochondrial Genome
Mitochondria are complex cell organelles found in all eukaryotes. They are involved
in respiration, oxidative phosphorylation, and fatty-acid biogenesis, and possess their
own genetic information (DNA) and protein synthesis machinery. 8,9,43
The mitochondrial DNA (mtDNA) of fungi is usually a covalently closed circular
(ccc) molecule. An exception is the linear mtDNA of the yeast Hansenula mrakii(Table
2). The length of mtDNA varies extremely from about 18.9 kbp in Torulopsis glabrata
to about 108 kbp in Brettanomyces custersii and, as may be seen from Table 2, even in
different strains of the same species the length of mtDNA may differ significantly.
Even though mtDNAs vary dramatically in size, only slight differences in coding
 239

Table 1
COMPILATION OF PLASMIDS IN FUNGI

Size of plDNA Molecular

Species Strain monomer (kbp) structure Ref.

MItochondria associated
Independent from mt
"chromosome"
Claviceps purpurea Wild strain KI 5.6 Linear II
6.8 Linear 11
Neurospora crassa Mauriceville Ic 3.6 ccc 12
N. intermedia Labelle 4.1 ccc 13
Fiji 5.2 ccc 13
N. tetrasperma Lihue ? ? 14
Hanalai 5.0 ccc 14
Waimea 14
Surinam 14
Homologue to parts of mt
"chromosome"
Aspergillus amstelodami ragged Varying ccc 15
Cochliobolus heterostrophus T40 1.9 ccc 16
Neurospora crassa stopper Varying ccc 17
poky Varying ccc 18
Podospora anserina senescence 2.5 ccc 19
senescence Varying ccc 20
Podospora curvicolla senescence Varying ccc 21
Nucleus associated
Dictyostelium discoideum 13-14 ccc 22
Saccharomyces cerevisiae Most strains 6.4 ccc 23
Unknown association
Ascobolus immersus 6.4 ccc 24
5.5 Linear 25
7.9 Linear 25
Cephalosporium acremonium 21.0 ccc 26
Fusarium oxysporum • 46.7 ccc 27
Kluyveromyces lactis 8.4 Linear 28
13.1 Linear 28
Morchella can cia 6.0 Lmear 29
8.0 Linear 29
Saccharomyces cerevisiae 9.4 ccc 30
Schizosaccharomyces pombe 6.4 ccc 31
9.6 ccc 32
Torulopsis glabrata 9.4 ccc 30
Rhizoctonia solani 2.6 linear 258

This "plasmid" is probably actually mtDNA."

capacItIes of different species are observed. In general, all mt genomes encode the
whole set of RNAs of the mitochondrial protein synthetic machinery, but only a lim-
ited number (5 to 10(70) of all mitochondrial proteins. The remaining proteins are gene
products of nuclear genes, synthesized on cytoplasmic ribosomes and transported into
mitochondria. 33
A compilation of components encoded by the mtDNA of different species is given
in Table 3. In this table the coding capacity of the mtDNA of three fungi with mt
genomes of different length are compared: Schizosaccharomyces pombe with a short
genome (19 kbp), Aspergillus nidulans with an intermediate one (33 kbp), and Saccha-
romyces cerevisiae with a rather large mt genome (78 kbp). The mtDNAs of these
species are all well characterized and extensive nucleotide sequences are avaiiable. 34 .• I .• 3
In addition, in S. cerevisiae the organization of the mtDNA has been extensively ana-
240 Fungal Virology

Table 2
COMPILATION OF SIZE AND STRUCTURE OF
FUNGAL MITOCHONDRIAL DNAs

Molwt
Species (kbp) Structure Ref.

Achlya ambisexualis 49.8 Circular 44

Allomyces macrogynus 49 ? 45
Aspergillus nidulans 32 Circular 46
Brettanomyces anomalus 56.5 Circular 47
B. custersii 108 ? 47
Candida parapsilosis ? Circular 48
C. tropicalis 52 Circular 49
Cephalosporium acremonium 26.7 Circular 26
Claviceps purpurea 45 Circular 11
Dekkera bruxellensis 75 47
D. intermedia 63.5 ? 47
Dictyostelium discoideum 35-40 Circular 37
Hanseniaspora vinea 26.7 ? 47
Hansenula mrakii 55 Linear 50
H. wingei 25.5 Circular 47
Fusarium oxysporum 46.5 Circular 51
Kloeckera africana 26.5 Circular 47
Kluyveromyces lactis 36.4 Circular 48
Neurospora crassa 60-73 Circular 52
Pachytrichospora transvaalensis 41.4 ? 47
Pichia lindneri 52 Circular 53
P. pinus 35.1; 76.9 Circular 53
Physarum polycephalum 41 Circular 54
Phytophthora infestans 36.2 Circular 55
Podospora anserina 95 Circular 56
P. curvicolla 55 Circular 57
Saccharomyces cerevisiae 68; 77.8 Circular 58
S. exiguus 23.7 Circular 47
S. teJ/uris 34.8 47
S. unisporum 27.4 Circular 47
S. unarum 61 Circular 49
Saccharomycopsis lipolytica 48.5 Circular 59
Saprolegnia sp. 42 Circular 60
Schizosaccharomyces pombe 19 Circular 61
Torulopsis glabrata 18.9; 20.3 Circular 47
Trichosporum cutaneum 34 Circular 49
Ustilago cynodontis 75 62

lyzed by classical genetic methods. Finally, the compact human mt genome (16.6 kbp)
is included in Table 3 because it is one of the most economically organized mtDNAs
for which the complete nucleotide sequence is known. 6' From Table 3 it may be seen
that the mtDNAs of all four species code for the three largest subunits of the cyto-
chrome c oxidase (COI,II,III),6s.66 one subunit of the cytochrome c reductase (Cyt b),
and the large and the small rRNAs. 34 In S. cerevisiae, A. nidulans, and most likely also
in S. pombe subunits 6, 8, and 9 (in man subunits 6 and 8) of the mitochondrial
ATPase are encoded by the mt genome. In S. cerevisiae one ribosomal-associated pro-
tein (var-l) is also encoded by the mtDNA.'4.6t.62.67 In addition, in S. cerevisiaeseveral
intervening sequences (introns) of mt genes were demonstrated by genetic and molec-
ular methods to code for polypeptides (RNA maturases) essential for correct RNA
splicing. 69 7S Whether such products are encoded also by other mt introns of other
species is uncertain even though introns of S. pombe, A. nidulans, Neurospora crassa,
and Podospora anserina contain long open reading frames, some of which code for
 241

Table 3
COMPILA TION OF MITOCHONDRIAL GENES OF DIFFERENT
ORGANISMS·

Mitochondrial gene product in

Mitochondrial component S. cerevisiae S. pombe A. nidulans H. sapiens

Cytochrome c oxidase
Subunit I + + + +
Subunit II + + + +
Subunit III + + + +
Cytochrome c reductase
Apocytochrome b + + + +
ATPase complex
Subunit 6 + + + +
Subunit 8 + +? + +
Subunit 9 + + +
Ribosomal components
Large rRNA + + + +
SmaJl rRNA + + + +
Ribosomal-associated + -? -?
protein
RNA-splicing enzymes
Intron 1 Cyt b maturase -? -?
Intron 2 Cyt b maturase + -?
lntron 4 Cyt b maturase +
lntron I COl maturase + -? -?
tRNAs 25 >22 28 22
Unidentified reading 7 IO 8
frames

From Borst et al.," modified. For references see text.

amino acid sequences significantly homologous to sequences of RNA maturases of S.

cerevisiae.61 •76 78 Finally, all mt genomes code for a different but complete set of
tRNAs and contain different numbers of unidentified reading frames (URFs).34.35.61
Even though there are minor differences in the number of mitochondrially encoded
gene products, these differences are not sufficient to explain the different length of the
mtDNAs of, for instance, Schizosaccharomyces pombeand Brettanomyces custersii.
Partial sequence data of the larger mt genomes indicate that there are several other
characteristics which are responsible for this extreme size variation.

1. In Saccharomyces cerevisiae it was demonstrated that the mt genes are scattered

throughout the genome with several large AT -rich noncoding regions (AT-
spacers) between individual genes. 34 .79
2. The mt genes usually contain extensive 5' -leader and 3' -trailer sequences, in con-
trast to the corresponding mt genes in mammals. These sequences are present in
the mRNA along with the coding sequence but are not translated. 34 The mRNA
of the apocytochrome b gene (Cyt b), for instance, has a 5' -leader of about 940
nucleotides and a 3' -trailer of about 100 nucleotides. 39
3. Different fungi contain a different set of introns. For this reason corresponding
genes in different species and even different strains of the same species sometimes
vary in size dramatically. The size of the gene coding for subunit I of cytochrome
c oxidase varies from about 1.7 kbp in Neurospora crassa to about 25 kbp in the
closely related Podospora anserina. Whereas the COl gene in one N. crassa strain
contains no introns (Figure 1, Table 4),80.81 the corresponding gene of Aspergillus
 242 Fungal Virology

Neurospora crassa
1kbp
~

•" " •"

Aspergillus nidulans
I
" E3• 13"
I II

,.,
I I I
E1 11 E2 12 "
E4

,
17 1819

• •..
Saccharomyces cerevislae (»Iong gene") )(
I
i
I
.. •
.. E5"
J J
.. I J
.. ,
I
Jl
-II
.;...;,wy
~
E1 "
11 E2 "
12 E3 13
~
E4 14 15 E6 16 E7 E8E9E10

FIGURE I. Comparison of the organization and size of the gene coding for subunit I of cytochrome c
oxidase (COl) of different fungi. The coding region for COl is indicated by dark areas. Exons (E) and introns
(I) of the discontinuous genes of A. nidulans and S. cerevisiae are numbered. '09

Table 4
COMPILA TION OF MITOCHONDRIAL
DISCONTINUOUS GENES IN FUNGI

No. of
Gene coding for Species introns Ref.

Cytochrome c oxidase, Aspergillus nidulans 63,84

subunit I (COl) Neurospora crassa 0,3,4 52,80,81
Neurospora intermedia 1,4 52
Saccharomyces cerevisiae 6,9 85,83
Schizosaccharomyces 2 86
pombe

Apocytochrome b Aspergillus nidulans 76,87

(Cyt b) Neurospora crassa 2 77
Saccharomyces cerevisiae 2,5 88,73
Schizosaccharomyces 2 61
pombe

ATPase, subunit 6 Neurospora crassa 2 89

rRNA, large subunit" Aspergillus nidulans 90

Neurospora crassa I 91
Podospora anserina 2 92
Saccharomyces cerevisiae 93

For the rRNA gene only, several examples for discontinuous genes are given.

nidulans 63,82 contains 3 introns, the "long" gene of Saccharomyces cerevisiae

(strain KL14-4A) 9 introns, and the COl gene of Podospora anserina a larger
but still not exactly determined number of introns. 84 This different organization
of mt genes can explain why some genes in several species are nearly as large as
or even larger than the whole mt genome of other species.

As may be seen from Table 4, at present only four genes have been demonstrated to
be discontinuous in fungi. These are the genes coding for subunit I of cytochrome c
oxidase (COl), for apocytochrome b (Cyt b), for subunit 6 of the ATPase complex,
and for the large subunit of the rRNA. Different strains of the same species may con-
tain or lack a specific intron ("optional" introns), so that in different strains one
specific gene may have a continous or a discontinous organization such as, for in-
stance, the COl gene in Neurospora crassa (Table 4).
 243

In addition to coding for a limited number of gene products, the mt genomes contain
at least one nucleotide sequence recognized by mitochondrial DNA polymerase as an
origin of replication ("ori" sequence). In contrast to mammalian mtDNAs which con-
tain only a single "ori" sequence on each strand,9' 96 the mtDNA of Saccharomyces
cerevisiae contains at least seven "ori" sequences. 97 This was concluded from a com-
parison of the nucleotide sequences retained in "petite" mutants in which, apart from
a short sequence stretch, large parts of the mtDNA are deleted. The retained mt se-
quence is amplified in this deletion mutant and is able to replicate. It must therefore
contain a functional origin of replication. A comparison of the nucleotide sequence of
the DNA segment retained in different "petite" mutants, which are derived from dif-
ferent regions of the mt genome, revealed that these sequences all contain a common
stretch of 265 bp with two short GC-rich clusters and an AT -rich palindrome of 23 bp.
A part of this sequence can be folded up into a secondary structure which resembles
the structure of mitochondrial origins of replication in mammalian cells, indicating
that the conserved sequence in S. cerevisiae may be part of a functional origin of
replication. 97 ,9"
Recently, putative "ori" sequences have been identified also in mtDNA from other
fungi, including Aspergillus amstelodami;' Podospora anserina,'8 and Candida
utilis!9 These sequences can all be folded up into a secondary structure which is very
similar to the one described above.

2. Expression of the Mitochondrial Genome

The expression of the genetic information proceeds in several steps which depend on
the gene product and on gene organization. The DNA sequence of a gene is first tran-
scribed to a RNA sequence, which is thereafter processed to the mature mRNA or to a
mature gene product (rRNA; tRNA). Finally, the mRNA must be translated into the
amino acid sequence of the corresponding polypeptide. In fungi these different steps
of mitochondrial gene expression are best analyzed in Saccharomyces cerevisiae. Most
of the information summarized in the next parts of this review will, therefore, be de-
rived from this organism.

a. Transcription
Mitochondrial RNA polymerases are encoded by the nuclear genome. Whereas this
enzyme of Neurospora crassa,100 Xenopus laevis,'o, and of rats'02 consists of a single
polypeptide, the active enzyme of Saccharomyces cerevisiae is probably a dimer con-
sisting of two subunits of a 45-kDa polypeptide. In contrast to the nuclear RNA poly-
merases, this mitochondrial enzyme is inhibited by manganese and rather resistant to
rifampicin.39
Using the vaccinia enzyme guanylyltransferase, which labels RNAs containing a di-
phosphate on the 5' terminus, it was demonstrated that mitochondrial transcripts are
initiated at several sites (at least 20) of the mitochondrial genome of S. cere-
visiae.'os '08 A comparison of the nucleotide sequences around these sites led to the
identification of a short characteristic consensus sequence just in front of transcription
initiation. This nonanucleotide (-AT AT AAGT A-) also precedes the rRNA genes of
Kluyveromyces lactis and may represent a part of a mitochondrial promoter sequence
in yeast. '06

b. RNA Processing
After transcription of a single gene, or DNA stretch containing several genes, the
mature transcript is formed through different processing steps. RNA splicing is best
analyzed in S. cerevisiae, which has been investigated in detail by several different
groups. Genetic and biochemical analyses led to the formulation of a model explaining
RNA splicing in yeast by the control of an intron-encoded enzyme (RNA maturase). 109
244 Fungal Virology

• I
"leader" E 1 11 E2 12 E3 13 E4 14 E5 15 E6 "trailer"

• r....... t • " • ~ •
Cytb
"long gene"

t,anscdpHon

pre mRNA

\: first splice
\ I (nuclear-encoded enzyme)
\\ iI
'J
0 circular RNA

1\
intermediate RNA

transla~ion I ! second splice

\ I (nuclear- and intron-
maturas~ I
\ I encoded enzyme)
\~
intermediate RNA 111f¥J?

""~
further splicin~

mRNA

gene product Cyt b

FIGURE 2. Expression of the Cyt b gene of Saccharomyces cerevisiae ("long" gene)

according to the maturase model of Lazowska et al. 69 Exons (E) are indicated by dark
areas. Those parts of introns (I) containing long open reading frames are hatched, those
with blocked reading frames are indicated by white areas. The nucleotide sequence of in-
tron 3 which has not been completely determined is marked by a question mark. Nontran-
slated sequences at the 5' ('leader') and 3' ("trailer") are indicated by dotted regions. For
details, see text (from Borst and Grivell, redrawn). 109

This model, first proposed for the apocytochrome b (Cyt b) gene in S. cerevisiae by
Slonimski and collaborators,·9 is diagrammed in Figure 2. As may be seen from this
figure, the Cyt b gene consists of six exons and five introns ("long" gene).73.88 Three
of these introns contain long open reading frames in phase with the preceding exons.
In the initial splicing step, which is controlled by nuclear factors, the first intron is
removed from the primary transcript, leading to a stable circular intron RNA and an
intermediate RNA. In this intermediate, the first two exons are precisely linked and a
long open reading frame is formed starting with the initiation codon of the Cyt b gene
at the 5' end of exon 1 and terminating at the first in phase termination codon in the
second intron. The translation of this reading frame results in the synthesis of a 423-
amino acid fusion protein, whose 143 amino-terminal residues are encoded by the first
two Cyt b exons. Together with nuclear-encoded factors, this protein, termed box 3
RNA maturase because of a cluster of mutations in the second Cyt b intron which
influence the function of this enzyme (box 3 locus), controls the correct removal of
intron 2. This intron partially contains the information for box 3 maturase. After the
 245

removal of intron 2 from the intermediate RNA, this information is destroyed. Thus,
the maturase controls its own synthesis, a phenomenon termed "splicing homeosta-
sis".69 In further splicing steps, which are also controlled in part by intron-encoded
RNA maturases, the mature mRNA is formed, which can be translated to apocyto-
chrome b.
Even though an enzyme with a splicing activity has not been isolated, various exper-
imental results prove its existence. At present it has been demonstrated by genetic com-
plementation tests and also by molecular analysis, that the introns 2, 3, and 4 of the
"long" Cyt b gene and the first intron of the COl gene in S. cerevisiae code for RNA
maturases.72-75 Most interestingly, intron 4 of the "long" Cyt b gene has a pleiotropic
effect. Point mutations in the open reading frame of this intron display the box phen-
otype (a simultaneous deficiency in apocytochrome b and cytochrome c oxidase).lIo 114
Whether mitochondrial introns of other fungi may code also for functional RNA
maturases is uncertain. Because most of these are obligate aerobic organisms, no non-
lethal mitochondrial mutants can be selected and thus no genetic analysis can be per-
formed. The only indication for RNA maturases in these organisms is the occurrence
of long open reading frames in specific introns of Aspergillus nidulans,'6 Neurospora
crassa,77 Podospora anserina,'8, 115 and Schizosaccharomyces pombe. 6I.86 Some of these
reading frames code for polypeptides with a clear homology with those of yeast RNA
maturases and may therefore code for such splicing enzymes. 76 ,78
In addition to the control of RNA splicing by intron-encoded RNA maturases and
nuclear factors, a correct RNA splicing depends on specific signal sequences localized
within the intron sequences (preferentially near the two splice points). 71 75,116 Mutations
in these signal sequences result in a splice deficiency. An analysis of such mutants
revealed that they effect the formation of a specific secondary structure of the intron
RNA, indicating that such a structure is a prerequisite for a correct RNA splicing. 117,118
This idea is confirmed by the fact that all mitochondrial and some nuclear introns can
be attributed to two distinct families. Members of the same family are characterized
by specific consensus sequences and can fold up into a characteristic secondary struc-
ture, which brings the two splice points into close proximity."9 121 In addition, most
interestingly, all so-called group II introns are spliced out of transcripts as stable cir-
cular RNAs. I22 ,I23
Another form of RNA processing has been demonstrated to occur in Neurospora
crassa.
Here the two rRNAs and most tRNAs are localized on a DNA fragment of about 20
kbp on only one strand. As demonstrated by hybridization experiments, one transcript
is formed from the whole region and the mature gene products arise from the process-
ing of this transcript. 124 Short 18-bp palindromes with two Pst I sites ("double Pst
palindromes"), which flank most tRNAs, are proposed to be signal sequences for proc-
essing enzymes. I25
A similar situation occurs in human mtDNAs, where a single main transcript is
formed from both strands of the mt genome. According to the tRNA-punctuation
model of Attardi and collaborators,126 these polycistronic messengers are processed by
processing enzymes capable of recognizing the tRNAs which flank nearly every gene
on the mtDNA. After cleavage of the primary messenger, the mature mRNAs, tRNAs,
and rRNAs are formed.

c. Translation
The mitochondrial protein synthetic machinery is partially encoded by the mitochon-
drial genome; however, most of the components of this apparatus are nuclear-encoded.
Examples of nuclearly encoded proteins are the initiation and termination factors, the
aminoacyl-tRNA-transferases, and nearly all ribosomal-associated proteins. Only the
246 Fungal Virology

Table 5
COMP ARISON OF THE MITOCHONDRIAL CODE OF
SACCHAROMYCES CEREVISIAE, HOMO SAPIENS, AND
NEUROSPORA CRASSA WITH THE 'UNIVERSAL CODE'

Mitochondrial code
Universal
Codon S. cerevisiae H. sapiens N. crassa code

UGA Tryptophan Tryptophan Tryptophan Termination

AGA Arginine Termination Arginine Arginine
AGG Arginine Termination Arginine Arginine
CUN Threonine Leucine Leucine Leucine
AUA Methionine Methionine/termination Isoleucine Isoleucine
AUU Isoleucine Methionine/initiation Isoleucine Isoleucine

For references see text.

large and small subunits of the rRNA, the whole set of mitochondrial tRNAs, and, in
S. cerevisiae and N. crassa, one ribosomal associated protein are encoded by the
mtDNA.
One surprising discovery was made when the first nucleotide sequences of mt genes
became available. A comparison of the amino acid sequences deduced from the nu-
cleotide sequences of the human and S. cerevisiae genes for subunit II of cytochrome
c oxidase (COIl) with the amino acid sequence of the bovine COIl protein revealed
several UGA termination codons in the nucleotide sequences of human and yeast genes
at positions where, in the amino acid sequence of bovine COIl, tryptophan is local-
ized. 128- 130
The isolation of tryptophan-binding tRNA of N. crassa and S. cerevisiae and the
determination of the corresponding nucleotide sequence showed that this tRNA pos-
sesses an anticodon with the sequence UCA. This anticodon is complementary to the
UGA termination codon of the "universal" code, indicating that, in contrast to the
cytoplasmic translation machinery, UGA represents in the mt system a sense codon
coding for tryptophan. 131.132
Similar analyses of the nucleotide sequence of other mt genes confirmed this conclu-
sion and revealed further that the mitochondrial code is not only different from the
"universal" one, but also differs from species to species (see Table 5).64.67.128.129.131 133
Apart from this unusual genetic code, the mitochondrial translation system is char-
acterized by another unusual feature: the number of tRNAs is much smaller than ex-
pected. According to the "wobble" hypothesis of Crick l34 a minimum of 32 different
tRNAs is necessary to decode the 61 different sense codons of the triplet code. As
shown by sequence analysis, the mtDNA of different species codes for a smaller num-
ber of tRNAs. On the other hand, no RNAs are imported from the cytoplasm. 34 To
solve this coding problem, it was proposed that mt tRNAs may in some way be able to
recognize the first two bases of a codon and ignore the third. This may be possible
because of the unusual features of mt tRNAs. Several tRNAs of N. crassa and S.
cerevisiae were demonstrated to contain a modified U at the first position of an anti-
codon ("wobble" position). This modification seems to inhibit the recognition of all
four synonymous codons, whereas these are recognized when the first base of an anti-
codon is an unmodified U .131.132 In contrast, in the mammalian system no modification
of nucleotides in anticodons was demonstrated. In this system, an unusual structure of
tRNAs is proposed to be responsible for an altered recognition of the codons of a
mitochondrial gene. 37 . US
 247

B. Transposition of Mitochondrial DNA

Transpositional events which occur within the same genetic compartment are well
characterized in both prokaryotic and eukaryotic systems (for reviews see Kleckner l36
and Shapiro I37 ). Recent findings demonstrate that in eukaryotes transposition of DNA
sequences can occur also between the different organelles which have their own genetic
information (reviewed by Barner I38 ). Independently Ellis I3 ' and Farrelly'40 called se-
quences which have been found in more than one of the genetic compartments of
eukaryotic cells "promiscuous DNA". In this section only examples of "promiscuous
DNA" detected in fungi are described. Up to now in fungi DNA sequences common
to mitochondria and the nucleus have been demonstrated in Neurospora crassa, Sac-
charomyces cerevisiae, and Podospora anserina.

}. Neurospora crassa
The smallest subunit of mitochondrial ATPase called the "DeeD-binding protein"
is part of the proton-translocating system in the inner mitochondrial membrane. In
Saccharomyces cerevisiae the DeCD-binding protein is encoded by mitochondrial
DNA and synthesized within mitochondria. 141 In contrast, in Neurospora crassa the
DC CD-binding protein is encoded by a nuclear gene, cytoplasmically synthesized, and
then transported to the inner mitochondrial membrane. 142
These findings are interesting because Saccharomyces and Neurospora are closely
related organisms which would be expected to have equivalent mitochondrial genomes.
An explanation for this apparent enigma was provided by hybridization studies which
demonstrated that Neurospora mtDNA contained a sequence which was homologous
to the yeast mt gene for the DCCD-binding protein. I43 Sequencing data 144 from this
region of Neurospora mtDNA revealed the existence of an open reading frame which
shows 65070 homology with the corresponding gene from yeast. I45 The amino acid se-
quence deduced from this open reading frame greatly resembles the known structures
of DCCD-binding proteins from mitochondria, chloroplasts, and bacteria. Although
transcripts homologous to the mitochondrial genes have been detected, no translation
product of the Neurospora mitochondrial gene coding for DCCD-binding protein
could be found. Thus, this gene seems to be "silent", i.e., it is not translated into a
polypeptide. 144
The endosymbiont hypothesis may explain the existence of genes for the DeeD-
binding protein in both the Neurospora nucleus and mitochondria (for reviews see
Gray and Doolittle,'8 and GrayI47). According to this hypothesis, mitochondria (and
chloroplasts) are derived from bacteria-like organisms which were taken up and main-
tained by ancient "eukaryotic" cells. During subsequent evolution nuclear genes en-
coding organelle-specific proteins could have been transferred from the organelles to
the nucleus.
For Neurospora, it can be suggested that the original gene for the DCCD-binding
protein located on mitochondrial DNA was duplicated. While one copy remained on
mitochondrial DNA, the other was transposed to the nucleus. The nuclear copy be-
came the active one, whereas the mitochondrial gene probably lost its function. 144

2. Saccharomyces cerevisiae
In baker's yeast, fragments of mitochondrial genes are integrated into the nuclear
genome. I40 The first evidence for this came from studies of the var-} gene which en-
codes an intramitochondrially translated protein of the mt small ribosomal subunit.
Althol,lgh sequencing data confirmed that the var-} gene is of mitochondrial origin,
hybridization experiments gave the first hints that there are nuclear copies of the var-}
gene in Saccharomyces cerevisiae. A cloned piece of the var-} gene was found to hy-
bridize with defined restriction fragments of nuclear DNA. In addition, the nuclear
248 Fungal Virology

DNA homologous to the mitochondrial var-l gene showed Mendelian segregation

when analyzed in crossing experiments. These data proved the integration of "mito-
chondrial" sequences into a nuclear chromosome.
Another interesting feature was revealed by DNA sequencing. Sequences homolo-
gous to the mitochondrial cytochrome b gene and to a mitochondrial sequence confer-
ring autonomous replication (orilrep) were found 5' to the nuclear var-l DNA. How-
ever, the rearrangements found in the nuclear sequences and the lack of any translation
product led to the assumption that the nuclear copies of the mitochondrial genes are
without function.
The pattern of rearranged "mitochondrial" sequences in the nucleus is reminiscent
of fused genes in "petite" mitochondrial genomes. This special organization led to the
suggestion that "mitochondrial" sequences in the nucleus originate from petite mito-
chondrial DNA which may have entered the nucleus.

3. Podospora anserina
In Neurospora crassa and Saccharomyces cerevisiae the presumptive gene transfer
from mitochondria to the nucleus took place during evolution. On the contrary, in
Podospora anserina there seems to be a regular transposition of mitochondrial se-
quences to the nucleus during ontogenesis. Wright and Cummings l4 " identified se-
quences homologous to the first intron of subunit I (pIDNA) and to coding sequences
of subunit III of mitochondrial cytochrome c oxidase (CO) in senescent nuclear DNA
by hybridization experiments. Data from these experiments indicated that the intron
sequence is integrated in an amplified multimeric "head-to-tail" form, whereas the
COllI gene seems to be integrated as a monomeric unit. Such homologies were never
detected in nuclear DNA derived from juvenile mycelia.
Additional evidence was brought about by investigation of the Podospora anserina
mutant mex-l. This mutant - which shows longevity - completely lacks the first
intron of the COl gene in mitochondrial DNA.14" Hybridization of mex-l nuclear DNA
with the cloned first intron of the COl gene, however, revealed a homologous nuclear
sequence. In this context it is noteworthy that another longlife mutant ex-l of Podos-
pora anserina was isolated in which COl sequences including pi DNA could not be
detected either in mitochondria or in the nucleus. 150 Thus it is possible that transposi-
tional events occur in the Podospora anserina mutant mex-l as well as in Podospora
anserina wild-type mycelia, but not in Podospora anserina ex-l. However, the possi-
bility cannot be ruled out that the hybridization signals in wild-type DNA are due to
contamination of senescent nuclear DNA with mitochondrial DNA. Cloning of nuclear
DNA homologous to mitochondrial sequences is necessary in order to address this
question further. Further experiments will show whether there is a correlation between
transposition and integration of mitochondrial sequences into nuclear DNA and the
onset of senescence in wild-type Podospora anserina mycelia.
There are several notions about the possible mechanisms of the exchange of genetic
information between organelles. In this context two ideas are interesting:

1. Mitochondria may lyse, thus releasing DNA into the cytoplasm. This DNA may
enter the nucleus and integrate into chromosomal DNA.
2. Transposition of mitochondrial genes may be facilitated by the action of transpo-
son- or plasmid-like sequences. In yeast, for instance, nuclear genes homologous
to parts of mitochondrial genes are flanked by yeast transposable elements in
some, but not all, strains. 140

Both possibilities appear reasonable, and at present it is impossible to decide between

them. Perhaps sometimes a combination of both (or other?) mechanism may lead to
common sequences in mitochondria and nucleus.
 249

It would be useful to demonstrate transposition as a process which still takes place

in present-day organisms. Investigation of such an example could provide better un-
derstanding of the interactions of nuclear and mitochondrial systems.

