Methods in

Molecular Biology 1068

Kazushige Touhara Editor

Pheromone
Signaling
Methods and Protocols
METHODS IN M O L E C U L A R B I O LO G Y

Series Editor
John M. Walker
School of Life Sciences
University of Hertfordshire
Hatfield, Hertfordshire, AL10 9AB, UK

For further volumes:

http://www.springer.com/series/7651
 Pheromone Signaling

Methods and Protocols

Edited by

Kazushige Touhara
Department of Applied Biological Chemistry, Graduate School of Agricultural
and Life Sciences, The University of Tokyo, Tokyo, Japan
Editor
Kazushige Touhara
Department of Applied Biological Chemistry
Graduate School of Agricultural and Life Sciences
The University of Tokyo
Tokyo, Japan

ISSN 1064-3745 ISSN 1940-6029 (electronic)

ISBN 978-1-62703-618-4 ISBN 978-1-62703-619-1 (eBook)
DOI 10.1007/978-1-62703-619-1
Springer New York Heidelberg Dordrecht London

Library of Congress Control Number: 2013947162

Springer Science+Business Media, LLC 2013

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is
concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction
on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation,
computer software, or by similar or dissimilar methodology now known or hereafter developed. Exempted from this
legal reservation are brief excerpts in connection with reviews or scholarly analysis or material supplied specifically for
the purpose of being entered and executed on a computer system, for exclusive use by the purchaser of the work.
Duplication of this publication or parts thereof is permitted only under the provisions of the Copyright Law of the
Publishers location, in its current version, and permission for use must always be obtained from Springer. Permissions
for use may be obtained through RightsLink at the Copyright Clearance Center. Violations are liable to prosecution
under the respective Copyright Law.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not
imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and
regulations and therefore free for general use.
While the advice and information in this book are believed to be true and accurate at the date of publication, neither
the authors nor the editors nor the publisher can accept any legal responsibility for any errors or omissions that may be
made. The publisher makes no warranty, express or implied, with respect to the material contained herein.

Printed on acid-free paper

Humana Press is a brand of Springer

Springer is part of Springer Science+Business Media (www.springer.com)
Preface

Many animal species utilize chemosensory systems to detect a variety of chemical sub-
stances in the external environment. When these substances convey information regard-
ing food, predators, species, or sex, it is considered chemical communication between
organisms. The notion of chemical communication was first proposed by Charles Darwin
in the late nineteenth century, in conjunction with the physical signal-mediated commu-
nication described for the visual and auditory systems. At approximately the same time,
Jean Henri Fabre described the attraction of male moths by conspecific females, and
proposed that the cue was not a visual signal but some kind of smell. The concept of
chemical interaction was scientifically recognized when Albrecht Beth proposed in 1932
the term ectohormone for substances that functioned like a hormone but were secreted
into the external environment.
The first molecular evidence for a substance that was able to mediate chemical com-
munication was obtained when Butenandt discovered bombykol, the sex pheromone of the
silk moth Bombyx mori, in 1959, which is released by females and elicits the full sequence
of sexual behavior in male moths [1]. The term pheromone was then defined by Karlson
and Luscher as substances which are secreted to the outside by an individual and received
by a second individual of the same species, in which they release a specific reaction, for
example, a definite behavior or developmental process [2]. Pheromone is a combined
term from the Greek words pherein for to bear or transfer and hormao for to excite.
Specific reactions by pheromones are generally categorized into two types: one that releases
a visible stereotyped behavior such as attraction or avoidance and a second that elicits endo-
crinological changes such as puberty acceleration and estrous cycle regulation.
A pheromone appears to be a type of smell recognized by some species; however, the
definition specifically avoids any use of the terms odor or odorant. It does not even
define pheromones as being volatile. Karlson and Luscher were shown to be insightful in
this regard, as future research was to find that pheromones can be nonvolatile substances
including relatively large organic compounds, peptides, and proteins. Recently, there has
been some contention in the field as to whether the definition of pheromones should be
extended, as pheromones appear to play far more diverse functional roles than previously
defined. Possible criteria for a more inclusive definition of a pheromone may be that (1)
pheromones are released by one individual and received by conspecifics; (2) pheromones
themselves may convey meaningful information including sex, strain, and species to the
receiver; and (3) pheromonal effects are stereotyped or innate. It is of note that the same
molecule may be utilized as a pheromone in different species. It is also important to note
that pheromones are distinct from individual chemical information that may be learned [3].
The mechanisms underlying pheromone action are also of significance. A receptor that
senses insect sex pheromone was first discovered in the silk moth. The bombykol receptor
was identified as a male-specific seven transmembrane protein that appeared to belong to
the olfactory receptor superfamily, and that which showed a highly specific and sensitive

v
vi Preface

response to bombykol [4]. It was later found that the mechanism of signal transduction was
distinct from the one observed in vertebrates where the insect pheromone receptors formed
a heteromeric ligand-gated cation channel complex [5, 6]. The electrical signal is transmit-
ted to specific glomeruli in the antennal lobe, the first center of the olfactory sensory path-
way, and integrated in the higher brain to elicit sexual behavior.
In vertebrates, volatile pheromones are detected by the main olfactory system in addi-
tion to the secondary olfactory system called the vomeronasal organ located at the bottom
of the nasal cavity [7]. The vomeronasal system was identified in amphibians, rodents, and
some primate species and detects not only volatile cues but also nonvolatile pheromones.
The old-world monkeys and humans do not possess a vomeronasal organ, but rather detect
pheromones in the main olfactory system, if at all. The receptors belong to the G protein-
coupled receptor superfamily including the olfactory receptor (OR), trace amine-associated
receptor (TAAR), and vomeronasal receptor (VR). The signal is further transmitted to the
main or accessory olfactory bulb and subsequently conveyed to the higher brain areas
responsible for specific behavioral and neuroendocrinological output.
The development of molecular biology and neuroscience techniques has permitted the
precise method of pheromone action from the molecule to the receptor to be defined [7].
The next aim of pheromone research is to further our understanding of how pheromone
information is integrated in the brain, and how specific behavioral output is regulated by
the neural circuitry. The pheromone system is an ideal model for understanding the neural
networks that lead to various adaptive behaviors. Another important area to investigate is
whether humans also impart chemical communication via pheromones. Given humans have
evolved visual and auditory senses as their major cue, it seems appropriate to suggest that
meaningful chemical communication does not exist in humans today. However, it is also
true that humans often feel appeased by the smell of infants. Menstrual synchrony in females
living closely together is also another example of potential chemical communication within
the human species [8]. In contrast, rodents appear to heavily rely on chemical communica-
tion throughout life. A full list of required compounds and their receptors, however, is yet
to be fully revealed. The mechanism underlying how a selective pressure occurred on sen-
sory systems so that each animal developed a sophisticated and individually tailored chemi-
cal communication strategy remains unknown. The strategy is thought to be optimal for
the survival of each organism under their living environment. In this regard, the evolution-
ary history of changes in pheromone molecules and genome related to pheromone recep-
tion will tell us where we came from and in which direction we are headed.
Over the last decade, a great deal of progress has been made into understanding the
molecular mechanisms underlying pheromone action, largely due to the discovery of recep-
tor genes and the advancement of imaging techniques. The major goal of Pheromone
Signaling is to provide experimental methods and protocols that allow us to perform pher-
omone research in a variety of organisms ranging from invertebrates to vertebrates. The
book will cover a wide spectrum of experimental approaches necessary for handling phero-
mone molecules, measuring receptor response and neural activation, and analyzing behav-
ioral output (Figure 1). Pheromone research requires multidisciplinary approaches including
aspects of organic chemistry, biochemistry, molecular biology, electrophysiology, and
behavioral science. On first glance, the transdisciplinary aspects of pheromone research may
 Preface vii

Pheromone Part I Chapter

-Insect (moth, fly) - 1, 2
-Mouse - 3, 4
-Fish -5
-Nematode -6

Receptor Part II
-Genomics - 7,
-Functional assay - 8, 9
-Gene knock-out - 10

Neural response Part III

-Electrophysiology - 11, 16, 17
-Ca-imaging - 12, 13, 14, 15
-Gene expression - 18

Behavior Part IV
Endocrine effect -Insect (fly) - 19
-Nematode - 20, 21
-Fish - 22
-Mouse - 23, 24, 25
-Rabbit - 26
-Human - 27

Fig. 1 The theme and a scheme for the book

seem overwhelming. This should not sidetrack investigators from delving into this field of
research. This book will describe in detail the methodologies and techniques utilized in
laboratories all over the world, making them accessible to those who want to begin investi-
gation in the area of pheromone research.

Tokyo, Japan Kazushige Touhara

References
1. Butenandt A, Beckmann R, Stamm D, Hecker E 5. Sato K, Pellegrino M, Nakagawa T, Nakagawa
(1959) Uber den Sexuallockstoff des T, Vosshall LB, Touhara K (2008) Insect olfac-
Seidenspinners Bombyx mori, Reindarstellung tory receptors are heteromeric ligand-gated ion
und Konstitution. Z Naturforsch 14b:283284 channels. Nature 452:10021006
2. Karlson P, Luscher M (1959) Pheromones: a 6. Wicher D, Schafer R, Bauernfeind R, Stensmyr
new term for a class of biologically active sub- MC, Heller R et al (2008) Drosophila odorant
stances. Nature 183:5556 receptors are both ligand-gated and cyclic-
3. Wyatt TD (2010) Pheromones and signature nucleotide-activated cation channels. Nature
mixtures: defining species-wide signals and vari- 452:10071011
able cues for identity in both invertebrates and 7. Touhara K, Vosshall LB (2009) Sensing odor-
vertebrates. J Comp Physiol A 196:685700 ants and pheromones with chemosensory
4. Nakagawa T, Sakurai T, Nishioka T, Touhara K receptors. Annu Rev Physiol 71:307332
(2005) Insect sex-pheromone signals mediated 8. McClintock MK (1971) Menstrual synchorony
by specific combinations of olfactory receptors. and suppression. Nature 229:244245
Science 307:16381642
Contents

Preface. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v
Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiii

PART I SAMPLING, CHROMATOGRAPHIC PURIFICATION, AND CHEMICAL

ANALYSIS OF PHEROMONE MOLECULES
1 Female Moth Pheromones . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Tetsu Ando
2 Pheromones in the Fruit Fly . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
Jacqueline S.R. Chin and Joanne Y. Yew
3 Analysis of Volatile Mouse Pheromones by Gas Chromatography
Mass Spectrometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
Milos V. Novotny and Helena A. Soini
4 Murine Nonvolatile Pheromones: Isolation of
Exocrine-Gland Secreting Peptide 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
Hiroko Kimoto and Kazushige Touhara
5 Chemical Analysis of Aquatic Pheromones in Fish . . . . . . . . . . . . . . . . . . . . . . 55
Michael Stewart, Cindy F. Baker, and Peter W. Sorensen
6 Analysis of Ascarosides from Caenorhabditis elegans Using
Mass Spectrometry and NMR Spectroscopy . . . . . . . . . . . . . . . . . . . . . . . . . . 71
Xinxing Zhang, Jaime H. Noguez, Yue Zhou, and Rebecca A. Butcher

PART II APPROACHES TO CHARACTERIZE GENE AND FUNCTION

OFPHEROMONE RECEPTORS
7 Identification of Chemosensory Receptor Genes
from Vertebrate Genomes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
Yoshihito Niimura
8 Functional Assays for Insect Olfactory Receptors in Xenopus Oocytes . . . . . . . 107
Tatsuro Nakagawa and Kazushige Touhara
9 A Protocol for Heterologous Expression and Functional Assay
for Mouse Pheromone Receptors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
Sandeepa Dey, Senmiao Zhan, and Hiroaki Matsunami
10 Genetic Manipulation to Analyze Pheromone Responses: Knockouts
of Multiple Receptor Genes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
Tomohiro Ishii

ix
x Contents

PART III METHODS IN MEASURING NEURAL RESPONSES TO PHEROMONES

11 Electroantennogram and Single Sensillum Recording in Insect Antennae. . . . . 157
Shannon B. Olsson and Bill S. Hansson
12 Calcium Imaging of Pheromone Responses in the Insect Antennal Lobe . . . . . 179
Susy M. Kim and Jing W. Wang
13 Live Cell Calcium Imaging of Dissociated Vomeronasal Neurons. . . . . . . . . . . 189
Angeldeep Kaur, Sandeepa Dey, and Lisa Stowers
14 Whole-Mount Imaging of Responses in Mouse Vomeronasal Neurons . . . . . . 201
Pei Sabrina Xu and Timothy E. Holy
15 Calcium Imaging of Vomeronasal Organ Response Using Slice Preparations
from Transgenic Mice Expressing G-CaMP2 . . . . . . . . . . . . . . . . . . . . . . . . . . 211
C. Ron Yu
16 The Electrovomeronasogram: Field Potential Recordings in the Mouse
Vomeronasal Organ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221
Trese Leinders-Zufall and Frank Zufall
17 Electrical Recordings from the Accessory Olfactory Bulb
in VNO-AOB Ex Vivo Preparations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237
Julian P. Meeks and Timothy E. Holy
18 Pheromone-Induced Expression of Immediate Early Genes in the Mouse
Vomeronasal Sensory System. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 247
Sachiko Haga-Yamanaka and Kazushige Touhara