C. Altered Mitochondrial DNA and Mitochondrial Plasmids

This chapter first deals with DNA, which is different from the normal mitochondrial
"chromosome" but which is in some way derived from it, including plasm ids which
are derived from standard mtDNA. Second, molecules are reviewed which are not
derived from the "chromosome" of the mitochondria but are localized within these
organelles.
Sometimes the term "mitochondrial plasmids" is restricted to small circular or linear
DNAs that are located within the mitochondria but the sequences of which are not
completely part of the standard mtDNA. This nomenclature apriori would exclude
plasm ids capable of stable integration into mtDNA as well as excision from it by any
mechanism. (But compare stable integration of molecules called plasmids into bacterial
chromosomes: e.g., Hfr from F factor in E. coli K 12151 or plasmid pMC 7105 of
Pseudomonas syringaepv. phaseolica.J52

1. Altered mtDNA and Plasm ids Derived from mtDNA

Apart from smaller alterations, for example mitochondrial mutations to drug resist-
ance reviewed by Dujon et al. 153 (probably point mutations or small deletions), or mit-
mutants of yeast!53 (point mutants deficient for one or a few specific functions), some
phenotypes are known in fungi, which are correlated with larger alterations of the
mitochondrial chromosome. These are the "rho" mutants of Saccharomyces cerevis-
iae, the "poky" and "stopper" mutants of Neurospora crassa, the "ragged" mutants
of Aspergillus amstelodami, the senescence of Podospora anserina and Podospora
curvicolla, the senescence (kalilo cytoplasms) of Neurospora intermedia, and possibly
a kind of senescence in Cochliobolus heterostrophus.

a. rho- Mutants in Yeast

The "petite" colony mutation in Saccharomyces cereV1Slae, first described by
Ephrussi,4 represents the beginning of extrachromosomal genetics in fungi and the
classical example for cytoplasmic inheritance. Countless papers have been published
on this subject, including several reviews.'3,34,!54 Although a lot of detailed information
is available, and models have been proposed, the molecular mechanisms leading to
petites are not understood.
rho- Mutants or cytoplasmic petites are respiratory deficient: in aerobiosis they form
the smaller "petite" colonies compared to the wild type "grands" colonies, consisting
of cells being capable of respiration. On non fermentable media, rho- cells are unable
to grow. In anaerobiosis, normal and rho- cells have identical growth rate.
The yeast Saccharomyces cerevisiae is a facultative anaerobe, an attribute which is
very rare among eukaryotes. But just this attribute makes yeast especially suitable for
the investigation of grossly altered mitochondrial DNA. In contrast to most other eu-
karyotes, which as obligate aerobes would die in consequence of respiratory incapabil-
ity, even complete loss of mtDNA reSUlting in loss of respiratory functions does not
lead to death of Saccharomyces cerevisiae. As long as appropriate carbon sources are
available, respiratory-deficient yeast cells can live using fermentative pathways, pro-
ducing ATP only by glycolysis.
The rho- mutation is pleiotropic: not only one or few specific respiratory functions
are affected, but the complete mitochondrial protein synthesis is missing, i.e., all mi-
tochondrially coded proteins are absent and all the remaining mitochondrial proteins
are coded for by the nucleus.
250 Fungal Virology

The rho- mutation is located extrachromosomally. In crosses between wild type and
rho- this phenotype undergoes non-Mendelian segregation (in contrast to the nuclear
petites, pee, reviewed by Gilham). ISS This clearly demonstrates the extrachromosomal
character of the mutation. The rho- mutants arise spontaneously with the very high
rate of about 1070 and this rate may even be drastically increased by chemical or phys-
ical induction. 3 • Revertants do not appear.
In contrast to other classes of respiratory deficient mutants, different petites do not
restore a wild type phenotype by complementation or recombination. The rho- mutants
are neutral or suppressive. With regard to the transmission of the rho- mutation to the
diploid progeny of crosses between rho- mutants and wild type, neutral petites are
distinguished from suppressive petites. The first are not transmitted; the second are
transmitted to the diploid progeny to a degree varying from one specific rho- to the
other (l to 99%). This degree of suppressiveness is inherited by the subclones of a
specific rho- mutant, but mutation to a different degree of suppressiveness is possible.
The stability (or vice versa, the mutability) of the degree of suppressiveness is a herit-
able and mutable character itself.
The place of genetical alteration is the mitochondrial DNA. But it is, in the words
of Wilkie,156 "not a mutation in the normal sense." Petites are the consequence of
large deletions of wild type mitochondrial DNA and amplification of the remaining
fragment. Each petite is characterized by its own specific deletion and, vice versa, its
own specific retained fragment of mitochondrial DNA. The extent of the retained frag-
ment ranges from 80% to often only 1% of the wild type mitochondrial chromosome
and even below (e.g., 66 bp'57). The retained fragment may be derived from any part
of the wild type mitochondrial DNA provided that the fragment includes an ori/rep
sequence.
Except for the degree of suppressiveness and the stability of this degree, all petites
have the same phenotype, although different fragments of the wild type mitochondrial
DNA are deleted and retained, respectively. The explanation is that the smallest dele-
tion known from petites is sufficient for the lack of an essential component of the
mitochondrial protein biosynthesis which results in the unique, pleiotropic phenotype.
The specific retained fragment is amplified. The mt DNA isolated from petites like
isolated wt mt DNA, mainly consists of linear fragments, being longer than the re-
tained fragment. In addition, mt DNA from most petites contains populations of small
circular molecules, the contour lengths of which conform to a series of multimers.
However, both kinds of molecules, linear and circular, consist of the same basic repeat
unit, characteristic for the specific petite but different in size and sequence for different
petites. However, in many petites, besides the main population, further minor popu-
lations of molecules exist, differing from the main component 15 • and some contain
complex mixtures of coexisting molecules which surprisingly do not segregate during
subcloning. '58
The repeat units are arranged in one of two main types of organization: 34
1. Direct tandem repeats (head to tail, generating one new junction (fc):
wt mt .... abc d e f g h ....
repeat unit cdef
amplification .... c d e f c d e f c d e f ....
new junction .... f c ....

2. Inverted tandem repeats (generating two new junctions (ff' and c'c):
wt mt DNA .... abc d e f g h ....
repeat unit c d e f f' e' d' c'
amplification .... c d e f f' e' d' c' c d e f f' e' d' c' c d e f f' e' d' c' ..
new junctions .... f f' .... and .... c' c ....
 0 00 0
251

excision by replication recombination

Illegitimate

~Q:::OO-d,,"1 0
site specific
recombination Inverted
------

_ _
8
~.~- _4I.~C>+----
C:-1I__
____

wtmtDNA rolling circle -~~1 • _ ___ / )

!
C~
FIGURE 3. Model of generation of linear and circular amplifications by illegitimate site-specific re-
combination and replication. Direct tandem duplication and inverted duplication:. direct repeats,
~ short inverted repeats (palindromes); for details see text.

Other petites contain molecules having both direct and inverted repeats.

Most petites contain ori or rep sequences. Although replication of petit mt DNA
lacking ori or rep sequences (repO rho-) seems to be possible, such DNAs are not able
to compete with molecules containing one of the (at least) seven ori sequences distrib-
uted over the wt mt DNA of yeast. 97.98.159 This feature and, in addition, possibly dif-
ferent efficiencies of the sequences functioning as origins of replication may at least
partly be responsible for the different degrees of suppressiveness 97 and unequal reten-
tion of markers of different map regions during mutation from rho+ to rho-.160.161
The molecular mechanisms generating petites, i.e., the mechanisms resulting in dele-
tion and retention, respectively, of wt mt DNA and amplification of the retained frag-
ment are not understood. However, on the basis of detailed information on the struc-
ture and sequence of wt mt DNA and petite DNA and similar processes known from
prokaryotes (lambda integration and excision, lambda dv plasmids), models have been
proposed. 154. 162.163
From experiments on petite induction, it seemed likely that DNA repair and repair-
linked recombination are of importance for the generation of petites. As the first step,
a monomeric circle might be excised from the wt mt DNA by illegitimate, but site-
specific, recombination l64 168 (Figure 3). Possible sites for such recombinations are the
(A+T) rich and (G+C) rich clusters scattered around the wt mt DNA.169
Short direct repeats have been found at the excision sites of wt mt DNA 170 and
molecules of some petites show Hae III or Hpa II site clusters at their junction se-
quences,171 but others lack such sequences. 157.'71 Amplification might be the result of
extended replication (for example by replicating rolling circles l67 ) and recombination
(Figure 3). The above-mentioned inverted tandem repeats might be generated by in-
verse recombination of monomers (and subsequent replication), provided these contain
a short inverted repeat (~ in Figure 3).
The question whether petite formation itself is of interest only for yeast cells or
whether it is an unavoidable byproduct of general fundamental mechanisms, the dis-
252 Fungal Virology

covery of which is more difficult in other organisms because of their obligate aero-
biosis, is still not answered.

b. Poky Mutants of Neurospora crassa

The original poky(= mi-l) mutant was isolated as a spontaneous mutant by Mitchell
and Mitchell in 1952.!72 Mutants related to poky with respect to the phenotype were
isolated by Bertrand, Pittenger, and colleagues (for references, see Reference 173). All
these, poky, exn-1, exn-2, exn-3, exn-4, 50-1, 50-3, and stp-B-1, represent the so-
called group I mutants which are characterized as follows: slow growth, gross deficien-
cies of cytochromes aa 3 and b, excess of cytochrome c, female fertility, extrachromo-
somal inheritance, suppressive behavior in heteroplasmons, and suppression by a nu-
clear suppressor mutation called f. Poky and other group I mutants are further
characterized by gross deficiencies of mt small ribosomal (sr) RNA and mitochondrial
small ribosomal subunits, leading to a deficiency of mitochondrial protein synthesis.
A 4-base pair deletion, occurring in all 4 analyzed group I mutants in the coding
sequence for the mt srRNA resulting in aberrant mt srRNAs missing 38 to 45 nucleo-
tides at their 5'ends, is proposed to be the primary defect of group I mutants, although
this does not explain the suppressive behavior of this class of mutants.!74
Other alterations of mtDNA in group I mutants, such as small or large deletions and
independent or integrated amplifications of parts of mtDNA, are not causally related
to the mutant phenotype because they are found neither in all group I mutants nor in
all subclones of the mutants, although the group I phenotype is not lost in such sub-
clones. On the contrary, they seem to have no obvious effect on the phenotype apart
from a 20070 increase for the small and 40 to 80070 increase for the large mitochondrial
ribosomal subunit reported in connection with an 18 MDa amplification, containing
the mt t and rRNA genes.!75,176 However, group I mutations seem to favor the gener-
ation of mtDNA alterations or at least to stabilize such modified mtDNAs.

c. Stopper Mutants of Neurospora crassa

The original stopper mutants of Neurospora crassa (stp, stpA, stpA18, stpB2, stpC)
were isolated and described by Pittenger and co-workers!73 and represent most of their
so-called group III mutants. One mutant (E35) isolated by de Vries et al.17? and another
(lAr155(II)2,107A herein abbreviated to lO7A) the origin of which is described in
Gross et. al., !78 were associated with the stopper mutants cited above, because of con-
formity with the stopper phenotype, which is as follows:

• Stop and start mode of growth (i.e., mycelial propagation is composed of irreg-
ular periods of growth and nongrowth)
• Deficiency of cytochromes band aa3
• Excess of cytochrome c
• Female sterility
• Non-Mendelian inheritance (if used as a paternal parent)

At the molecular level, modifications of mtDNA (partial deletions and/or amplifica-

tions) are described for each stopper mutant in relation to the stopper phenotype (see
Figure 4).
Bertrand et al.!7 reported a 16 MDa (24 kb) amplification of the tRNA-rRNA region
in stp. On the other hand, some restriction fragments are nonamplified and the restric-
tion patterns of these differ slightly from subisolate to subisolate. Some fragments
cannot be detected in stp, e.g., EcoRI-3. On the other hand, new fragments appear
that do not comigrate with fragments of wt mtDNA. Either the nondetectable frag-
ments are part of a deletion or they are part of the new fragments. Other stopper
 253

----------.- ~01~
circle in growing

FIGURE 4. Map of type II mitochondrial DNA of Neurospora crassa. The map is derived from Gross
et al.'" The ribosomal sequences are noted by open boxes (intron by a jagged line). The two tRNA met
direct repeats are noted by boxes, all other tRNA sequences are noted by small circles. Regions of dele-
tions, underrepresentation, or amplifications correlated to stopper mutants are noted by thick lines. For
details see text.

mutants are characterized by other alterations of mtDNA: 17 stpA shows a 5-kb deletion
in the region of EcoRI-2 and -10, while stpA18 has a 0.35-kb deletion in EcoRI-7b,
and stpB2 has a 4-kb insertion in EcoRI 2. The stop and start mode of growth is
proposed to be the result of competition between predominating defective mtDNA
molecules and underrepresented less defective mtDNA molecules.
In E35 a permanent deletion of 3.5 kb in the region of EcoRI-2 and 10 is described
as the primary event, leading to the stopper phenotype.'79.'80 The fact that complete wt
mtDNA was never found in E35 seemed to be incompatible with the "competition
hypothesis" of Bertrand et al.'7
During subsequent growth on agar for several months, the mutant showed accumu-
lation (amplification) of an 8 /-Im-(24 kb) circular molecule containing the tRNA -
rRNA region. The formation of the aberrant molecules is found to be the result of
recombination at GC-rich palindromic sequences. '77.'78.'80 The fact that the mutant in
the beginning does not show any detectable amplification would classify the 24-kb
circle as a consequence rather than a cause of the stopper phenotype. But, as shown by
254 Fungal Virology

Gross et al." 78 it is the actual growth phase of the inoculum used for the culture from
which the DNA is isolated, not the overall time of vegetative growth, which is impor-
tant.
A further typical feature of E35 is the absence of an ll-kDa mt translation product.
Because the mutant is deficient in cyt band aa 3, although cyt b apoprotein and all
subunits of cyt aa3 are present in the mt inner membrane, it is assumed that the 3.5-kb
deletion concerns a mitochondrial gene coding for a protein (perhaps the 11 kDa-pro-
tein) responsible for the proper assembly of the respiratory chain.179,180 Sequencing of
part of the wtDNA which is deleted in E35revealed an URF coding for over 375 amino
acid residues and showing homology to human URF 2.'80 However, these results do
not explain the typical stop and start mode of growth.
Gross et al. 178 investigated mtDNA isolated from different physiological growth
states of the stopper mutant lO7A. DNA from mitochondria of the stop phase consists
of a heterogeneous population of molecules: the predominant molecule is a 7.2 /-Im-(21
kbp) circular amplification of the tRNA-rRNA region of the wt mtDNA. From restric-
tion patterns it was concluded that the adjacent region (EcoRI-6 to -7a) is present in a
normal amount. The remaining regions of wt mtDNA are more and more underrepre-
sented, with increasing distance from around the junction of EcoRI-7a and -5 up to
appearing deleted in the region of EcoRI-lO and -2 (Figure 4). Consequently, the non-
amplified part of mtDNA is suggested to be represented by a heterogeneous population
of molecules extending from a point in EcoRI-l to different points in the underrepre-
sented region.
The region appearing deleted overlaps the deletion of E35 which possibly contains a
gene responsible for the assembly of the respiratory chain. The adjacent, severely un-
derrepresented region contains the COl and the cyt b genes. Both features could ex-
plain the cytochrome deficiencies of "stopper".
DNA from mitochondria of the growth phase mainly consists of two types of mole-
cules: The most prominent again is the 21-kbp circle. The next most frequent is a 14.6-
/-1m (43 kb) circular molecule, which is exactly complementary to the 21-kbp circle and
includes even those sequences appearing deleted in the stop phase. In wt mitochondria,
both circles are found at low amounts in addition to the complete mtDNA circle.
The two complementing circles are supposed to have originated by intramolecular
reciprocal recombination at direct repeats of the complete wt mtDNA. Possible sites
of frequent homologous recombination leading to the 7.2- and 14.6-/-Im reciprocal cir-
cles are the two direct repeats of tRNA met. However, these are surely not the only
sites for intramolecular recombination, because the mtDNA contains many repeated
sequences.
These results allow speculation about the molecular steps involved in the control of
the cyclical mode of growth of stopper mutants. One, but not the only plausible way,
is the following:
Primary event: prevention of normal replication of the complete mtDNA circle, per-
haps by generation of a kind of stop signal for normal replication (as hypothesized by
de Vries et al. 179 ) or destruction or deletion of the most efficient origin.
Molecules of the existing population of heterogeneous circles, generated by intra-
molecular recombination and usually severely underrepresented, gain replicative sig-
nificance. Depending on the distribution of sites of recombination and the distribution
and efficiency of functional origins of replication, some regions of wt mtDNA will
appear to be amplified or at least replicated to a moderate extent, while others will
become drastically underrepresented (functionally deleted). Functional deletion of
genes, e.g., functional deletion of those responsible for the assembly of the respiratory
chain or structural genes of proteins of the respiratory chain, finally leads to cessation
of growth.
 255

Us.rRNA

FIGURE 5. Map of the mtDNA of Aspergillus amstelodami

(derived from Lazarus and Kuntzel"), showing the location of
rRNA genes (open boxes) and the two regions giving rise to
amplifications in ragged mutants (solid bars). E=EcoRI,
P=Pstl, B=BamHI, S=SalI. For details see text.

Physiological conditions during stoppage of growth somehow might allow compen-

sation, leading to sufficient representation of all essential sequences. Possibly this is
achieved simply by better viability of those hyphal compartments which fortuitously
contain mitochondria with adequate proportions of all required sequences.
An interesting unresolved question in connection with this phenotype is, what is the
causative molecular difference between "stoppers" which are suppressive (e.g., stpA),
and those which are not suppressive (e.g., J07A), in heteroplasmons.

d. Ragged Mutants of Aspergillus amstelodami

Mutants of Aspergillus amste10dami (a member of the Aspergillus glaucus species
group) in many respects resembling the vegetative death (vgd) condition of Aspergillus
glaucus,181 were first isolated and described by eaten. 182 They are characterized by
irregular growth of colonies, low conidial viability, deficiency of cytochrome a, excess
of cytochrome c, suppressiveness, and cytoplasmic inheritance, but in contrast to the
vegetative death condition of Aspergillus glaucus, by "stop and start" growth behavior
and the possibility of indefinite maintenance by subculturing. Taking these differences
into account, they were called "ragged' (rgd).
Only the rgdmutants of Aspergillus amste10damiand the rgd4mutant of Aspergillus
heterocaryoticus (a natural variant of Aspergillus amstelodami) were used for the mo-
lecular investigations of Kuntzel and co-workers: 15.46.183.184
All six rgd mutants investigated contain amplifications of parts of wt mtDNA, ar-
ranged in circular tandem repeats. The amplifications occur in addition to the complete
intact wt mt genome and are mutant specific, e.g., differing in size of the monomer
(rgdl/O,9 kb; rgd312.7 kb; rgd4/1.7 kb; rgd512.2 kb; rgd6/1.5 kb; rgd7/1. 7 kbp).
However, they could be located to only two regions of wt mt DNA (Figure 5). In
addition, only the amplification in rgdl arises from region 1 (located around the EcoRI
115 junction), whereas all other amplifications proved to be homologous to sequences
of region 2 (located at EcoRI-4a, -7, and -2). The latter, according to their different
sizes, have different excision sites, but overlap, sharing a common sequence of 215 bp.
This common sequence is suggested to carry an origin of replication, because part of
it can be arranged into a hairpin loop similar to origin-structures of other organisms.
As far as investigated, no homologies exist between heads or tails of the amplifications
256 Fungal Virology

and the corresponding flanking sequences in normal mtDNA. Thus, the generation of
these amplifications by homologous recombination seems unlikely.
The predominance of region 2 amplifications in ragged strains could, perhaps, be
due to a replicative advantage of region 2 amplifications resulting in suppression of
other amplifications. This view is supported by the occurrence of a region 2 amplifi-
cation in a rgdlline (containing the region 1 amplification), which on subculturing led
to enrichment of the region 2 amplification and loss of the region 1 amplification.
Although the mechanisms generating amplifications of mt sequences in stopper mu-
tants of Neurospora crassa and Aspergillus amstelodami probably are different (com-
pare direct repeats in Neurospora crassa l07A), even if these are not variable in one
organism, mechanisms causing stop and start mode of growth may be similar, i.e.,
competition between different members of a population of more-or-Iess defective mt
DNA molecules. In "stopper", however, competition takes place between comple-
menting parts of normal mt DNA, in "ragged" between parts of and complete
mtDNA. Investigation of different growth stages of "ragged" could decide whether
this speculation proves to be true.

e. Senescence in Podospora anserina

Senescence or strain aging in Podospora anserina was first reported in 1953 185 and
subsequently investigated by Rizet and co-workers (for references, see References 186,
187, and 193).

Each strain of Podospora anserina is characterized by a limited capability of vegetative

growth: 186.187
When Podospora anserina is cultured asexually, after a period of linear propagation
a progressive decrease l88 of growth occurs, resulting in complete arrest of growth and
finally death of the hyphae. The cessation of growth is accompanied by a reduction in
the amount of aerial hyphae, vacuolization of hyphae and bursting of their tips, dark
pigmentation of the mycelium, and decrease l88 of cytochrome aa3.

Timing of senescence depends on both environment and genotype: 186.187

Darkness and low temperature postpone the onset of senescence, whereas optimal
growth conditions favor aging. Different strains (races) of Podospora anserina, that is
different genotypes, show a different but strain-specific life span (median length of
growth).

Senescence is under nuclear and mitochondrial control:

(1) Mutations in nuclear genes influence senescence: the mating type +189 as well as
mutations in morphogenetic genes, the latter especially when acting synergistically in
certain combinations,19o.191 may postpone the onset of senescence; and (2) further, mi-
tochondrial point mutations affect life span. 192

Senescence is maternally inherited:

(1) In reciprocal interracial crosses the life span of the progeny is according to the
maternal parent;189 and (2) in reciprocal crosses between juvenile and senescent myce-
lia, the aging stage of the paternal parent is without importance. A juvenile proto-
perithecial parent results in juvenile progeny only, a senescent maternal parent leads to
senescent (900/0) as well as juvenile (10%) progeny, demonstrating discontinuous dis-
tribution of a cytoplasmic factor. 193 Thus, indefinite maintenance of the strains is pos-
sible by regular regeneration by sexual propagation.
 257

Table 6
COMPILA TION OF DATA CHARACTERIZING THE DIFFERENT
AMPLIFICA TIONS CORRELATED WITH SENESCENCE IN
PODOSPORA ANSERINA

Designation Density Size of Number of

(synonyms) (g/cm) Conformation monomer Genetic location events Ref.

pIDNA 1,699 ccc 2539 bp Intron I of COl Many 19,20,56,78,

(asenDNA) Tandemly amplified 115,148,149,
188,194-209
(JsenDNA 1,694 Circular 9.8 kb COllI 2" 20,148,196,201
Tandemly amplified
ysenDNA 1,694 ?- ?- 1rRNA 20,196,201
dsenDNA 1,694 ?- ?- Near COIl 2 20,196,201
(JsenDNA Varying Circular Varying COl Many 195,199,201,
(0senDNA) 1,693-1,696 Tandemly amplified I.l kb 203,207,210-
Common 212
sequence

There are published only two events (3 preparations) of (JsenDNA. First event preparation 4 and 7,
second event preparation 5, reported in Cummings et al. ". These show differences in Hae III restriction
pattern. Therefore, it seems unclear whether all (JsenDNAs are absolutely identical.
Only a fragment of the Hae III restriction pattern of the DNAs of the senescence events was cloned and
analyzed in each case. In both cases the cloned fragments are probably not new junction fragments but
Hae III or Hae III/EcoRI fragments occurring in standard mtDNA too. Correlation between the circular
molecules found in the preparations and the cloned fragments was not shown.

Senescence is infectious:
The senescence factor is not only transmitted by sexual propagation, but also
through hyphal fusion (anastomoses) of vegetative hyphae. Once transmitted, the fac-
tor rapidly spreads through the mycelium. '.3

Senescence is postponed by certain inhibitors:

Effective inhibitors are either agents intercalating with DNA or affecting mitochon-
drial functions. In addition, by means of inhibitor effects, growth of Podospora an-
serina can be divided into three phases:'"

1. A juvenile phase in which all effective substances postpone the onset of senes-
cence;
2. A presenescent phase which is morphologically indistinguishable from the juve-
nile phase but in which only DNA-intercalating agents act to postpone senes-
cence; and
3. A senescent phase, correlated with morphological alterations and failure of any
tested inhibitor to produce a life-prolonging effect.

Subculturing of the inhibitor-treated mycelia on inhibitor-free medium displays senes-

cence. Thus the inhibitors postpone the onset of senescence only as long as they are
present in the medium, but they do not cure senescence.
Search for the infective cytoplasmic factor led to the discovery of small circular
DNAs correlated to senescence, the first of which was reported by Stahl et al." The
DNA species were characterized with respect to their densities, conformation, size,
origin, and location (Table 6).

Density:, •. 20.1.5 "7

The piDNA was isolated as an extra band in CsCI density centrifugation, having a
258 Fungal Virology

density different from nuclear and standard mitochondrial DNA: 1,699 g/cm 3. All
other reported senescence-correlated DNA species have the same density as standard
mtDNA, which is 1,694 g/cm 3.

Conformation: 19,20194,197,199
plDNA, (JsenDNA, and 8senDNA each were shown by electron microscopy (EM) to
consist of circular molecules of multimeric sets of a repeat unit (monomer), the sizes
of which correspond to the values determined by restriction analysis. For plDNA,
cccDNA molecules were shown by EM. Restriction analysis, in addition, revealed the
head-to-tail arrangement of the repeat units and led to circular physical maps. For y
and dsenDNA, from which only fragments were analyzed in detail, it is reported that
the DNA preparations contained circular molecules, but these were not directly corre-
lated with the fragments investigated.

Sizes of monomers:
For plDNA, the size of the monomer was determined by electron microscopy, re-
striction analysis, and recently by sequencing: plDNA consists of 2539 bp.78
The size of (JsenDNA monomers was evaluated by electron microscopy and restric-
tion analysis to be 9.8 kb,z° but in view of different HaeIII restriction patterns of two
preparations containing (JsenDNA,195 it seems unclear whether all events including the
"(J-region" are absolutely identical.
8senDNA is a family of different, independently arising amplifications, varying in
the size of the monomer, but all sharing a common sequence of 11 00 bp.212 Sizes of y-
and dsenDNA were not determined, but both seem to be very large. 20 ,196

Origin and location:

By hybridization studies, all DNA species correlated to senescence proved to origi-
nate from the standard mitochondrial genome: they each are an integral part of normal
mtDNA of juvenile mycelia. The different DNA species could be located to different
nonoverlapping parts of the physical map of the standard mtDNA. In addition, mt
genes localized on the restriction map, in some cases allowed genetic attachment. The
physical map of the standard mtDNA, showing the location of the different DNA
species, is presented in Figure 6.

Replication and expression:

plDNA, 13, y, and 8 sequences are reported to contain autonomously replicating
sequences (ARS) functioning in yeast. 78,201,203
Original plDNA as well as hybrid molecules (plDNA inserted into pBR322) were
able to transform juvenile mycelia to senescence, clearly demonstrating a cause and
effect relationship. 198 In addition, the hybrid molecules were shown to replicate auton-
omously in Podospora anserina: 213 They were reisolated unchanged from fungal trans-
formants and isolated unchanged from retransformed bacteria. Furthermore, Podo-
spora transformants not only express senescence but also a prokaryotic gene «(3-lacta-
mase) present on the hybrid plasmid in Podospora anserina. Thus these molecules
behave like true plasmids.
Sequencing data revealed a sequence in plDNA which may fold into a secondary
structure reminiscent of yeast and human ARS of mtDNA.78

Nuclear control:
An analysis of two nuclear, morphogenetic double mutants (i viv, gr viv) identical in
standard mtDNA, revealed that nuclear control concerns at least two steps in the onset
of senescence. 191 ,202,206
 259

kb
950
90
LrRNA
=--=r- 70

80

70

60

40

FIGURE 6. Physical map of mtDNA of Podospora anserina (derived from Klick et aI. "0), showing the
location of the rRNA genes, the COl and cyt b genes, plDNA and (3, y, d, and BsenDNA. For details see
text.

1. Liberation of plDNA from standard mtDNA. The mutant gr vivdoes not liber-
ate plDNA from mtDNA, but when plDNA is introduced into gr viv by trans-
formation, replication and expression of plDNA is not hindered.
2. Expression of the replicated pIDNA. In contrast, the mutant i viv liberates the
plDNA from the standard mtDNA and allows its replication but never becomes
senescent.

Mitochondrial deletion mutants:

mexl Is a slow-growing but long-living mitochondrial deletion mutant, which has
escaped from a senescent culture. In this case the plDNA is more-or-Iess precisely
deleted from standard mtDNA. 14 9.214 However, Wright and Cummings 148 reported that
in this mutant plDNA has integrated into nuclear DNA.
exl Is a mitochondrial mutant deficient for the entire COl gene, which includes the
plDNA. The mutant is slow growing and has increased longevity. ISO
260 Fungal Virology

This clearly demonstrates that neither the plDNA sequence nor the COl gene is
essential for the life of Podospora anserina, but rather that life is prolonged when the
sequence is deleted from mtDNA. Presence of the sequence in nuclear DNA as in
mex-l does not seem to hinder the life-prolonging effect of its deletion from standard
mtDNA. Other long-living mt mutants are reported to contain rearrangements in re-
gion 9 or y or both.209

Transposition to nuclear DNA:

plDNA and (JsenDNA are reported to be transposed to the nucleus and integrated
into nuclear DNA during senescence, possibly causing instability of the nuclear gen-
ome; 148 but not all senescent isolates showed nuclear integration of an amplified
mtDNA sequence. Thus transposition to the nucleus seems to be correlated to senes-
cence but not essential for the phenomenon. This is also demonstrated by the mex-l
mutant, which has plDNA integrated into nuclear DNA 148 but displays increased lon-
gevity.214

Molecular investigation of rejuvenation:

The occurrence of circular amplifications in senescent cultures is accompanied by
disappearance of the standard mtDNA as far as this can be concluded from gel electro-
phoresis. 56 ,188,189 Nevertheless, small amounts of complete standard mtDNA must be
present in senescent mycelia. Koll et a1. 207 have rejuvenated senescent mycelia which
contained plDNA and 9senDNA by treatment with ethidium bromide (EB). Whereas
in the senescent mycelia no standard mtDNA was detected by gel electrophoresis, the
rejuvenated mycelia showed the standard mtDNA pattern without any detectable
plDNA or 9senDNA.
This observation fits very well to the possibility of rejuvenation by sexual propaga-
tion (see above).

Predominance of pIDNA.
There is no doubt that plDNA is the molecule occurring in most of the senescence
events. It was even detected at low amounts in some juvenile cultures 195,199 but never
in juvenile long-living mutant strains.'49 Furthermore, plDNA is often present in se-
nescent cultures, in which another molecule (e.g., (1-, y-, d-, or 9-sen DNA) is the
mainly amplified molecule. 2o ,207 Thus plDNA is somehow a particularity, either being
favored in replication, perhaps due to a very effective origin of replication, or being
favored in generation, perhaps due to a special, more effective excision mechanism,
which may be connected with the fact that plDNA is an intron.

plDNA is a mitochondrial intron:

From sequencing data,'8,115,20S,206,208 it was concluded that plDNA is precisely the
intron 1 of the COl gene:

1. plDNA sequences show correspondence with consensus sequences characteristic

for mt introns of the group 11.'20
2. The RNA secondary structure which may be derived from the plDNA sequence
corresponds to a model for group II introns" 9 and brings the two splicing points
(corresponding to the excision points of plDNA) into close proximity.
3. The mtDNA sequences adjacent to the integrated plDNA show homology to the
coding sequences of the COl gene of Saccharomyces cerevisiae and Neurospora
crassa. The coding sequence is precisely interrupted by the plDNA.
4. The plDNA contains a long URF in phase with the preceding exon. The ami-
noacid sequence derived from the plDNA sequence shows significant homology
 261

COI gene

-;--. ..
Expression of mt DNA

QI
I ex 1/mex 1 I IIIlntron 1 ' IE Intron 2
IL-_ _ _- - - " I - Matu rase?
U Mitochondrial
t:
QI
U
Deficient in
II)
QI plDNA ~-r----"'''''I----- pre mRNA

...-
t:

o
~----l
QI
II)

o
QI sscRNA I II mRNA

t
IJ)
t:
o
Cl

... I,,,', I~E"O" "3"""PIIO"

t:
t:

Q-....g
QI
>
~ Nuclear

gg
0-
...
Il)
Ligation pi DNA Protein synthesis
...
t:
(1)

B -1
::J
~

L---_N_uclear
Mitochondrion ®y
~A((,~\
\~y Nucleus
Essential functions blocked t Cytochrome -c-oxidase
Senescence Subunit I

FIGURE 7. Model proposed by Kuck et al."o showing generation and targets of plDNA and inter-
vention of nuclear and mitochondrial mutations in the onset of senescence in Podospora anserina.
For details see text.

to the aminoacid sequences coded by the introns 1 and 2 of the Saccharomyces

COl gene, for which a RNA maturase function is postulated. 215

In addition sequencing data revealed that (1) at both excision sites of the integrated
piDNA short nonidentical interrupted palindromes are present; (2) a few basepairs up-
and downstream of the integrated pIDNA, 10 bp palindromes are localized which are
nonidentical, but the 8 inner bp of which represent an inverted repeat; (3) a 5-bp se-
quence at excision site 1 is directly repeated 8 bp downstream. Involvement of these
palindromes and repeats in the generation of free pIDNAs, possibly by DNA splicing,
has still to be proved. Precise excision of an intron sequence has been reported by
Gargouri et a1. 216
Because so far all group II iIftron sequences found in yeast are released during RNA
processing as stable single-stranded circular (ssc) RNAs,122.123 another speculation is
that piDNA may be generated by reverse transcription of such stable sscRNAs of the
intron sequence. This hypothesis is supported by the fact that in Northern hybridiza-
tions, two transcripts hybridize with pIDNA. This hybridization pattern either repre-
sents two different pIDNA homologous transcripts with different initiation sites or
represents the linear and circular configuration of a single transcript.150.204 However,
so far, reverse transcriptase has not been proved to exist in Podospora anserina.
Concerning the piDNA, which is the most prominent among the amplifications oc-
curring in correlation with senescence, a model (Figure 7) is proposed,150 showing the
different possible ways of generation of pIDNA, intervention of nuclear and mito-
chondrial mutations, and the possible targets of pIDNA.
To what extent other amplifications correlated to senescence can be integrated into
262 Fungal Virology

40

FIGURE 8. Physical map of mtDNA of Podospora curvicolla showing the

location of the IrRNA gene and the region giving rise to different amplifications
(pI-I, pI-2, pI-3) correlated to senescence.'" For details see text.

the model is still unclear; but taking into account that all other amplifications are not
exactly introns and often have varying excision sites (e.g., the 9 family of amplifica-
tions), probably different mechanisms of generation are working. Possibly the other
amplifications arise by site-specific recombination similar to the mechanism leading to
rho- mutants in yeast.

f. Senescence in Podospora curvicolla

Podospora curvicolla. a species which is closely related to Podospora anserina, also
shows symptoms of senescence. In contrast to Podospora anserina, it is able to resume
growth after a period of nongrowth, resembling the stopper mutants of Neurospora
crassa or the ragged mutants of Aspergillus amstelodami.
The plDNA of Podospora anserina was never found in Podospora curvicolla, either
as plamid or integrated in standard mtDNA. Thus, this special intron is not present in
Podospora curvicolla. This is not surprising in view of the size of the standard mtDNA
of Podospora curvicolla. which is only 55 kbp. However, in independent senescence
events different circular tandem amplifications occur in addition to very low amounts
of standard mtDNA. The different monomers are 9.0 kbp, 10.6 kbp, and 10.9 kbp in
size. By comparison of the restriction maps of the amplifications and standard mtDNA
as well as by DNA-DNA hybridizations, all were found to be homologous to the large
ribosomal (lr) RNA region of the mt genome (Figure 8), the genetic region which in
Podospora anserina gives rise to the y senDNA. All three amplifications overlap, hav-
ing a large common central part, but differ in their excision sites. Thus they represent
a family of amplifications like 9senDNA of Podospora anserina. 21 •217
 263

g. Senescence of Cochliobolus heterostrophus?

One of the 23 investigated strains of Cochliobolus heterostrophus, the isolate T40,
contains a tandemly amplified plasmid in addition to the normal mitochondrial
DNA.218 The monomer of the plasmid is 1.9 kbp in size. It is homologous to a distinct
part of the normal mtDNA, where the plasmid sequences occur as a single integrated
copy, not only in T40 but also in the mtDNA of strains not containing the free plasmid.
Homology could not be ascertained to the nuclear DNA of the organism, to the Neu-
rospora plasmids "Labelle", "Fiji", and "Mauriceville", or to the mtDNA of Podos-
pora anserina. The plasmid sequences function as an ARS in yeast.
Although there is no proof of causative relation, in contrast to other (plasmid free)
isolates, T40 showed cessation of growth during long time growth tests after 30 to 50
cm. Thus a phenotypic relationship to senescence of Podospora may be possible.

h. Senescence ("Kalilo Cytoplasms") in Neurospora intermedia

Rieck et al. 219 reported five variants of Neurospora intermedia collected from Kauai
(Hawaii) showing strong analogy to the stopper mutants of N. crassa: stop and start
mode of growth, cytochrome irregularities, deficiencies of mt ribosomes, and altered
mtDNA correlated to the altered phenotype.
Subsequently, Griffith and Bertrand220.221 investigated a large sample of Neurospora
intermedia strains from Hawaii, including most of the normally growing strains re-
ported by Rieck et al.:'19 26 strains showed senescence; at some specific point in serial
subculturing the strains by conidial mass transfer, growth ceases mostly irreversibly.
Random ascospore analysis of reciprocal crosses showed that senescence is maternally
inherited, indicating the cytoplasmic location of the senescence factor. In addition ran-
dom ascospore analysis, ascus analysis, and conidial analysis indicated heterogeneity
of the cytoplasm with respect to the senescence factor. Subculturing by conidial mass
transfer seemed to have an averaging effect.
Molecular studies'19 revealed that senescence is correlated with progressive deficien-
cies of cytochrome band aa 3 and excess of cytochrome c, alterations of mtDNA (re-
duction of some fragments of standard mtDNA, appearance of new fragments), and
occurrence of a unique fragment, which shows no homology to standard mtDNA.
This is of special interest in comparison with the main plasmid causing senescence in
Podospora anserina, but being part of the standard mtDNA in juvenile mycelia. Pos-
sibly the new fragment occurring in Neurospora intermedia during senescence will
prove to be a plasmid, too. However, because it has no homology to standard mtDNA,
it will be reviewed in the next part of this chapter.

2. Other Mitochondrial Plasmids

Plasmids reported in the following paragraphs are those intramitochondrially lo-
cated plasmids that are usually not an integral part of standard mtDNA and as yet are
not correlated to a specific altered phenotype. However, not all of these plasmids are
completely unrelated to the standard mtDNA (see below, canonical sequence of Maur-
iceville-Ic plasmid of Neurospora crassa). Usually they are found in wild-type strains
recently isolated from nature but not from laboratory strains.

a. Mitochondrial Plasmids in the Genus Neurospora

The three plasmids of this kind characterized first were found in wild-type strains:
Mauriceville-Ic of Neurospora crassa and P405-Labelle and Fiji N6-6 of Neurospora
intermedia, respectively!,·13 The plasmids were designated "Mauriceville", "La-
belle", and "Fiji" with respect to their origin.
According to their occurrence in wt strains, they were discovered not by a specific
altered phenotype but at the molecular level by strong bands appearing in addition to
264 Fungal Virology

Hinell
Bgill

Mauriceville Hinell
EeoRI Labelle plasmid
plasmid
3.6kb BamHI 4.1kb HineR

HineR
Pst!

HineR
Fiji plasmid
HineR
5.2kb
HineR
Bgill

Bgill

FIGURE 9. Physical maps of the Mauriceville, Labelle, and Fiji plasmids. showing the relative position of
different cleavage sites and the location of restriction site clusters, respectively (derived from References 13
and 223). For details see text.

standard mtDNA restriction patterns. Purification by EB CsC! density centrifugation,

EM, and restriction analysis proved these bands to be parts of oligomeric series of ccc
molecules with the repeat units arranged tandemly. The monomers were 3.6 lim (Maur-
iceville), 4.1 lim (Labelle), and 5.2 lim (Fiji) in size. Restriction maps of the monomeric
circles are shown in Figure 9.
By DNA-DNA hybridization experiments strong sequence homology could not be
detected either to the corresponding mtDNAs or to each other.
Intramitochondrial location of the plasmids was proved by their isolation from nu-
clease-treated mitochondria. This location is in agreement with maternal inheritance of
the Mauriceville and Labelle plasmids in reciprocal crosses, or at least typical cyto-
plasmic inheritance of the Fiji plasmid (the Fiji strain could only be used as male parent
because of inefficiency in the formation of protoperithecia).
No major transcripts of the Labelle and Fiji plasmids could be detected by DNA-
RNA hybridization. However, it was concluded from similar experiments that the
Mauriceville plasmid is transcribed in vivo and that the predominant transcript is close
to the monomer length of the plasmid (see also below).
Translation products encoded by the plasmids could not be detected by comparative
analysis of mitochondrially coded proteins from plasmid-containing and plasmid-free
 265

strains. This is perhaps due to their low concentration or to comigration with major
mtDNA translation products.
An additional polypeptide was reported by Collins et al., 12 but it was unclear
whether it was due to the Mauriceville plasmid or to differences between the standard
mtDNAs of the strains used. However, it was no longer detected in two additional
experiments. 13
Sequencing of the complete 3581 bp Mauriceville plasmid and transcript mapping
gave the following results: 222 • 223 the PstI site cluster of the plasmid contains 8 Pst! sites
organized in five palindromic elements (two contain only one PstI site). The two out-
ermost elements are identical with the canonical 18-nucleotide GC-rich sequence,
which is found repeated many times throughout the Neurospora crassa mt genome. 125
The remaining three elements differ from the canonical sequence by only a few nucleo-
tides. This difference contrasts with the strong conservation of the canonical sequence
in mtDNA. In addition, the PstI cluster contains a 162 and 168 bp, respectively, direct
imperfect repeat (identical in 143 positions) which could reflect a duplication event,
including two of the 5 palindromic elements mentioned above. In principle these ele-
ments are potential sites for homologous recombination (integration, excision) with
standard mtDNA, but integration has not been observed so far.
The EcoRI site cluster is located within the long ORF of the plasmid (see below) and
does not appear to be correlated with any special feature such as longer palindromes
or repeats.
Using the Neurospora crassa mtDNA genetic code, the plasmid contains a long
ORF. This ORF is expressed in the major transcript and could encode a polypeptide
of 710 amino acids. SI nuclease and primer extension experiments revealed that the
major transcripts are full length, colinear molecules with major 5' and 3' ends imme-
diately adjacent to each other and in some cases slightly overlapping. Some minor
transcripts have identical 3' but different 5' ends. The plasmid seems to be related to
group I mtDNA introns (according to the classification of Michel and Dujon 120 ), be-
cause a set of sequence elements shows good correspondence to the consensus sequence
elements of this class of introns. A fourth plasmid which was isolated from Neurospora
intermedia Varkud-lc, has a monomer size of 3.8 kbp and has some homology to the
Mauriceville plasmid of Neurospora crassa (e.g., the direct repeat of about 160 bp is
also present in the Varkud-lc plasmid). Like the Mauriceville plasmid, the Varkud
plasmid is transcribed. 223 • 22 '
Further plasm ids were found in wild strains of Neurospora tetrasperma designated
according to the geographic origin of the strains from which they were isolated: "Han-
alei", "Lihue", "Waimea Falls", and "Surinam" ." The first three plasmids, all iso-
lated from strains from Hawaii, were indistinguishable by Hinc II digest pattern and
hybridization with cloned Hanalei plasmid (recombinant plasmid pNMT22: Hanalei
plasmid linearized by cleavage of the single EcoRI site and cloned into the EcoRI site
of pBR325). The Surinam plasmid showed a similar but slightly different HincH re-
striction pattern and also strong hybridization with pNMT22.
Thus the four plasmids are related to each other, and the three plasmids from Ha-
waiian isolates may be identical. They appear to exist predominantly as ccc head-to-
tail amplifications. The monomer of the Hanalei plasmid is 5.0 kbp in size. The Lihue
and Waimea Falls plasm ids probably are the same.
In addition, homology was detected between the Hanalei plasmid and the Fiji plas-
mid of Neurospora intermedia, but not between the Hanalei plasmid and the Labelle
plasmid from Neurospora intermedia or the Mauriceville plasmid from Neurospora
crassa.

b. Linear Plasmids of Claviceps purpurea

The linear plasmids of Claviceps purpurea were discovered by the appearance of low
 266 Fungal Virology

molecular bands in the electrophoresis pattern of undigested mtDNA preparations of

one of 6 investigated strains: two strong bands at 6.6 kb (pH) and 5.3 kb (pI2), respec-
tively, and two faint bands at 10 kb (p13) and 1.1 kb (pI4), respectively." Because these
DNAs have the same buoyant density as mtDNA (1699 g/cm3), purification was only
possible by elution from agarose gels. Because of the small amounts of the DNAs
available only pH, p12, and pl4 were characterized further. 11 ,225 The pI D N As are lo-
cated intramitochondrially, because they can be isolated from nuclease-treated mito-
chondria.
From the fact that in EM preparations only linear molecules were present, the aver-
age lengths of which (2.1, 1.7, and 0.4 ~m) correspond very well to the values deter-
mined by gel electrophoresis, it was concluded that they are linear molecules. Physical
maps were constructed confirming the linearity of the molecules. By Southern hybrid-
ization it was shown that that the three plDNAs share common sequences but none
share homologies with standard mtDNA of the strain. Thermal denaturation and ren-
aturation of pH as well as pl2 led to stem loop formations, indicating 300 bp inverted
repeats at both ends of the molecules. The function of the plasmids is still unknown,
but similarities with mt plasm ids of higher plants 22• and occurrence only in wild strains
suggest that they have a function in the parasite-host relationships.

III. NONMITOCHONDRIAL PLASM IDS

The term "nonmitochondrial plasmids" comprises plasmids of unknown origin as

well as those occurring in the nucleus. Plasmids with recognized nuclear association
are the 2-~m DNA of the baker's yeast Saccharomyces cerevisiae and the Ddp 1 plas-
mid of the cellular slime mold Dictyostelium discoideum.

A. Nuclear Associated Plasmids

1. 2-j,lm DNA of Saccharomyces cerevisiae
The 2-~m DNA was the first eukaryotic plasmid to be discovered and is probably
the most famous and best characterized of the fungal plasmids. It was shown to be
present in most of the strains analyzed in 30 to 100 copies per cell, comprising about
2.4 to 30/0 of the total cellular yeast DNA. 227 ,228

a. Biophysical Properties and Structure

The plasmid is a circular molecule with a contour length of 1.8 to 2.0 /-I m227 233 and
a buoyant density of 1.698 g/ml, thus banding with nuclear DNA in cesium chloride
gradients. Despite its chromatin-like structure,>3.,23' it was never found to be integrated
into chromosomes as revealed by Southern experiments. 229 It might therefore be con-
sidered as an additional type of chromosome except for its lack of a centromere region
and its high copy number.
Homoduplex analysis by electron microscopy231,236 and sequencing data 237 revealed
the presence of two inverted repeats with a length of 599 bp, separated by segments of
unique sequences 2346 bp and 2774 bp in length. These inverted repeats are known to
be the site of intramolecular recombination events giving rise to two different forms of
the 2 /-1m plasmid, called A and B form (Figure 10).238,239 They differ in the orientation
of one unique region with respect to each other and occur in equal amounts. Intermo-
lecular recombination at these sites might further result in the formation of oligomeric
plasmids, which were found to exist beside the two forms just mentioned. 2•

b. Replication
The origin, from which replication starts and continues bidirectionally, encompasses
a 350-bp sequence of one of the inverted repeats and a 100-bp region of the adjacent
 267

A - Form

B - Form

FIGURE 10. Diagram of 2 I'm DNA. The 599 inverted repeats are
shown as parallel lines separating a large (L) from a small (S) unique
segment. A reciprocal cross-over event mediated by the FLP product
occurs near the two XbaI sites and converts the A form to the B form.
The genes FLP, REP 1, REP2, and REP3 are shown in thick lines with
a taper at the 3' ends of the genes indicating the direction of transcrip-
tion. The dotted region shows the origin of replication. 234 239

large unique sequence (Figure 10).238,240 The part within the unique region has a high
A T content (80070), a feature which it shares with yeast chromosomal replication sites.
Replication is mediated by enzymes encoded by the 2 J.lm circle itself. The replication
loci REP I, REP2, and REP3 ensure stable propagation and maintenance of a high
copy number in the yeast cell. 241 The genes REP! and REP2 are trans-acting and cor-
respond to two open coding regions, whereas REP3, located several hundred bp away
from the origin, is active in cis and consists of direct repeats of a 62-bp sequence. It is
thus somewhat reminiscent of the structure and situation of the simian virus 40-enhan-
cer region, which is also contiguous to the origin and consists of 72-bp direct re-
peats. 242 ,243
These three genes are not solely responsible for replication: chromosomally encoded
functions, like the chromosomal replication machinery, are required as well!" Repli-
cation is further under strict cell cycle control. 245 In cells with normal plasmid titers,
replication takes place only during the early stage of the S-phase so that each molecule
duplicates once every cell division. 24 •
Nonetheless, evidence has accumulated that REP-loci constitute a copy control sys-
tem that is able to override restriction on plasmid replication by cell cycle control and
to amplify the plasmid when copy number is too low. 241 Sigurdson et al!" demon-
strated that after acquisition of one plasmid by cytoduction, copy number in cells
originally lacking the 2-l-Im circle will rise to normal levels.
268 Fungal Virology

It should be stressed that the REP genes constitute only an amplification system and
do not have the capacity to promote autonomous replication on their own.

c. Interconversion
In addition to replicative functions, the plasmid also encodes a protein that enables
the circle to undergo recombination. This event is, at least in part, catalyzed by a
product of the trans-acting FLP gene.238.239.248 FLP-mediated recombination is site-
specific and occurs in a region which has been limited to a 60-bp sequence spanning
the XbaI sites in the inverted repeats (Figure 10). As shown recently, it even promotes
recombination in the bacterial transposon Tn5. 249
Functional aspects of interconversion are still unknown. Broach 250 discussed a role
in maintenance of the plasmid in the cell. Recombination might lead, e.g., to alterna-
tive gene expression of two sets of genes as described for the H2 locus of Salmonella 251
or the G-loop of phage Mu. 252 It might further serve as a replication aid allowing
resolution of catenated oligomers which normally arise during replication of circular
molecules.

d. Function
The role of the 2-/-Im plasmid in the yeast cell remains obscure. It appears that it is
involved solely in its own expression and propagation, thus stating a prime example of
selfish DNA!S3 However, there are several lines of evidence that it might be associated
with oligomycin resistance and lethal sectoring. Guerineau and colleagues 253 and Guer-
ineau23 described a correlation between the loss of oligomycin resistance and the ab-
sence of 2-/-Im circles in mutant strains originally harboring the plasmid and exhibiting
multiple antibiotic resistance. These results, however, do not establish a causal rela-
tionship. Oligomycin resistance cannot be ascribed to 2-/-Im DNA unless it will be
shown that reintroduction of the circle reestablishes the resistance phenotype.
Lethal sectoring 254 is a phenomenon exhibited by strains of Saccharomyces carlsber-
gensis which carry the 2-/-Im DNA and the recessive chromosomal nibl allele. Cultures
with this genetic background produce two types of cells: small cells which will give rise
to the so called "nibbled colonies" and large cells that contain twice as much 2-/-Im
DNA as the small ones and which are nonviable. The elevated plasmid titer apparently
causes a defect in DNA replication or nuclear division. Holm 254 proposed that the NIB
locus controls the 2-/-Im circle copy number: the wild-type product normally represses
amplification of 2-/-Im DNA, whereas the nib allele is defective in its function.

2. The Ddp 1 Plasmid in Dictyostelium discoideum

In contrast to the 2-/-Im DNA or Saccharomyces cerevisiae only little is known about
the plasmid in Dictyostelium. Most of the work on this plasmid has been done by Metz
and co-workers. 22 The data available may be summarized as follows.
It has a covalently closed circular structure with a length of 4 /-1m. The buoyant
density has not been determined yet, but restriction enzyme analysis indicates a high
AT-content.
It does not appear to encode for essential functions, since some wild strains ob-
viously do not carry the plasmid, and long term growth in axenic medium results in its
loss. Nevertheless, it may be involved in cobalt resistance, as suggested by the fact that
cobalt-resistant strains contain the plasmid while cured cobalt-sensitive derivatives of
these strains do not contain it or at least do so in a considerably lower copy number.

B. Plasmids of Unknown Association

Plasmids with an undetermined origin and location have been described in several
fungal species. A compilation of these plasmids is given in Table 1. The best analyzed
 269

among these plasm ids are the two linear plasm ids pGKl and pGK2, occurring in strains
of the yeast Kluyveromyces lactis. They have molecular weights of 8.8 kbp and 13.4
kbp, respectively, and identical buoyant densities (1.687 g/m£).28 Strains harboring
these plasmids secrete a toxic protein that is lethal to sensitive Kluyveromyces strains
as well as to a variety of other yeast species, e.g., Saccharomyces cerevisiae, Saccha-
romyces italicus, Saccharomyces rouxii, Torulopsis glabrata, and Candida utilis.28
The smaller plasmid pGKl encodes for killer and resistance functions while the
larger one, pGK2, is supposed to be involved in replication and maintenance. 28 They
both carry inverted repeats at their ends with lengths of 182 bp and 202 bp, respect-
ively255 (for further details, see Chapter 2 of this book).
The other plasmids have been characterized to a lesser extent. In the edible mycor-
rhiza-forming fungus Morchella conica two linear plasmids with lengths of 6 kbp and
and 8 kbp have been detected. At least the smaller one carries terminal 750-bp inverted
repeats as revealed by homoduplex analysis. 29 Their possible role in the mycorrhizal
behavior of Morchella is being investigated!56
A 2l-kbp plasmid of covalently closed circular structure with a heterogeneous buoy-
ant density (ca 1.699 g/ml) and restriction pattern has been isolated from Cephalospo-
rium acremonium!6,257
Recently, Hashiba et aU 58 reported the presence of a linear double-stranded DNA
plasmid, 2.6 kbp in length, in weakly pathogenic isolates of the phytopathogenic fun-
gus Rhizoctonia solani. According to the authors, this plasmid might be related to the
pathogenic properties of this fungus (see Chapter 4).
In general, however, functional aspects of most of these plasmids have not been
clarified to date. But the fact that they are widely distributed suggests that they might
play an essential role in their hosts. In any case, they offer us potential tools for mo-
lecular cloning and genetic engineering with fungi.

IV. GENETIC ENGINEERING WITH FUNGAL

EXTRACHROMOSOMAL DNA

A. Yeasts
The most advanced fungal transformation systems have been developed with yeasts.
To date, successful transformation systems have been reported for the yeast species
Saccaromyces cerevisiae: 59 ' 261 Schizosaccaromyces pombe: 62 Kluyveromyces lac-
tis: 63 ,26" and Kluyveromyces iragilis!65 The advantage of yeast cells with respect to
genetic manipulation is their bacteria-like ease of handling. There is rapid growth in
genetically homogeneous populations, a defined alteration of sexual stages, and a large
variety of extrachromosomal genetic elements. Protoplasts are obtained readily and
even intact cells can be transformed when treated with alkali cations such as lithium
ions!66 A further advantage of the yeast system is the comparatively stable mainte-
nance and expression of the introduced plasmids, which can be achieved by some plas-
mid modifications.
Hybrid plasmids consisting of Escherichia coli plasm ids and selectable markers of
yeast chromosomal DNA (like the LEU2, HIS3, and URA3 gene) integrate into the
homologous chromosomal region. 261 ,262 Since in most cases the entire plasmids gets
integrated, this event will lead to duplication of homologous sequences. The plasmid
YIP 1 (pBR 322 and the yeast HIS3 fragment) integrated in the vicinity of the chro-
mosomal his3 locus, thus duplicating the his3 and HIS3 sequences. Other plasmids
have been reported to integrate also into several unlinked chromosomal regions, like
the pYeleu 10 plasmid (recombinant plasmid of ColEl and the yeast LEU2 frag-
ment).267
Transformation frequency of these plasmids, however, is low (less than 10 trans-
270 Fungal Virology

formants per lAg DNA) and somewhat unstable. With a frequency of 1070 per cell divi-
sion they get looped out of the genome by reversion of the integration process. 260 ,261
Transformation efficiency is greatly improved (up to 103 to 10 4 transformant cells per
lAg DNA) by inserting the 2-lAm DNA origin or chromosomal or mitochondrial ars-
sequences from homologous or heterologous sources. This manipulation converts the
integrating plasm ids to autonomously replicating ones.
Various heterologous DNA fragments, such as origins of Tetrahymena thermophila
ribosomal DNA and of Xenopus laevis mitochondrial DNA as well as putative origins
from chromosomal DNA of Neurospora crassa, Dictyostelium discoideum, Droso-
phila meianogaster, Zea mays,>68 Trypanosoma brucei,>69 Vinca rosea, ChI orella ellip-
soideus 270 and from mitochondrial DNA of Cephalosporium acremonium 271 have
been shown to have ars functions in yeast. Even telomere sequences from nonyeast
sources, such as Drosophila 272 or vaccinia virus 273 are able to function as ars. Though
transformation efficiency is considerably enhanced, mitotic and meiotic stability of
these plasmids is still unsatisfactory. Under nonselective conditions 10 to 30070 of the
transform ant cells will lose the plasmid after 15 to 20 generations (for references, see
References 270 and 274). Plasmid losses can be significantly reduced through addi-
tional insertion of yeast centromere sequences. These artificial minichromosomes are
maintained and inherited relatively stably in the absence of selective pressures (for
review and details see Blackburn 275 ).
Another aspect that makes the yeast an ideal candidate for gene transfer is its im-
pressive capacity to express heterologous genes from bacterial276.280 and mammalian
sources. 281 286 Application of this property might culminate in commercial production
of human interferon and insulin with yeasts. 287 Stepien and co-workers 281 reported the
expression of a synthetic human pro insulin gene ligated with the promoter and protein
leader sequence of the GALl gene in the Escherichia coli-yeast hybrid vector pYT781O.
Hitzeman 282 and Dobson 283 and their colleagues succeeded in expressing human inter-
feron genes by fusing them to the yeast ADHI gene promoter.
Experiments related to heterologous gene cloning, however, revealed that yeast in
some cases does not provide the adequate environment for proper expression of the
foreign genes; e.g., genes encoding rat growth hormone,27o rabbit (J-globin,>88 or Dro-
sophila alcohol dehydrogenase 289 have been shown to be expressed abnormally because
of posttranscriptional barriers. Several reports demonstrated differences in splicing
behavior between yeasts and other eukaryotes; the yeast failed to excise the foreign
introns. 288- 292
Some of these difficulties might be circumvented by using more highly evolved fun-
gal species, which are more likely to process transcripts in the same way as do mam-
malian cells.