PART IV PROTOCOLS IN ANALYZING BEHAVIORAL AND ENDOCRINE

CHANGES ELICITED BY PHEROMONES
19 Insect Pheromone Behavior: Fruit Fly . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 261
Daisuke Yamamoto, Soh Kohatsu, and Masayuki Koganezawa
20 Quantitative Assessment of Pheromone-Induced Dauer Formation
in Caenorhabditis elegans. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 273
Scott J. Neal, Kyuhyung Kim, and Piali Sengupta
21 Acute Behavioral Responses to Pheromones in C. elegans
(Adult Behaviors: Attraction, Repulsion) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285
Heeun Jang and Cornelia I. Bargmann
22 Behavioral Analysis of Pheromones in Fish . . . . . . . . . . . . . . . . . . . . . . . . . . . 293
Peter W. Sorensen
23 Analysis of Male Aggressive and Sexual Behavior in Mice. . . . . . . . . . . . . . . . . 307
Takefumi Kikusui
24 Assessment of Urinary Pheromone Discrimination, Partner Preference,
and Mating Behaviors in Female Mice . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 319
Olivier Brock, Julie Bakker, and Michael J. Baum
 Contents xi

25 Assessing Postpartum Maternal Care, Alloparental Behavior,

and Infanticide in Mice: With Notes on Chemosensory Influences . . . . . . . . . 331
Kumi O. Kuroda and Yousuke Tsuneoka
26 Testing Smell When It Is Really Vital: Behavioral Assays of Social Odors
in the Neonatal Mouse . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 349
Benoist Schaal, Syrina Al An, and Bruno Patris
27 An Assay for Human Chemosignals. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 373
Idan Frumin and Noam Sobel

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 395
Contributors

SYRINA AL AN Centre des Sciences du Got, CNRS (UMR 6265), Universit de

Bourgogne, Dijon, France
TETSU ANDO Graduate School of Bio-Applications and Systems Engineering, Tokyo
University of Agriculture and Technology, Tokyo, Japan
CINDY F. BAKER National Institute of Water & Atmospheric Research Ltd., Hamilton,
New Zealand
JULIE BAKKER Groupe Interdisciplinaire de Genoproteomique Appliquee Neurosciences,
University of Liege, Liege, Belgium
CORNELIA I. BARGMANN Howard Hughes Medical Institute, The Rockefeller University,
New York, NY, USA
MICHAEL J. BAUM Department of Biology, Boston University, Boston, MA, USA
OLIVIER BROCK Netherlands Institute of Neuroscience, Amsterdam, The Netherlands
REBECCA A. BUTCHER Department of Chemistry, University of Florida, Gainesville, FL, USA
JACQUELINE S.R. CHIN Temasek Life Sciences Laboratories, Singapore, Singapore;
Department of Biological Sciences, National University of Singapore, Singapore
SANDEEPA DEY Department of Molecular Genetics and Microbiology, Duke University
Medical Center, Durham, NC, USA; Department of Cell Biology, The Scripps Research
Institute, La Jolla, CA, USA
IDAN FRUMIN Department of Neurobiology, Weizmann Institute of Science, Rehovot, Israel
SACHIKO HAGA-YAMANAKA Stowers Institute for Medical Research, Kansas City, MO, USA
BILL S. HANSSON Department of Evolutionary Neuroethology, Max Planck Institute for
Chemical Ecology, Jena, Germany
TIMOTHY E. HOLY Department of Anatomy and Neurobiology, Washington University in
St. Louis, St. Louis, MO, USA
TOMOHIRO ISHII Department of Cell Biology, Graduate School of Medical and Dental
Science, Tokyo Medical and Dental University, Tokyo, Japan
HEEUN JANG The Rockefeller University, New York, NY, USA
ANGELDEEP KAUR Department of Cell Biology, The Scripps Research Institute, La Jolla,
CA, USA
TAKEFUMI KIKUSUI School of Veterinary Medical Sciences, Azabu University,
Sagamihara, Japan
KYUHYUNG KIM Department of Brain Science, Daegu Gyeongbuk Institute of Science and
Technology (DGIST), Daegu, South Korea
SUSY M. KIM Neurobiology Section, Division of Biological Sciences, University of
California San Diego, La Jolla, CA, USA
HIROKO KIMOTO Biological Research Laboratories, R&D Division,
Daiichi Sankyo Co., LTD., Tokyo, Japan
MASAYUKI KOGANEZAWA Department of Developmental Biology and Neurosciences,
Tohoku University Graduate School of Life Sciences, Sendai, Japan

xiii
xiv Contributors

SOH KOHATSU Department of Developmental Biology and Neurosciences, Tohoku

University Graduate School of Life Sciences, Sendai, Japan
KUMI O. KURODA Unit for Affiliative Social Behavior, RIKEN Brain Science Institute,
Saitama, Japan
TRESE LEINDERS-ZUFALL Department of Physiology, School of Medicine, University of
Saarland, Homburg, Germany
HIROAKI MATSUNAMI Department of Molecular Genetics and Microbiology, Duke Institute
for Brain Sciences, Duke University Medical Center, Durham, NC, USA; Department of
Neurobiology, Duke Institute for Brain Sciences, Duke University Medical Center,
Durham, NC, USA
JULIAN P. MEEKS Department of Neuroscience, UT Southwestern Medical Center, Dallas,
TX, USA
TATSURO NAKAGAWA Biological/Pharmacological Research Laboratories, JAPAN
TOBACCO INC., Central Pharmaceutical Research Institute, Osaka, Japan
SCOTT J. NEAL Department of Biology, Brandeis University, Waltham, MA, USA;
National Center for Behavioral Genomics, Brandeis University, Waltham, MA, USA
YOSHIHITO NIIMURA Department of Bioinformatics, Medical Research Institute, Tokyo
Medical and Dental University, Tokyo, Japan
JAIME H. NOGUEZ Department of Chemistry, University of Florida, Gainesville, FL, USA
MILOS V. NOVOTNY Department of Chemistry, Institute for Pheromone Research, Indiana
University, Bloomington, IN, USA
SHANNON B. OLSSON Department of Evolutionary Neuroethology, Max Planck Institute
for Chemical Ecology, Jena, Germany
BRUNO PATRIS Centre des Sciences du Got, CNRS (UMR 6265), Universit de
Bourgogne, Dijon, France
BENOIST SCHAAL Centre des Sciences du Got, CNRS (UMR 6265), Universit de
Bourgogne, Dijon, France
PIALI SENGUPTA Department of Biology, Brandeis University, Waltham, MA, USA;
National Center for Behavioral Genomics, Brandeis University, Waltham, MA, USA
NOAM SOBEL Department of Neurobiology, Weizmann Institute of Science, Rehovot, Israel
HELENA A. SOINI Department of Chemistry, Institute for Pheromone Research, Indiana
University, Bloomington, IN, USA
PETER W. SORENSEN Department of Fisheries, Wildlife, & Conservation Biology,
University of Minnesota, St. Paul, MN, USA
MICHAEL STEWART National Institute of Water & Atmospheric Research Ltd.,
Hamilton, New Zealand
LISA STOWERS Department of Cell Biology, The Scripps Research Institute, La Jolla,
CA, USA
KAZUSHIGE TOUHARA Department of Applied Biological Chemistry, Graduate School of
Agricultural and Life Sciences, The University of Tokyo, Tokyo, Japan; ERATO Touhara
Chemosensory Signal Project, JST, The University of Tokyo, Tokyo, Japan
YOUSUKE TSUNEOKA Unit for Affiliative Social Behavior, RIKEN Brain Science Institute,
Saitama, Japan
JING W. WANG Neurobiology Section, Division of Biological Sciences, University of
California San Diego, La Jolla, CA, USA
PEI SABRINA XU Department of Anatomy and Neurobiology, Washington University
in St. Louis, St. Louis, MO, USA
 Contributors xv

DAISUKE YAMAMOTO Department of Developmental Biology and Neurosciences,

Tohoku University Graduate School of Life Sciences, Sendai, Japan
JOANNE Y. YEW Temasek Life Sciences Laboratories, Singapore, Singapore; Department of
Biological Sciences, National University of Singapore, Singapore
C. RON YU Stowers Institute for Medical Research, Kansas City, MO, USA
SENMIAO ZHAN Department of Molecular Genetics and Microbiology, Duke University
Medical Center, Durham, NC, USA; Boston University School of Medicine, Boston,
MA, USA
XINXING ZHANG Department of Chemistry, University of Florida, Gainesville, FL, USA
YUE ZHOU Department of Chemistry, University of Florida, Gainesville, FL, USA
FRANK ZUFALL Department of Physiology, School of Medicine, University of Saarland,
Homburg, Germany
 Part I

Sampling, Chromatographic Purification, and Chemical

Analysis of Pheromone Molecules
 Chapter 1

Female Moth Pheromones

Tetsu Ando

Abstract
The sex pheromone, a volatile secreted by a female moth, is stored in the pheromone gland and can be
easily extracted with hexane. The extract is effectively analyzed using a gas chromatography combined with
an electro-antennogram detector (GC-EAD) and a mass spectrometry (GC-MS), both of which are
equipped with a capillary column. GC-EAD analysis indicates the number of pheromone components that
have a different chromatographic behavior. The mass spectrum measured by GC-MS suggests the outline
of the chemical structure. In addition to a comparison with chemical data of authentic synthetic com-
pounds, micro-chemical reactions reveal a precise structure of the natural pheromone. Finally, the chemical
structure is confirmed by field evaluation of the synthetic pheromone.

Key words Insect, Lepidoptera, Electro-antennogram, GC-MS, Diagnostic ion, Chiral HPLC,
Mating communication, Male attractant

1 Introduction

Lepidoptera is the second largest insect group, including about

160,000 described species, which are currently divided into 47
supper families and further approximately 130 families (see Note 1).
As a consequence of the diversity of Lepidoptera, interesting
species-specific sex pheromone systems are exhibited in this insect
group. The quite varied pheromones, which have been identified
from female moths of nearly 630 species around the world [1, 2],
are classified into groups of Type I (75 %), Type II (15 %), and
miscellaneous (10 %) according to their chemical structures [3].
Figure 1 shows the structures of typical pheromone components.
Type I pheromones are composed of unsaturated compounds with
a C10C18 straight chain and a terminal functional group, such as a
hydroxyl, an acetoxyl, or a formyl group. On the other hand, Type
II pheromones are composed of unsaturated hydrocarbons and
epoxy derivatives with a C17C23 straight chain. The biosynthetic
pathways of pheromones in these two types are quite different
(see Note 2).

Kazushige Touhara (ed.), Pheromone Signaling: Methods and Protocols, Methods in Molecular Biology, vol. 1068,
DOI 10.1007/978-1-62703-619-1_1, Springer Science+Business Media, LLC 2013

3
4 Tetsu Ando

a b c
Type I Type II Others
O
OH
O
O O
H

Fig. 1 Representative lepidopteran sex pheromones: (a) Type I pheromone of the diamond back moth (Plutella
xylostella, Plutellidae), (b) Type II pheromone of the Japanese giant looper (Ascotis selenaria cretacea,
Geometridae), and (c) a methyl-branched pheromone of a lichen moth (Miltochrista calamine, Arctiidae)

Type I pheromones have been identified from the species in

various families, such as Crambidae, Tortricidae, and Noctuidae,
which include many agricultural pest insects. For example, females
of the diamond back moth (Plutella xylostella, Plutellidae) secrete
a mixture of (Z)-11-hexadecenyl acetate and (Z)-11-hexadecenal [4].
Each lepidopteran species produces a species-specific pheromone
with a difference of the chemical structure, i.e., functional group,
carbon chain length, number of double bond(s), unsaturated
position and configuration. The diversity of lepidopteran sex pher-
omones is also generated by blending multiple components. Type
II pheromones are mainly produced by the species in highly
evolved insect groups, such as Geometridae, Lymantriidae, and
Arctiidae. For example, females of the Japanese giant looper
(Ascotis selenaria selenaria, Geometridae) secrete a mixture of
(3Z ,6Z,9Z)-3,6,9-nonadecatriene and (3R,4 S)-3,4-epoxy-
(6Z,9Z)-6,9-nonadecadiene [5]. In addition to the unbranched
compounds, several novel pheromones with a methyl branch have
been identified. Females of a lichen moth (Miltochrista calamine,
Arctiidae) secrete a (5R,7R)-5-methylheptadecan-7-ol [6].
Furthermore, sex attractants have been reported for the male moths
of about 1,200 species, which were found by screening tests of
known synthetic pheromones and their analogues in the field [2].
Since we have extensive knowledge of the chemical communi-
cation systems developed in lepidopteran insects, new identifica-
tion should be planned with reference to previous works carried
out with a species taxonomically related to a new target moth.
Although sex pheromones play an important role in reproductive
isolation, closely related species possibly produce similar phero-
mone components (see Note 3). Generally, a pheromone titer is
very low (<100 ng/female) (see Note 4), and mass rearing of pest
insects is not easy. Additionally, a considerable number of species
are univoltine. In these cases, every experiment has limitations.
Male antennae, however, respond to a pheromone quite sensi-
tively, and, thus, not only a major component but also a minor
 Female Moth Pheromones 5

component can be detected using an electrophysiological technique,

namely, by measurement with an electro-antennogram (EAG) [7].
Gas chromatography combined with an EAG detector (GC-EAD)
is the most useful instrument to identify pheromone candidates in
a crude extract of pheromone glands [8]. The sensitivity of the
EAD against a pheromone component is higher than that of a
flame ionization detector (FID), as shown in Fig. 2. The chemical
structure of each EAG-active component can be estimated by the
analysis of the mass spectrum measured by GC combined with
mass spectrometry (GC-MS). Epoxy components in Type II pher-
omones include chiral centers. Even in an unsuccessful case of
optical resolution by chiral GC columns, enantiomeric separation
of the epoxides has been accomplished by high-performance liquid
chromatography (HPLC) with a chiral column [3]. Finally, field
evaluation of synthetic pheromone candidates reveals the phero-
mone with an attractive activity for males of the target species. This
chapter deals with the above techniques and some related experi-
mental procedures.