B. Filamentous Fungi
Genetic engineering procedures for filamentous fungi are by far much more difficult
and laborious. As a reflection of their higher structural complexity and their coenocytic
organization, protoplasts, once obtained, are heterogeneous with respect to their num-
ber of nuclei, organelle constitution, and biochemical function. 293 High levels of nu-
cleases and polysaccharides not seldom complicate purification and characterization
of the nucleic acids. 294 295 Specific procedures have to be developed for each particular
species.
To facilitate isolation, analysis, and manipUlation of fungal genes, efforts are cur-
rently centered on the development of shuttle vectors that are able to replicate in fila-
mentous fungi as well as in Escherichia coli or yeasts. Promising and efficient systems
in this regard have been elaborated with Aspergillus nidulans, Neurospora crassa, Po-
dospora anserina, and Cephalosporium acremonium.
 271

Timberlake and co-workers 296 succeeded in transforming an Aspergillus nidulans

trpC- strain to trpC+ with a complete wild-type copy of the Aspergillus nidulans trpC
gene that was inserted into the bacterial vector pBR 329. This hybrid plasmid, either
circular or linear in structure, transforms Aspergillus nidulans at a rate of more than
20 stable transformants per Jig DNA.
Tilburn et a1. 297 reported transformation of a mutant that was defective in the ace-
tamidase-Iocus (amdS) by a recombinant plasmid of pBR 322 and the wild-type copy
of the amdS gene. Heterologous gene expression in Aspergillus nidulans has also been
achieved. Ballance et al!98 converted a pyrG- mutant of AspergiIlus nidulans to pro-
totrophy by using a plasmid carrying the pyr-4 gene of Neurospora crassa.
In all cases the plasmids became integrated into the resident gene sites. Integration
events are analogous to those described for integrating plasmids in yeasts.
Autonomous replication of hybrid plasmids has been reported in strains of Neuro-
spora crassa. Stohl and Lambowitz 299 stably transformed qa-2- strains to qa-2+ with a
vector consisting of the Neurospora qa-2+ gene, the Escherichia coli plasmid pBR 325,
and a mitochondrial plasmid from Neurospora crassa P405-Labelle. Transformation
frequencies were 5- to lO-fold higher than those described for integrating plasmids.
The recombinant plasmid was shown to be present in nuclear as well as in cytosolic
fractions of transformants. Parts of it could also be detected in mitochondria. How-
ever, many of the plasm ids recovered from the Neurospora transformants carried dele-
tions of the entire Labelle insert. It was concluded that pBR 325 plus the qa-2+ segment
probably constitutes a Neurospora replicon. 107
Similar results have been obtained by transforming Neurospora crassa strains defec-
tive in the glutamate dehydrogenase gene (am-) to am+. In this case the authors even
postulated formation of oligomeric derivates of the transforming plasmid. 'oo
Giles and co-workers ,ol published work on a pBR 322 vector with an inserted qa-2+
gene that showed free replication as well as integration in stably transformed qa-2-
strains. The yield of transformants was about 100 per Jig DNA. Plasmid DNA reiso-
lated from Neurospora crassa turned out to be apparently methylated. A substantial
part of the plasmid was resistant to cleavage by Bcll, an enzyme that does not cut
sequences which have been modified by Escherichia coli dam methylase. Genomic
DNAs of Neurospora crassa and the inserted plasmid sequences originating from Neu-
rospora, however, were completely digested. On the assumption that Neurospora
crassa does not possess a methylation enzyme like that of Escherichia coli, these results
were interpreted as evidence for plasmid contamination and the paper was retracted. 102
Bull and W ootton '03 very recently however demonstrated that Neurospora crassa heav-
ily methylates incoming foreign DNA. The cellular function of this methylation system
in Neurospora is still unknown. It does not appear to affect gene expression but is
more likely to be involved in recombination or repair processes of amplified or re-
arranged DNA.'o,
The first hybrid plasmid freely replicating in a filamentous fungus and in bacterial
hosts was described by Esser and co-workers.211 Using pBR 322 and sequences of the
senescence-inducing p1DNA of Podospora anserina, they succeeded in transforming
the long-lived double-mutant gr vivof Podospora anserina (see Section II.C.l.e) to
senescence.
The transformant strains were further able to express the prokaryotic vector part as
well: they produced (3-lactamase. Since pBR 322 by itself is neither replicated nor ex-
pressed in Podospora anserina, the eukaryotic vector part was identified as responsible
for self-replication of the hybrid vector; it could be replaced by homologous regions
of native mitochondrial DNA which also proved to function as a replicon.
Encouraged by their own data and examples given above, Esser et a1. 104 suggested
the use of mitochondrial plasmids or replicons as a base to construct self-replicating
272 Fungal Virology

shuttle vectors for eukaryotes. The potential and practicability of this new concept in
genetic engineering has been impressively demonstrated by recent work of Tudzynski
and Esser. 3os ,30o They constructed a shuttle vector (pCP2) by inserting a mitochondrial
DNA fragment from Cephalosporium acremonium into the yeast/bacterial hybrid
plasmid pDAMI which lacks a eukaryotic origin of replication. This shuttle vector was
shown to replicate autonomously in Saccharomyces cerevisiae, Cephalosporium acre-
monium, and Podospora anserina.
This concept is of particular importance for strain improvement of industrially used
species that are not accessible to sexual genetics. Nevertheless, it should be noted that,
in order to broaden the avenues for genetic engineering in fungi and to fully exhaust
the potential provided by nature, it might become necessary to use fungal RNA viruses
as gene vectors as well.

ACKNOWLEDGMENT

We thank Prof. Dr. Dr. h. c. Karl Esser for his support and advice, Tom Elthon
and Dr. Ulrich Kuck for reading parts of the manuscript, and H. Rathke for doing the
artwork.

REFERENCES

l. Correns, c., Vererbungsversuche mit bla(J (gelben), grunen und buntblattrigen Sippen bei Mirabilis,
Urtica und LunarIa, Z. Vererbungsl., I, 291, 1909.
2. Correns, C., Zur Kenntnis der Rolle von Kern und Plasma bei der Vererbung, Z. Vererbungsl.,2,
331,1909.
3. Baur, E., Das Wesen und die Erblichkeitsverhaltnisse der 'Varietates albo-marginatae hort' von Pe-
Iargonium zonaIe, Z. Vererbungsl., I, 330,1909.
4. Ephrussi, B., Hottinguer, H., and Chimenes, A. M., Action de I'acriflavine sur les levures. I. La
mutation 'petite colonie', Ann. [nst. Pasteur Paris, 76, 351, 1949.
5. Ephrussi, B., Hottinguer, H., and Tavlitzki, J., Action de I'acriflavine sur les levures. II. Etude
genetique du mutant 'petite colonie', Ann. [nst. Pasteur Paris, 76,419, 1949.
6. Ephrussi, B., L'Hertier, P., and Hottinguer, H., Action de I'acriflavine sur les levures. VI. Analyse
quantitative de la transformation des populations, Ann. [nst. Pasteur Paris, 77, 419, 1949.
7. Sionimski, P. P. and Ephrussi, B., Action de l'acriflavine sur les levures. V. Le systeme des cyto-
chromes des mutants 'petite colonie', Ann. [nst. Pasteur Paris, 77,47, 1949.
8. Nass, M. M. K. and Nass, S., Intramitochondrial fibers with DNA characteristics. I. Fixation and
electron staining reaction, J. Cell. BioI., 19, 595, 1963.
9. Nass, M. M. K. and Nass, S., Intramitochondrial fibers with DNA characteristics. II. Enzymatic and
other hydrolytic treatments, J. Cell. BioI., 19, 613, 1963.
10. Borst, P. and Grivell, L. A., The mitochondrial genome in yeast, Cell, 15, 705, 1978.
II. Tudzynski, P., Duvell, A., and Esser, K., Extrachromosomal genetics of Claviceps purpurea. I. Mi-
tochondrial DNA and mitochondrial plasmids, Curro Genet., 7, 145, 1983.
12. Collins, R. A., Stohl, L. L., Cole, M. D., and Lambowitz, A. M., Characterization of a novel
plasmid DNA found in mitochondria of N. crassa, Cell, 24,443, 1981.
13. Stohl, L. L., Collins, R. A., Cole, M. D., and Lambowitz, A. M., Characterization of two new
plasmid DNAs found in mitochondria of wild-type Neurospora intermedia strains, Nucleic Acids
Res., 10, 1439, 1982.
14. Natvig, D.O., May, G., and Taylor, J. W., Distribution and evolutionary significance of mitochon-
drial plasm ids in Neurospora spp., J. Bacteriol., 159, 288, 1984.
15. Lazarus, C. M. and Kuntzel, H., Anatomy of amplified mitochondrial DNA in 'ragged' mutants of
Aspergillus amstelodami: excision points within protein genes and a common 215 bp segment con-
taining a possible origin of replication, CurT. Genet., 4, 99, 1981.
16. Garber, R. C., Turgeon, B. G., Yoder, O. C., A mitochondrial plasmid from the plant pathogenic
fungus Cochliobolus heterostrophus, Mol. Gen. Genet., 196, 301, 1984.
 273

17. Bertrand, H., Collins, R. A., Stohl, L. L., Goewert, R. R., and Lambowitz, A. M., Deletion mutants
of Neurospora crassa, mitochondrial DNA and their relationship to the 'start-stop' growth pheno-
type, Proc. NatI. Acad. Sci., U.S.A., 77,6032, 1980.
18. de Vries, H., de Jonge, J. C., Van't Sant, P., Aggsteribbe, E., and Amberg, A., A 'stopper' mutant
of Neurospora crassa containing two populations of aberrant mitochondrial DNA, Curr. Genet., 3,
205, 1981.
19. Stahl, U., Lemke, P. A., Tudzynski, P., Kuck, U. and Esser, K., Evidence for plasmid like DNA in
a filamentous fungus, the ascomycete Podospora anserina, Molec. Gen. Genet., 162, 341, 1978.
20. Wright, R. W., Horrum, M. A., and Cummings, D. J., Are mitochondrial structural genes selectively
amplified during senescence in Podospora anserina?, Cell, 29,505, 1982.
21. Bockelmann, B. and Esser, K., A plasmid of mitochondrial origin in the ascomycete Podospora
curvicolla correlated with the onset of senescence, in: 3rd European Congr. Biotechnol., VoL 3,
Verlag Chemie, Weinheim, 1984,247.
22. Metz, A., Ward, T. E., Welker, D. L. and Williams, K. L., Identification of an endogenous plasmid
in Dictyostelium discoideum, EMBO J., 2, SIS, 1983.
23. Guerineau, M., Plasmid DNA in yeast, in Viruses and plasmids in fungi, Lemke, P., Ed., Marcel
Dekker, New York, 1979,539.
24. Francou, F., Isolation and characterization of a linear DNA molecule in the fungus Ascobolus im-
mersus, Mol. Gen. Genet., 184, 440, 1981.
25. Kempken, F., personal communication, 1984.
26. Minuth, W., Tudzynski, P., and Esser, K., Extrachromosomal genetics of Cephalosporium acremon-
ium. L Characterization and mapping of mitochondrial DNA, Curro Genet., 5,227, 1982.
27. Ballario, P., Filetici, P., Junacovic, N. and Pedone, F., Tyl extrachromosomal circular copies in
Saccharomyces cerevisiae, FEBS Lett., 155, 225, 1983.
28. Gunge, N., Tamaru, A., Ozawa, F. and Sakaguchi, K., Isolation and characterization of linear de-
oxyribonucleic acid plasm ids from Kluyveromyces lactis and the plasmid associated killer character,
J. Bacteriol., 145, 382, 1981.
29. Meinhardt, F. and Esser, K., Linear extrachromosomal DNA in the morel Morchella conica, Curro
Genet., 8, 15, 1984.
30. Clark-Walker, G. D. and Azad, A. A., Hybridizable sequences between cytoplasmic ribosomal RNAs
and 3 micron circular DNAs of Saccharomyces cerevisiae and Torulopsis glabrata, Nucleic Acids
Res., 8, 1009, 1980.
31. Del Giudice, L., Wolf, K., Sassone-Corsi, P., and Mazza, A., 2/-lm covalently closed non-mitochon-
drial circular DNA in the petite-negative yeast Schizosaccharomyces pombe, Mol. Gen. Genet., 172,
165, 1979.
32. Fournier, P., Gaillardin, C., de Louvencourt, L., Heslot, H., Lang, B. F., and Kaudewitz, F., r-
DNA plasmid from Schizosaccharomyces pombe: cloning and use in yeast transformation, Curro
Genet., 6,31, 1982.
33. Evans, I. H., Molecular genetic aspects of yeast mitochondria, in Yeast Genetics, Fundamental and
Applied Aspects, Spencer, J. F. T., Spencer, D. M., and Smith, A. R. W., Eds., Springer Verlag,
New York, Berlin, 1983,269.
34. Dujon, B., Mitochondrial genetics and functions, in The molecular biology of the yeast Saccharo-
myces, Life cycle and inheritance, Strathern, J. N., Jones, E. W. and Broach, 1. R., Eds., Cold
Spring Harbor Laboratory, Cold Spring Harbor, N.Y., 1981,505.
35. Borst, P., Grivell, L. A. and Groot, G. S. P., Organelle DNA, TIBS, 100, 128, 1984.
36. Michaelis, G. and Pratje, E., Mitochondrial genetics, Progr. Bot., 44,286, 1982.
37. Sedoroff, R. R., Structural variation in mitochondrial DNA, Adv. Genet., 22, I, 1984.
38. Gray, M. W. and Doolittle, E. F., Has the endosymbiont hypothesis been proven?, Microbiol. Rev.,
46, I, 1982.
39. Tabak, H. F., Grivell, L. A. and Borst, P., Transcription of mitochondrial DNA, Crit. Rev.
Biochem., 14,297, 1983.
40. Kroon, A. M. and Saccone, C., Eds., The organization and expression of the mitochondrial genome,
VoL 2, Elsevier/North-Holland. Amsterdam, New York, Oxford, 1980.
41. Slonimski, P. P., Borst, P. and Attardi, G., Eds., Mitochondrial Genes, Cold Spring Harbor Labo-
ratory, Cold Spring Harbor, N.Y., 1982.
42. Schweyen, R. J., Wolf, K., and Kaudewitz, F., Eds., Mitochondria 1983, Nucleo-Mitochondrial
Interactions, de Gruyter, Berlin, 1983.
43. McLean, J. R., Cohn, G. L., Brandt, J. K., and Simpson, M. V., Incorporation of labelled amino
acids into the protein of muscle and liver mitochondria, J. BioI. Chem., 233,657, 1958.
44. Hudspeth, M. E. S., Shumard, D. S., Bradford, C. J. R., and Grossman, L. J., Organization of
Achlya mtDNA: A population with two orientations and a large inverted repeat containing the rRNA
genes, Proc. Natl. Acad. Sci. U.S.A., 80, 142, 1983.
274 Fungal Virology

45. Dizikes, G. J. and Burke, D. D., Isolation and characterization of the mitochondrial DNA of Allo-
myces macrogynus, Biochim. Biophys. Acta, 518,2471,1978.
46. Kuntzel, H., Kochel, H. G., Lazarus, C. M. and Lunsdorf, H., Mitochondrial genes in Aspergillus,
in Mitochondrial Genes, Slonimski, P. P., Borst, P. and Attardi, G., Eds., Cold Sprmg Harbor
Laboratory, Cold Spring Harbor, N.Y., 1982, 391.
47. Clark-Walker, G. D. and Skriprakash, K. S., Size diversity and sequence rearrangements in mito-
chondrial DNAs from yeast, in Mitochondrial Genes, Slonimski, P. P., Borst, P. and Attardi, G.,
Eds., Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y., 1982,349.
48. O'Connor, R. M., McArthur, C. R., and Clark-Walker, G. D., Closed-circular DNA from mito-
chondrial-enriched fractions of four petite-negative yeasts, Eur. f. Biochem., 53, 137, 1975.
49. Stahl, U., personal communication, 1984.
50. Wesolowski, M. and Fukuhara, H., Linear mitochondrial deoxyribonucleic acid from the yeast Han-
senula mrakii, Molec. Cell. BioI., I, 387, 1981.
51. Mariott, A. C., Archer, S. A. and Buck, K. W., Mitochondrial DNA in Fusarium oxysporum is a
46.5 kilobase pair circular molecule, f. Gen. Microbiol., 130, 3001,1984.
52. Collins, R. A. and Lambowitz, A. M., Structural variations and optional introns in the mitochondrial
DNAs of Neurospora strains isolated from nature, Plasmid, 9,53, 1983.
53. Chaiprasert, A., personal communication, 1983.
54. Bohnert, H. H., Size and structure of mitochondrial DNA from Physarum polycephalum, Exp. Cell.
Res., 106, 426, 1977.
55. Klimczak, L. J. and Prell, H. H., Isolation and characterization of mitochondrial DNA of the oom-
ycetous fungus Phytophthora infestans, Cur. Genet., 8,323, 1984.
56. Kuck, U., Stahl, U., and Esser, K., Plasmid-like DNA is part of mitochondrial DNA in Podospora
anserina, Curro Genet., 3,151,1981.
57. Bockelmann, B., personal communication, 1984.
58. Sanders, J. P. M., Heyting, C., Verbeet, M. P., Meijlink, C. P. W., and Borst, P., The organization
of genes in yeast mitochondrial DNA. III. Comparison of the physical maps of the mitochondrial
DNAs from three wild-type Saccharomyces strains, Mol. Gen. Genet., 157, 239,1977.
59. Kuck, U., Stahl, U., Lhermitte, A., and Esser, K., Isolation and characterization of mitochondrial
DNA from the alkane yeast Saccharomycopsis lipolytica, Mol. Gen. Genet., 157, 239, 1977.
60. Clark-Walker, G. D. and Gleason, F. H., Circular DNA from the water mold Saprolegnia, Arch.
Mikrobiol., 92,209, 1973.
61. Lang, B. F., Ahne, F., Distler, S., Trinkl, H., Kaudewitz, F., and Wolf, K., Sequence of the mito-
chondrial DNA, arrangement of genes and processing of their transcripts in Schizosaccharomyces
pombe, in Mitochondria 1983, Schweyen, R. J., Wolf, K., and Kaudewitz, F., Eds., de Gruyter,
Berlin, 1983, 313.
62. Mery-Drugeon, E., Crouse, E. 1., Schmitt, J. M., Bohnert, H. -J. and Bernardi, G., The mitochon-
drial genomes of Ustilago cynodetis and Acanthamoeba castellanii, Eur. f. Biochem., 114, 577,
1981.
63. Scazzacchio, D., Brown, T. A., Waring, R. B., Ray, J. A., and Davies, R. W., in Mitochondria
1983, Schweyen, R. J., Wolf, K., and Kaudewitz, F., Eds., de Gruyter, Berlin, 1983,303.
64. Anderson, S., Bankier, A. T., Barrell, B. G., de Bruijn, M. H. L., Coulson, A. R., Drouin, J.,
Eperon, J. C., Nierlich, D. P., Roe, B. A., Sanger, F., Schreier, P. H., Smith, A. J. H., Staden, R.,
and Young, J. G., Sequence and organization of the human mitochondrial genome, Nature (Lon-
don), 290,457, 1981.
65. Padmanaban, G., Hendler, F., Patzer, J., and Rabinowitz, M., Translation of RNA that contains
polyadenylate from yeast mitochondria in an Escherichia coli ribosomal system, Proc. Natl. Acad.
Sci. U.S.A., 72,4293, 1975.
66. Cobon, G. S., Groot-Obbink, D. J., Hull, R. M., Maxwell, R., Murphy, M., Rytka, J., and Linnane,
A. W., Mitochondrial genes determining cytochrome b (complex III) and cytochrome oxidase func-
tion, in Genetics and Biogenesis of Chloroplasts and Mitochondria, Bucher, T., Neupert, W., Sebald,
W., and Werner, S., Eds., Elsevier/North-Holland, Amsterdam, 1976,453.
67. Macreade, J. G., Novitski, C. E., Maxwell, R. J., John, U., Beng-Guat Ooi, McMullen, G. L.,
Lukins, H. B., Linnane, A. W., and Nagley, P., Biogenesis of mitochondria: the mitochondrial gene
(aapl) coding for mitohondrial ATPase subunit 8 in Saccharomyces cerevisiae, Nucleic Acids Res.,
11,4435,1983.
68. Hudspeth, M. E. S., Ainley, W. M., Shumard, D. S., Butow, R. A., and Grossman, L. J., Location
and structure of the var 1 gene on yeast mitochondrial DNA: nucleotide sequence of the 40.0 allele,
Cell, 30,617, 1982.
69. Lazowska, J., Jacq, C., and Slonimski, P. P., Sequence of introns and flanking exons in wild-type
and box3 mutants of cytochrome b reveals an interlaced splicing protein coded by an intron, Cell,22,
333, 1980.
 275

70. Church, G. M. and Gilbert, W., Yeast mitochondrIal mtron products reqUlred introns for RNA
splicing, in Mobilization and Reassembly of Genetic Information, Scott, W. A., Werner, R., Joseph,
D. R., and Schultz, J., Eds., Academic Press, New York, 1980,379.
71. Anziano, P. Q., Hanson, D. K., Mahler, H. R., and Perlman, P. S., FunctIOnal domains in mtrons:
trans-acting and cis-acting regions of mtron 4 of the cob gene, Cell, 30,925, 1982.
72. Jacq, C., Pajot, P., Lazowska, J., Dujardin, G., Claisse, M., Groudinsky, 0., de La Salle, H.,
Grandchamp, C., Labouesse, M., Gargouri, A., Guiard, B., Spyridakis, A., Dreyfus, M., and Slon-
imski, P. P., Role of introns in yeast cytochrome b gene. cis- and trans-acting signals, intron manip-
ulation, expression and intergemc communications, in Mitochondrial Genes, Slonimski, P. P., Borst,
P. and Attardi, G., Eds., Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y., 1982.
73. Mahler, H. R., Hanson, D. K., Lamb, M. R., Perlman, P. S., Anziano, P. Q., Glauss, K. R., and
Haldi, M. L., Regulatory interactions between mitochondrial genes: expressed introns - their func-
tion and regulation, in Mitochondrial Genes, Slommski, P. P., Borst, P. and Attardi, G., Eds., Cold
Spring Harbor Laboratory, Cold Spring Harbor, N.Y., 1982.
74. WeiSS-Brummer, B., Rodel, G., Schweyen, R. J., and Kaudewitz, F., Expression of the split gene
cob in yeast: evidence for a precursor of a 'maturase' protein translated from intron 4 and preceding
exons, Cell, 29, 527, 1982.
75. Carignani, G., Groudinski, 0., Frezza, D., Schiavon, E., Bergantino, E., and Slonimski, P. P., An
mRNA maturase is encoded by the first intron of the mitochondrial gene for the subunit I of cyto-
chrome oxidase m S. cerevisiae, Cell, 35,733, 1983.
76. Waring, R. B., Davies, R. W" Scazzocchio, C. and Brown, T. A., Internal structure of a mitochon-
drial intron of Aspergillus nidulans, Proc. Natl. Acad. Sci. U.S.A., 79,6332, 1982.
77. Helmer Citterich, M. H., Morelli, G., and Macino, G., Nucleotide sequence and intron structure of
the apocytochrome b gene in Neurospora crassa mitochondria, EMBO J., 2, 1235, 1983.
78. Osiewacz, H. D. and Esser, K., The mitochondrial plasmId of Podospora anserina: a mobile intron
of a mitochondrial gene, Curro Genet., 8,299, 1984.
79. Dujon, B., Mitochondrial genes, mutants and maps: a review, in Mitochondria 1983, Schweyen, R.
1., Wolf, K., and Kaudewitz, F., Eds., de Gruyter, Berlin, 1983, 1.
80. Burger, G., Scriven, C., Machleidt, W., and Werner, S., Subunit 1 of the cytochrome oxidase from
Neurospora crassa: nucleotide sequence of the coding gene and partial amino acid seque'1ce of the
protein, EMBOJ., I, 1385, 1982.
81. de Jonge, 1. and de Vries, H., The structure of the gene for subunit I of cytochrome c oxidase in
Neurospora crassa mitochondria, Curro Genet., 7, 21, 1983.
82. Waring, R. B., Brown, T. A., Ray, 1. A., Scazzocchio, C., and Davies, R. W., Three variant introns
of the same general class in the mitochondrial gene for cytochrome oxidase subunit I in Aspergillus
nidulans, EMBO J., 3,2121, 1984.
83. Hensgens, L. A. M., Amberg, A. C., Roosendaal, E., van der Horst, G., van der Veen, R., van
Ommen, G. -J. B., and Grivell, L. A., Variation, transcription and circular RNAs of the mitochon-
drial gene for subunit I of cytochrome c oxidase, J. Mol. BioI., 164, 35, 1983.
84. Kuck, U., personal communication, 1984.
85. Bontiz, S. G., Coruzzi, G., Thalenfeld, B. E., and Tzagoloff, A., Assembly of the mitochondrial
membrane system. Structure and nucleotide sequence of the gene coding for subunit I of yeast cyto-
chrome oxidase, J. BioI. Chern., 255, 11927, 1980.
86. Lang, B. F., The mitochondrial genome of the fission yeast Schizosaccharornyces porn be: highly
homologous introns are inserted at the same position of the otherwise less conserved coxl genes in
Schizosaccharomyces porn be and Aspergillus nidulans, EMBO J., 3,2129, 1984.
87. Waring, R. B., Davies, R. W., Lee, S., Grisi, E., McPhail Berks, M., and Scazzocchio, C., The
mosaic organization of the apocytochrome b gene of Aspergillus nidulans revealed by DNA sequenc-
ing, Cell,27, 4,1981.
88. Nobrega, F. G. and Thalenfeld, A., Assembly of the mitochondrial membrane system. DNA se-
quence and organization of the cytochrome b gene in Saccharomyces cerevisiae 0273-1013, J. BioI.
Chern., 255,9828, 1980.
89. Helmer Citterich, M., Morelli, G., Nelson, M. A., and Macino, G., Expression of split genes in the
Neurospora crassa mitochondrial genome, in Mitochondria 1983, Schweyen, R. J., Wolf, K., and
Kaudewitz, F., Eds., de Gruyter, New York, 1983, 357.
90. Lazarus, C. M., Lunsdorf, H., Hahn, U., Stepien, P. P., and Kuntzel, H., Physical map of Asper-
gillus nidulans mitochondrial genes coding for ribosomal RNA: intervening sequence in the large
rRNA cistron, Mol. Gen. Genet., 177, 389,1980.
91. Hahn, U., Lazarus, C. M., Lunsdorf, H., and Kuntzel, H., Split gene for mitochondrial 24 S rit-o-
somal RNA of Neurospora crassa, Cell, 17, 191, 1979.
92. Wright, R. W. and Cummings, D. 1., Mitochondrial DNA from Podospora anserina. IV. The large
ribosomal RNA gene contains two long intervening sequences, Curro Genet., 7,151,1983.
276 Fungal Virology

93. Bos, J. L., Heyting, L., Borst, P., Amberg, A. C., and Van Bruggen, E. G. J., An insert in the single
gene for the large ribosomal RNA in yeast mitochondrial DNA, Nature (London), 275,336, 1978.
94. Crews, S., Ojala, D., Posakony, 1., Nishigucki, J., and Attardi, G., Nucleotide sequence of a region
of human mitochondrial DNA containing the precisely identified origin of replication, Nature (Lon-
don},277, 192, 1979.
95. Martens, P. A. and Clayton, D. A., Mechanism of mitochondrial replication in mouse L-cells: local-
ization and sequence of the light-strand origin of replication, 1. Mol. BioI., 135, 327, 1979.
96. Clayton, D. A., Replication of animal DNA, Cell, 28,693, 1982.
97. de Zamaroczy, M., Marotta, R., Faugeron-Fonty, G., Goursot, R., Mangin, M., Baldacci, G., and
Bernardi, G., The origins of replication of the yeast mitochondrial genome and the phenotype of
suppressivity, Nature, 292, 75, 1981.
98. Bernardi, G., Baldacci, G., Colin, Y., Faugeron-Fonty, G., Gourset, R., Huygard, A., Le van Kim,
C., Mangin, M., Marotta, R., and de Zamaroczy, M., The origins of replication on the mitochondrial
genome in yeast, in Manipulation and Expression of Genes in Eucaryotes, Nagley, P., Linnane, A.
W., Peacock, W. J., and Pateman, 1. A., Eds., Academic Press, Australia, 1983,279.
99. Tikhomirova, L. P., Kryukov, V. M., Strizkov, N. 1., and Bayev, A. A., mtDNA sequences of
Candida utilis capable of supporting autonomous replication of plasmids in Saccharomyces cere vis-
iae, Mol. Gen. Genet., 189, 479, 1983.
100. Kuntzel, H. and Schafer, K. P., Mitochondrial RNA polymerases from Neurospora crassa, Nature
(New BioI.) (London), 231,265, 1971.
101. Dawid, J. B. and Wu, G. J., Transcription of mtDNA by mitochondrial RNA polymerase from
Xenopus laevis, in The Biogenesis of Mitochondria, Kroon, A. M. and Saccone, C., Eds., Academic
Press, New York, 1979,79.
102. Gallerani, R. and Saccone, C., The DNA dependent RNA polymerase from rat liver mitochondria,
in The Biogenesis of Mitochondria, Kroon, A. M. and Saccone, C., Eds., Academic Press, 1974,59.
103. Levens, D., Lustig, A., and Rabinowitz, M., Purification of mitochondrial RNA polymerase from
Saccharomyces cerevisiae, 1. BioI. Chern., 256, 1474, 1981.
104. Levens, D., Morimoto, R., and Rabinowitz, M., Mitochondrial transcription complex from Saccha-
romyces cerevisiae, 1. BioI. Chern., 256, 1466, 1981.
105. Levens, D., Christianson, T., Edwards, J., Lustig, A., Ticho, B., Locker, J., and Rabinowitz, M.,
Transcription initiation of yeast mltochondrial RNA polymerase, in Mitochondrial Genes, Slonimski,
P. P., Borst, P., and Attardi, G., Eds., Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.,
1985,295.
106. Osinga, K. A., De Haan, M., Christianson, T., and Tabak, H. F., A nonanucleotide sequence in-
volved in promotion of ribosomal RNA synthesis and RNA priming of DNA replication in yeast
mitochondria, Nucleic Acids Res., 10,7993, 1982.
107. Christianson, T. and Rabinowitz, M., Identification of multiple transcription initiation sites on the
yeast mitochondrial genome by in vitro capping with guanylyltransferase, 1. BioI. Chern., 258, 14025,
1983.
108. Edwards, J. C., Osinga, K. A., Christianson, T., Hensgens, L. A. M., Janssens, P. M., Rabinowitz,
M., and Tabak, H. F., Initiation of transcription of the yeast mitochondrial gene coding for ATPase
subunit 9, Nucleic Acids Res., II, 8269, 1983.
109. Borst, P. and Grivell, L. A., One gene's intron is another gene's exon, Nature, 289,439, 1981.
110. Sionimski, P. P., Claisse, M. L., Foucher, M., Jacq, C., Kachko, A., Lamouroux, A., Pajot, P.,
Perrodin, G., Spyridakis, A., and Wambier-Kluppel, M. L., Mosaic organization and expression of
the mitochondrial DNA region controlling cytochrome c reductase and oxidase. III. A model of
structure and function, in Biochemistry and Genetics of Yeasts, Bacilla, B., Horecker, B. L., and
Stoppani, A. O. M., Eds., Academic Press, New York, 1978,391.
111. Claisse, M. L., Spyridakis, A., Wambier-Kluppel, M. L., Pajot, P., and Sionimski, P. P., Mosaic
organization and expression of the mitochondrial DNA region controlling cytochrome c reductase
and oxidase. II. Analysis of proteins translated from the box region, in Biochemistry and Genetics of
Yeasts, Bacilla, B., Horecker, B. L., and Stoppani, A. O. M., Eds., Academic Press, New York,
1978, 369.
112. Church, G. M., Sionimski, P. P., and Gilbert, W., Pleiotropic mutations within two yeast mitochon-
drial cytochrome genes block mRNA processmg, Cell, 18, 1209, 1979.
1l3. Dhawale, S., Hanson, D. K., Alexander, N. J., Perlman, P. S., and Mahler, H. R., Regulatory
interactions between mitochondrial genes: interactions between two mosaic genes, Proc. Natl. Acad.
Sci. U.S.A.,78, 1778, 1981.
114. Dujardin, G., Jacq, C., and Slonimski, P. P., Single base substitution in an intron of oxidase com-
pensates splicing defects of the cytochrome b gene, Nature (London), 298, 628, 1982.
liS. Osiewacz, H. D. and Esser, K., DNA sequence analysis of the mitochondrial plasmid of Podospora
anserina, Curro Genet., 7,219, 1983.
 277