a
electrodes

EAD antenna
injection b
FID I III I III
Expt. 1 II Expt. 2 II

0.1 mV
EAD

FID

0.0 5.0 10.0 Rt (min) 10.0 Rt (min)

Fig. 2 GC-EAD analysis: (a) an instrument equipped with a capillary column and a handmade EAD (b) data of
three synthetic epoxides stimulating a male antenna of the mulberry looper (Menophra atrilineata, Geometridae).
Compounds; I = 9,10-epoxy-(Z)-6-octadecene (a minor pheromone component), II = 9,10-epoxy-(3Z,6Z)-3,6-
octadecadiene (a major pheromone component), and III = 9,10-epoxy-(3Z,6Z)-3,6-nonadecadiene (a pheromone
homologue). Expt. 1; I (10 ng), II (1 ng), and III (100 ng). Expt. 2; I (10 ng), II (0.001 ng), and III (1 ng)
6 Tetsu Ando

2 Materials

2.1 Collection 1. Organic solvents: Hexane and diethyl ether.

2. Solid phase micro-extraction (SPME) fiber: A fused-silica fiber
coated with poly(dimethylsiloxane).
3. Porous polymer beads: Porapak Q or Tenax TM.

2.2 Bioassay 1. Flask (100 ml).

2. Filter paper.
3. Red light.
4. Wind tunnel: An acrylic tube (0.3 m ID 2 m) with a steady
airflow of 0.4 m/s.
5. Sticky board trap with a 30 27 cm bottom plate with a roof.
6. Dispenser of synthetic pheromone: Rubber septum (OD 8 mm).

2.3 GC 1. GC instrument: Agilent 7890A GC system equipped with FID.

2. Gas: He (carrier gas, >99.9995 % purity), H2 (>99.9995 %
purity) and dry air (>99.999 % purity) for FID, and N2 (make-
up gas, >99.9995 % purity).
3. Capillary columns: A low polar column such as DB-5 and a
high polar column such as DB-23 (0.25 mm ID 30 m,
0.25 m film). A compound is eluted from the column with a
specific retention time (Rt), depending on its volatility (i.e.,
molecular weight and chain length) and polarity. The Rt is also
changed by the polarity of a stationary phase in the column.
Therefore, analyses with both low and high polar columns are
informative. By comparing the Rts measured by two columns,
the number of double bonds and conjugation of two double
bonds are indicated. Separation of geometrical and positional
isomers of unsaturated compounds is successfully accomplished
by the high polar column.
4. EAD: The effluent from a capillary column is split equally into
two lines, one leading to FID, and the other, to the EAD
(Fig. 2a). An antenna is excised at the base from the male moth,
and a few distal segments are cut off. Insect saline is used for the
electrodes. Each end of the antenna is attached to a droplet of a
saline solution on an electrode of the EAD device so that the
sensilla faces the airflow from the GC. The rate of the airflow
carrying eluted compounds to the antenna is about 8 cm/s.

2.4 GC-MS 1. GC-MS instrument: Electron impact GC-MS is conducted

with a quadrupole mass spectrometer (Agilent 5973 mass
spectrometer system) or a magnetic sector mass spectrometer.
The carrier gas is helium, and the ionization voltage is 70 eV.
2. He gas: Carrier gas (>99.9995 % purity).
 Female Moth Pheromones 7

2.5 Chiral HPLC 1. HPLC instrument: Jasco LC-2000 series equipped with
ultraviolet (UV) and refractive index (RI) detectors.
2. Solvents: Hexane, 2-propanol, methanol, and water.
3. Chiral columns: A normal-phase column such as Chiralpak AD
and Chiralpak AS, and a reversed-phase column such as
Chiralcel OJ-R (4.6 mm ID 25 cm). In the case of epoxy
pheromones, no chiral columns have the ability to separate
enantiomers of many epoxides universally [3]. A reasonable
resolution has been accomplished by selection of the best chi-
ral column after trial and error.

2.6 Reagents for 1. Florisil.

Purification and 2. Organic solvents: Hexane, diethyl ether, CH2Cl2, ethanol,
Chemical Reaction diethylene glycol, THF.
3. Dimethyl disulfide (DMDS).
4. 4-Methyl-1,2,4-triazoline-3,5-dione (MTAD).
5. Reduction: Hydrazine hydrate, H2O2, KOH, LiAlD4.
6. Mesylation: Methanesulfonyl chloride, triethylamine,
N,N-dimethyl-4-aminopyridine.

3 Methods

3.1 Pheromone 1. Pheromone gland extraction: Sex pheromone components are

Collection secreted from a pheromone gland, which is located at an
abdominal tip (eighth and ninth abdominal segments) of a vir-
gin female in many lepidopteran species. With reference to
observation of a mating behavior, the gland of a target species
is cut off at a calling time and then immersed in hexane (50 l/
tip) for 15 min to extract the pheromone components that are
stored in the gland. It is important to avoid contamination of
obstructive lipids extracted with a polar solvent for a long-time
immersion. This solvent extraction is the most convenient
method for pheromone collection.
2. Absorption by SPME: SPME is used to collect airborne phero-
mone components emitted by a female. The fiber is placed a few
mm from the abdominal gland of a calling female for 23 h [9].
Collection is also accomplished by gently rubbing the fiber on
a tegument of a glandular area for 5 min [10]. SPME samples
are injected by exposing the fiber in the hot injector of GC for
5 min before the start of a chromatogram.
3. Collection by an adsorbent tube: A small glass tube is filled
with adsorbent (porous polymer beads, 200 mg), which is suf-
ficiently washed with diethyl ether [11]. To collect an airborne
pheromone by the adsorbent, the tube is connected to a gas
8 Tetsu Ando

scrubber bottle including female moths and supplied air

through an activated charcoal filter. The adsorbed materials are
eluted from the tube with diethyl ether.

3.2 Bioassay 1. Laboratory bioassays: In the case of nocturnal species, responses

of the males placed in a flask are observed in darkness under a
red light (10 lx). A piece of filter paper is loaded with a natural
pheromone (less than one female equivalent (1 FE)) or other
test materials dissolved in hexane, and the biological activity is
examined after evaporation of the solvent. Mating behavior is
divided into four categories: vigorous antennal swing, orienta-
tion flight, mating dance, and abdominal contact with the
pheromone source [12].
2. Wind tunnel: Male responses are also observed in a wind tun-
nel. The filter paper with a test material is hung 15 cm above
the floor and 15 cm upwind of the wind tunnel. Males are
released at the leeward side, and the following male behaviors
are recorded for 3 min: upwind flight, approach to the source,
contact with the source, and attempt of copulation [11].
3. GC-EAD: Differently from the case of mating behavior, a male
antenna responds to each constituent of a multicomponent
pheromone. Injection of a crude pheromone extract (1 FE)
indicates the number of included components and approxi-
mate length of their carbon chains. A natural component
(>10 pg) can be sensitively detected by a male antenna, and
structure-related analogues of a pheromone also stimulate the
antenna as shown in Fig. 2b. The double-bond positions of a
dienyl pheromone can be presumed by the EAG-activities of a
series of synthetic monoenyl compounds [13].

3.3 Partial If a crude extract includes many impurities, column chromatogra-

Purification by phy of the extract (100 FE) is conducted on Florisil (500 mg)
a Mini-Column using 5, 10, 20, and 30 % ether in hexane (3 ml each) as an eluent.
Acetates and aldehydes of Type I pheromones are recovered in the
10 % ether fraction, and alcohols are recovered in the 30 % ether
fraction [14].

3.4 GC-MS Analysis A mass spectrum of each EAG-active component (>10 ng/gland) is
obtained by an injection of the crude extract (110 FE) (see Note 4).
The EI-MS analysis usually shows a molecular ion (M+) and charac-
teristic fragment ions indicating a summary of the chemical struc-
ture, i.e., chain length, functional group, and degree of unsaturation
[15]. M+ of monoenyl compounds of the Type I pheromone is
scarcely detected, but ions at m/z M-18 of alcohols and aldehydes
and ions at m/z M-60 and 61 of acetates indicate the structure. It
is not easy to determine a double-bond position by the mass spec-
trum, except for some conjugated dienyl compounds that show
 Female Moth Pheromones 9

diagnostic fragment ions reflecting the positions (see Fig. 3 and

Note 5). After determination of the position (see next section),
configuration of the double bond is determined by comparing with
Rts of authentic geometrical isomers. The positions of the epoxy
ring and double bond(s) of Type II pheromones can be estimated
by some diagnostic fragment ions, which have been published with
many spectral data (see Note 6). Methyl-branched hydrocarbons
show characteristic secondary fragment ions indicating the position
of the branch [3]. In the case of branched alkenes, ketones, and
alcohols, positional isomers show similar mass spectra [15].

3.5 Determination of 1. DMDS derivatization (Fig. 4a): The double-bond position of

Double-Bond Positions a monounsaturated Type I pheromone is revealed by derivatiza-
tion with DMDS [16]. A crude extract including a monoene
(about 250 ng, more than 5 FE) is treated with DMDS (50 l)
and a diethyl ether solution of iodine (60 mg/ml, 50 l) after
removal of the solvent. The mixture is warmed at 40 C over-
night, and a 5 % sodium thiosulfate solution (0.5 ml) is added
to it. Products are extracted with hexane and analyzed by
GC-MS. The DMDS adduct shows two characteristic frag-
ment ions produced by cleavage of the bond between the
sulfur-substituted carbons indicating the position of the origi-
nal double bond.

a
100 82 95
Relative intensity (%)

67 95 M+
OH 182
82
50 7,9-diene

164
0
50 100 150 200

b
100 68
Relative intensity (%)

81
81
M+
OH 182
50 68
8,10-diene

164
0
50 100 150 200

Fig. 3 Mass spectra of C12 dienyl pheromones indicating two characteristic fragmentations: (a) (7E,9Z)-7,9-
dodecadien-1-ol of the grapevine moth (Lobesia botrana, Olethreutinae) and (b) (8E,10E)-8,10-dodecadien-
1-ol of the codling moth (Cydia pomonella, Olethreutinae)
10 Tetsu Ando

a b CH3
175 CH3
145 O N O N O
5-ene 5,7-diene O
H3CS SCH3 N N
DMDS N N
OAc OAc OH OH
MTAD 238
222
c 14-ene
10,14-diene OAc
H2NNH2/ H2O2 DMDS
OAc
10-ene
OAc

d e
O OH
98 198

H2NNH2/ KOH 1) CH3SO2Cl/ DMAP D

185C 2) LiAlD4 85
197

Fig. 4 Chemical derivatization of moth pheromones and characteristic fragment ions of the derivatives: (a) a
DMDS adduct of 5-dodecenyl acetate, (b) an MTAD adduct of 5,7-doecadien-1-ol, (c) partial diimide reduction,
(d) WolffKishner reduction of a methyl-branched 2-ketone, and (e) mesylation and LiAlD4 reduction of a
methyl-branched secondary alcohol

2. MTAD derivatization (Fig. 4b): The double-bond positions of

a conjugated diene are determined by derivatization with
MTAD [17]. A crude pheromone (about 0.5 g) dissolved in
CH2Cl2 (10 l) is treated with a CH2Cl2 solution of MTAD
(10 mg/ml, 40 l) for 30 min at room temperature. After
changing the solvent to hexane, products are analyzed by
GC-MS. The MTAD adduct, which is formed by DielsAlder
cycloaddition between the conjugated diene and the dieno-
phile, gives mass spectrum including abundant fragment ions
that are diagnostic for the unsaturated positions in the parent
compound.
3. Diimide reduction (Fig. 4c): The unsaturated positions of an
unconjugated polyene are confirmed by combination of a par-
tial diimide reduction and DMDS derivatization [18]. A crude
pheromone (about 1 g) dissolved in hexane (10 l) is treated
with 10 l each of a hydrazine and a hydrogen peroxide solu-
tion in ethanol (N2H4: 0.3 ml of hydrazine hydrate in 10 ml of
ethanol, H2O2: 0.04 ml of 31 % H2O2 in 10 ml of ethanol). The
mixture is maintained at 55 C for 1 h for partial hydrogena-
tion. Since the double bonds of a polyene are unselectively
reduced to make a mixture of monoenyl derivatives, the crude
products are directly treated with DMDS and adducts are ana-
lyze by GC-MS to determine the double-bond positions.

3.6 Determination 1. WolffKishner reduction (Fig. 4d): Carbonyl compounds are

of Branched-Methyl converted into hydrocarbons to reveal carbon skeletons [19].
Positions A crude pheromone (about 1 g) dissolved in diethylene gly-
col (0.2 ml) is treated with hydrazine hydrate (1 g) and KOH
 Female Moth Pheromones 11

(0.2 mg) and heated at 185 C for 4 h. After cooling, water

(1 ml) is added to the mixture. Crude products are extracted
with hexane and analyzed by GC-MS.
2. LiAlD4 reduction (Fig. 4e): Secondary alcohols are also con-
verted into hydrocarbons [6]. A crude pheromone (about
1 g) dissolved in CH2Cl2 (0.1 ml) is mixed with triethylamine
(10 mg), N,N-dimethyl-4-aminopyridine (1 mg), and meth-
anesulfonyl chloride (20 mg) and stirred at 0 C for 2 h and at
room temperature for 1 day. After the usual work-up, the pro-
duced mesylate is dissolved in dry THF (0.1 ml) and treated
with LiAlD4 (5 mg) under an argon atmosphere. The reaction
mixture is acidified with 1N HCl (2 ml) after 3 h of stirring
and then extracted with hexane. The extract is analyzed by
GC-MS.

3.7 Determination Absolute configuration of the pheromone with a chiral center is

of Stereochemistry assigned by the biological activities of both enantiomers, which
have been synthesized by starting from a chiral compound or by a
chiral reaction. The stereochemistry of a chiral natural pheromone
is determined by comparison with the chromatographic behaviors
of synthetic enantiomers on a chiral GC or HPLC column. In the
case of HPLC analysis, compounds with a conjugated system are
highly sensitive to a UV detector. Otherwise Rts of the two syn-
thetic enantiomers must be measured by an RI detector, and then
the natural pheromone is analyzed. After the injection of the pher-
omone, eluted materials are collected every 30 s around the Rt and
quantitatively analyzed by GC-MS.

3.8 Field Evaluation Each synthetic pheromone candidate (1 mg) dissolved in hexane
of Synthetic (100 l) is incorporated into a rubber septum, which is used as a
Pheromone dispenser. The lure is placed at the center of a sticky board trap.
Traps are hung separately in the field at intervals of 10 and 1.5 m
height. Captured males are counted weekly.

4 Notes

1. (Taxonomy of Moths): Nearly 99 % of lepidopteran species

belong to Ditrysia, which has two separate genital openings (a
copulatory orifice and an ovipore). Ditrysia includes two
superfamilies of butterflies and 31 superfamilies of moths.
For the most part, female sex pheromones have been identified
from the species of the following large superfamilies;
Tortricoidea (167 species), Pyraloidea (87 species),
Geometroidea (48 species), and Noctuoidea (174 species) [2].
2. (Biosynthesis of Moth Pheromones): The variation of the
chemical structures results from differences in both a starting
12 Tetsu Ando

material and a biosynthetic enzyme system [3]. The Type I

compounds are biosynthesized from saturated fatty acids pro-
duced by de novo synthesis starting from acetyl-CoA. The
Type II compounds are biosynthesized starting from dietary
linoleic and linolenic acids, which are changed into unsatu-
rated hydrocarbons (6,9-dienes and 3,6,9-trienes) outside of
the pheromone gland (in oenocytes). Some females are able to
introduce an additional double bond and produce further
modified polyenes, such as 4,6,9-trienes and 1,3,6,9-tetraenes.
The hydrocarbons are transported to the gland via hemolymph
and emitted into the atmosphere directly and/or after
epoxidation.
3. (Common Pheromone Structures in Related Species): It is
expected that taxonomically related species that have devel-
oped from a common ancestor also exhibit similarity at the
pheromone level [3]. The family of Tortricidae in Tortricoidea
is divided into subfamilies, such as Olethreutinae and
Tortricinae. The species in Olethreutinae usually secrete C12
chain compounds unsaturated at the 8- or 9-position. On the
other hand, the species in Tortricinae utilize C14 chain com-
pounds mainly unsaturated at the 11-position. Noctuoidea,
the largest and highly evolved superfamily, is divided into
several families, such as Noctuidae, Lymantridae, and Arctiidae.
In the family of Noctuidae, species in one group (Trifinae)
produce Type I pheromones, while species in the other group
(Quadrifinae) produce Type II pheromones. Many species of
Lymantridae and Arctiidae also utilize Type II pheromones.
Although no sex pheromone has been reported from female
butterflies, a chemical cue is an important factor for the mating
communication of diurnal moths, such as clearwing moths in
Sesiidae. Almost all of the Sesiidae pheromones are composed
of 2,13- and 3,13-dienyl compounds with a C18 chain.
4. (Pheromone Titer): The titer fluctuates with a circadian
rhythm, and the most abundant pheromone is usually detected
at a calling time. The pheromone content is also different
depending on the species. Although a quantitative correlation
between the size of a female moth and its pheromone titer is
not always observed, the gland of a large insect, such as a
Noctuidae species, usually contains around 10100 ng of a
main pheromone component. On the contrary, the phero-
mone content in micro-lepidopteran species, such as leafminer
moths, is very limited, sometimes lower than a 1-pg order.
5. (Diagnostic Ions of Conjugated Dienes): Compounds unsatu-
rated at a side of the terminal methyl group [n,(n + 2)-diene,
n = 15] show characteristic fragment ions, [CH2CH=CHCH=
CHCn1H2n1]+ and [CH2=CHCH=CHCn1H2n1]+, reflecting
the position of unsaturations [15]. For example, the positional
 Female Moth Pheromones 13

isomers of C12 dienyl compounds are distinguished by the

following fragment ions: m/z 123 and 110 of 5,7-dienes
(5,7-diene), m/z 109 and 96 of 6,8-dienes (4,6-diene),
m/z 95 and 82 of 7,9-dienes (3,5-diene), m/z 81 and 68 of
8,10-dienes (2,4-diene), and m/z 67 and 54 of 9,11-dienes
(1,3-dienes) (Fig. 3). In particular, ions at m/z 67 of 1,3-
dienes, m/z 81 or 68 of 2,4-dienes, and m/z 95 or 82 of
3,5-dienes are recorded as the characteristic base peaks.
Although it is noteworthy that 4,6-dienal shows a characteristic
base peak at m/z 84 [12], the spectral patterns of the dienes
unsaturated at a side of the functional group are similar.
6. (Diagnostic Ions of Type II Pheromones): While a characteris-
tic fragment ion at m/z 79, [H(CH=CH)3]+, is detected for
many compounds with a homoconjugated dienyl structure,
each pheromone component can be identified by some diag-
nostic ions, which reflect the positions of a double bond and an
epoxy ring as follows (underlined numbers mean base peaks)
[15]; ions at m/z M-56, 108 and 79 for 3,6,9-trienes, m/z
M-85, 163, 110 and 79 for 6,9,11-trienes, m/z M-98, 150
and 79 for 6,9,12-trienes, m/z M-54, 106, 91 and 79 for
1,3,6,9-tetraenes, m/z M-82, 108 and 79 for 3,6,9,11-tetraenes,
m/z M-96, 148 and 79 for 3,6,9,12-tetraenes, m/z M-58,
M-72 and 79 for 3,4-epoxy-6,9-dienes, m/z M-69, M-83,
111, 97 and 67 for 6,7-epoxy-3,9-dienes, m/z M-109, M-123,
122, 108 and 79 for 9,10-epoxy-3,6-dienes, and m/z M-107,
M-121, 120, 106 and 79 for 9,10-epoxy-1,3,6-trienes.

References

1. El-Sayed AM. Internet database 2012. http:// 7. Schnaider D (1969) Insect olfaction: decipher-
www.pherobase.com/ ing system for chemical messages. Science
2. Ando T. Internet database 2012. http://www. 163:10311037
tuat.ac.jp/~antetsu/LepiPheroList.htm 8. Arn H, Stadler E, Rauscher S (1975) The elec-
3. Ando T, Inomata S, Yamamoto M (2004) troantennographic detectora selective and
Lepidopteran sex pheromones. Top Curr sensitive tool in the gas chromatographic analy-
Chem 239:5196 sis of insect pheromone. Z Naturforsch
4. Tamaki Y, Kawasaki K, Yamada H, Koshihara 30c:722725
T, Osaki N, Ando T, Yoshida S, Kakinohana H 9. Mozuraitis R, Borg-Karlson A-K, Buda V,
(1977) (Z)-11-Hexadecenal and (Z)-11- Ivinskis P (1999) Sex pheromone of the spot-
hexadecenyl acetates: sex-pheromone compo- ted tentiform leafminer moth Phyllonorycter
nents of the diamondback moth (Lepidoptera: blancardella (Fabr.) (Lep., Gracillariidae).
Plutellidae). Appl Entomol Zool 12:208210 J Appl Entomol 123:603606
5. Ando T, Ohtani K, Yamamoto M, Miyamoto 10. Frrot B, Malosse C, Cain A-H (1997) Solid-
T, Qin X-R, Witjaksono (1997) Sex phero- phase microextraction (SPME): a new tool in
mone of Japanese giant looper, Ascotis selenaria pheromone identification in Lepidoptera.
cretacea: identification and field tests. J Chem J High Resolut Chromatogr 20:340342
Ecol 23:24132423 11. Fujii T, Nakano R, Takubo Y, Qian S, Yamakawa
6. Yamakawa R, Kiyota R, Taguri T, Ando T R, Ando T, Ishikawa Y (2010) Female sex
(2011) (5R,7R)-5-Methylheptadecan-7-ol: a pheromone of a lichen moth Eilema japonica
novel sex pheromone component produced by (Arctiidae, Lithosiinae). J Insect Physiol 56:
a female lichen moth, Miltachrista calamine, in 19861991
the family Arctiidae. Tetrahedron Lett 12. Naka H, Vang LV, Inomata S, Ando T, Kimura
52:58085811 T, Honda H, Tsuchida K, Sakurai H (2003)
14 Tetsu Ando

Sex pheromone of the persimmon fruit moth, trometry of dimethyl disulfide adducts. Anal
Stathmopoda masinissa: identification and labo- Chem 55:818822
ratory bioassay of (4E,6Z)-4,6-hexadecadien- 17. Young DC, Vouros P, Holick MF (1990) Gas
1-ol derivatives. J Chem Ecol 29:24472459 chromatography-mass spectrometry of conjugated
13. Roelofs W, Comeau A, Hill GM (1971) Sex dienes by derivatization with 4-methyl-1,2,4-tri-
attractant of the codling moth: characterization azoline-3,5-dione. J Chromatogr 522:295302
with electroantennogram technique. Science 18. Wakamura S, Ohno S, Arakaki N, Kohama
174:297299 T, Haraguchi D, Yasui H (2010)
14. Do ND, Ohbayashi K, Naka H, Nakada K, Identification and field activity of the sex
Ando T (2011) Identification and field evalua- pheromone component of the sweetpotato
tion of sex pheromone components of the pear vine borer moth Omphisa anastomosalis
barkminer moth, Spulerina astaurota. J Chem (Lepidoptera: Crambidae). Appl Entomol
Ecol 37:12221230 Zool 45:635640
15. Ando T, Yamakawa R (2011) Analyses of lepi- 19. Yamamoto M, Kamata T, Do ND, Adachi Y,
dopteran sex pheromones by mass spectrome- Kinjo M, Ando T (2007) A novel lepidopteran
try. Trends Anal Chem 30:9901002 sex pheromone produced by females of a
16. Buser H-R, Arn H, Guerin P, Rauscher S Lithosiinae species, Lyclene dharma dharma, in
(1983) Determination of double bond position the family of Arctiidae. Biosci Biotechnol
in mono-unsaturated acetates by mass spec- Biochem 71:28602863
 Chapter 2

Pheromones in the Fruit Fly

Jacqueline S.R. Chin and Joanne Y. Yew

Abstract
The identification of pheromones (chemical communication cues) is critical to our understanding of
complex social behavior in insects and other animals. In this chapter, we describe analytical methods for
the purification of lipid pheromones by thin layer chromatography and the quantification and determina-
tion of their elemental composition by mass spectrometry.