116. Lamb, M. R., Anziano, P. Q., Glauss, K. P., Hanson, D. K., Klapper, H. 1., Perlman, P. S., and
Mahler, H. R., Functional domains in introns, f. BioI. chem., 258,1991,1983.
117. Davies, R. W., Waring, R. B , Ray, 1. A., Brown, T. A., and Scazzocchio, C., Making ends meet: a
model for RNA splicing in fungal mitochondria, Nature (London), 300,719,1982.
118. Weiss-Brummer, B., Holl, 1., Schweyen, R. 1., Rodel, G., and Kaudewitz, F., Processing of yeast
mitochondrial RNA: involvement of intramolecular hybrids in splicing of cob IJ1tron 4 RNA by mu-
tation and reversion, Cell, 10, 177, 1983.
119. Michel, F., Jacquier, A., and Dujon, B., Comparison of fungal mitochondrial introns reveals exten-
sive homologies in RNA secondary structure, Biochemie, 64, 867, 1982.
120. Michel, F. and Dujon, B., Conservation of RNA secondary structures in two intron families including
mitochondrial-, chloroplast- and nuclear encoded members, EMBO f., 2,33, 1983.
121. Waring, R. B., Scazzocchio, C., Brown, T. A., and Davies, R. W., Close relationship between certain
nuclear and mitochondrial introns, f. Mol. BioI., 167, 595, 1983.
122. Amberg, A. C., Van Ommen, G. 1. B., Grivell, L. A., Van Bruggen, E. F. 1., and Borst, P., Some
yeast mitochondrial RNAs are circular, Cell, 19, 313, 1980.
123. Grivell, L. A., Mitochondrial gene expression 1983, in Mitochondria 1983, Schweyen, R. 1., Wolf,
K., and Kaudewitz, F., Eds., de Gruyter, Berlin, 25, 1983.
124. Green, M. R., Grimm, M. F., Goewert, R. R., Collins, R. A., Cole, M. D., Lambowitz, A. M.,
Heckmann, J. B., Yin, S., and RajBhandary, U. L., Transcripts and processing patterns for the
ribosomal RNA transfer RNA region of Neurospora crassa mitochondrial DNA, f. BioI. Chem.,
256,2027, 1981.
125. Yin, S., Heckmann, J. and Raj Bhandary, U. L., Highly conserved GT-rich palindromic DNA se-
quences flank tRNA genes in Neurospora crassa mitochondria, Cell, 26, 325, 1981.
126. Ojala, D., Montoya, 1. and Attardi, G., tRNA punctuation model of RNA processing in human
mitochondria, Nature (London), 290,470, 1981.
127. Piechulla, B. and Kuntzel, H., Mitochondrial polypeptide elongation factor EF-TU, Eur. f.
Biochem., 132, 235, 1983.
128. Fox, T. D., Five TGA 'stop' condons occur within the translated sequence of the yeast mitochondrial
gene for cytochrome c oxidase subunit II, Proc. Natl. Acad. Sci. U.S.A., 76,6534, 1979.
129. Barrell, B. G., Bankier, A. T., Drouin, J., A different genetic code in human mitochondria, Nature
(London), 282,613,1979.
130. Steffens, G. 1. and Buse, G., Studies on cytochrome c oxidase. IV. Primary structure and function
of subunit II, Hoppe-Seyler's Z. Physiol. Chem., 360,613,1979.
131. Heckman, J., Sarnoff, 1., Alzner-De Weerd, B., Yin, S., and RajBhandary, U. L., Novel featUles in
the genetic code and coding reading patterns in Neurospora crassa mitochondria based on sequences
of six mitochondrial tRNAs, Proc. Nat!. Acad. Sci. U.S.A., 77,3159, 1980.
132. Martin, R. P., Sibler, A. P., Bordonne, R., Canaday, 1., and Dirheimer, G., Nucleotide sequence
and gene localization of yeast mitochondrial initiator tRNA and UGA - decoding tRNA, in The
Organization and Expression of the Mitochondrial Genome, Kroon, A. M. and Saccone, C., Eds.,
Elsevier/North-Holland, Amsterdam, 1980,311.
133. Bonitz, S. G., Berlani, R., Coruzzi, G., Li, M., Macino, G., Nobrega, F. G., Nobrega, M. P.,
Thalenfeld, B. E., and Tzagoloff, A., Codon recognition rules in yeast mitochondria, Proc. Natl.
Acad. Sci. U.S.A., 77,3167, 1980.
134. Crick, F., Codon-anticodon pairing: the wobble hypothesis, f. Mol. BioI., 19, 548,1966.
135. Roe, B. A., Wong, J. H. F., Chem, E. Y., Armstrong, P. W., Stankiewicz, A., Ma, D. -P., and
McDonough, 1., Mammalian mitochondrial tRNAs: a modified nucleotide 3' to the anticodon may
modulate their codon response, in Mitochondrial Genes, Sionimski, P. P., Borst, P., and Attardi,
G., Eds., Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y., 1982,45.
136. Kleckner, N., Transposable elements in prokaryotes, Annu. Rev. Genet., 15, 341, 1981.
137. Shapiro, 1. A" Mobile genetic elements, Academic Press, 1983.
138. Borner, T., Intercompartmental gene transfer - a review, BioI. Zbl., 103, 485, 1984.
139. Ellis, 1., Promiscuous DNA-chloroplast genes inside plant mitochondria, Nature (London), 299,296,
1982.
140. Farrelly, F. and Butow, R. A., Rearranged mitochondrial genes in the yeast nuclear genome, Nature
(London),301, 296,1983.
141. Orian, J. M" Murphy, M., and Marzuki, S., Mitochondrially synthesized protein subunits of the
yeast mitochondrial adenosine triphosphatase, Biochem. Biophys. Acta, 652,234, 1981.
142. Jackl, G. and Sebald, W., Identification of two products of mitochondrial protein synthesis associ-
ated with mitochondrial adenosine triphosphatase fom Neurospora crassa, Eur. f. Biochem., 54, 97,
1975.
278 Fungal Virology

143. Agsteribbe, E., Samallo, J., De Vries, H., Hensgens, A. M., and Grivell, L. A., Sequence homologies
between the mitochondrial DNAs of yeast and Neurospora crassa, in The Organization and Express-
sian of the Mitochondrial Genome, Kroon, A. M., Saccone, C., Eds., Elsevier/North-Holland, Am-
sterdam, 1980, SI.
144. Boogaart van den, P., Samallo, 1., and Agsteribbe, E., Similar genes for a mitochondrial ATPase
subunit in the nuclear and mitochondrial genomes of Neurospora crassa, Nature, 298, 187, 1982.
145. Hensgens, L. A. M., Grivell, L. A., Borst, P., and Bos, J. L., Nucleotide sequence of the mitochon-
drial structural gene for subunit 9 of yeast ATPase complex, Proc. Nat!. Acad. Sci. U.S.A.,76, 1663,
1979.
146. Sebald, W., Hoppe, J., and Wachter, E., in Function and Molecular Aspects of Biomembrane Trans-
port, Quagliariello, E., Ed., Elsevier, Amsterdam, 1979,63.
147. Gray, M. W., The bacterial ancestry of plastids and mitochondria, BioScience, 33,693, 1983.
148. Wright, R. M. and Cummings, D. J., Integration of mitochondrial gene sequences within the nuclear
genome during senescence in a fungus, Nature (London), 302,86, 1983.
149. Vierny, C., Keller, A. M., Begel, 0., and BeIcour, L., A sequence of mitochondrial DNA is associ-
ated with the onset of senescence in a fungus, Nature, 297, IS7, 1982.
150. Kuck, U., Osiewacz, H. D., Schmidt, U., Kappelhoff, B., Schulte, E., Stahl, U., and Esser, K., The
onset of senescence is affected by DNA rearrangements of a discontinous mitochondrIal gene, Curro
Genet., 9,373, 1985.
151. Holloway, B. W., Plasmids that mobilize bacterial chromosomes, Plasmid,2, I, 1979.
152. Szabo, L. 1., Characterization of eight excision plasm ids of Pseudomonas syringaepv. phaseolica,
Mol. Gen. Genet., 19S, 90,1984.
153. Dujon, B., Colson, A. M., and Slonimski, P. P., The mitochondrial genetic map of Saccharomyces
cerevisiae: a literature compilation towards a unique map, in Mitochondria 1977, Genetics and Bio-
genesis of Mitochondria, Bandlow, W., Schweyen, R. J., Wolf, K., and Kaudewitz, F., Eds., de
Gruyter, Berlin, 1977,579.
IS4 Locker, J., Lewin, A., and Rabinowitz, M., The structure and organizatIOn of mitochondrial DNA
from petite yeast, Plasmid, 2, 155, 1979.
ISS. Gilham, N. W., Genetic analysis of the chloroplast and mitochondrial genomes, Annu. Rev. Genet.,
8,347, 1974.
156. Wilkie, B., Genetic and functional aspects of yeast mitochondria, in Yeast Genetics, Fundamental
and Applied aspects, Spencer, 1. F. T., Spencer, D. M., and Smith, A. R., W., Eds., Springer, New
York, 1983, 25S.
IS7. Van Kreijl, C. F., and Bos, J. L., The repeating nucleotide sequence in the repetitive mitochondrial
DNA from a "low density" petite mutant of yeast, Nucleic Acids Res., 4,2369, 1977.
IS8. Lazowska, J. and Slonimski, P. P., Electron microscopy analysis of circular repetitive mitochondrial
DNA molecules from genetically characterized rho- mutants of Saccharomyces cerevisiae, Mol. Gen.
Genet., 146, 61, 1976.
IS9. Blanc, H. and Dujon, B., Replicator regions of the yeast mitochondrial DNA responsible for sup-
pressiveness, Proc. Nat!. Acad. Sci. U.S.A., 77,3942, 1980.
160. Fukuhara, H. and Wesolowski, M., Preferential loss of a specific region of mitochondrial DNA by
rho-mutation, in Mitochondria 1977, Genetics and Biogenesis of Mitochondria, Bandlow, W.,
Schweyen, R. J., Wolf, K., and Kaukewitz, F., Eds., de Gruyter, Berlin, 1977, 123.
161. Mathews, S., Schweyen, R. J., and Kaudewitz, F., Preferential loss or retention of mitochondrial
genes in rho- clones, in Mitochondria 1977, Genetics and Biogenesis of Mitochondria, Bandlow, W.,
Schweyen, R. 1., Wolf, K., and Kaudewitz, F., Eds., de Gruyter, Berlin, 1977, 133.
162. Lazowska, J. and Slonimski, P. P., Site-specific recombination in "petite colony" mutants of Sac-
charomyces cerevisiae. I. Electron microscopic analysis of the organization of recombinant DNA
resulting from end to end joining of two mitochondrial segments, Mol. Gen. Genet., 156, 163, 1977.
163. Bernardi, G., Ba1dacci, G., Bernardi, G., Faugeron-Fonty, G., Gaillard, C., Goursot, R., Huyard,
A., Mangin, M., Marotta, R., and de Zamaroczy, M., The petite mutation: excision sequences,
replication origins and suppressivity, in The Organization and Expression of the Mitochondrial Gen-
ome, Kroon, A. M. and Saccone, C., Eds., Elseiver/North-Holland, 1980,21.
164. Piperno, G., Fonty, G., and Bernardi, G., The mitochondrial genome of wild type yeast cells. II.
Investigations on the compositional heterogeneity of mitochondrial DNA, 1. Mol. BioI., 6S, 191,
1972.
16S. Prunell, A. and Bernardi, G., The mitochondrial genome of wild-type yeast cells. IV. Genes and
spacers, 1. Mol. BioI., 86,825, 1974.
166. Locker, 1., Rabinowitz, M., and Getz, G. S., Tandem inverted repeats in mitochondrial DNA of
petite mutants of Saccharomyces cerevisiae, Proc. Natl. Acad. Sci. U.S.A., 71, 1366, 1974.
167. Locker, J., Rabinowitz, M., and Getz, G. S., Electron microscopic and renaturation kinetic analysis
of mitochondrial DNA of cytoplasmic petite mutants of Saccharomyces cerevisiae, 1. Mol. BioI., 88,
489, 1974.
 279

168. Clark-Walker, G. D. and Miklos, G. L. D., Mitochondrial genetics, circular DNA and the mecha-
msms of the petite mutation in yeast, Genet. Res., 24,43, 1974.
169. Prunell, A. and Bernardi, G., The mitochondrial genome of wild-type yeast cells. VI. Genome orga-
nization, f. Mol. BioI., 110,53, 1977.
170. Gaillard, C., Strauss, F., and Bernardi, G., Excision sequences in the mitochondrial genome of yeast,
Nature (London), 283,218, 1980.
171. Faugeron-Fonty, G., Culard, F., Baldacci, G., Goursot, R., Prunell, A., and Bernardi, G., The
mitochondrial genome of wild-type yeast cells. VIII. The spontaneous cytoplasmic "petIte" muta-
tion, f. Mol. BioI., 134, 493,1979.
172. Mitchell, M. B. and Mitchell, H. K., A case of "maternal" inheritance in Neurospora crassa, Proc.
Natl. Acad. Sci. U.S.A., 38, 442, 1952.
173. Bertrand, H. and Pittenger, T. H., Isolation and classification of extranuclear mutants of Neurospora
crassa, Genetics, 71, 521, 1972.
174. Akins, R. A. and Lambowitz, A. M., The poky mutant of Neurospora contains a 4-base-pair deletion
at the 5' end of the mitochondrial small rRNA, Proc. Natl. Acad. Sci. U.S.A., 81,3791,1984.
175. Collins, R. A. and Lambowitz, A. M., Characterization of a variant Neurospora crassa mitochon-
drial DNA which contains tandem reiteratIOns of a 1.9 kb sequence, Curro Genet., 4, 131, 1981.
176. Mannella, C. A., Goewert, R. R., and Lambowitz, A. M., Characterization of variant Neurospora
crassa mitochondrial DNAs which contain tandem reiterations, Cell, 18, 1197, 1979.
177. De Vries, H., De Jonge, J. C., and Van't Sant, P., Characterization of a mitochondrial "stopper"
mutant of Neurospora crassa, Deletions and rearrangements in the mitochondrial DNA result in
disturbed assembly of respiratory chain components, in The Organization and Expression of the
Mitochondrial Genome, Kroon, A. M. and Saccone, C., Eds., Elsevier/North-Holland, 1980,333.
178. Gross, S. R., Hsieh, T., and Levine, P. H., Intramolecular recombination as a source of mitochon-
drial chromosome heteromorphism in Neurospora, Cell, 38,233, 1984.
179. De Vries, H., De Jonge, J. C., van't Sant, P., Agsteribbe, E" and Amberg, A., "Stopper" mutant
of Neurospora crassa containing two populations of aberrant mitochondrial DNA, Curr. Genet., 3,
205, 1981.
180. De Vries, H. and De Jonge, J. C., Deletion of a protein gene on mitochondrial DNA causes the
stopper phenotype in Neurospora crassa, In EMBO Workshop, Gene Expression in Filamentous
Fungi, Rhenen, Netherlands, April 1984, 51.
181. Jinks, J. L., Lethal suppressive cytoplasms in aged clones of Aspergillus glaucus, f. Gen. Microbiol.,
21,397,1959.
182. Caten, C. E., Vegetative incompatibility and cytoplasmic infection in fungi, f. Gen. Microbiol., 72,
221,1972.
183. Lazarus, C. M., Earl, A. J., Turner, G., and Kuntzel, H., Amplification of a mitochondrial DNA
sequence in the cytoplasmically inherited 'ragged' mutant of Aspergillus amstelodami, Eur. f.
Biochem., 106, 633, 1980.
184. Lazarus, C. M. and Kuntzel, H., Amplification of a common mitochondrial DNA sequence in three
new ragged mutants of Aspergillus amstelodami, in The Organization and Expression of the Mito-
chondrial Genome, Kroon, A. M. and Saccone, C., Eds., Elsevier/North-Holland, 1980, 87.
185. Rizet, G., Sur I'impossibilite d'obtenir la multiplication vegetative interrompue et iIlimitee de I'asco-
mycete Podospora anserina, C.R. Acad. Sci. Paris, 237,838, 1953.
186. Esser, K. and Tudzynski, P., Senescence in fungi, in Senescence in Plants, Thimann, K. V., Ed.,
CRC Press, Boca Raton, Florida, 67-83, 1980.
187. Esser, K. and Tudzynski, P., Genetic control and expression of senescence in Podosproa anserina, in
Viruses and Plasm ids in Fungi, Lemke, P. A., Ed., Marcel Dekker, New York, 1979,595.
188. Cummings, D. J., Belcour, L., and Grandchamp, C., Mitochondrial DNA from Podospora anserina.
II. Properties of mutant DNA and multimeric circular DNA from senescent cultures, Mol. Gen.
Genet., 171, 239,1979.
189. Smith, J. R. and Rubenstein, I., CytoplasmIc inheritance of the timing of 'senescence' in Podospora
anserina, f. Gen. Microbiol., 76,297, 1973.
190. Esser, K. and Keller, W" Genes inhibiting senescence in the ascomycete Podospora anserina, Mol.
Gen. Genet., 144,107,1976.
191. Tudzynski, P. and Esser, K., Chromosomal and extrachromosomal control of senescence in the as-
comycete Podospora anserina, Mol. Gen. Genet., 173, 71, 1979.
192. Belcour, L. and Begel, 0., Life-span and senescence in Podospora anserina:effect of mitochondrial
genes and functions, f. Gen. Microbiol., 119, 505,1980.
193. Marcou, D., Notion de longevitee et nature cytoplasmique du determinant de la senescence chez
quelques champignons, Ann. Sci. Nat, BioI. Bot. Veg., 12, 653, 1961.
194. Cummings, D. J., Belcour, L., et Grandchamp, C., Etude au microscope electronique du DNA mi-
tochondrial de Podospora anserina et presence d'une serie multimerique de molecules circulaires de
DNA dans des cultures senescentes, C.R. Acad. Sc. Paris, 287, 157, 1978.
280 Fungal Virology

195. Vierny, C., Begel, 0., Keller, A. -M., Raynal, A., and BeIcour, L., Senescence specific DNA of
Podospora anserina' its variability and its relation with mitochondrial DNA, m The Organization
and Expression of the Mitochondnal Genome, Kroon, A. M. and Saccone, c., Eds., Elsevier/North-
Holland 1980, 91.
196. Cummings, D. J., Laping, 1. L., and Nolan, P. E., Cloning of senescent mitochondrial DNA from
Podospora anserina: a beginning, in The OrganIzation and Expression of the Mitochondrial Genome,
Kroon, A. M. and Saccone, C., Elsevier/North-Holland, Amsterdam, 1980,97.
197. Stahl, U., Kuck, U., Tudzynski, P., and Esser, K., Characterization and cloning of plasmid like DNA
of the ascomycete Podospora anserma, Mol. Gen. Genet., 178, 639, 1980.
198. Tudzynski, P., Stahl, U., and Esser, K., Transformation to senescence with plasmid like DNA in the
ascomycete Podospora anserina, Curro Genet., 2, 181, 1980.
199. BeIcour, L., Begel, 0., Mosse, M. -0., and Vierny, C., Mitochondrial DNA amplification in senes-
cent cultures of Podospora anserina: variability between the retained, amplified sequences, Curro
Genet., 3, 13, 1981.
200. Kuck, U. and Esser, K., Genetic map of mitochondrial DNA in Podospora anserina, Curro Genet.,
5, 143, 1982.
20 I. Lazdins, I. B. and Cummings, D. J., Autonomously replicating sequences in young and senescent
mitochondrial DNA from Podospora anserina, Curro Genet., 6,173,1982.
202. Tudzynski, P., Stahl, U., and Esser, K., Development of a eukaryotic cloning system in Podospora
anserina. I. Long-lived mutants as potential recipients, Curro Genet., 6,219, 1982.
203. Sainsard-Chanet, A., Vierny-Jamet, C., and Guerineau, M., The senescence-specific DNAs from
Podospora anserina carry replication origins which are functional in yeast, BioI. Cell, 45, 225, 1982.
204. Wright, R. M. and Cummings, D. J., Transcription of a mitochondrial plasmid during senescence in
Podospora anserina, Curro Genet., 7, 457, 1983.
205. Cummings, D. 1. and Wright, R. M., DNA sequence of the excision sites of a mitochondrial plasmid
from senescent Podospora anserina, Nucleic Acids Res., 2111, 1983.
206. Esser, K., Osiewacz, H. D., Stahl, U., and Tudzynski, P., Nuclear-mitochondrial interactions cause
senescence in the filamentous fungus Podospora anserina , in Mitochondria 1983 - Nucleo-Mito-
chondrial Interaction, Schweyen, R., Wolf, K., and Kandewitz, Eds., de Gruyter, Berlin, 1983,251.
207. Koll, F., Begel, 0., Keller, A. -M., Vierny, C., and Belcour, L., Ethidium bromide rejuvenation of
senescent cultures of Podospora anserina: Loss of senescence-specific DNA and recovery of normal
mitochondrial DNA, Curro Genet., 8, 127, 1984.
208. Osiewacz, H. D. and Esser, K., The Podospora plasmid: A selfreplicating mobile intron, EMBO
Workshop, Gene Expression in Filamentous Fungi, Rhenen, Netherlands, 1984,39.
209. Belcour, L., Koll, F., Sainsard, A., Begel, 0., and Vierny, C., Mitochondrial DNA rearrangements
in senescent cultures and in some mutants escaped from senescence in Podospora anserina, EMBO
Workshop, Gene Expression in Filamentous Fungi, Rhenen, Netherlands, 1984, II.
210. Jamet-Vierny, C., Begel, 0., and Belcour, L., Senescence in Podospora anserina:amplification of a
mitochondrial DNA sequence, Cell, 21,189,1980.
211. Vierny, C., Begel, 0., and Belcour, L., Localization of the genes col and cob on the mitochondrial
DNA of Podospora, EMBO Workshop, Gene Expression in Filamentous Fungi, Rehnen, Nether-
lands, 1984,50.
212. Koll, F., Begel, 0., Belcour, L., and Vierny, C., Study of a region of mitochondrial genome involved
in senescence process in Podospora anserina, EMBO Workshop, Gene Expression in filamentous
Fungi, Rehnen, Netherlands, 1984,32.
213. Stahl, U., Tudzynski, P., Kuck, U., and Esser, K., Replication and expression of a bacterial-mito-
chondrial hybrid plasmid in the fungus Podospora anserina, Proc. Natl. Acad. Sci. U.S.A., 79,3641,
1982.
214. Belcour, L., Begel, 0., Keller, A. -M., and Vierny, C., Does senescence in Podospora anserinaresult
from instability of the mitochondrial genome?, in Mitochondrial Genes, Sionimski, P., Borst, P., and
Attardi, G., Eds., Cold Spring Harbor Laboratory, Cold Spring Harbor, New York, 1982,415.
215. Grivell, L. A., Hensgens, L. A. M., Osinga, K. A., Tabak, H. F., Boer, P. H., Crusius, 1. B. A.,
Van der Laan, J. C., De Haan, M., Van der Horst, G., Evers, R. F., and Arnberg, A. C., RNA
processing in yeast mitochondria, in Mitochondrial Genes, Sionimski, P., Borst, P., and Attardi, G.,
Eds., Cold Spring Harbor Laboratory, Cold Spring Harbor, N. Y., 1982, 225.
216. Gargouri, A., Lazowska, 1., and Sionimski, P. P., DNA-splicing of introns in the gene: a general
way of reverting intron mutations, in Mitochondria 1983, Nucleo-mitochondrial interactions,
Schweyen, R. J., Wolf, K., and Kaudewitz, F., Eds., De Gruyter, Berlin, 1983,259.
217. Bockelmann, B., unpublished, 1984.
218. Garber, R. C., Turgeon, B. G. and Yoder, O. C., A mitochondrial plasmid from the plant pathogenic
fungus Cochliobolus heterostrophus, Mol. Gen. Genet., 28, I, 1984.
219. Rieck, A., Griffiths, A. J. F., and Bertrand, H., Mitochondrial variants of Neurospora intermedia
from nature, Can. 1. Genet Cytol., 24,741, 1982.
 281

220. Griffiths, A. 1. F. and Bertrand, H., Unstable cytoplasms in Hawaiian strains of Neurospora inter-
media, Curro Genet., 8,387, 1984.
221. Griffiths, A. 1. F., Mitochondrially-based senescence in Hawaiian isolates of Neurospora intermedia,
EMBO Workshop, Gene Expression in Filamentous Fungi, Rhenen, Netherlands, 1984, 24.
222. Nargang, F. E., Bell, 1. B., Stohl, L. L., and Lambowitz, A. M., A family of repetitive palindromic
sequences found in Neurospora mitochondrial DNA is also found in a mitochondrial plasmid DNA,
J. Bioi. Chern., 258,4252, 1983.
223. Nargang, F. E., Bell, 1. B., Stohl, L. L., and Lambowitz, A. M., The DNA sequence and genetic
organization of a Neurospora mitochondrial plasmid suggests a relationship to introns and mobile
elements, Cell, 38,441, 1984.
224. Nargang, F. E., 12th Neurospora Information Conference, Neurospora Newsl., 31(Abstr.), 12,
1984.
225. Duvell, A., pers. communication, 1984.
226. Howell, S. H., Plant molecular vehicles: potential vectors for introducing foreign DNA into plants,
Annu. Rev. Plant. Physiol., 33,609, 1982.
227. Gubbins, E. 1., Newlon, C. S., Kann, M., and Donelson, 1. E., Sequence organization and expression
of a yeast plasmid DNA, Gene, 1,185,1977.
228. Clark-Walker, G. D. and Miklos, G. L. G., Localization and quantification of circular DNA in yeast,
Eur. J. Biochem., 41, 395,1974.
229. Cameron, 1. R., Philippsen, P. and Davis, R. W., Analysis of chromosomal integration and deletions
of yeast plasmids, Nucleic Acids Res., 4, 1429, 1977.
230. Guerineau, M., Grandchamp, C., Paoletti, C., and Slonimski, P. P., Characterization of a new class
of circular DNA molecules in yeast, Biochem. Biophys. Res. Commun., 42, 550, 1971.
231. Hollenberg, C. P., Degelmann, A., Kustermann-Kuhn, B., and Royer H. D., Characterization of 2
,..m DNA of Saccharomyces cerevisiae by restriction fragment analysis and integration in an Esche-
richia coli plasmid, Proc. Nat!. Acac. Sci. U.S.A., 73,2072, 1976.
232. Livingston, D. M. and Klein, H. L., Deoxyribonucleic acid sequence organization of a yeast plasmid,
J. Bacteriol., 129, 472, 1977.
233. Sinclair, 1. H., Stevens, B. I., Sanhavi, P., and Rabinowitz, M., Mitochondrial-satellIte and cIrcular
DNA filaments in yeast, Science, 156, 1234, 1967.
234. Livingston, D. M. and Hahne, S., Isolation of condensed intracellular form of the 2,..m DNA plasmid
of Saccharomyces cerevisiae, Proc. Nat!. Acad. Sci. U.S.A., 76,3727, 1979.
235. Nelson, R. G. and Fangman, W. C., Nucleosome organization of the yeast 2 ,..m DNA plasmid: a
eucaryotic minichromosome, Proc. Nat!. Acad. Sci. U.S.A., 76,6515, 1979.
236. Guerineau, M., Grandchamp, C., and Slonimski, P. P., Circular DNA of a yeast episome with two
inverted repeats: structural analysis by a restriction enzyme and electron microscopy, Proc. Nat!.
Acad. Sci. U.S.A., 73,3030, 1976.
237. Hartley, 1. C. and Donelson, 1. E., Nucleotide sequence of the yeast plasmid, Nature (London), 286,
860, 1980.
238. Broach, 1. R. and Hicks, 1. B., Replication and recombination functions associated with the yeast
plasmid 2,..m circle, Cell, 21,501, 1980.
239. Vetter, D., and Andrews, B. 1., Roberts-Beatty, L., and Sadowski, P. D., Site-specific recombination
of yeast 2,..m DNA in vitro, Proc. Natl. Acad. Sci. U.S.A., 80,7284, 1983.
240. Kikuchi, Y., Yeast plasmid requires a cis-acting locus and two plasmid proteins for its stable main-
tenance, Cell, 35,487,1983.
241. layaram, M., Li, Y. Y., and Broach, J. R., The yeast plasmid 2 ,..m circle encodes components
required for its high copy propagation, Cell, 34, 95, 1983.
242. Varshavsky, A. J., Sundin, 0., and Bohn, M., A stretch of late SV 40 viral DNA about 400 bp long
which includes the origin of replication is specifically exposed in SV 40 minichromosomes, Cell, 16,
453, 1979.
243. DeVilliers, J., Schaffner, W., Tyndall, C., Lupton, S., and Kamen, R., Polyoma virus DNA repli-
cation requires an enhancer, Nature (London), 312,242, 1984.
244. Petes, T. D., Williamson, D. H., Replicating circular DNA molecules in yeast, Cell, 4,249, 1975.
245. Livingston, D. M. and Kupfer, D. M., Control of Saccharomyces cerevisiae 2 ,..m DNA replication
by cell division cycle genes that control nuclear DNA replication, J. Mol. BioI., 116,249, 1977.
246. Zakian, V. A., Brewer, B. J., and Fangman, W. L., Replication of each copy of the yeast 2 micron
DNA plasmid occurs during the S phase, Cell, 17, 923, 1979.
247. Sigurdson, D. C., Gaarder, M. E., and Livingston, D. M., Characterization of the transmission
during cytoductant formation of the 2 micron DNA plasmid from Saccharomyces, Mol. Gen. Genet.,
183,59, 1981.
248. Broach, J. R., The yeast plasmid 2,..m circle, Cell, 28, 203, 1982.
249. Jayaram, M. and Broach, 1. R., Yeast plasmid 2,..m circle promotes recombination within bacterial
transposon Tn5, Proc. Natl. Acad. Sci. U.S.A., 80,7264, 1983.
282 Fungal Virology