Key words Pheromones, Drosophila melanogaster, Lipids, Hydrocarbons, Thin layer chromatogra-
phy, Mass spectrometry

1 Introduction

Insect pheromones can consist of a mixture of compounds. Many of

these signaling molecules are lipids composed of branched and
straight-chain alkanes and alkenes and oxygen-containing hydrocar-
bons. Lipid pheromones are usually found on the surface of the
insect cuticle or are secreted by specialized glands [1, 2]. In order to
study pheromone profiles, the cuticle or glands are chemically
extracted and the extract is analyzed using mass spectrometry or
nuclear magnetic resonance spectroscopy. One disadvantage of
whole extract analysis is that less abundant compounds are usually
masked by more abundant components. To address this shortcom-
ing, chromatographic methods are used to separate complex mix-
tures into simpler fractions and allow for a more thorough analysis.
One chromatographic method, thin layer chromatography
(TLC), has been proven to be useful in identifying novel lipid
pheromones. TLC plates consist of glass or aluminum coated with
an adsorbent layer of silica gel. Using TLC, components of the
chemical extract can be separated into discrete fractions according
to hydrophobicity. Different molecules will have different degrees
of polarity depending on their molecular weight and functional
groups. The Drosophila melanogaster male sex pheromone, cis-
vaccenyl acetate (cVA), was discovered by Brieger and Butterworth

Kazushige Touhara (ed.), Pheromone Signaling: Methods and Protocols, Methods in Molecular Biology, vol. 1068,
DOI 10.1007/978-1-62703-619-1_2, Springer Science+Business Media, LLC 2013

15
16 Jacqueline S.R. Chin and Joanne Y. Yew

[3, 4] using TLC together with gas-chromatography mass

spectrometry (GCMS). Comparing D. melanogaster extracts from
males and virgin females, a male-specific spot on the TLC plate was
identified. A similar strategy was also used to identify a female aph-
rodisiac, heptacosadiene [5]. In this case, a behavioral assay was
used to test individual fractions for their ability to induce high
levels of courtship from males.
Recent technological advances have been useful for detecting
compounds directly from the insect cuticle, forgoing chemical
extraction altogether. These techniques allow for spatial expression
of a compound to be preserved and in some cases, for the insect to
remain alive during sampling. One method makes use of solid
phase micro-extraction (SPME), wherein flies are probed with a
thin fiber coated with adsorbent compounds and the fiber is ana-
lyzed with GCMS [6, 7]. Recent work using SPME led to the
identification of 17 new Drosophila hydrocarbons [8].
In addition to improved methods of sampling, developments
in mass spectrometry instrumentation have also been impor-
tant in revealing new chemical cues. Ultraviolet laser desorption/
ionization orthogonal time-of-flight mass spectrometry (UV-LDI-
o-TOF-MS) allows detection of novel polar compounds from the
cuticles of individual insects [9]. Use of this method led to the
discovery of a male D. melanogaster pheromone, CH503, a long-
lived inhibitor of male courtship [10]. In addition, a number of
novel triacylglyceride molecules were found in the anogenital
region of male Drosophila mojavensis and Drosophila arizonae [11].
These newly identified compounds were undetected by previous
GCMS analysis due to the high molecular weight and polarity of
the molecules. Direct Analysis in Real Time (DART) MS is another
recently introduced method that has been useful for the rapid anal-
ysis of cuticular lipids [12, 13]. Both UV-LDI MS and DART MS
provide high sensitivity detection of polar and apolar lipids with
higher molecular weight (m/z 5001,000). However, unlike
GCMS, neither method is capable of detecting saturated hydrocar-
bons nor elucidating branched vs. unbranched structures under
standard conditions. For the most comprehensive profile of chemi-
cal compounds on the cuticular surface, multiple analytical meth-
ods are necessary.

2 Materials

2.1 Drosophila 1. Hexane spiked with 10 g/mL hexacosane (Sigma-Aldrich;

Cuticular Hydrocarbon St. Louis, Missouri, USA).
Extraction and 2. Gas-tight glass syringe (Hamilton, Reno, Nevada, USA).
Semiquantitative
3. Wash solution: chloroformmethanol (66:33; per vol).
Analysis
 Chemical Analysis of Drosophila Pheromones 17

4. 1.8 mL glass vials with Teflon-lined caps (s/n 224740;

Wheaton, Millville, New Jersey, USA).
5. Nitrogen evaporator or rotary evaporator.

2.2 Separation 1. Silica gel 60 TLC plates (Merck, Darmstadt, Germany).

of Cuticular Lipids 2. Flat bottom TLC chamber for 20 20 cm plates, with glass lid
by Thin Layer (Camag, Muttenz, Switzerland).
Chromatography
3. TLC solvent systems:
(a) For neutral lipid analysis (triacylglycerides, alkanes, alkenes,
fatty acid methyl esters, acetate esters such as cVA): hexane
diethyl etheracetic acid (90:9:1; per vol). See Fig. 1a.
Alternative system: petroleum/ether (70:30; per vol).
(b) For polar lipid analysis (secondary and primary alcohols,
oxoaldehydes, free fatty acids, acetoxy esters such as
CH503): hexanediethyl etheracetic acid (66:33:1; per
vol). See Fig. 1b. Alternative solvents: hexanemethanol
diethyl ether (85:10:5; per vol) or etherpetroleum
(70:30; per vol).

a b
Alkanes
Alkanes Alkenes
Alkenes TAGs
cVA

cVA

TAGs FAs

CH503
FAs

CH503
origin
origin

Fig. 1 A comparison of solvent systems for TLC separation of cuticular extract

from Drosophila. Both plates were stained with the neutral lipid dye primuline
(0.02 % in 20 % acetone) and visualized under UV illumination. Extract from
approximately 50 flies (a mixture of males and females) is loaded onto each lane.
(a) Hexanediethyl etheracetic acid (90:9:1; per vol) provides separation of
neutral lipids. Alkanes, alkenes, triacylglycerides (TAGs), and the male sex phero-
mone cVA are clearly resolved, while more polar components of the cuticle
such as fatty acids (FAs) and CH503 only moved a short distance from the origin.
(b) Hexanediethyl etheracetic acid (66:33:1; per vol) provides separation of
polar lipids. Alkanes, alkenes, TAGs, and cVA are no longer separated, but CH503
and FAs are clearly resolved
18 Jacqueline S.R. Chin and Joanne Y. Yew

4. Lipid and pheromone standards can be obtained from Sigma-

Aldrich, Cayman Chemical Company (Ann Arbor, Michigan,
USA), Pherobank (Wageningen, Netherlands), and Avanti
Polar Lipids (Alabaster, Alabama, USA).
5. Chromatography paper (Whatman, Springfield Mill, UK).
6. Sample applicator (Linomat IV; Camag, Muttenz, Switzerland)
or microsyringe (Hamilton).
7. Staining reagents:
(a) Primuline (Sigma-Aldrich): 0.02 % in 20 % acetone in
aerosol spray or shallow glass staining dish.
(b) Sulfuric acid charring: 10 % sulfuric acid in aerosol spray.
(c) Iodine crystals (Sigma-Aldrich): sealed glass container
containing a few crystals of iodine.
(d) Water: Milli Q-purified water in aerosol spray.
8. UV-light for visualization of primuline stain.
9. Phosphorimager imager for quantification of lipid signals from
TLC plates.

2.3 Purification 1. Clean razor or scalpel blade.

of Compounds 2. Glass cutter or diamond knife.
from TLC Plates
3. Weighing paper.
4. Narrow glass funnel.
5. Chloroformmethanol (66:33; per vol).
6. Flash chromatography column:
(a) Stuff a small amount of glass wool fiber (Pall Corporation,
Ann Arbor, Michigan, USA) into the tapered end of a dis-
posable borosilicate glass Pasteur pipette (15 cm length).
The wool will serve as a stopper.
(b) Suspend the pipette vertically using a ring stand and clamp.
7. Nitrogen evaporator.

2.4 Analysis of 1. 0.1 mL micro inserts with spring bottom for GCMS vials
Extracts by GCMS (Supelco, St. Louis, Missouri, USA).
2. Glass syringe (Hamilton).
3. Lipid and pheromone standards (see Subheading 2.2).
4. Gas chromatography mass spectrometer. We use a GCMS
QP2010 system from Shimadzu (Kyoto, Japan).

2.5 Analysis of 1. Double-sided tape. We use self-adhesive tabs from Plano

Intact Insects by (Plano GmbH, Wetzlar, Germany).
UV-LDI MS 2. Glass coverslips (22 22 mm, VFM, Wales, UK).
3. Fine stainless steel forceps.
 Chemical Analysis of Drosophila Pheromones 19

4. External calibrants consisting of synthetic peptide standards

such as Substance P, Bradykinin, Renin substrate (all are avail-
able from Sigma-Aldrich). Calibrants are used in 1 mM con-
centration, dissolved in ddH2O.
5. Matrix for analysis of calibrants: 10 mg/mL 2,5-Dihydroxy-
benzoic acid (DHB; Sigma-Aldrich) in 70 % methanolwater.
6. Time-of-flight mass spectrometer with a MALDI source and
equipped with a UV-laser (either N2 or Nd:YAG). The ion
source must be able to provide a 13 mbar buffer gas environ-
ment. We use an orthogonal time-of-flight MS with a modified
oMALDI2 source (ABI Qstar Elite, Framingham, MA, USA).

2.6 Analysis of TLC 1. Double-sided tape. We use self-adhesive tabs from Plano
Plates by UV-LDI MS (Plano GmbH, Wetzlar, Germany).
2. Aluminum-backed TLC plates (Sigma-Aldrich).
3. External calibrants consisting of synthetic peptide standards
such as Substance P, Bradykinin, Renin substrate (all are avail-
able from Sigma-Aldrich). Calibrants are used in 1 mM con-
centration, dissolved in ddH2O.
4. Matrix for analysis of calibrants and TLC plates:
(a) 10 mg/mL 2,5-Dihydroxybenzoic acid (DHB; Sigma-
Aldrich) in 70 % methanolwater.
(b) Graphite pencil.
5. Time-of-flight mass spectrometer with a MALDI source and
equipped with a UV-laser (either N2 or Nd:YAG). The ion
source must be able to provide a 13 mbar buffer gas environ-
ment. We use an orthogonal time-of-flight MS with a modified
oMALDI2 source (ABI Qstar Elite, Framingham, MA, USA).

2.7 Analysis 1. Fine stainless steel forceps.

of Intact Insects 2. Undiluted polyethylene glycol (Sigma-Aldrich).
by DART MS
3. 5 M ammonium acetate solution in water.
4. Melting point tubes, 1.02.0 mm OD (World Precision
Instruments, Sarasota, Florida, USA).
5. Atmospheric pressure ionization time-of-flight mass spectrometer
(AccuTOF-DART; JEOL USA, Inc., Peabody, MA, USA) equipped
with a DART interface (IonSense Inc., Saugus, MA, USA).

3 Methods

Carry out all procedures at room temperature unless otherwise

stated.
Use gas-tight glass syringes for transfer of solvents.
20 Jacqueline S.R. Chin and Joanne Y. Yew

3.1 Drosophila 1. Anesthetize flies with cold or CO2 and place in a glass vial with
Cuticular Lipid a Teflon cap. For quantitative analysis by GCMS, prepare five
Extraction and replicate sets for each genotype (see Notes 1 and 2). Place vials
Semiquantitative on ice so that the flies remain anesthetized.
Analysis 2. Using a glass syringe, add enough hexane to cover the flies
(see Note 3). For quantitative analysis, add 120 L of hexane
with hexacosane standard into each vial.
3. Cap the vial and incubate at room temperature for 20 min.
4. Transfer all of the solvent into a new vial (see Notes 4 and 5).
For quantitative analysis, transfer 100 L of the solvent from
each vial. Only a portion of the total extraction solvent is
removed in order to ensure consistent volumes from sample to
sample.
5. Add additional hexane to quickly rinse residual lipids and pool
together with the first extraction. This step is omitted for
quantitative analysis.
6. Evaporate the extract under a gentle stream of nitrogen
( see Note 6). If a nitrogen stream is not available, the solvent
can be evaporated by leaving the vials uncapped in the fume
hood for 46 h at RT.
7. Store extract at 20 C until analysis.

3.2 Separation 1. A tank with dimensions (17.5 cm 16.0 cm 6.2 cm) can fit
of Cuticular Lipids two 10 10 cm TLC plates at once.
by Thin Layer 2. Clean the inside of the tank with methanol and allow the tank
Chromatography to dry.
3. Prepare 100 mL of the solvent system.
4. Place a sheet of filter paper approximately 30 cm 15 cm into
the tank in order to allow the tanks environment to equili-
brate. Seal the tank lid with silicone gel in order to prevent
evaporation of solvent. Place a heavy weight on the lid. Allow
equilibration for at least 5 h (see Note 7).
5. While waiting for the tank to equilibrate, prepare the TLC
plate: Redissolve the extract by adding 10 L of hexane for
every 50 flies into the vial. To load the extract on the TLC
plate, use an automated system such as the Linomat (Camag).
Alternatively, use a pencil to draw a line across about 1.5 cm
above the bottom of the plate. Use a ruler to measure and
mark points at even intervals with a pencil. Using a capillary or
glass syringe, dot 10 L of the extract on a marked point,
allowing the hexane to dry at intervals in order to prevent the
size of the dot from spreading too much (see Note 8).
6. Place the plate into solvent in tank, taking care that the edges
of the plate touch the bottom of the tank evenly. Cover tank and
replace the weight. Allow solvent front to reach approximately
 Chemical Analysis of Drosophila Pheromones 21

1 cm from top of the plate. This should take about 1420 min
(see Note 9).
7. Remove plate. Allow the plate to dry in the fumehood for at
least 15 min (see Note 10).
8. Stain the plate with any of the following methods:
(a) Spray with primuline using aerosol spraying device or by
dipping into a solution of primuline for approximately
20 s. Let dry for 1 h. Bands can be visualized under UV
light (see Note 11). Relative quantification can be per-
formed using a phosphorimager to measure the brightness
of bands.
(b) Spray with sulfuric acid. Visualize by placing on a thermal
plate heated to 100 C for 10 min. Bands will appear as
dark brown or black spots.
(c) Iodine crystals can also be used to visualize lipid bands and
is generally considered to be nondestructive. Place the
TLC plate in a sealed chamber with a few crystals of potas-
sium iodine. After 3060 min, dark brown lipid stains can
be observed. Once the plate is removed from the con-
tainer, the position of these bands should be marked
quickly with a pencil since the stain fades rapidly.
(d) If lipids are highly concentrated, spraying water on the
plate will allow visualization of the lipid bands. This is the
least destructive method of detection but also the least
sensitive.

3.3 Purification 1. Run a test plate with one lane in order to determine the opti-
of Compounds mal solvent conditions and optimal loading amount (see Note
from TLC Plates 8 regarding signs of overloading). Scale up to multiple plates
with ten lanes each. The number of plates will depend on the
natural abundance of the target compound and the amount of
product desired.
2. After developing the TLC plates, use a glass cutter or diamond
knife to cut out the first lane. Stain only this lane with the
method of choice. If nondestructive dyes such as iodine vapor
or water are used, the entire plate can be used for purification.
3. After drying, align the stained, cut lane with the rest of the
plate. Using a clean razor blade, draw two lines across marking
the top and the bottom of the band of interest. Extend these
lines onto the unstained plate. Scrape the silica gel onto weigh-
ing paper.
4. Prepare a flash chromatography column. Wash the column
with 10 volumes of chloroformmethanol in order to remove
contaminants.
22 Jacqueline S.R. Chin and Joanne Y. Yew

5. Place a clean glass vial under the column for collection of the
solvent. Make sure the volume of the vial will be sufficient for
at least 5 the volume of the scraped silica gel.
6. Carefully add the scraped gel from the TLC plate to the column
using a glass funnel.
7. To elute compounds from the silica gel, add solvent slowly into
the column using a glass syringe (see Note 12). Use the syringe
to wash residual silica dust from the funnel and walls of the
pipette. To ensure thorough elution, use a volume of solvent
5 the volume of the silica gel.
8. Push out residual solvent by aspiration using a rubber bulb
placed over the top opening of the column.
9. Evaporate the eluant under a gentle stream of nitrogen.

3.4 Analysis of 1. Prior to analysis of the experimental samples, establish the

Extracts by GCMS minimal concentration and sample volume that is suitable for
the instrument. These parameters will vary depending on the
instrument (see Notes 13 and 14).
2. Establish the retention times of compounds of interest by first
analyzing synthetic standards. The retention time will change
depending on the instrument gradient conditions and
column.
3. Standard gradient conditions for analysis of D. melanogaster
hydrocarbons are as follow:
(a) Temperature gradient: column temperature begins at 50 C
for 2 min and increases to 300 C at a rate of 15 C/min.
(b) Column: 5 % phenyl-methylpolysiloxane (DB-5, 30 m
length, 0.32 i.d., 0.25 m film thickness, Agilent, Santa
Clara, CA, USA).
4. Redissolve the prepared extracts in hexane (without hexaco-
sane standard) and transfer to the appropriate sample vials
specific for the GCMS instrument (see Notes 15 and 16).
5. For quantification, prepare a dilution series of synthetic stan-
dards in the appropriate concentration range (see Note 17).
6. Run all samples (e.g., experimental genotypes vs. control)
under the same instrument conditions.
7. Align retention times of peaks in sample with standards. If elec-
tron ionization (EI) information is available, match the EI spec-
trum with the National Institute of Standards and Technology
(NIST) Mass Spectral Library (see Note 18) or with the EI
spectrum obtained from analysis of the synthetic version.
8. Quantification can be done by normalizing the area under the
compounds of interest to the area under the peak of hexacosane
 Chemical Analysis of Drosophila Pheromones 23

or to the total area under all of the peaks (the total ion chromato-
gram). The relative abundance amount of the compound is the
average of the three to five replicates (see Note 19).

3.5 Analysis 1. We operate under 2 mbar buffer gas in the ion source (see
of Intact Insects Note 20). Under these conditions, mostly potassiated mole-
Using UV-LDI MS cules ([M + K]+) and on occasion, sodiated molecules
([M + Na]+) are observed.
2. Anesthetize flies with cold or CO2. Mount flies on adhesive
tape on glass cover slips. Mount the cover slip on the custom-
ized target plate, making sure the area to be analyzed is parallel
to the same plane on which the laser beam is focused. Stack
adhesive tape and glass cover slips to achieve desired height.
3. Add a calibrant to the sample plate by mixing 0.5 L of a syn-
thetic peptide standard with 0.5 L of DHB matrix directly on
the edge of the target plate. Allow to dry.
4. Calibrate the instrument using the monoisotopic peak for the
peptide standard and the signal for DHB. Internal calibrants
that can be detected from the surface of the fly include dihex-
ose ([M + K]+ 381.07) and a triacylglyceride ([M + K]+ 815.65).
5. Raster the laser beam over areas of interest such as the ano-
genital region and the dorsal surface of the fly. Parameters such
as laser pulse duration, firing rate, fluence, and irradiation time
should be kept consistent from sample to sample.
6. Use an elemental composition calculator to determine all pos-
sible chemical formula for a given mass, instrument accuracy,
and charge (see Note 21).
7. Relative quantitation of compound abundance can be per-
formed by normalizing individual signal intensities to the total
area underneath all detected signals (see Note 19).

3.6 Analysis 1. For detection of compounds from TLC plate without extrac-
of TLC Plates tion, use a TLC plate with aluminum backing. Run the plate
Using UV-LDI MS using the same method and conditions as would be used for a
glass-backed TLC plate (see Subheading 3.3) making sure to
include a duplicate lane that will be used for staining.
2. Excise one lane with a pair of scissors and stain the lane with a
lipophilic dye (see Subheading 3.3).
3. Mark corresponding spots/bands of interest on the unstained
portion of the plate with a graphite pencil. Color over a por-
tion of the band with the pencil. Overlay the colored band
with 0.5 L of DHB solution (see Note 22).
4. Apply the graphite/DHB matrix to a portion of the TLC plate
that does not contain any bands. This region will serve as the
negative control.
24 Jacqueline S.R. Chin and Joanne Y. Yew

5. Adhere the TLC plate to the sample plate using sticky tabs or
double-sided tape, making sure that the area to be analyzed
is parallel to the same plane on which the laser beam is
focused.
6. Add a calibrant to the sample plate by mixing 0.5 L of a syn-
thetic peptide standard with 0.5 L of DHB matrix directly on
the edge of the target plate. Allow to dry.
7. Calibrate the instrument using the signal for the peptide stan-
dard and the signal for DHB.
8. Raster the laser over the individual bands, making note of
which signals are detected in which position on the plate.
9. Obtain a mass spectrum from the spot serving as a negative
control. DHB and graphite will produce signals as well.
This control will facilitate identification of signals belonging
to the biological sample vs. signals arising from the matrix
or silica.

3.7 Analysis 1. DART MS analysis is performed using the following ion source
of Intact Insects settings: the gas heater is set to 200 C, the glow discharge
Using DART MS needle is set at 3.5 kV. Electrode 1 is set to +150 V and elec-
trode 2 is set to +250 V. Helium gas flow is set to 2.5 L/min.
Under these conditions, mostly protonated ([M + H]+) mole-
cules are observed (see Note 23).
2. Anesthetize flies with cold or CO2.
3. Using forceps that have been cleaned with 70 % ethanol, pick
up a fly by both its wings, making sure not to poke or squish
the fly (see Note 24).
4. Put the fly in the stream of charged helium gas. Hold the fly in
the stream until peaks of triacylglycerides start to appear
(see Note 25).
5. Hold the sample in the DART stream in approximately the
same location for the same amount of time each time in order
to produce reproducible spectra from sample to sample.
6. Remove the fly and immediately after, place a capillary tube
dipped into PEG into the stream (see Note 26).
7. Calibrate mass spectra using PEG as a standard.
8. Relative quantitation of compound abundance can be per-
formed by normalizing individual signal intensities to the total
area underneath all detected signals (see Note 19).
9. Use an elemental composition calculator to determine all pos-
sible chemical formulae for a given mass, instrument accuracy,
and charge (see Note 21).
 Chemical Analysis of Drosophila Pheromones 25

4 Notes

1. The number of flies needed will depend on the sensitivity of

the instrument used for detection. Each adult D. melanogaster
contains approximately 11.5 g of cuticular lipids. For GCMS,
six to ten flies is adequate. The limit of detection of the instru-
ment should be assessed prior to experiments by testing extrac-
tions using different quantities of flies, e.g., from 1 to 10.
2. Flies are isolated as pupae to prevent mating and hydrocarbon
exchange from contact. Flies will adsorb cuticular hydrocar-
bons from other flies even in the absence of mating. Extraction
should be performed around the same time each day since
quantities of cuticular hydrocarbon varies with circadian
rhythm.
3. Hexane is the standard solvent used for extraction. A chloro-
formmethanol (66:33; per vol) solution will extract more
lipids, particularly triacylglycerides. However, because this sol-
vent seems to break the cuticle down faster, we recommend a
shorter incubation time of 5 min. Ethanol may be used when
selectively extracting polar compounds such as sugars.
4. Be careful not to damage the fly tissue by piercing the cuticle
with the syringe tip since doing so will result in internal stores
of triacylglycerides and other lipids to be released.
5. In between each transfer, clean the glass syringe by aspirating
from a solution of chloroformmethanol (66:33; per vol) and
dispensing into a waste container. Repeat three times. Repeat
this process with a second bottle of chloroformmethanol
(66:33; per vol). Hereafter, all cleaning of syringes should pro-
ceed by using each bottle of washing solvent in the same order.
6. When evaporating volumes above 15 mL, it is more efficient to
use a rotary evaporator. Vacuum evaporators such as a SpeedVac
should be avoided since more volatile compounds such as fatty
acid methyl esters are likely to evaporate.
7. Waiting for too short a time causes the solvent front to be
curved and bands to be misaligned.
8. Usually a maximum of ten lanes can be plated for a 10 10 cm
TLC plate. With each lane containing a concentration corre-
sponding to approximately 100 flies, the maximum number of
flies that can be extracted for each plate is 1,000. Having too
high a concentration for each lane causes obstruction to the
separation of the compounds where the larger compounds
such as the triacylglycerides block the smooth migration of
small compounds such as dienes.
9. The solvent in the tank can be reused for ten plates though the
retention index (distance from the origin) will change with
26 Jacqueline S.R. Chin and Joanne Y. Yew

each repeated use. Other factors such as temperature, humidity

can also affect separation. Therefore, it is important to use
one lane for staining or to have a standard on each plate in
order to determine the retention index of the target
compound.
10. Fumes from the TLC tank are hazardous. Avoid inhaling and
use tanks only in the fumehood.
11. Wear goggles to prevent UV light from going into the eyes.
12. Elute the column with the same solvent that was used for
extraction.
13. Use the minimal volume allowed for the GCMS instrument.
By using smaller volumes, the extract will be more concen-
trated and fewer flies may be needed for analysis.
14. The volume of sample solvent can be reduced by using a glass
micro insert into the sample vial. A sample volume of 100
200 L is normally needed for standard GCMS autosamplers.
The insert allows volumes as small as 2050 L to be used.
15. Make sure the sample vials are compatible with the brand of
instrument. Using the wrong vials can damage the autosam-
pler syringe.
16. If amounts are scarce, the extract can be dissolved in small
volumes (less than 10 L) and directly injected into the
instrument.
17. When performing absolute quantification, it is necessary to
generate a standard curve using use the synthetic version of the
compound of interest rather than a generic hydrocarbon. The
signal intensity for a compound depends on both its abun-
dance and ionization efficiency.
18. The NIST MS library is available by subscription and can be
found on most GCMS instruments.
19. The total area under a subset of peaks can be selected for nor-
malization rather than the total area under all detected peaks.
This method is preferred if there is variation in the number or
intensity of background or contamination peaks from sample
to sample. It is important to note that these methods of nor-
malization will provide only relative quantification and do not
indicate an absolute increase or decrease in amounts of a par-
ticular compound.
20. Optimal conditions for ion source pressure, extraction voltages,
and quadrupole voltages will differ for each instrument [10].
21. A freeware elemental composition calculator can be down-
loaded here: http://www.wsearch.com.au/Tools/elemental_
composition_calculator.htm. Typically, we limit the elements
to C, H, O, Na, and K.
 Chemical Analysis of Drosophila Pheromones 27

22. In our experience, combining both DHB and graphite results

in more intense signals when analyzed by UV-LDI MS.
Graphite and pencil lead have been shown to be effective
matrices for MS analysis of small molecules and lipids [14, 15].
23. Placing a vial of concentrated ammonium acetate beneath the
DART stream can facilitate the formation of NH4 adducts.
This modification is useful for triacylglycerides and higher
molecular weight lipids.
24. Damaging the fly tissue will cause compounds from internal
stores to be detected.
25. Hold the fly tightly to prevent it from being sucked into the
instrument. Detection of triacylglyceride peaks indicates that
the cuticle has been broken and compounds from internal
stores are detected.
26. Only a small amount of PEG is needed for detection. Be care-
ful not to touch any part of the inlet or DART source with the
capillary tube containing PEG. Doing so can result in contami-
nation peaks for subsequent analyses.