250. Broach, J. R., The yeast plasmId 2 !lm Circle, in The Molecular Biology of the Yeast Saccharomyces,
Life Cycle and Inheritance, Strathern, J. N., Jones, E. W., Broach, J. R., Eds., Cold Spring Harbor
Laboratory, Cold Spring Harbor, N. Y., 445, 1981.
25 I. Zeig, J., Silverman, M., Hilmen, M., and Simon, M. I., Recombinational switch for gene expression,
Science, 196, 170, 1977.
252. Bukhari, A. and Ambrosio, L., The invertIble segment of bacteriophage Mu DNA determines the
adsorption properties of phage Mu particles, Nature (London), 271,575,1978.
253. Guerineau, M., Slonimski, P. P., and Avner, P. R., Yeast episome: oligomycin resistance associated
with a small covalently closed non-mitochondrial circular DNA, Biochem. Biophys. Res. Commun.,
61,462,1974.
254. Holm, C., Clonal lethality caused by the yeast plasmid 2!lm DNA, CeIl, 29,585,1982.
255. Sor, F., Wesolowski, M. and Fukuhara, H., Inverted terminal repetitions of the two linear DNA
associated with the killer character of yeast, Nucleic Acids Res., 11, 5037,1983.
256. Meinhardt, F., personal communication, 1984.
257. Tudzynski, P., personal communication, 1983.
258. Hashiba, T., Homma, Y., Hyakumachi, M. and Matsuda, T., Isolation of a DNA plasmid III the
fungus Rhizoctonia solani, f. Gen. Microbiol., 130, 2067, 1984.
259. Beggs, J. D., Transformation of yeast by a replicating hybrid plasmid, Nature (London), 275, 104,
1978.
260. Hinnen, A., Hicks, J. B. and Fink, G. R., Transformation of yeast, Proc. Natl. Acad. Sci. U.S.A.,
75, 1929, 1978.
261. Struhl, K., Stinchcomb, D. T., Scherer, S. and Dawis, R. W., High frequency transformation of
yeast: autonomous replication of hybrid molecules, Proc. Natl. Acad. Sci. U.S.A., 1035, 1979.
262. Beach, D. and Nurse, P., High frequency transformation of the fission yeast Schizosaccharomyces
pombe, Nature (London), 290, 140, 1981.
263. Das, S. and Hollenberg, C. P., A high frequency transformation system for the yeast Kluyveromyces
lactis, Curro Genet., 6, 123, 1982.
264. DeLouvencourt, L., Fukuhara, H., Heslot, H., and Wesolowski, M., Transformation of Kluyvero-
myces lactisby killer plasmid DNA, f. Bacteriol., 154, 737, 1983.
265. Das, S., Kellermann, E. and Hollenberg, C. P., Transformation of Kluyveromyces fragilis, f. Bac-
teriol., 158, 1165, 1984.
266. Ito, H., Fukuda, Y., Murata, K., and Kimura, A., Transformation of intact yeast cells treated with
alkali cations, f. Bacteriol., 153, 163, 1983.
267. Kingsman, A. J., Gimlich, R. L., Clarke, L. J., Chinault, A. C., and Carbon, J., Sequence variation
in dispersed repetitive sequences in Saccharomyces cerevisiae, f. Mol. BioI., 145, 619, 1981.
268. Stinchcomb, D. T., Thomas, M., Kelly, J., Selker, E., and Davis, R. W., Eucaryotic DNA segments
capable of autonomous replication in yeast, Proc. Natl. Acad. Sci. U.S.A., 77,4559,1980.
269. Davison, J. and Thi, V. A., The Trypanosoma bruceimaxi circle DNA contains ars elements active
in Saccharomyces cerevisiae, Curt. Genet., 6, 19, 1982.
270. Gunge, N., Yeast DNA plasmids, Annu. Rev. Microbiol., 37,253, 1983.
27 I. Tudzynski, P. and Esser, K., Nuclear association in yeast of a hybrid vector containing mitochondrial
DNA, Curro Genet., 7,165,1983.
272. Danielevskaya, O. N., Kurenova, E. V., Liebovitch, B. A., Shevelev, A. Y., Bass, I. A., and Khesin,
R. B., Telomeres and P-element of Drosophila melanogastercontain sequences that replicate auton-
omously in Saccharomyces cerevisiae, Mol. Gen. Genet., 197, 342, 1984.
273. DeLange, A. M., Futcher, B., Morgan, R., and McFadden, G., Cloning of the vaccinia virus telomere
in a yeast plasmid vector, Gene, 27, 13, 1984.
274. Hsiao, C. L. and Carbon, J., Direct selection procedure for the isolation of functional centromeric
DNA, Proc. Nat!. Acad. Sci. U.S.A., 78, 3760, 1981.
275. Blackburn, E. H., The molecular structure of centro meres and telomeres, Annu. Rev. Biochem., 53,
163, 1984.
276. Reipen, G., Erhardt, E., Breuning, K. D., and Hollenberg, C. P., Non-selective transformation of
Saccharomyces cerevisiae, Curro Genet., 6, 189, 1982.
277. Cohen, J. D., Eccleshall, T. R., Needleman, R. B., Fedoroff, H., Buchferer, B. A., and Marmur,
J., Functional expression in yeast of the Escherichia coli plasmid gene coding for chloramphenicol
acetyl transferase, Proc. Natl. Acad. Sci. U.S.A., 77, 1078, 1980.
278. Cohen, J. D., Abrams, E., Eccleshall, T. R., Buchferer, B., and Marmur, J., Expression of a pro-
caryotic gene in yeast: isolation and characterization of mutants with increased expression, Mol. Gen.
Genet., 191, 451,1983.
279. Webster, T. D. and Dickson, R. C., Direct selection of Saccharomyces cerevisiae resistant to the
antibiotic G418 following transformation with a DNA vector carrying the kanamycin resistance gene
Tn903, Gene, 26,243, 1983.
 283

280. Feher, Z., Kiss, A., ami Venetianer, P., Expression of a bacterial modification methylase gene in
yeast, Nature (London), 302, 266, 1983.
281. Stepien, P. P., Brousseau, R., Wu, R., Narang, S., and Thomas, D., Synthesis of a human msulin
gene, Gene, 24,289, 1983.
282. Hitzeman, R. A., Hagie, F. E., Levine, H. L., Goeddel, D. V., Ammerer, G., and Hall, B. D.,
Expression of a human gene for interferon in yeast, Nature (London), 293,717,1981.
283. Dobson, M. 1., Tuite, M. F., Mellor, 1., Roberts, N. A., Burke, D. C., Kingsman, A. 1., and
Kingsman, S. M., Expression in Saccharomyces cerevisiae of a human interferon alpha directed by
the TRP1 5' region, Nucleic Acids Res., 11, 2287, 1983.
284. McNeil, J. B. and Friesen, J. D., Expression of the Herpes simplex virus thymidine kinase gene in
Saccharomyces cerevisiae, Mol. Gen. Genet., 184, 386, 1981.
285. Valenzuela, P., Medina, A., Rutter, W. J., Ammerer, G., and Hall, B. D., SynthesIs and assembly
of hepatitis B virus surface antigen particles in yeast, Nature (London), 298, 347, 1982.
286. Miyanohara, H., Toh, E. A., Nozaki, C., Hamada, F., Ohsoma, N. and Matusbara, D., Expression
of the hepatitis surface antigen gene in yeast, Proc. Natl. Acad. Sci. U.S.A., 89, 1, 1983.
287. Dickson, D., Mixed welcome for genes on Wall Street, Nature (London), 290,77,1981.
288. Beggs, J. D., van den Berg, J., van Ooyen, A., and Weissmann, C., Abnormal expression of chro-
mosomal rabbit {J-globin gene in Saccharomyces cerevisiae, Nature, 283,5750, 1980.
289. Watts, F., Castle, C. and Beggs, 1., Aberrant splicing of Drosophila alcohol dehydrogenase tran-
scripts in Saccharomyces cerevisiae, EMBO J., 2,2085, 1983.
290. Gallwitz, D., Construction of a yeast actin gene intron deletion mutant that is defective in splicing
and leads to the accumulation of precursor RNA in transformed yeast cells, Proc. Nat!. Acad. Sci.
U.S.A., 79,3493, 1982.
291. Langford, C. 1. and Gallwitz, D., Evidence for an intron contained sequence required for the splicing
of yeast RNA polymerase II transcripts, Cell, 33, 519, 1983.
292. Langford, C. 1., Nellen, W., Niessing, 1., and Gallwitz, D., Yeast is unable to excise foreign inter-
vening sequences from hybrid gene transcripts, Proc. Nat!. Acad. Sci. U.S.A., 80, 1496, 1983.
293. Peberdy, J. F., Fungal protoplasts: isolation, reversion and fusion, Annu. Rev. Microbiol., 33,21,
1979.
294. Bennett, J. W., Genetics of mycotoxin production with emphasis on the aflatoxins, in Overproduc-
tion of Microbial Products, Krumphanzl, V., Sikyta, B., and Vamek, Z., Eds., Academic Press,
London, 1982,549.
295. van den Broek, J., personal communication, 1983.
296. Yelton, M. M., Hamer, 1. E. and Timberlake, W. E., Transformation of Aspergillus nidulans by
using a trpC plasmid, Proc. Nat!. Acad. Sci. U.S.A., 81, 1470, 1984.
297. Tilburn, J., Scazzocchio, C., Taylor, G. C., Zabicky-Zissman, J. H., Lockington, R. A., and Davies,
R. W., Transformation by integration in Aspergillus nidulans, Gene, 26,205,1984.
298. Ballance, D. 1., Buxton, F. P., and Turner, G., Transformation of Aspergillus nidulansby the oro-
tidine -5- phosphate decarboxylase gene of Neurospora crassa, Biochem. Biophys. Res. Commun.,
112,284, 1983.
299. Stohl, L. L. and Lambowitz, A. M., Construction of a shuttle vector for the filamentous fungus
Neurospora crassa, Proc. Natl. Acad. Sci. U.S.A., 80, 1058, 1983.
300. Grant, D. M., Lambowitz, A. M., Rambosek, J. A., and Kinsey, J. A., Transformation of Neuro-
spora crassa with recombinant plasm ids containing the cloned glutamate dehydrogenase (am) gene:
evidence for autonomous replication of the transforming plasmid, Mol. Cell. Bio!., 4,2041, 1984.
301. Hughes, K., Case, M. E., Geever, R., Vapnek, D., and Giles, N. H., Chimeric plasmid that replicates
autonomously in both Escherichia coli and Neurospora crassa, Proc. Nat!. Acad. Sci. U.S.A., 80,
1053, 1983.
302. Hughes, K., Case, M. E., Geever, R. E., Vapnek, D., and Giles, N. H. (retraction), Proc. Nat!.
Acad. Sci. U.S.A., 80,7678, 1983.
303. Bull, J. H. and Wootton, 1. C., Heavily methylated amplified DNA in transformants of Neurospora
crassa, Nature (London), 310, 701, 1984.
304. Esser, K., Kuck, U., Stahl, U., and Tudzynski, P., Cloning vectors of mitochondrial origin for
eucaryotes: a new concept in genetic engineering, Curro Genet., 7,239,1983.
305. Tudzynski, P. and Esser, K., Extrachromosomal genetics of Cephalosporium acremonium. II, Curro
Genet., 6, 153, 1982.
306. Tudzynski, P., personal communication, 1984.
307. Stohl, L. L., Akins, R. A., and Lambowitz, A. M., Characterization of deletion derivatives of an
autonomously replicating Neurosporaplasmid, Nucleic Acids Res., 12, 6169, 1984.
 285

INDEX

A ccc head-to-tail. 265

In Podospo/"{/ amerllla, 257
AMY, see Alfalfa mO"lIc ViruS
Aberrant growth, 60
a-Amylase. 20
Abn nlutant" 31-33
Anamorph" 21
Abnormalltie'>, 35
Anastomo'>es (hyphal fU'lOn), see also
Ach/ya, 39, 240
Heterokaryons
ACllllform rod YLP, 13
d-factor trammi"lon. 211-212, 216
Acremonium (hry.logenum, 16
ggt trammlSSlon, 229-231
Actinomycin, 95, 130
hYPclVIrulence transml'>Slon, 145-146. 149, 157
Acuba strain of TMY, 25
,enescence factor tran'>mls"on. 257
S-AdenosylmethlOnIne (SAM), 95
ViruS tramml"lon. 7, 44--45, 152
Adenovlru, DNA, 101
Animal vlru,es. ,ee also speCIfic types, 62--63
Adenovlru,es, 40, 41
AnnelId,>, II
Adenyl cycla,e, 102
Anthraquinone, 157
Afflmty chromatography, 9
Antibiotics
Aflatoxin blosynthe'ls, 57
dlsea,e-Induced. 164
AgariclIs
resistance to, 6
blSporus, 46
Antibodies, see also speCific types,S, 9
dbea,e" 7, 25, 58-61
Antigemc vanatlOn, In trypano,omes, 5
infectIon, 45
Antigen" see also '>peCifle types
spore transmission of ViruS, 43--45
large T, 63
viruses charactenzed, 7. 12-14. 28. 31, 50---
Antisense RNA, 62
51, 153
Antl,era, see also specific types
vlrus/ho,t relationship, 54--55
to poly U and poly I poly C, 10
bitorquis. 44
for 1905 ViruS, 166
brunescens, 12
campestris, 13, 14, 30, 46 Anti-toxin gamma-globulin, 93
hortensls, 12 Antitumor agents. see also speCific type,. 8
species. 21 Antiviral agents, see also speCific types. 8, 46
Agarm,e gel electrophoresIs, 10, 117 AphehdlUm species. 19.40
AgglutinatIOn. 9 AphId transmissible VIrw,es, 49
Aggressive strainS of Ceratocystls u/ml. 178.201, Apis melilfera (honey bee), 28, 29
203,210, 216 Apocytochrome b, 241, 242, 244, 245
d-factors in, 205 ApoproteIns, see also specific types
Agrobacterium, 36,214 cytochrome b, 254
tumefaciens, 5 ApothecIa, 25
Agropyron repens, 222 Appressona, 24, 26
A/bugo candida, 18,40 Arenaviruses, 3, 32
Alcohol dehydrogenase, 21, 270 Armlilana mel/ea, 13. 15, 30
Alfalfa mosaic Vlru, (AMY), 28, 92, 96 Armillanella, see Armillaria mellea
Algae, see also specific types, II, 40 ARS. see Autonomously replicatIng sequences
Allen Viruses, infection with, 62-64 Arthrobotrys speCies, 16
Alkali cations, 269 Arthropods, II, 29
Alkaline phosphatase, 103 ArtifiCial VS. natural disease control, 204
Allomyces Asci, 25, 35
arbu.lcula, 18, 20, 37, 44, 51, 54, 62 Ascobolus immenu5, 239
macrogynus, 20, 240 Ascomycetes, 21, 44
species, 21 AscomycotIna, 21, 25, 26. 37. 42
Alpha-amanitIn, 95 Ascospores, see abo speCific fungi
Alpha-beta dimer, 95 d-factor transmiSSIOn, 196-197. 202, 210---212
Alpha factor, 89, 93, 94, 100, 103 Gaeumannomyces and Phwlophora, 222-223,
Alpha-D-mannoslde, 133 230, 232
Altemaria tenuis, 16 virus transmission, /09-114, 228-229
Alu-type sequences,S Asexual ;pore;, 42, 43, 53
Amencan chestnut tree (Castanea dentata), 155 Aspergillus
Ammonium sulfate, 132 amstelodaml, 238, 239, 243
AMP, cyclIc, 102 mutants of, 255-256, 262
AmplificatIOns, 251, 255 awamori, 16
286 Fungal Virology

flal'lls, 16,51,57 Botrydi.1 sp., 19

joetidu.l, 8, 16,46----49,51,54-55. 131 Bovine serum albumm, 9
jllllligallllls, 19 Brachlopod~, II
glllllCU.I, 16, 255 Brettallom) (es
nidulans, 240, 242, 245, 270, 271 anomalus, 240
niger, 16, 44, 46, 49 eU.ltersii, 23H, 240, 241
oehraeeous, 16,46 Bnos/{/ (Coremlellaj ell/JlSpOra, 16
orwae, 136 Brome mO~a1c Vlru" 6
parasitieus, 57 Bromoviru~, 39
specie" 7-9, 43, 57 Brown di~ea~e (La France d,<;ea,e), 7, 58---61
Assays, ~ee also ~pecific type~ Bryophytes, I I
enzyme-linked immuno~orbent (ELISA), 9, 59, Budding (exocyto",), 3
166---168,171,174 Bul/era sp., 134
fluorescent antibody, 9 Bullet-shaped partlcle~, 13, 22, 28-31
for virus detection, 166---167 Bunyaviruses, 32
Aster dlliensis, 24
Asynchronous replication, 54
ATPase, 241, 242 c
AU-rich sequence, 87, 97
Autographa caliJomica, 29 Caliciviruses, 39
Autonomou~ly replicating sequences (ARS), mito- Cambium tis~ues, 155
chondrial DNA, 258 Candida
Autonomous replication (ori/rep), 248, 250 albicans, 16,37,58,62, 134
of hybrid plasmids, 271 parapsilosis, 240
A venll satii'a, 24 ~p., 93, 102
Axial canal, 26 tropicali.I, 13, 16, 35, 240
utilis, 16, 243, 269
Cankers, 155
B Capping activity, 97
Capsid polypeptides, see also Encapsldatlon, mOl
Bacilliform particles, 7, 22, 28-30, 60 phology of specIfic groups, 3, 24, 91
Bacteria, see also speCific types, II recognition "te of, 99
DeCD-binding proteins from, 247 Capsids, see abo Encapsldation, 4, 91-93
phage-infected, 10 double-shelled, 48
Bacterial endosymbionts, 36 single-shelled, 47
Bacteriocins, 58 Capsomers, 36
Bacteriophages, see also specific types, 13, 86, 135 Cardiolipin, 31
DNA filamentous, 3, 4 Carlavirus, 26, 27
MS2,8 Carnation latent ViruS, 27
from Penicillium species, 35-36 Carpophores, 43
RNA,4 Cassava latent virus, 36
Baculoviruses, 28, 29 Castanea
Bark phase of Dutch elm disease, 200, 201 delllata (American chestnut tree), 155
Barley mosaic virus, 6, 23 sativa (European chestnut tree), 155
Barley powdery mildew, see Erysiphe graminis Catenaria angulilulae, 40
Basidiomycetes, 21 Cauliflower mosaic virus, 40
Basidiomycotina, 21, 25, 26, 37, 42, 43 Caulimoviru~, 41
Batch cultures, 51 Ccc, DNA, 135, 238
Bean rust, see Uromyces phaseoli cDNA, 5, 214, 229
Beetle phase, of Dutch elm disease, 200---202 Cell division, 51
Beetle pheromone lures, 206 Cell-free infection, by dsRNA, 149
Beet necrotic yellow vein virus, 6, 23 Cell-free infectivity studies, 150
Bentonite, 131 Cell-free transmission, of Viruses, 164
Beta peptides, 89, 93, 94, 100 Cell-free virus preparatIOns, 45--46
Bioassays, see also Assays; specific types, 24 Cellular organisms VS. Viruses, 3
Bimaviruses, 52 Cellulose, 10
Blastocladiella emersonii, 20, 21 Cell walls, 8, 45, 46, 63
Blue-green algae, II breakdown of, 39
Boletus chitinous, 21
edulis, 12, 15, 26 fungal,35
species, 15 galactosamine, 57
 287

receptor of. 45. 95 ChytndlOmycetc" 20. 37-39. 62

'porophore. 30 CIrcular DNA. 135
CEN pla;,mld, 42 Circular ssDNA. 22. 36
Central canal. 26 Cisternae of endoplasmIC reticulum. 34
Centromeres, yeast, 10] Cla,sificatlOn of fungi, 21
Cephalosporium C/llI'ICep.l' purpurea, 239, 240
acremonium, 16,35,27.2 plasm ids of, 265-266
DNA of, 239-240, 270 Climates. 206
plasmids, 239. 269 Cloning vectors, 103
chrysogenum, 57 Closterovlruses, 26, 27
Ceratocystis u/mi, 44, 58, 61,144 Club-shaped particles, 13,31, 153
ascospores, 196---197, 20 I Clusters. 53
progeny, 211-212 Cms-S. see Cytoplasmically tran;,miued male-
conidia, 192-193 sterility
culture, 183, 186---189, 199 Coat proteins, 6
d-facton" see d-Factors Coch/iobo/u.l'
ds RNA, 218-219 hererostrophus, 239, 263
growth rates, 186---189 miyabeanus, 14. 26
i.olates, nomenclature, 179 Coelenterates, II
lIfe cycle, 200---203 Coe/omomyces punctarus, 20
plasmids.218-219 Coelomycidium simu/ii, 20
subgroups, 178 COl genes, 241. 245, 248, 254
up-mut factor, 179, 191 Cold sensitivity, 58-61
v-c groups, 193-194,201,203,206 Coleoptera, 37
yeast .tage, 195-196 Co/esporium, 12,24
Cerato-ulmin (CU) production, 218 Colicin E2, 136
Cesium chloride, 10, 20 Colletorrichum
density centrifugation, 117, 118, 166, 257 arramentarium, 16
EB,264 fa/catum, 19
Cesium sulfate, 28 gramico/a, 19
Cha/era ele gans, 18 lindemuthianum, 16.44,51
Chenopodium species, 43
amarantic%r, 22, 25 Col/ybia peronata, 12,26
quinoa, 22 Comovirus, 39
Chestnut blight, see Endothia parasitica Compatibility, 44, 55-56, III, 124, 229-230
Chestnut trees, ]45, 155 v-c groups, see v-c groups
Chitin, 21 Competition, intermolecular, 116
ChItin synthetase, 21 Complementation tests, 124, 134, 135
Chitosome-like particles, 21 Concanavalin A, 133
Chitosomes, 21 Conidia, 22, 43, 55, 144
Chlamydiae, 4 formation of, 20 I, 203
Chlamydospores, 43 germination of, 192
Chloramphenicol, 216 phialidic, 223
in culture medium, 181, 185 reduced transmission of d-factor by, 194---196
resistance, 196 transmission of virus in, 228
tolerance (CR), 183, 185 types I to III, Ceratocystis, 212-216
Ch/orella viability of, and d-factor, 192-193
e/lipsoideus, 270 Conidial isolates
species, 36 type I single, 212-213
Chloris striate mosaic virus, 36 type 11 single, 213-215
Chloroplasts, 4, 5 type 1II single, 215-216
DCCD-binding proteins from, 247 Conidiogenesis
Chromatography. see also specific types transmission of d'-factors. 212-216
affinity, 9 transmission of dsRNA, 212-216
Chrom%sporium (Ostracoderma) sp., 12 Conidiophores, 43, 144
Chromosomes, see also Nuclear genes, 5, 112 Convalently closed circular (ccc) head-to-tail ampli-
Chromosome walking techniques, 36 fications, 265
ChrY.I'osporium sp., 19 Conversion, 112
Chymotrypsin-like protea;,es, 94 Copia particle;" 42, 62
Chytnd fungus, 6, 26 Copnnu;
Chvrndia/e.l', 6 comarUI, 55
288 Fungal Virology

cOllg' ~gatlO. 55 S(l( rhamlll\,( 1'1 'pp . 241-242

leIKoPUS. 15 COli, 246
Copnnw> DNA polymerase. 62 COllI, 248
Copy number, 100 Cytochrome c reducta,c, 240. 241
d,RNA. 57, 59 Cytochromcs, see also speclf/(: t) pes. 32
Coremlel/a (Bnos/(/) rublsportl. 16 Cytoductlon. 42. III
Corn, 5, 110, 134 CytoplasmiC ImmunIty. 113, 114. 124
chlorotic dwarf Vlrm, 39 CytoplasmiC Inhentance, see also Extrachromosorr
chromosomal DNA of, 270 inhentance, 5. 58, 232
Corn rw,t (Pu((in/{/ sorghi), 12, 15,24 of plasmId,. 101
Corn ,mut, 86 In Ust"ago lllaydi.l. 110, 112
Corn streak ViruS, 36 CytoplasmiC location of d-factors, 179-186
Cortical tissue. 155 CytoplasmiC membranes, 34
Corflclltlll roljlli. 15 Cytoplasms. 31, 40
Cortlcovlruse" 41 Infectlou" 145
Corynebuctenum dlphthenae. 5 kalilo, 263
Countenmmunoelectrophoresi<;, 9 CytOtOXICity, 8
Covalently closed Circular DNA. 135,238
Cowpea rust, see Uromyees pha.leoh
CR (chloramphenicol tolerance), 183, 185 D
C-reactIon, 193
ere locus EndothlQ parulltiw, 158 Da/dll1/(/ sp , 19
C-nch end, 89 DCCD-bIndIng protein, 247
CroSSing Ddp I plasmid, 268
Ceratocystls ulml. 196--197 Deban'omyees sp .. 93, 102
Gaeumanllomyce.l. 232 DefectIve dsRNA, 42
Sarrharomyce.1 cereVI.lwe. 97-98 Defective Interfenng (DI) particles, 99
Usttlago maydis, 110---116, 123 Defective RNA, 42, 47, 49
Cross-protectIon tests, 22 Degenerative disease, 57, 61, 149
Crotch feeding, 20 I Dekkera bruxel/ellsls. 240
Crown gall (tumor), 5 Deletion mutants, 123, 259
Cr)pholleetna paraslliea, see Endoth/(/ pllra.Hllcli Deletions
CryptIc dsRNA, 230 of genes, 254
CryptIc viruses, 49, 52 Internal, 102
CU (cerato-ulmin), 238 Delta sequences, 42
Cucumber Denaturation, 130
coty ledons, 25 Density components, 118
necrosIs ViruS, 6 Dermocystldlum manllum, 18
Cucumovirus, 39 Detergent, non-lome, 32
Cueumus satlvuS, 25 Deuteromycotina, 21, 25, 26, 37
Culuvated mushrooms, see Agartcus bisporu.l; Lell- d-factors, see abo d'-factors; d'-factors, 179
tlllUS edodes biological characteristics of, 178-199
Cycads, II Ceratocystis ulmi, see Ceratocystls ulml
CycliC AMP, 102 COnidial germInalIOn and, 192
Cycloheximide, 57, 124, 148, 174 conidial transmisSion, reduced, 194--196
Cylllldrociadium scopanum, 20 conidIal ViabIlity and, 192-193
Cystovlruses, 48, 52 cytoplasmic locatIon of, 179-186
Cyst wall, 21 In disease control, 204--207
Cyt b, see Cytochrome b during disease cycle, 200-203
Cytochrome a, 255 dlstnbullon of, 178-179
Cytochrome aa, 252, 254, 263 frequency of, 202-204
Cytochrome b (cyt b). 245, 252, 263 growth and, 186--192
apoproteIn, 254 growth ~tablhty and, 190-191
gene, 244--245 impact of, on Dutch elm disease, 199-207
mltochondnal, 248 nature of, 216, 217
Cytochrome c. 252, 255, 263 nonaggresslve Isolates to, 190
Cytochrome c OXidase genes pathogemc behavior of Infected Isolates, 197-
COl 199
Aspergtlhs mdulans. 241-242 pentheclal formatIOn and, 192-193
Neurospora cras.la. 241, 254 regulation of transmisSion of, 193-197
Podospora allsenna. 241-242 reproduction and, 186--192
 289