References

1. Wyatt T (2003) Pheromones and animal practical and methodologic aspects. J Am Soc
behaviour. Cambridge University Press, Mass Spectrom 22:12731284
Cambridge, MA 10. Yew JY, Dreisewerd K, Luftmann H et al
2. Howard RW, Blomquist GJ (2005) Ecological, (2009) A new male sex pheromone and novel
behavioral, and biochemical aspects of insect cuticular cues for chemical communication in
hydrocarbons. Annu Rev Entomol 50:371393 Drosophila. Curr Biol 19:12451254
3. Butterworth FM (1969) Lipids of Drosophila: 11. Yew JY, Dreisewerd K, de Oliveira CC et al
a newly detected lipid in the male. Science (2011) Male-specific transfer and fine scale spa-
163:13561357 tial differences of newly identified cuticular
4. Brieger G, Butterworth FM (1970) Drosophila hydrocarbons and triacylglycerides in a
melanogaster: identity of male lipid in repro- Drosophila species pair. PLoS One 6:e16898
ductive system. Science 167:1262 12. Cody RB, Larame JA, Durst HD (2005)
5. Antony C, Jallon J (1982) The chemical basis Versatile new ion source for the analysis of
for sex recognition in Drosophila melanogas- materials in open air under ambient conditions.
ter. J Insect Physiol 28:873880 Anal Chem 77:22972302
6. Tentschert J, Kolmer K, Hlldobler B et al 13. Yew JY, Cody RB, Kravitz EA (2008) Cuticular
(2001) Chemical profiles, division of labor and hydrocarbon analysis of an awake behaving fly
social status in Pachycondyla queens using direct analysis in real-time time-of-flight
(Hymenoptera: formicidae). mass spectrometry. Proc Natl Acad Sci U S A
Naturwissenschaften 88:175178 105:71357140
7. Bland JM, Osbrink WL, Cornelius ML et al 14. Black C, Poile C, Langley J et al (2006) The
(2001) Solid-phase microextraction for the use of pencil lead as a matrix and calibrant
detection of termite cuticular hydrocarbons. for matrix-assisted laser desorption/ionisa-
J Chromatogr A 932:119127 tion. Rapid Commun Mass Spectrom 20:
8. Everaerts C, Farine JP, Cobb M et al (2010) 10531060
Drosophila cuticular hydrocarbons revisited: 15. Langley GJ, Herniman JM, Townell MS
mating status alters cuticular profiles. PLoS (2007) 2B or not 2B, that is the question: fur-
One 5:e9607 ther investigations into the use of pencil as a
9. Yew JY, Soltwisch J, Pirkl A et al (2011) Direct matrix for matrix-assisted laser desorption/
laser desorption ionization of endogenous and ionisation. Rapid Commun Mass Spectrom
exogenous compounds from insect cuticles: 21:180190
 Chapter 3

Analysis of Volatile Mouse Pheromones

by Gas Chromatography Mass Spectrometry
Milos V. Novotny and Helena A. Soini

Abstract
High-precision quantitative profiling of volatile organic constituents in rodent physiological fluids and
glandular secretions is needed to relate olfactory signals to physiology and behavior. Whereas capillary gas
chromatography-mass spectrometry (GC-MS) analysis has become the most widely applied in such
investigations, the extraction and preconcentration of volatile organics is arguably the most critical step in
the overall analytical task. In this chapter, we describe technical details of two main sample extraction
procedures used in our laboratory: dynamic headspace trapping, and stir bar sorptive extraction (SBSE).
They have been demonstrated here for the chromatographic analysis of mouse urine, serum, saliva, and
preputial gland specimens.

Key words Dynamic headspace trapping, Stir bar sorptive extraction, Gas chromatography-mass
spectrometry, Mouse urine, Mouse serum, Mouse saliva, Mouse preputial glands

1 Introduction

1.1 General Among the rodent species, the house mouse (Mus domesticus) has
Considerations become a most popular and widely studied mammal in terms of
chemical (olfactory) communication. Although olfactory commu-
nication is ubiquitous in nature, ranging from microbial systems to
insects, lizards, birds, etc., to mammals including primates, the
house mouse has evolutionarily developed a most sophisticated
system of chemical messengers. Some of the structurally identified
substances in this species truly deserve denotation as phero-
mones, the term initially and primarily coined for the insect world
[1]. The unique scientific opportunities to study chemical com-
munication systems of rodents, in general, and the house mouse,
in particular, were noted already in the 1950s and 1960s during
the first observations of the importance of olfactory signals in the
Lee-Boot effect for estrus suppression [2, 3] and the Whitten effect
in estrus synchronization [4, 5]. Many years after these basic bio-
logical discoveries, the interest of scientific community has further

Kazushige Touhara (ed.), Pheromone Signaling: Methods and Protocols, Methods in Molecular Biology, vol. 1068,
DOI 10.1007/978-1-62703-619-1_3, Springer Science+Business Media, LLC 2013

29
30 Milos V. Novotny and Helena A. Soini

been reinforced through the structural identification of the first

primer mouse pheromones such as dehydro-exo-brevicomin and
2-sec-butyl-4,5-dihydrothiazole [6], 2,5-dimethylpyrazine [7],
and the farnesene isomers that are dominance-signaling factors [8]
in male mice.
Following the groundbreaking discovery by Buck and Axel [9]
in decoding the basic mechanisms in mammalian olfaction, the
rodent species have become increasingly popular as appropriate
models for pheromone-receptor investigations. This, in turn, has
increased the needs for structural identification of additional con-
stituents of rodents biological fluids and glandular secretions as
the potential ligands for neurobiological studies in the vomero-
nasal organ [1012] and the major olfactory epithelium [1315].
Additionally, the knowledge of chemical structure of such olfac-
tants has been important in elucidating the relevant biochemical
mechanisms such as protein-pheromone binding (complex forma-
tion and its relevant structural and functional attributes at the
molecular level) [1622].
The unique role of laboratory-bred mice for numerous bio-
logical, biomedical, and pharmacological investigations will further
necessitate improved knowledge of the composition of complex
metabolite and protein mixtures excreted by these most valuable
experimental animals. Simultaneously, new bioanalytical tech-
niques in the fields of genomics, proteomics, and metabolomics are
now capable yielding a nearly complete knowledge of the systems
biology of the house mouse from which some parallels can be
drawn concerning more ecologically significant attributes of the
wild mice. The pertinent aspects of bioanalytical quantitative pro-
filing of volatile metabolites (relatively small molecules) will be
detailed below.
While chemical communication in mice can be principally
mediated by either volatile or nonvolatile molecules, the mouse has
on its body a variety of known and potential signal sources including
urine, saliva, preputial glands, plantar eccrine glands, Harderian
glands, exorbital lacrimal glands, coagulating glands, submaxillary
glands, and non-specialized sebaceous glands [23]. The best stud-
ied sources of volatile chemosignals seem urine and preputial
glands, where a variety of both primer and releaser pheromone
signals were initially located. These two sources are also easily avail-
able for the profiling studies by modern analytical procedures such
as capillary gas chromatography-mass spectrometry (GC-MS) and
liquid chromatography-mass spectrometry (LC-MS).
Whereas our methodological descriptions in this chapter relate
extensively to volatile organic compounds as chemical messengers,
the unique capability of the house mouse and some other rodents
has been to excrete into their environments a series of lipocalin
proteins [24, 25], which can interact with the volatile pheromones
and facilitate their ecological impact. The best known of these
 Analysis of Volatile Mouse Pheromones by Gas Chromatography 31

biomolecules are so-called major urinary proteins (MUPs), which

had been noticed as endocrinologically dependent urinary constit-
uents for a number of decades [26], but could be studied in their
molecular details only recently through the availability of modern
structural tools, such as mass spectrometry [27, 28], NMR spec-
trometry [18, 19, 29] and X-ray crystallography [30]. The current
studies seem to endorse the hypothesis that the associations
between the volatile molecules and such proteins are at the heart of
the complex chemical communication system that the rodent
uniquely utilize in their environment, although the MUPs may
have unique roles of their own in the chemical communication
events [31, 32].

1.2 Methodological The current success of the volatile pheromone profiling not only
Considerations reflects the instrumental capabilities of todays GC-MS in terms of
measurement sensitivity and resolution of complex biological mix-
tures, but it is also dependent on the modern techniques of solute
preconcentration and sampling. Different sampling conditions will
be described in this article, representing primarily the procedures
practiced in the authors laboratory, while other modified or
optimal conditions may need to be sought for other specialized
investigations of volatile constituents of rodent urines and other
related biological materials.
The metabolomic profiling is usually related to a series of
quantitative measurements where precision (repeatability) is of a
primary concern, such as the analytical runs aiming at a comparison
between male and female animals, the experiments involving endo-
crinological or behavioral manipulations, genetic studies, etc. [7,
3339]. As an example, the biological relevance of such compara-
tive analyses is represented in Fig. 1. Here, the urinary output of
the male Mus domesticus and the male (monogamous) Mus spicile-
gus is being compared with the possible correlation of their different
chemistry [39] and behavioral traits [40, 41].
Sampling pheromones from different biological materials for
the GC-MS analysis requires efficient enrichment of the com-
pounds of interest, as certain biologically active components may
act at trace levels. Enrichment techniques, such as solvent extrac-
tion [4244], solid-phase microextraction, SPME [4548],
adsorption methods using porous polymers [34, 37, 4951], and
the sorptive stir bar extraction (SBSE) [38, 39, 52, 53] are among
the commonly used approaches. Solvent extraction is more
applicable to less volatile compounds due to the potential losses of
volatiles during the solvent evaporation steps. Volatile compounds,
which are suitable for solvent extraction need to be present at rela-
tively high levels to be detected using this method, since only a
fraction of the final sample volume will be introduced to the
analytical instrument. Solid-phase microextraction (SPME) has
relatively low sample capacity, while displacement of some low-level
32 Milos V. Novotny and Helena A. Soini

TIC
900000

800000

700000

600000
Mus spicilegus
Intact (I)
500000

400000
-hexalactone
300000

200000

100000

5.00 10.00 15.00 20.00 25.00 30.00 35.00

900000 2-coumaranone Mus spicilegus

Castrated ( C )
800000

700000

600000

500000

400000

300000

200000

100000

5.00 10.00 15.00 20.00 25.00 30.00 35.00

2-sec-butyl-4,5-
dihydrothiazole Mus domesticus
900000
Intact
800000

700000

600000

500000

400000

300000

200000

100000

5.00 10.00 15.00 20.00 25.00 30.00 35.00

Time (min)

Fig. 1 Comparative urinary profiles (GC-MS total ion current) of intact and castrated male Mus spicilegus and
intact Mus domesticus. Reprinted with permission from ref. [39]. Method as described in Subheading 3.2.1,
except the urine sample volume was 200 l
 Analysis of Volatile Mouse Pheromones by Gas Chromatography 33

Abundance
190000 4
5 6
180000
170000
160000
150000
140000
130000
120000
110000
100000
90000
80000
70000
60000
50000
1
40000
30000
2
20000 3

10000

10.00 20.00 30.00 40.00 50.00 60.00 70.00 80.00

Time (min)

Fig. 2 Volatile compound profile (GC-MS total ion current) measured from male ICR mouse serum as described
in Subheading 3.2.3 below. Arrows indicate 1: and 2: dihydrofuran decomposition products of 6-hydroxy-
6-methyl-3-heptanone; 3: dehydro-exo-brevicomin; 4: 2-sec-butyl-4,5-dihydrothiazole; 5: geraniol; 6: farnesol.
Modified and reprinted from ref. [38]

or low molecular weight compounds by the more concentrated

sample constituents or higher molecular weight compounds can
take place within a limited polymer volume of the SPME fiber
[54]. Therefore, we generally prefer the sorptive stir bar extraction
(SBSE) with higher PDMS (polydimethylsiloxane) sorptive poly-
mer volume (24 l) as compared with 0.6 l polymer volumes in
typical SPME fibers [54].
In this chapter, we will emphasize two methods being actively
used in our laboratory: (1) Tenax sorbent-aided dynamic head-
space procedure used for the mouse urine samples; and (2) the
sorptive stir bar extraction (SBSE) method used for mouse urine,
preputial glands, serum and saliva samples. In addition to urine,
investigation of the volatile compounds in the preputial glands are
important for the mouse chemical communication, since these
glands are the original source of important pheromone constitu-
ents in urine, such as the farnesenes, which are linked to dominant
and aggressive behavior in males [8]. Measuring volatile com-
pound levels in serum, on the other hand, provides information on
the metabolic routes involved in the biosynthetic pathways leading
to the chemical signaling compounds [55]. Figure 2 (Subheading
3.2.3 described below) illustrates the volatile compound profile
extracted from mouse serum [38].
Saliva, like urine, can be collected without sacrificing the
animal. Anesthetized mice can produce enough saliva to prove the
presence of some of the key male mouse pheromones in this material,
34 Milos V. Novotny and Helena A. Soini

Abundance
3
200000

180000

160000

140000

120000

100000

80000
2
60000

40000

20000 1

0
5.00 10.00 15.00 20.00 25.00 30.00 35.00 40.00 45.00 50.00 55.00 60.00 65.00 70.00
Time (min)

Fig. 3 Volatile compound profile (GC-MS total ion current) from male ICR mouse saliva as described in
Subheading 3.2.4 below. Arrows indicate selected identified compounds. 1: limonene; 2: 2-sec-butyl-4,5-
dihydrothiazole; 3: nonanal [56]

which are also present in serum and urine, as shown in Fig. 3

(Subheading 3.2.4 described below) [56].
While this article is confined to documentation of the ana-
lytical strategies to investigations on the mouse chemosignals,
similar strategies have served our laboratory in other studies in
the field of chemical ecology, albeit sometimes with procedural
modifications.