spread of, 179-11I6 of SaC( haromyces cereVl.llae, 266--268

transmIssIon to ascospores, 196--197, 210 satellite, 20
variabdlty in, 205 selfish, 61
d'-factors, see also d-factors, 216--217 sequency analysIs of, 36
vs d'-factors, 191-192 topology of. 40
d'-factors, see also d-factors, 216 transposlllOn, 5
acqUlsIllon of, 210-211 DNA-binding proteins, 62
host protein for, 214 DNA filamentous bacteriophages, 3, 4
latency of, 213-215 DNA polymerase, 55, 243
loss of, 211-212 copnnus, 62
transfer of, 212 DNA provlruses, 42, 55
transmIssion by comdla, 212-216 DNase, 95
DI particles, see Defecllve mterfenng partIcles DNA topoisomerases, 62, 100
Dlanthovlrus, 39 DNA viruses, 4, 62
Dlaporthine, 156 double-stranded, 37--41
Diatoms, II Dommant (DET) genes, 100
Dictyosteilum d,sco,deum, 239, 240 Dot hybndlzatlOn, 152, 167
chromosomal DNA of, 270 Double anllbody sandwich method, 9
ddp I plasmid m, 268 Double-helIcal structure, 9
DIe-back (La France) disease, 7, 58---{)1 Double-membrane-bound veSicles, 53
Dlfferenllal host mdexing, 22 Double-membrane envelope, 31
DiffractIon Double-shelled capsid, 48
optical, 25, 28 Double-stranded DNA (dsDNA), 37--40
X-ray, 25 Icosahedral, 41
Dlkaryotic hyphae, III Imear, 269
DI mutants, 101 plasmlds, 58
DI partIcle RNA, 99 Imear, 101
Diphenylamme, 116 Double-stranded RNA (dsRNA)
Diphtheria toxin,S, 136 acquisllion of, 210-211
Dlp/ocarpon rosae, 14, 46 cell-free mfectlOn by, 149
DiplOIds, see also Heterokaryons, 99, I 13, 114 in CeralOc}'~tis u/mi, 218-219
colonies, 58, 238, 249 comdlOgenesls of transmission of, 212-216
mllospores, 44 copy number of, 57, 59
vlflons, 54 crypllc, 230
Diptera, 37 defective, 42
Direct repeats,S, 250, 251 d-factors and, 210-217
Disease control distribution of, 225
artificIal vs. natural, 204 encapsldation of, 117-120
d-factors in, 204--207 Ggt, see Gaeumal1110myces viruses
DIseased fungi, 9, 45, 60 hypovlrulence and. 144--149
Disease-induced antibIOtics, 164 Interferon Inducers, 7-9
Disease-resistant elms, 206 Isometnc, see lsometnc dsRNA mycoviruses
DisorganizatIOn of mitochondna, 34 L,218
Displacement mechamsm, 54 loss of. 148-149, 211-212
DisulfIde bonds, 95 M,21X
DNA, see also Double-stranded DNA, Single- male stenlIty and, 49
stranded DNA nature of. 150-154
adenOVirus, 101 negallve regulation of syntheSIS, 59
chromosomal,S Northern blots of, 126
covalently closed circular, 135, 238 P particles, of Usn/ago maydls, 116--120
endosymblOnt, 36 properties of, 52
extrachromosomal, see Extrachromosomal DNA relatedness of segments, 126--130
linear double-stranded, 269 replIcation of, 51,53,61
mitochondrial, see MItochondrial DNA replIcative form, 10
nuclear, see Nuclear DNA restrIctIOn enLymes of, 135
plasmid, 101-102 rod-shaped, 48--49
promiscuous, 247 satellite, 42,57,60, 130
prophage, 4 ScV" see Saccharomwe,l cerel'i~Ule viruses
rearrangements of,S segment patterns of, 151
repair of, 251 selfIsh, 61
replicatIOn, 56, 61 suppreSSive, 99
290 Fungal Virology

tramcnpllon of, 54 d,RNA ellects. 158-159

tran,fer of, 148, 212 h-factor, see H strain,
viral funcllon, and ,egments of, 120--126 host-pathogen relatIOn> of. 155-156
yea,t, 129,215 H strains of. 150--154.206
d-reactlOn, ,ee abo d-factor" 179-186 hypovlrulence In, 61, 144---149, 150--154. 156.
Dnft, genetic, 121, 126 217
Drosophila virulence, expre,Slon, 154---158
melanogaster, 270 EndotOXin, 8
,p , 5, 42 Entomopara"tlc fungu" 30
alcohol dehydrogenase of, 270 Envelope. 53
COPIa particle, in, 62 Enveloped pleomorphiC parllcles. 31-33
Dual speclflcllle" 115 Enzyme-linked Immunmorbent as;ay (ELISA). 9.
Duchesnea Indica, 24 59, 16&-168. 171, 174
Dutch elm dI>ease, see abo Cerato()stIV uln/l Enzymes, ,ee also ,peClflc types
control, 204---207, 217 d,RNA restrIction, 135
cycle, pha,es, 200--203 lytiC, 50
d-factors, see d-factors proteolytic, 110, 132
expenmental condillons, 197-199 RNA-splICing, 241
In nature, 197 Epidemiology, of Vlru, and host, see abo individual
pathogens, 230--231
Eryslphe
E gramll/lS, 12, 14. 22, 24
polygollI. 12,24,26
EAN I>olates, 178, 187-191, 194, 196.203,205, ESA (elm sapwood agar), 179
206 EscherIchlll colt, 35, 36, 269-271
EB ce,lUm chlOrIde denSity centrIfugatIOn, 264 Ethldium bromide, 10,95
Echmoderms, II Eukaryotes, 5, I I, 42
Electron microscopy (EM), 101, 130, 153,266 Eumycota (true fungi), 21
electron-dense cores, 20, 31, 33, 169 EupatOrium. sp., 2
electron-opaque cores, 33 European chestnut tree (Castanea saliva), 155
electron-transparent core, 33 European EAN, 203
Helmil1fhosporium 1905 virus, 165-166 EvolutIOn, of dsRNA "ometnc viruses, 61--62
Immunospeclfic (ISEM), 9, 59, 229 ExclUSIOn, 97-98, 100, 115, 116
particle dimension measurement, II, 20, 22, Exocytosls (budding), 3
24---25, 117, 119, 258 Exons, 61
virus detectIOn, 43, 173-174 Extrachromosomal DNA, see also Mltochondnal
ElectrophoresIs, gel, see Gel electrophoresIs DNA
ELISA, see Enzyme-lInked Immunosorbent assay extrachromosomal inherItance, see Extrachromo-
Elm, see Dutch elm disease somal Inhentance
Elm bark beetles, 178, 197 genetIc engineerIng With, 269-272
EM, see Electron microscopy plasmlds, nonmltochondnal, 31, 32
Emodin, 157 Extrachromosomal Inhentance, see also Cytoplasmic
Empty particle molecular weight, 119 Inheritance; Killer ,ystems
EncapsldatlOn, 120 mltochondnal genes
ofdsRNA,117-120 apocytochrome b (cyt b), 241-242, 244---245,
Encephalomyocarditis, 62 248
Encephalopathies, 5 ATPase complex, 241-242
Encystment, 21 cyt b, see Apocytochrome b
Endogenous retroviruses, 4 cytochrome c OXidase genes, see Cytochrome c
Endomyces geornchum, 16 oxidase genes
Endonucleolyllc activity, 135 cytochrome c reductase, 241-242
EndopeptIdase, 95 delellon mutants, see also by name, 259-260
Endoplasmic retIculum, 34, 153 discontinuous, 242
cisternae of, 34 locatIon, 241-242
rough, 20 mex-J, 248, 259
Endosymblonts, 36 Neurospora crassa group I mutants, 252
hypothesis of, 247 ori sequences, 243, 248, 250--251
prokaryotic, 35 poky, 31. 32. 57, 252
Endothia parasitica ragged (rgd), 57, 255-256, 262
compatIbilIty genes, 44 rho-, 249-252
ere locus of, 158 nbosomal components, 241
 291

RNA-,piIttmg enzyme" 241, 243, 245 Ga{'um{/nnomy(eJ, 10

,ene~cence, ,ee Sene,ccnce n'illJdrOlporus, 222
'topper, 32, 33, 57, 252-256, 262 grmnlllLl, 222
tRNA" 241 var w'e/we (Ggg), 222
unIdentIfied readIng frame, (URF,), 241, 254 var gramll1l5 (Ggg), 222
var-/, 247 gramm/.\ var. trtllci (Ggt)
mItochondnal mtrons, 5, 61, 241, 260--262 blOlogy of, 222-223
mltochondnal pla,mld" 266--269 d,ver,ny 01, 222, 229
nonmltochondnal plasmlds, 266--269 host/VIrus epIdemIOlogy, 230--231
Extranuclear mutants, 31, 32 Isolates, 228, 229, 232-233
pathogenIcIty, 231-233
Q-factors, 231
F take-all d"ease" 222
VIruses of, see Gaeum{/nl1omyces, virw,es
Fannw canicu/ari~, 30 Gaeumannomyce,\, VIruses, 223
FeedIng groove pha,e, of Dutch elm dIsease, 200, Gga, 223, 227-228
202, 204 Ggg, 223, 227-228
Fibnllar structures, 39 Ggt
Fibou, matenal, 53 effects on pathogenIcity. 231-234
FIbrous proJectIons, 40 group" 223-228
Fiji plasmId, 263, 265 host/VIrus epIdemIOlogy, 230--231
FIlamentous partIcles, 25, 26 protoplast mfectlon, 231
genetIc engIneenng of, 270--272 tranSmIS5l0n, 43, 228-229
FIngerpnnt analYSIS, 87 GaeumannomyceslPhw/ophora complex, 222
jlatlocl,158,215 Gag regIon, 42
Flexuous rod" 12, 25-28 Gametes, 44
narrow-diameter (NFR), 26 Gamma partIcles, 20
WIde-dIameter (WFR), 26 Gamma regIons, 89
FLP gene, 268 Gamma reSIstance, 94
Fluorescent antibody assays, 9 Gel electrophoreSIS, 215
Foot and mouth dIsease, 2 agarose, 10, 117
Frame,hlfts, translatlOnal, 42 polyacrylamIde (PAGE), 10, 91, 117, 166
Frommea obtusa, 24 SDS polyacrylamIde, 166
FruIting bodIes, 156 Gel ImmunodIffusion, 10
F sex factors, 4 GemInate partIcles, 13, 31, 36
Fungal cell walls, 35 GeminIvlrus, 2
Fungal veSIcles, 149, 153 Genes, see also specIfic genes, 112, 158
Fungi, see also genera and ,ubgroups by name COL, 241, 245, 248, 254
cell walb, 35, 43 cyt b, 245
hyphae, see Hyphae deletIOn of, 254
mterferon inducers from, 7-9 dommant (DET), 100
reproductIve structures, ,ee by name ImmunIty, 60
ultrastructure, 53-54 mak (MAK), 55, 59, 96, 98-100
as vectors, 6 mitochondnal cytochrome b, 248
FungIcIde" 210, 211 mitochondnal, discontInuous, 242-243
FungI ImperfectI, see DeuteromycotIna mkt, 97, 98, 100
Furoviruse" 22, 23 N,4
FU5arium nUclear, see Nuclear genes
momltjorme, 16 pathogenicIty, 144, 232
oxysporum, 16,218,239, 240 sec, 94, 100
roseum, 16, 46 ski, see Superkiller mutants
species, 43 tOXin, 101
FUSIOn trans-acting FLP, 268
hyphal, see Anastomoses Genehc drift, 121, 126
mcompatiblhty to, 44 Genehc elements, mobile, 5
protoplast, 44, 101 Genetic engIneenng, 103
WIth extrachromosomal DNA, 269-272
of filamentous fungI, 234, 270--272
G of Yeasts, 269-270
GenetIc markers, 145
G I arrest, 102 Genetics of interstraIn inhibitIOn, 110--114
292 Fungal Virology

Genome" see abo specIfIc type, Heads and tails of VLP" 13, 34--36
expression of. 91-97 Heat lnaclivatlOn tests, 22
human. 240 Heat sensitivity. 22
mitochondrial. 238-243 Helemne, 8
expressIOn, 243-246 Helical nucleocapslds, 26
orgamzation. 238-243 HelIcal parlJcles, 26
RNA processing, 243-245 Helical ,ymmetry. 28
transcnption, 243 Helmtnthosporrum
translatIOn, 245 carbonum, 16, 134
nuclear, 243 maydls, 17,47,48,51,134,163
organization of, 46-49 orystae, 17
reduced, 123 sacchan, 12, 17, 26
structure of, 46-49 furcccum, 19, 134
taxonomy of, 46-49 vagans, 134
Geotnchum vlctorcae, 7,57,58,61
candidum. 16, 49 assay methods, 166--167
lactis, 43 colony morphology, 168-171
Germ tubes, 24, 26, 44, 46 pathogenicity of diseased Isolates, 169-171
Gga, ,ee Gaeumannomyces gramims var avenae Victoria bhght, 164--165
Ggg, see Gaeumannomyces gramtnis var. gramlnis viruses. morphology, 17,49-50, 165-166
Ggt, see Gaeumannomyees gramtnls var tnllel viruses, transmissIOn, 45, 171-174
GliocladlUm sp., 19 Helper viruses. 47, 56
Gliomasllc sp., 19 satellites of, 58
GlIotoXIn, 46 HemagglUl1nal1on, 62
I3-GlobIn, 270 Hemcleca vasatrcx, 15
u-GlobuIIn, 93 Hepatitis B virus, 40
13-( I ,6)-D-Glucan, 93 Herpes-like VLPs, 13,39
Glutamate dehydrogenase, 271 Herpesviruses, 33-34, 41
Glycogen granules, 20 Heteroduplex analysis, 87, 128, 129
GlycoproteIn, 28 Heterokaryons, 32, 42, 45
Glycosylatlon, 94, 102 formalion, 110, 113, 124
Golgi bodies, 21, 94, 95 In Gaeumannomyces, 232-233
vesicles related to, 34 transfer experiments, 110, 113-114
Gomphrena globosa, 25 transmission by, 44-45, 112-114
Gonatobotrys sp , 16 Heteroplasmons, 32
Graft-transmissible viruses, 49 transmission by, 44-45
Grande (diplOid) colomes, 238. 249 Heterozygote diploids, 114
Group I mutants, 32 Hexamer,28
Group III mutants, 32 H-factor, in Endothca paraslfiea, 206
Growth Histoplasma eapsulatum, 17, 37
d-factor effect on, 186--192 HomeostasIs, splicing, 245
mycelIal, 186--190 Homokaryotic hyphae, 232
poor, 146 Homology sequence, 87. 121, 126
Growth hormone, 270 Honey bee (Apis melli/era), 28, 29
Growth phase, 130 Hordeiviruses, 22, 23
Growth rates, 31, 145 Hordeum vulgare, 24
Guanylyltransferase, 97, 243 Hormones, see also specifiC types
Guttultnopsis vulgaris, 18, 40 rat growth, 270
Gymnosperms, II Host-pathogen interactions, 99-101
in Endothca parasttlcu, 155-156
host phenotype affectIng. 56---61
H mixed Infections, 55-56
virus replication and host growth, 49-53
Hanalei plasmid, 265 ViruS replication cycle, 54--55
Hanseniaspora vtnea, 240 Host proteins for d'-factor, 214
Hansenula Hosts, see also Vectors; specific plants
mrakli, 238, 240 defenses of, 154, 155
species, 93, 102 epidemiology of, 230--231
wingei,240 pathogen interaction, see Host-pathogen
Haploid meiospores, 44 Interactions
Haploids, segregation, 97-98 pathology of, 61
 293

m phage Mu, 5 IncompatIbIlIty, 55-56, 115-116

phenotype, 01, 56--D I allele, for, 122
range of, 5, 22, 46, 63 fu,IOn, 44
Host-'peClflC toxms, 156, 164, 169 pm,tfuslOn, 44
Human ARS of mitochondrial DNA, 251l Mlmatlc,20
Human mltochondnal DNA, 245 vegetatIve, 44, 193-194,203
Human mItochondrial genome, 240 IndIrect agglulInalIon, 9
Human URF, 2, 254 InfeclIon" see abo speCIfIc types
Hybndizatlon cell-free, 149
dot, 152, 167 of hyphae, 24, 26
of fungI, see Crossing latent, 45, 59
Northern, 153, 261 mIxed, 42, 55-56, 152
nucleIC aCId, 30,46,48, 121, 126-127, 129, mushroom, 45
132, 152 of protoplasts, 231
solId phase, 226 InfeclIous cytoplasmIC agents, 145
solulIon, 226 Influenza VIruS, 56, 62
Hybnd plasmlds, 269 Inheritance
autonomous replIcatIon of, 271 cytoplasmic, see Cytoplasmic inheritance
Hydra VlTld/,\, 36 maternal, 32, 210
Hydrodynamic propertIes, 118 non-Mendelian, 238, 250, 252
HydrophobIc regIon, 95 nuclear, see Nuclear inhentance
Hymenoform prolIferatIOns, 30 InhIbitIOn, 123, 124, 129
Hymenomycetes sp , 30 mterstram, 110---116
Hymenoptera, 29 phenotypes for, 126
Hypersem,llIve-IIke lesIOns, 22 potentIal for, 121
Hyphae, ,ee also Heterokaryons, 31, 222 of replicalIon, 254
compartment5 of, 9 zone of, 133
fusIon, see Anastomo,es Inner nuclear membrane, 34
growth, tran5mlS5IOn durmg, 42--43 lnocybe dulcamara, 13, 15
mfectIon of, 24, 26 Inovlruses, 25, 27, 29
mycelIal growth, 186-190 InsulIn, 270
tip culture, 121, 124 Integrase, 42
virus translation by, 186-190 InteraclIons among vIruses, 97-99
Hypholoma sp" 19 Interconverslon, 268
Hypochromlclty, 116 Interferon, 7-9, 270
Hyposoter ex/guue, 29 inducers, 7-9
Hypovlrulence Intermolecular competitIOn, 116
dsRNA associated with, 150 Internal deletions, 99, 102
Endothta paraslllca, 61, 144-149 Internal rearrangements, 99
JR type, 144-/49 Interstrain inhibitIOn, 112, 114-116
Rhiz(}ctoma so/ani, 144, 149-150 genetics of, 110---/14
Hypoxylon Intervening sequence5, see Introns
mult/forme, 14 Intracellular (vertIcal) transmiSSIon, 34, 39
species, 19 IntracistronIc complementation, 134
Introns, 5, 61, 260-----262, 241
Invertable DNA segments, 5
I Invertase, 103
Invertebrates, 29
Icosahedral VIruses, 28, 41, 47 Inverted repeats, 87, 101,250,251
Ig-reactIOn, 193 In VItro translatIon, 124, 129
Ilarvlrus, 39, 91 of dsRNA, 126
Immune ,trams, 114 In VIVO mRNA, 97
ImmunIty, see Resi5tance lon-permeable channels, 93
ImmunodIffUSIon, 9, 10 I-reaclIon, 193
ImmunoelectrophoresIs, 9 Indovlruses, 40, 4 I
Immunofluorescence bmdmg, 95 Iris xiphioi<ies, 24
ImmunogcnIclty, 8 IrradIatIOn
ImmunologIcal tes!)" 127 UV,47
Immunospeclflc electron microscopy (ISEM), 9, 59, IS elements, 5
228, 229 ISEM, see Immunospeclflc electron mICroscopy
ImmunosuppreSSIOn, 8 Isolate~, see al,o ,peclflC types
294 Fungal Virology

aggressive, see also EAN isolates; NAN isolates, K/oeckera

201,203,210 africalla, 240
asco~pore, 232 species. 19
of Ceratocystis u/ml, 179, 183, 185, 187~199, K/uyveromyceJ
202~203, 205, 210--212 fragi/is, 269
d-infected, 197~199 /aois, 93, 240, 269
of Gaeumannomyces, 228, 229, 233 plasmids of, 58, 86, 101-102, 239
of Helminthosporium, 167~171, 174, 215~217 toxins, 101-102
nonaggressive, 190, 196 tRNA genes, 243
single conidial, 174 species, 86, 93, 102
type I single conidial, 212~213 krb mutants, 100
type II single conidial, 213~215 kre mutants, 95, 100, 101
type III single conidial, 215-216 K unkelia lliTells, 24
Isometric dsRNA mycoviruses, see also specific
types by name
L
evolution, 61----62
fungal hosts, 39, 46
Rhizidiomyces, 37, 39---40 Labelle ~train, 263, 271
structure and organization, 46-49, 52 Labyrinthomyxa marina. 40
taxonomy, 48-51 Laccuria speCIes, 13, 15
virus/host interactions J3-Lactamase, 271
infections and host phenotypes, 55----61 La France disease. 7, 58----61
replication, 49-55 Large ribosomal (lr) RNA, 241, 242, 262
Isometric particles, 37--40 Large-scale virus extraction, 229
hosts and morphology, 13-18,37 Large T antigen, 63
Isometric single-stranded RNA, 39 Late log phase cell;, 130
Latency
d'-factor, 213-215
J disease, 61
infection, 45, 59
JR type of hypovirulence, 144 virus, 49-53
VLP,34
LathyrinThomyxa manna. 18
K L dsRNA, see also Saccharomyces cerev/sllle
viruse~, L partIcles

Kalilo cytoplasms, 263 Leaf dip method. 22

kex mutants, 94, 95, 100 Leafhoppers, 36
Killer mutants, 9, 60 Lentinus
suicide, 61 edodes, 12, 43, 48, 51
morphological types, IS, 20, 26
Killer plasmids, 238
/ipideus, 12, 15, 26
Killer proteins, see also Killer toxins, 57~58, 61,
Lepidoptera, 37
132-136
Lethal sectoring, 268
Killer systems
Lettuce big veIn VIruS, 6
of Saccharomyces cerevisiae, 57-58, 61
Lettuce necrotic yellow virus, 29
compared with Ustilago, 128-131,134
LeuC(}sporidium sp., 134
dsRNAs, 87-91
Light scattering, 119
expression of viral genome, 91-97
Lignified barrier, 155. 156
host relationships, 58, 99-100 Lignitubers, 222
interactions, viral, 97-99 Lihue plasmid, 265
of Usti/ago maydis Linear DNA, 265-266
mode of action, 134--136 double-stranded, 101, 269
nature of, 132-133 mitochondrial, 238
specificity, 133-134 Lipids, see also speCIfIC type~, 26, 32
Killer toxins, 60, 93, 102, 103 Lithium chloride, 10, 131
genes, 101 Lithium ions, 269
mode of action, 134--136 Logarithmic growth of cells, 131,134
nature of, 132-133 Log phase cells, 130, 131
specificity, 133---134 Long narrow-diameter flexuous rods. 26
Kinetics Long wide-diameter flexuou~ rods, 26
reassociation, 35 Loop structures. 87
of toxin-target cell interactions, 135 Low temperature growth, 59
 295

IrRNA gene, 262 nce downy, see Scleropi1lhura mau()spora

Luteovirus, 39 MIld mutagenesIs, 121
Lyw,,34 Mimchromosomes, 101
Lysogemc converSIon, 5 Mites, 201
Lysogemc phages, 35 Mitochondna, 4, 5
Lytic enzymes, 50 DCCD-bindmg proteins from, 247
LytIc plaques, 33, 58---{'1 dlsorgamzallon of, 26, 34, 39
genetIcs of, see also Extrachromosomal Inhent-
ance, 210
M genome, see MItochondrIal genome
membranes, 31-32
MadLU saliva, 24 plasmlds, see MItochondrial plasm Ids
MaIze, see Com MItochondrial DNA, see also Extrachromosomal In-
MAK (mak) genes, 55, 59, 96, 98-100 heritance, 57, 61, 197
Male bacteria, 4 alterations of, 263
Male steniity altered
cytoplasmIcally transmItted (cm-S), 5 poky mutants (Neurospora crassa), 252
dsRNA assocIated wIth, 49 ragged mutants (Aspergillus amste[odaml),
MammalIan mitochondnal DNA, 243 255-256
MammalIan vIruses, 62 rho mutants, 249-252
Manganese, 243 senescence, see Senescence
Mapping, 124 stopper mutants (Neurospora crassa), 252-255
heteroduplex, 87, 128, 129 ARS of, 258
restnclIon, 101, 258, 264 coding, 246
of viral functIons, 123 human, 245
Marasmlus androsaceous, 45 lInear, 102, 238, 265-266
Mastigomycotma, 21, 25, 40, 44 los; of, 249
Maternal mheritance, 32, 210 mammalIan, 243
Mallng plasmlds, see MItochondnal plasmids
alleles for, 110 Clavlceps purpurea, 265-266
pairs for, 46 Neurospora, 263-265
Matnx protein, 28 of Saccharomyces cerevisiae, 243, 246
MauncevIlle plasmid, 263, 265 sIze of, 240
MBC fungICIde, 210, 211 structure of, 240
tolerance, 181, 183, 185 transposition of, 247-249
M dsRNA, see also ScV VIruses, M partIcles Mitochondnal genome
Mechanically transmissIble viruses, 49 expression, 243-246
MeIosIs, 55 organizalIon, 238-243
Meiospores, 44 RNA processing, 243-245
Membrane-bound VLPs, 33, 154 transcnptlOn, 243
Membranes, 31, 33, 34, 39, 40 translalIon, 245
cell, 100 Mltochondnal plasmids, see also Mitochondnal
Inner nuclear, 34 DNA, altered
mltochondnal, 32 In Clavlceps purpurea, 265-266
nuclear, see Nuclear membrane In Neurospora sp , 263-265
plasma, 21, 34 MItochondna1 proteIn syntheSIS, 32
Messenger RNA, 55, 245 MitOchondrial RNA polymerases, 243
in VIVO, 97 Mltochondnal transcriplIon RNA, 246
splIced, 63 Mitospores, dIplOId, 44
subgenomlc, 87 Mitotic cell cycle, 53
MetabolItes, 57 Mixed InfectIons, 42, 55-56, 152
Methylated anthraqUInone, 157 mkt genes, 97, 98, 100
mex-I mutants, 248, 259, 260 Mobile genetIc elements, 5
Mlcroflbnls, 21 Molecular selection, 123
Mlcrophaera Molecular weIght of empty particles, 119
mougeotlt, 13, 14,30 Molluscs, 11
pOllgOnt, 14 Monoclonal antIbodIes, 9
Mlcrovlruses, 35 Morchella COntca, 239, 269
MIldews, see also speCIfIC types, 22, 25 MorphologIcal types of VIruses, 11---42
barley powdery, see Erysiphe gramtnlS MosaIC VIruses, see speCIfic types
oak powdery, see Sphaerorheca lanesrns mRNA, see Messenger RNA
296 Fungal Virology

mtDNA, see Mitochondrial DNA Narrow-diameter flexuous rod~ (NFR), 26

Mucor ~p., 21,45 Natural VS. artificial dl,ease control, 204
Multiple buds, 33, 35 NOV, see Newcastle dlsea~e ViruS
Mu~hrooms, see also specific types, 25, 30, 55, 59 Necroviruses, 39
cultivated, see also Agaricus bisporus, 7 Negative regulatIOn, 51, 54
die-back disease of, 58--61 of dsRNA synthesis. 59
infection of, 45, 60 of plasmid DNA replIcatIon, 56
shiitake, see Lentinus edodes Nematodes, II
spawns, 45, 59 Neomycin, 35
Mushroom virus I (MV!), 7, 30, 59 Nepovirus, 39
Mushroom virus 2 (MV2), 7, 59 Neurospora. 21,32,241-242
Mushroom virus 3 (MV3), 7, 28, 30 crassa
Mushroom virus 4 (MV4), 7, 30, 59 abn-I mutant, 31-33
Mushroom virus 5 (MV5), 7, 59 extranuclear mutant~, see also specific mutants.
Mushroom viruse~, see also specific type~, 43, 44, 32
60 geminate particles, 36
Mutagenesis, 121 genetic engineerIng, 270---271
Mutants, see also specific types, 100 mitochondrial code, 246
abn-I,31-33 mtDNA, 240, 247, 249, 256
deletion, 123 plasmids, 239, 263, 265
DI,IOI poky mutants, 31-33, 57, 249, 252, 256
extranuclear, 31 RNA polymerase, 243, 245
hypovirulent, see Hypovirulence rRNA of, 32-33
kex, 93-95, 100 stopper mutants, 32, 33, 57, 252-256, 262
killer, 9, 60 VLPs, 13, 14
KRB.IOO intermedia, 239, 242
kre, 85, 100, 101 plasmids of, 239, 265
mak. 55, 59, 96, 98-100 senescence In, 263
mitochondrial, see Extrachromosomal inheritance species, 21
nonkiller, 135 mitochondrial plasmlds In, 263-265
petite, 243 tetrasperma, 239, 265
pig. 157-158, 215 Newcastle disease virus (NOV), 62, 63
poky. 31,32,57,252 NEX, 98
ragged, 57, 255-256, 262 NFR, see Narrow-diameter flexuous rods
respiratory deficient, 31 N gene, 4
rex. 100 Nibbled colomes, 268
rho - , 249---252 Nicotiana
sec. 94, 100 g/utinosa, 25
slow-growing, see Slow-growing mutants tabacum, 22, 25
stopper, 32, 33, 57, 252-256, 262 Nodaviruses, 37, 39
suicide, 60, 61 Nonaggressive strains, o!Ceratocystls ufml, 196,
superkiller, 51, 59, 97, 100 218
MV, see Mushroom virus d-factors in, 190, 205
Mycelial fan, 155 Non-ionic detergent, 32
Mycelial growth, 186-190 Nonkiller mutants, 135
Mycogone perniciosa, 12, 17,43,49 Non-Mendelian inheritance, 238, 250, 252
Mycophenolic acid, 46 Nonmitochondrial plasmlds, fungal, 239
Mycoplasms, 4, 22, 29 nuclear-associated, 266-268
Mycoplasma virus MVL51, 28, 29 Nonperrnissive conditIons, 34
Mycoviruses, see Viruses, fungal Nontransmissible hypovirulence, 146
Myxomycota (slime molds), 21, 40 Non-VLP-producing condItIOns, 34
Myxoviruses, 62 Northern blots, 126, 153
Northern gel analysIs, 98, 102
Northern hybridizations, 261
N Nuclear-associated plasmlds, 266-268
Nuclear cap, 44
NAN isolates, d-factor infected, 178, 210, 216 Nuclear determinants, 232
mycelial growth, 182-183, 187-191, 193---194 Nuclear DNA, see also Nuclear genes, 5, 214
in nature, 203, 205, 206 transposition to, 260
pathogenic behavior, 197-199 Nuclear envelope, 26, 39
spread of, 201-203 Nuclear factor, 179
 297

Nuclear genes, 97, 170 Paramecium sp , 36

of Gaeumannomyces, 232 Paramoebldium arcuatum, 18, 40
of Saccharomyces cerevisiae, 58, 100---101 Paramyxoviruse>, 26, 32, 62
of Ustilago maydis, 110-111, 232 Parasitoid Hymenoptera, 29
Nuclear genome, 243 Partitiviruse> (PartitIvindae), 48-50, 52, 54, 225,
Nuclear membranes, lamellae, 34 230
Nucleases, 87, I 10, 129 Pathogenicity, 31, 231-233
Nuclei, 4, 5, 34, 39, 40, 43, 101 of diseased Isolates of Helmllllhosponum \'IClo-
Nucleic acid hybridization, 30, 46, 48, 152-153 riae, 169-171
Nucleic acids, see also specific types, 24, 116---118 genes, 144, 232
Nucleocapsids, 28, 34 of phtyopathogemc fungI, 57
Nucleoid, electron-dense, 31 Pathogenic (xylem) behavior, of d-mfected isolates,
Nucleolus, 34 197-199
Nucleoside, terminal, 129 Pathogenic (xylem) phase of Dutch elm dlsea<;e,
Nudaurelia virus, 39 200, 202-203
Patulin, 46

o Pea enation mosaIc VIruS, 39

Peanut clump VIruS, 6
Penicillin, 57
Oak powdery mildew (Sphaerotheca lanestris), 12, Penicillium
22,24 brevicompaclum
Oat mosaic virus, 6 bacteriophages of, 35, 36
Oat seedlings, 169 morphology of viruses and VLPs, 13, 17,50
Oat Victoria blight, see also Helminthosporium vic- transmiSSIOn of viruses and VLPs, 43, 46
toriae, 164 caneo-fulvum, 50
Oligo (dA), 5 chrysogenum, 46, 52, 57, 165
Oligo-dT, 89, 97 bacteriophages of, 35, 36
Oligoisoadenylate synthetase, 8 lytic plaques, 58-59
Oligonucleotide fingerprinting, 226 virus genomes m, 48, 52
Oligopurine tract, 42 virus and VLP morphology In, 8, II, 17,50,
Olpidium 52
brassicae, 6, 20 virus structure in, 47, 53
radicale, 6 virus transmiSSIOn In, 43-46
Oomycetes, 33, 37, 40 citrinum. 17, 37, 58
Oospores, 37 claviforme. 17
Open reading frames (ORF), 89, 244, 247 cyaneo-fub'um. 8. 17, 46. 47
Kluyveromyces lactis, 102 cyclopium, 19, 35
Optical diffraction, 25, 28 funiculosum. 46, 53
Orcinol, 116 virus and VLP morphology, 7-8, 17, 19
Organic acid, 156 multicalor. 17
Orifrep, see Autonomous replication nigricans, 35
ori sequences, 243, 248, 250-251 notalum. 17. 57
Orthomyxoviruses, 32 purpurogenum, 17, 46
Osmotic shock, 118 species, 7-9, 43
Ostracoderma sp., 12, 17 bacteriophages from, 35-36
Oxalate, 156 stoloniferum, 166
Oxalis pes-caprae, 24 bacteriophages of, 35
Oxyskyrin, 157 isometric partIcles In, 37, 38
mixed virus infectIOns. 55-56
PsV-S, 38, 47-48, 50-51, 54
p virus genomes Ill, 47-48
virus morphology m, 8, 17,50
Pachytrichospora transvaalensis, 240 virus replicatIon cycle, 54
Paecilomyces sp., 17, 19 virus transmIssion in, 43-46
PAGE, see Polyacrylamide gel electrophoresis virus ultrastructure, 53
Pale phenotype, 55 variabile, 17, 58
Palladium shadowing, 22 Pentamer, 28
Pancreatic fingerprints, 89 Peptides, see also speCIfIC types
Papovaviruses, 41, 62 alpha, see Alpha factor
SV40,63 beta, 89, 93, 94, 100
Parallel arrays, 26 immunity, 93. 100
298 Fungal Virology

tryptic, 93 plllll,l, 240

Per/conla ctrCllwra, 17, 51 Plcornavlruse,. 39. 62, 63
Penderm, 155, 156 Pigment gene (PIR). 157-158,215
Pentheclal formatIon, 20 I P!lIns. 4
and d-factor, 192-193 Pitch, 25, 26
PelarRrmium domestlcum, 24 Plant pathogens, ,ee also by name. 144
Pen nuclear space, 31 Plant viruses. see al,o plant genera by name, 5. 6,
Perml,)lve conditions, 34 10, 37,41
PETI8, 55, 98 cryptIc vlruse" 49, 52
Petiole, 22 Plaques. lytIc. 33. 58---61
PetIte colomes, 238, 243, 249 Plasmalemma. 26. 28. 43, 53
Peziza Pla,ma membranes. 21, 34
oSlracocierma, 12, 14, 17,25 Plasmid DNA, 101.
species, 25 Plasmid, (fungal), 53
Phage 'A, 4, 35 ARobaclerium, 214
Phage fd, 27 CEN.42
Phage-Infected bactena, 10 In CeratocysIIS ulml, 218-219
Phage Mu host range vanation, 5 In Clavlceps purpurea, 265-266
Phages, ,ee also Bactenophages, 35 ddp I. 268
Iysogemc, 35 dsDNA. 58, 61, 86
Phage T2, 35 evolutIon of, 4
Phage T7, 35 hybnd, 269, 271
Phaseolus vulgar/s, 22, 24, 25 killer, 238
Phase variatIon, 5 Kluyveromyces lacli,I, 101-102
Phenotypes, see also specIfic type, lInear, 265-266
host, 56----61 mltochondnal, ,ee Mitochondnal plasmlds
InhIbitory, 126 from mltochondnal DNA, 249-263
NEX, 98 molecular ,tructure, 239
pale, 55 In Neurospora tetrasperma, 265
poky, 57 nonmitochondnal, 266---269
ragged, 57 nuclear assOCiated, 238, 266---268
slow-growIng, 31 types of, 238, 249
stopper, 57 of unknown assocIation. 238, 268-269
In UstilaRo maydls, 110-116 vs. vIruses, 3
Phialidlc conidia, 223 yeast, 86
Phwlophora, 222 Plasmlds (plant), 5, 36, 214
RramtnlCola, 17,46,227 Plasmodia, 21
radlclcola, 222 Plasmodiophora brasslcae, 18,40
species, 10 Pla,mogeny, 45, 46
hosts, 46 Plalanus acenfolta, 24
ViruS and VLP morphology, 17, 227-228, 230 PlatyhelmInths, II
virus discovery in, 223 Plectovlrus, 22
PhtalophoralGaeumannomyces complex, 222 Pleomorphic particles, 13
PhIal os pores , 223 enveloped, 31-33
Phloem, 155 Plicana fulva, ,ee PeZlza ostracocierma
Phlyctochytnum irregulare, 20 Podospora, 245
Phond flies, 45 ansenna
PhospholIpids composition, Neurospora crassa, 31 cytoplasmiC tram,misslOn, 5, 57
Phosphotungstic aCid, 26, 30 genetIc engineenng potential, 270---272
Phragmldlum sp., 24 mitochondrial DNA, tramposition, 248-249
Phycomycetes, 21 mltochondnal genomes, 238-243
Phyllactinia corylea, 24 mitochondnal plasmlds, 238-240, 243
Physarum polycephalum (,lIme mold), 20, 240 senescence, 5, 256---262, 271
Phytopathogemcity, 57, 61 curvicolla, 239, 240
Phytophthora senescence in, 262
tnfestans, 13, 18,31,40,240 Podovlruses, 35
parasitlca, 33 Poky mutants, 31, 32, 57, 252
Phytoreovirus, 48 PolIovirus, 62
Plchia, 93, 102 Polyacrylamide gel electrophoresis (PAGE), 10,91,
kluyveri, 93 117
Lindneri, 240 SDS, 166
 299

Polyadenylic acid:polyuridylic acid. 9 host, see Ho" protem

PolyamIne synthesis, 100 Interaction, among, 30
Poly(A) ,yntheta,e, 97 killer, 57-58.61. 132-136
Polycistronic mes,ages. 132 matrix, 28
Polycytidylic acid, 9 mitochondrial. 32
Polydnaviruse\, 28, 29 ribosomal. 32. 100
Polyethylene glycol (PEG), 45 ,ecreted. I 03
Polygonum m'icalare, 24 viral. 91-95
Polymerases, 31 Protein tOXInS. ,ee abo ,peclfIc types
mitochondrial, 243 diphtheria. 5
RNA, see RNA polymerases Proteolytic enzymes. 110. 132
Polymorphic particles, 31 Protonophores, 93, 102
Polymyxa species, 6 Protoplasts, 40, 45. 47, 149
Polyoma virus, 39, 62, 63 fusion of. 44, 10 I
POIY-l-ornithine (PLO), 45 mfection of. 231
Polypeptides regeneratIon of. 157. 171
capsid, 3, 24, 91, 99 Prototrophic thallI. 45
toxin, 89, 233 Protoxin. 93. 94
Polyphenol, 187 Protozoa. II. 40
Polyphosphate granules, 21 Proviruses, see also specIfIc types. 4, 121
Polyporus sp., 19 DNA, 42, 55
Polyribocytidylic acid, 8 retroviral, 40
Polyriboinosinic acid:polyribocytidylic acid, 8 Pseudogenes. 5
Polyribonucleotides, 9 Pseudomona~ synngae, 249
Poly U and poly I:poly C antisera, 10 PseudoperonD.lpora cuben.Hs. 24--25
Polyuridylic acid, 9 Pseudoplasmodm, 21
Poly(U)-sepharose, 97 PsV-S (Penicillium stoionijerum Vlru, S). see Pew-
Polyzoa, II cillium stoiollljerum
Positive regulators, 53 Pteridophytes. II
of plasmid DNA replication, 56 Puccmia
Postfusion incompatibility, 44 alli. 15
Posltranscriptional addition, 129 coronata. 15. 164
Postlranscriptional barriers, 270 graminis, 15, 43
Potato mop top virus, 6 heiianthi. 12, 15, 24
Potato virus X, 6, 26, 27 hordei, 19
Potato virus Y, 27 horiana, 15
Potato wart disease, 26 iridis, 24
Potato yellow dwarf virus, 32 madiae, 12
Potexvirus, 27 mal"acearum, 15
Potyvirus, 2, 26, 27 miscanthi, 15
Powdery mildew, see also specific types, 24 oxalidis, 24
Poxviruses, 4 pelargonii, 24
Preinhibitory substance, 124 recondita, 15, 19
Preprotoxin sorghi (com ru,t), 12, 15,24
KluJ"eromyces lactis, 102, 103 striiformis. 15, 19
Saccharomyces cerevisiae, 89, 93, 94, 97 suaveoians, 15
Pretoxin, 133 triticina, 15
Primer extension, 87 Purification, 21
Primers, 96 Pustules, 24
Prokaryotes, 5, II Pyricularia
endosymbiontic, 35-36 grisea. 17
Promiscuous DNA, 247 oryzae, 18,20,44,49,57
Prophage DNA, 4 species, 43
Proteases, 42 Pyriculol, 57
chymotrypsin-like, 94 Pyrophosphate, 95
Protein coat, see Capsid Pythium
Protein kinase, 8 arrhenomonas, 63
Proteins, see also specific types, 10 I butleri, 19
coat, 6 debaryanum, 63
DCCD-binding, 247 species, 63
DNA-binding, 62 sylvaticum. 63
 300 Fungal Virology

ultlmum. 63 to antibIOtiC, 6
to chloramphemcol, 196
cytoplasmiC. 113. 114, 124
Q to dl~ea"e, 206
factors, 94--95
Q-beta rephcatlOn. 100 gamma factor, 94
Q-factors. 231 mhentance, 60, 110-114, 124--125
Quercus agrifolla. 24 toklllertoxms,94--95, 101, 110, 112
speCificities of, 93
Resl,tance peptldes, 93, 100
R Respiratory-deficient mutants, 31
Restnction analysl~, 10 I, 258, 264
Rabbit f3-globm. 270 Reticulocyte lysate system, 30, 124
Race specificity. 31 Reticulum, endoplasmic, see EndoplasmiC reticulum
Ragged (rgd) mutants, 57, 255-256, 262 Retrovlral provlruses, 40
Rat growth hormone, 270 Retroviruses, see also specifIC types, 4
ROY, see Rice dwarf virus Retrovlru,-hke particle" 19,40----42,62
Readmg frames Reverse transcnptase, 40, 42
open, 89, 244, 247 Reverse transcription, of RNA, 5
umdentified (URF), 241, 254 Reversion, 214
Reassociation kmetIcs, 35 rex mutants, 100
Recombination, 251 Rhabdovlruses, 28-30, 32
Red-brown pigment, 187 RhizidlOmyces sp" 18, 37~0
Red clover necrotic mo~aic virus, 6 Rhizoctonia solam. 18,44,51,58,61
Reduced d-factor transmiSSIOn by comdla, 194--196 hypovlrulence of, 144, 149-150
Reduced genomes, 123 plasmlds, 239, 269
Reduction of virulence, 156---158 Rhodosporidlum sp , 134
Regeneration of protoplast, 157, I71 Rhodotorula, 13,35, 134
RegulatIon Rho- mutants in Saccharomyces cerevislae. 249-
of d-factor transmissIOn, 193-197 252
of vegetative mcompatIbllity, 193-194 Ribose, 133
Reiterations of sequence, 152 Ribosomal proteins. 32, 100
Reovlruses, 48, 52, 54, 96 Ribosomal RNA, 32, 33, 135,243,245
dsRNA of, 8 large, 241, 242, 246
Repair of DNA, 251 small, 241, 246
Repeats Rice downy mildew, see Sclerophthora macrospora
direct, 5, 250, 251 Rice dwarf ViruS (ROY), 9
mverted, 87, 101,250,251 Rice necrosIs ViruS, 6
termmal (LTR), 42 Rice plants, 37
Rephcase, 54, 89, 96, 123 Rice stripe vlruse~, 25, 27
RNA, 61 Rifamplcm, 243
Replication (ViruS), 97, 101, 149 Rigid rod particle~, 12, 21-25
asynchronous, 54 Ri plasmids, 36, 214
autonomous, 248, 258, 271 RNA, 87-91
control of, 99-100 antisense, 62
cycle of, 54--55 defective, 47, 49
DNA, 56, 61 DI particle, 99
dsRNA, 51, 53, 61 double-stranded, see Double-stranded RNA
host growth and, 49-54 large nbosomal (lr), 262
mhibitIon of, 254 messenger, see Messenger RNA
nonmitochondnal plasmids, 266---268 processmg of, 243-245
Q beta, 100 replicatIOn of, 61, 116
RNA, 61---62, 116 reverse transcnptIon of, 5
RNA polymerases and, 54--56 nbosomal, see Ribosomal RNA
runaway, 53, 60 satelhte, 47, 49, 120
self-, see Self-replicatmg smgle-Mranded Circular (sse), 261
semlconservatlve, 54, 131 transcnption, see TranscriptIon RNA
signals in, 99 RNA bacteriophages, 4
of USlilago viruses, 128, 131, 133 RNA maturase, 243-245
of viruses m fungi, 6---11 RNA polymerase, 47, 55, 56, 63
Resistance fungal vesicles, 149, 153
 301

mitochondrial, 243 morphological types, 14, 19

U,~tilllgo maydis viruses. 130--132 M particles. 59. 62. 86--91,93-94, 100
virIon-associated. 48. 54--55, 95 replication, 51. 54. 99-100
RNA replica~e, 54, 89. 96 retroviru~-like particle~, 40, 42
proof-readmg activities in, 61 Sc V transcriptase, 95-97
RNase, 131. 135 S particles, 86, 91, 95
RNA-splicing enzymes. 241 toxin" 93-95
RNA viruses. 62 transcription, 55
replicative forms of. 116 transcripts, 97
single-stranded. 37 Ty elements, 40, 42
Rocket immunoelectrophoresis, 9 viral dsRNAs, 49, 87-91
Rod-~haped dsRNA viruses. 48-49 viral proteins, 91-95
Rolling circle mechanism, 5 VLPs, see also specific particles
Rosa sp .. 24 chodati, 33
Rotifers, II diastaticu,I', 19, 33
Rough endoplasmic reticulum, 20 exiguus, 240
Rozella allomycis, 20 iralicus, 269
rRNA, see Ribosomal RNA ludwigii, 14
Rubus vitiJolia, 24 microellipsoideus, 33
Rugulosin, 157 rouxii. 269
Runaway replication, 53, 60 species, 14,21,33,93,102
Runner hyphae, 222 tel/uris, 240
Rusts, see also specific types, 22, 24, 25 unarum, 240
bean, see Uromyces phaseoli unisporum, 240
com, see Puccinill sorghi uvarum, 19
sunflower, see Puccinia helianthi Saccharomycopsis, see Yarrowia lipolytica
S-adenosylmethionine (SAM), 95

s Salmonella, 5. 268
typhimurium, 4
SAM, see S-Adenosylmethionine
Saccharomyces Sanitation, 206
capensis, 19 Saprolegnia sp" 39, 240
carisbergensis, 13, 21, 33, 35, 268 Saprophytic ability, 61
cerevisiae, 56, 62, 128,269 Satellite DNA, 20
cold sensitivity, 58-61 Satellite RNA, 47, 49, 120
cultures, 53 double-stranded, 42, 57, 60, 130
extrachromosomal inheritance, see also Mito- Satellites, 56, 215
chondrial genome, 238 of helper viruses, 58
host/virus interactions, 99-101 Satellite viruses, 47, 48
killer systems, see Killer systems tobacco necrosis, 6
large T antigen, 63 Scenedesmus armatus, 40
mating pairs, 46 Schizochytrium aggregatum, 13, 33, 34
mitochondrial DNA, 243, 247-249 Schizophyllum commune, 15, 20
mitochondrial gene mutants, 238, 243, 249- Schizosaccharomyces pombe, 239-242, 245, 269
252 Sclerophthora macrospora, 18. 37, 40
mitochondrial genome, 238, 239. 243-246 Sclerotia. 43, 149
nuclear associated plasmids, 266 Sclerotium cepivorum, 18, 43, 46
nuclear genome. 55. 247-248 Scolytidae, 178
particle transmission, 44 Scolytus scolytus. 202
respiration, 249-250 Scopulariopsis sp" 19
RNA processing, 243-245 Screening for fungal viruses, see also specific types,
transmission, 46 9-11
tRNAs, 245-246 ScV, see Saccharomyces cerevisiae viruses
2 f.lm plasmid, 218-219, 238, 266 ScV transcriptase, 95-97
viruses, see Saccharomyces cerevisiae viruses SDS polyacrylamide gel electrophoresis, 91, 166
(ScV) sec mutants, 94, 100
cerevisiae viruses (ScV) Secondary metabolItes, 57
capsid proteins, 91-93 Secretion vesicles, 94
host interaction, 99-100 Secretory proteins, 103
L particles, 47, 48, 54, 86, 87, 89-91 Sectoring, 146
messenger RNAs in vivo, 97 Segmentation patterns, 1/7, 121
302 Fungal Virology

Segregatlon, 114, 232 Spheroplasts, 93, 95

non-MendelIan, 250 Spikes, 37
Self-inhibition, 231 Spliced mRNA, 63
Selfish DNA, 61 Splicing homeostasis, 245
Selfish dsRNA, 61 Sponges, 11
Self-replicating shuttle vectors, 271-272 Spongiform encephalopathies, 5
Semiconservative replication, 54, 131 Spongospora sp" 6
Senescence, 5, 26, 57 Sporangia, 26, 31, 39
in Cochlioholus heterostrophus, 263 zoospores, 33-34
in Neurospora intermedia, 263 Spore-bearing structure, 21
in Podospora anserina, 256--262 Spores, 6, 44
in Podospora curvicolla, 262 asexual, see Asexual spores
~-Senescence DNA, 260 sexual, see Sexual spores
SensitivIty Sporogenesis, 20
to cold, 58--61 Sporophores, 45, 59, 60
to heat, 22 cell walls of, 30
to temperature, 98, 133 Sporulation, 145, 154
Sephadex GIOO, 22 ssc, see S ingle-stranded circular
Septa, 43 ssDNA, see Single-stranded DNA
Septal pores, 43 ssRNA, see Single-stranded RNA
Sequence homology, 87, 121, 126 Starvation medium, 34
Sequence reiterations, 152 Stationary phase of growth, 51. 130, 134
Serological comparisons, 24 Statolon, 8
Sexual spores, 21, 42, 53 Stemphylium hotryosum, 18
transmission by, 43-44 Stem structures, 87
Shiitake mushroom, see Lentinus edodes Stopper mutants, 32, 33, 57, 252-256, 262
Shock, osmotic, 118 Strand switching, 56
Shuttle vectors, self-replicating, 271-272 Stroma, 156, 157
Simian virus 40 (SY40), 135 Strongwellsea magna, 13, 30
Single COnIdial isolates, 174 Styloviruses, 35
Single-membrane-bound vesicles, 53 Subcellular organelles vs. viruses, 3
Single-shelled capsids, 47 Subgenomic mRNA, 87
Single-stranded DNA, 27 Substomatal vesicles, 24, 26
circular, 22, 36 Subviral particles, 54, 96
Single-stranded RNA, 29, 37 Sucrose density gradients, 20, 130
circular (ssc), 261 Sugar beet yellow virus, 27
isometric, 39 Suicide mutants, 60, 61
transcripts of, 30 Sunflower rust, see Puccinia heliaflthi
ski, see Superkiller Superkiller mutants, 51, 59, 97,100
Skyrin, 156, 157 Suppression, 99, 214
SlIme molds, 20, 21, 40, 240 Suppressive dsRNA, 99
Slow-growing mutants, 31, 36 Suppressive sensitive strains (Saccharomyces cere-
Smallpox, 2 visiae), 99
Small rRNA, 241 Surinam plasmid, 265
Sobemovirus, 39 Switching of strands, 56
Soil-borne wheat mosaic virus, 6, 23 Synchytrium endohioticum, 6, 12, 26
Solid phase hybridization, 226 Synnematal formation, 20 I
Solution hybridization, 226 Synnematospores, 201
Somatic incompatibility, 20
Southern experiments, 266
S particles, 86, 91, 95, 99 T
SPE2 gene, 100
Specificity Take-all diseases, see also Gaeumannomyces; Phi-
dual, 115 alophora, 222-223, 231
race, 31 Take-all fungi, see Gaeumannomycej'; Phialophora
toxin, 133-134 Tandem duplication, 99
virus-vector, 6 T-DNA, 5, 36
Sphaerotheca Tectiviruses, 41
juliginea, 14, 25 Teleomorph~, 21
lanestrts, 12, 22, 24 Temperate Viruses, 49
S phase, 53 Temperature sensitivity, 98, 133
 303

cold,58-61 primer bmding .,ite for. 42

heat. 22 Tran~fer
Terminal nucleoside, 129 d'-factor, 212
Terminal repeats. 42 dsRNA, 148.212
Tetrads. 110. 125 heterokaryon. 110. /13. 114
Tetrahymena hypovirulence. 148
telomeres, 101 Transformation, 101
thermophila. 270 Transformed strain, 125
Thallus. 6 Translation. 136, 245-246
Thanatephorus cucumeris, 19 dsRNA. 126
Thielaviopsis basicola, 18, 25, 49 in vitro, 124. 129
Thraustochytrium product of, 124
aureum, 18 Translational frameshifts, 42
species, 13, 33, 34, 39 Transmissible diseases, 57-61, 146
Tilletiopsis sp., 15 of Helminthosporium I'ictoriae, 163-175
Ti plasmids, 5. 214 Transmission, 42-46, 228-231
Tissue tropisms, 63 into ascopores. 228-229
TMV. see Tobacco mosaic virus with cell-free virus preparation~, 45-46
TNV, see Tobacco necrosis virus in conidia. 228
Tobacco mosaic virus (TMV). 2, 22-24 of d-factor, 193-197
aucuba strain of. 25 of d2-factor, 212-216
Pythium as host, 63 of dsRNA. 212-216
Tobacco necrosis virus (TNV), 6, 23, 25 of Heiminthosporium victoriae, 171-174
Pythium as host, 63 by heterokaryons, 44-45
Tobacco stunt virus, 6, 23, 52 by heteroplasmons, 44-45
Tobraviruses, 22. 23 by hyphal anastomosis, 229-230
Tolerance
during hyphal growth, 42-43
chloramphenicol (CR). 183, 185
by seeds, 49
MBC, 181, 183, 185
by sexual spores, 43-44
Tomato bushy stunt virus. 10.25,37,39
vertical, 34, 39
Tomato golden mosaic virus, 36
Transposable elements, 5,40, 42, 61
Tonoplasts, 26
Transposition
Topology of DNA, 40
DNA, 5, 247-249
Torulopsis
to nuclear DNA, 260
glabrata, 19, 134,239,240,269
plasmids, 239, 269 in yeast, 40-42
species, 93, 102 Transposons, 5
Totiviruses (Totiviridae), 48, 49, 52, 54, 55, 225 Traustochytrium aureum, 40
Toxins, see also specific types, 9,57, 126, 156 Trichop/usia nigranulosis, 29
alpha, see Alpha factor Trichorsporum cutaneum, 240
beta, 89, 100 Tricomaviruses, 29, 30
diphtheria protein, 5 Tricothecium
host-specific, 156, 164, 169 roseum, 18
killer, see Killer systems; Killer toxins species, 19
polypeptide, 89, 233 Trilamellar unit membrane, 31
protein, see Protein toxins tRNA, see Transcription RNA
resistance factor and, 93-95 True fungi (Eumycota), 21
secretion of, 134 Trypanosoma brucei, 270
specificity of, 133-134 Trypanosomes, antigenic variation, 5
TPCK,94 Tryptic peptides, 91, 93
Trans-acting FLP gene, 268 Tubular particles. 26, 28, 31
Transcriptase, 89, 96, 123, 131 Tubular viruses, 28
reverse, 40, 42 Tulip breaking virus, 2
ScV,95-97 Tumor (crown gall), 5
Transcription, 54, 62, 99,102,121,131,243 Two-dimensional immunodiffusion, 9
dsRNA,54 tya, 42
reverse, 5 tyb,42
RNA, 5 Ty elements, see Transposable elements
ssRNA,30 Tymovirus, 39
Transcription RNA, 10,241,243,245,246 Tyrosinase. 60
mitochondrial, 246 Ty-VLPs, 40, 42
304 Fungal Virology

u Vertebrates. II. 29
Vertical (intracellular) transnllSslOn, 34. 39
Ultra,tructural studies. 53-54 Verticil/ium
Unidentified reading frames (URF), 241, 254 dahlia. 18
Universal code, 246 jimgicola. 13,18.30
Up-mut factor, 179, 190, 191 malthousei. 18
Uranyl acetate, 26 Vesicles, see abo ,peclflc type" 43
Uredospores, 24, 43 fungal. 149, 153
URF, see Unidentified reading frames
membrane-bound, 53
U-rich end, 91, 96
secretion. 94
Uromyces
duras. 15 substomatal, 26
labae. 12, 24 Vesicular ,omalltls virus, 29
lopecuri. 15 Vesiculate structures, 30
phaseoli. 12, 15,22,24-26 Vicia laba. 24
polygoni. 24 Victoria blight of oats, see also Helmll1rhosporillm
Ustilaginales, 134 \'ictorwe, 164---165
Ustilago Victorin, 164. 169
cynodotztls. 240 Vigna sine/l.li.I, 22, 24
maydis. 15,43
Vinca rosea. 270
genetics of interstrain inhibition, 110---114
Virion-associated RNA polymerases, 54---55
incompatibility, 115-116
viruses of, see Ustilago maydis viruses Viroids. charactenzed, 5
virus/host interaction, 137-138 Virulence, see abo PathogenicIty, 232
specIes, 57 expression of, 154---159
replication of, 131 mechanisms, 156----158
Ustilago maydis viruses tests, 149--150
dsRNA Viruses, fungal
encapsidation, 93, 117, 120
characterized, 3-------D
relatedness of segments, 126----130
discovery, 2-3, 7
,egments characterized, 120----126
isometric double-stranded RNA, see hometric
interaction with host, 137-138, 233
killer proteins (toxins) dsRNA mycovlruses
action in vivo and in vitro, 134-136 morphology, see abo speCIfIC types
nature of, 132-133, 137-138,231 bacteriophages, 35-36
specificity, 133--134 characteri,lIcs, 12-19
killer strains, 57-58, 86, 215 club-shaped partIcles, 31
mixed infections, 56 herpes-like particles, 33-34
morphological types, 15, 49, 60 pleomorphic, enveloped partIcle" 31-33
nucleic acids, see also specific kinds, 116----117 rods, bacilliform. 28-31
RNA polymerase, 54, 130----132
rods, flexuous, 25-28
transcription, 55
rods, rigid, 21-25
transmission through spores, 43, 109--114
as vectors, 62-------D4
UV irradiation, 47
virus-like particles (VLPs), see Virus-like
particles
v Virus-like lesions, 22
Virus-like particles (VLPs), see also speCIfic types
Vacuoles, 25, 26, 53 characterized, 12-19
v-c groups cores, 20, 26, 31, 33, 169
Ceratocystis, 193-194,201,203,206 geminate partIcles, 13, 31, 36
Gaeumannomyces. 229--230 heads and tails, 34---35
Vectors, see also Hosts, 49, 62--64, 101, 103
isometric. 37-40
Dutch elm disease, 197,200----202,206,217
in lower fungI, 40
Endothia parasitica. 217
in Neurospora crassa, 31-33
Gaeumannomyces. 230
Vegetative compatibility, 44, 229--230 in Saccharomyce.1 species, 33
Vegetative death, 57 in yeasts, 34-35
Vegetative incompatibility, 44, 203 Virus-specific sera, 9
regulation of, 193--194 VLP, see Virus-like particles
 305

w y

Waimea Falls plasmid, 265 Yarrowia lipoIYf/ca, 14, 49, 86, 91, 93
Wall receptor, 93 killer ~ystem~, 86, 91, 93
Watery stipe (La France disease), 7, 58---61 Yeast-like state, 202
Western blots, 91 Yeast plasmlds, see abo genera by name, 86, 101,
WFR, see Wide-diameter flexuous rods 239, 269
Wheat-germ translation system, 91 Yeasts, see also genera by name, 42, 62, 63
Wheat mosaic virus, 6, 23
cells of, 21
Wheat spindle streak virus, 6
centromeres of, to I
Wheat take-all fungus, see Gaeumannomyces
d~RNA of, 129,215
graminis
Wheat yellow mosaic virus, 6 genetic engineering of, 269-270
Whiteflies, 36 killer sy~tems, ,ee also Ktller systems, 86, 93
Wide-diameter flexuous rods (WFR), 26 transposition in,S, 40---42
Wound periderm, 156 vectors, JO I
W-reaction, 193 VLPs from, 34---35
W2 to II isolate, see Ceratocystis ulmi, isolates Yeast-stage, 197
Yellow fever, 2
x
z
X-disease (La France disease), 7, 58---61
Xenopus laevis, 243, 270
Zea mays, see Corn
X-ray diffraction, 25
Zoospores, 20,21,31,39
Xylem (pathogenic) behavior of d-infected isolates,
sporangial release, 33-34
197-199
Xylem (pathogenic) phase of Dutch elm disease, Zygomycotina, 21, 40
200, 202-203, 217 Zymogen, 21