2 Materials

All used materials need to be at the highest purity to avoid co-

extracting exogenous impurities that can interfere with the com-
pounds of biological interest. Rinsing glassware and ceramics with
acetone and drying them at 8095 C in the oven is highly recom-
mended. When choosing the plastic materials as sample vials, inert
polypropylene is our recommended choice.

2.1 Dynamic 1. Ultrapure water (OmniSolv EM Science, Gibbstown, NJ).

Headspace Trapping 2. Methanol (J.T. Baker, Baker Analyzed HPLC Solvent,
of Urinary Volatiles Mallinckrodt Baker, Phillipsburg, NJ).
3. Internal standard, 7-tridecanone (Aldrich Chemical Company,
Milwaukee, WI) (see Note 1).
4. Helium gas (Ultrahigh purity, Airgas, Inc., Radnor, PA).
5. Glass condenser with the ice water cooling loop (see Fig. 4 and
Note 2).
 Analysis of Volatile Mouse Pheromones by Gas Chromatography 35

Dynamic Head-Space Sampling

Glass GC injector liner

TenaxTM TA (8 mg)
Glass
wool

Ice water cooling

Helium 70 ml/min

Urine

Water bath 40 C

Fig. 4 Schematic of the dynamic head space sampling setup and a photograph of the used glassware

6. Tenax sorbent (TA, 35/65 mesh, Alltech, Deerfield, IL).

7. Splitless GC injector liner 78.5 6.0 mm, 2 mm, i.d. (Grace
Davison Discovery Sciences, Deerfield, IL), to pack Tenax
polymer between glass wool plugs (glass wool, DMCS treated,
Alltech Associates, Deerfiled, IL) (see Fig. 4).
8. A modified GC injector port.
9. A loop of uncoated deactivated silica tubing (30 cm 0.25 mm,
i.d., Restek) attached with the universal Press-Tight Connector
(Restek Corporation, Bellefonte, PA) to the separation column.
10. Plastic 200-ml liquid nitrogen container.
11. The separation capillary column DB-5MS (30 m 0.25 mm,
i.d., 0.25 m film thickness) (Agilent, J&W Scientific, Folsom,
CA).
12. Finnigan MAT Magnum ion trap GC-MS system (Finnigan
MAT, San Jose, CA).

2.2 Stir Bar 1. Ultrapure water (OmniSolv EM Science, Gibbstown, NJ).

Extraction of Urine 2. Methanol (J.T. Baker, Baker Analyzed HPLC Solvent,
Mallinckrodt Baker, Phillipsburg, NJ).
36 Milos V. Novotny and Helena A. Soini

Fig. 5 Schematic of the stir bar sorptive extraction (SBSE) setup. In the photograph, samples are being
extracted in a batch mode

3. Internal standard, 7-tridecanone (Aldrich Chemical Company,

Milwaukee, WI).
4. Stir bars (Twister PDMS polymer-coated stir bar, dimensions
10 mm, 0.5 mm film thickness, 24 l PDMS volume, Gerstel
GmbH, Mlheim an der Ruhr, Germany) (see Note 3).
5. Wheaton disposable glass scintillation vials, 20-ml (VWR
International, West Chester, PA) (see Fig. 5 and Note 4).
6. Stir plate (15-place) Variomag Multipoint HP 15 stirplate
(H+P Labortechnic, Oberschleissheim, Germany).
7. Kimtech Kimwipes (Kimberly Clark, Roswell, GA).
8. TDS Desorption tubes (Gerstel GmbH) (see Note 5).
9. The gas chromatograph-mass spectrometer (GC-MS) instru-
ment: Agilent 6890N gas chromatograph connected to 5973i
MSD mass spectrometer (Agilent Technologies, Inc.,
Wilmington, DE).
10. The separation capillary column DB-5MS (30 m 0.25 mm,
i.d., 0.25 m film thickness) (Agilent, J&W Scientific,
Folsom, CA).
11. Thermal Desorption Autosampler and Cooled Injection
System (TDSACIS4) (Gerstel GmbH, Mlheim an der Ruhr,
Germany).

2.3 Stir Bar 1. Miltex disposable safety scalpel (VWR International, West
Extraction of Preputial Chester, PA).
Glands 2. CoorsTek porcelain 65-ml mortar and pestle (VWR
International, West Chester, PA).
3. Liquid nitrogen (see Note 6).
4. Ultrapure water (OmniSolv EM Science, Gibbstown, NJ).
 Analysis of Volatile Mouse Pheromones by Gas Chromatography 37

5. Methanol (J.T. Baker, Baker Analyzed HPLC Solvent,

Mallinckrodt Baker, Phillipsburg, NJ).
6. Internal standard, 7-tridecanone (Aldrich Chemical Company,
Milwaukee, WI).
7. Ammonium sulfate (99.95 %, Aldrich Chemical Company,
Milwaukee, Wisconsin) (see Note 7).
8. See, items 811 in Subheading 2.2 above.

2.4 Stir Bar 1. Nonspecific protease enzyme (type XIV), Pronase E from
Extraction of Serum Streptomyces griseus, 5.3 units/mg (Sigma-Aldrich, St. Louis,
MO).
2. 50 mM ammonium bicarbonate (NH4HCO3), pH 8.2 (J.T.
Baker, Phillipsburg, NJ).
3. Ultrapure water (OmniSolv EM Science, Gibbstown, NJ).
4. Methanol (J.T. Baker, Baker Analyzed HPLC Solvent,
Mallinckrodt Baker, Phillipsburg, NJ).
5. Internal standard, 7-tridecanone (Aldrich Chemical Company,
Milwaukee, WI).
6. See items 811 in Subheading 2.2 above.

2.5 Stir Bar 1. Enzyme -amylase from Bacillus licheniformis (Sigma-Aldrich,

Extraction of Saliva St. Louis, MO).
2. Enzyme Glyko Sialidase A from Arthobacter ureafaciens
(Prozyme, Hayward, CA).
3. Ultrasonication bath FS30 (Fisher Scientific, Waltham, MA).
4. Ammonium sulfate (99.95 %, Aldrich Chemical Company,
Milwaukee, Wisconsin).
5. 30 mM sodium phosphate (Na2HPO4) (Mallincrodt Chemicals,
Austin, TX), pH to 6.5 with o-phosphoric acid (Sigma-Aldrich,
St. Louis, MO).
6. Methanol (J.T. Baker, Baker Analyzed HPLC Solvent,
Mallinckrodt Baker, Phillipsburg, NJ).
7. Internal standard, 7-tridecanone (Aldrich Chemical Company,
Milwaukee, WI).
8. See items 811 in Subheading 2.2 above.

3 Methods

The used sample preparation and analysis methods aim to release

maximally small organic volatile compounds from the matrix com-
pounds and to do so reproducibly and quantitatively. These attri-
butes are extremely important in the biological studies comparing
38 Milos V. Novotny and Helena A. Soini

compound levels in different study subject groups in the exact

statistical manner. These methodological attributes should also be
applicable at the low-concentration levels, since some of the phero-
monally active sample components may exist at low levels. Different
matrix materials are known to influence markedly the success in
isolating small organic molecules from different media. This is
taken into consideration by using appropriate additives such as
ammonium sulfate, digestive enzymes, co-solvents, etc., to improve
recoveries of the analytes of interest. Verification of the compound
identities is generally performed using either commercially avail-
able or synthesized standard materials, with a subsequent record-
ing of retention time index data [57], through high-resolution
mass spectrometry measurements [55], and element-specific detec-
tor responses for verifying the presence of nitrogen- and sulfur-
containing compounds in our samples [38].

3.1 Dynamic The current procedure used in our laboratory has its origin in the
Headspace Trapping analytical studies of the 1970s [5860], in which various volatile
of Urinary Volatiles trace constituents of aqueous media were purged and successfully
preconcentrated on a small precolumn (capsule) containing a ther-
mostable porous polymer. Following analytically adequate precon-
centration time (optimized during the method development), the
trapped volatiles are released at high temperature. Typically, this
takes place at the modified hot injection port of a gas chromato-
graph. Released (desorbed) volatiles are directed to a cryogenically
cooled portion of the capillary precolumn [34, 37, 52] to cryofo-
cus the compound mixture into a narrow frozen band. Actual
injection onto the separation column occurs when the frozen sec-
tion of the column is rapidly heated and the compounds from the
cryofocused band transfers back into the gas phase and the mixture
is separated in the capillary column, followed by detection in
the MS for quantitative evaluations. As described in the original
studies of the analytical aspects of this preconcentration procedure
[5860], the purpose of the preconcentration step and the subse-
quent recording of the profiles of volatiles is the capability of
comparative studies of different samples. Thus, the reproducibility
of trapping and desorption, rather than a complete solute adsorp-
tion, is the primary objective. Our dynamic headspace sampling
apparatus is shown in Fig. 4. As an alternative, there are available
commercial purge-and-trap autosamplers dedicated to GC-MS
instruments, such as an autosampler device from Teledyne-Tekmar
(Mason, OH).

3.1.1 Dynamic Head 1. Place 1.0 ml of urine and 2 ml of ultrapure water in a glass ves-
Space Extraction of Urine sel connected to a helium flow to purge the volatiles from the
urine sample.
2. Place the glass vessel in the water bath at 40 C.
 Analysis of Volatile Mouse Pheromones by Gas Chromatography 39

3. Place the condenser with the ice-water cooling flow on the top
of urine vessel.
4. Pack 8 mg of Tenax TA in the 7.8-cm long glass injector liner
and close Tenax section ends with a plug of glass wool.
5. Place the injector liner containing Tenax section on the top of
the condenser.
6. Turn cold water stream on to reduce the amount of condensed
water in the gaseous volatile mixture and turn helium flow on
(70 ml/min) for collecting the sample headspace as shown in
Fig. 4 (see Note 2).
7. Let the helium flow purge (bubble) into the sample for 30 min,
while volatiles are being trapped in the Tenax adsorbent.
8. Remove the injector liner with the Tenax section and place it
inside the GC injector (carrier gas flow disconnected) and fol-
low the GC-MS instructions in Subheading 3.3 below.

3.2 Stir Bar During the late 1990s, Sandra and coworkers found the polydimeth-
Extraction of Urine, ylsiloxane (PDMS) polymer that was well suited for preconcentrat-
Preputial Glands, ing organic molecules from dilute aqueous and gaseous
Serum and Saliva environmental samples [61, 62]. When the PDMS film (thickness
0.5 or 1.0 mm) is placed around a magnetic stir bar (1-cm length),
sorptive extraction of small organic molecules into the PDMS layer
can be performed in the aqueous phase, while the sample placed in
a flat bottom vial is stirred with the immersed PDMS stir bar, with
the vial being placed on the top of the magnetic stir plate (see
Fig. 5). The PDMS polymer action is stated as sorptive rather
than adsorptive. Small organic molecules dissolve inside the
polymer instead of adsorbing on the surface. While using 4560 min
extraction time, the sorptive process reaches an equilibrium, which
improves the methods reproducibility greatly. In early 2000,
Twister stir bars based on this idea became commercially available
from the Gerstel Company (Germany). Precise manufacturing of
stir bars of identical dimensions allows to extract multiple samples
simultaneously as illustrated in Fig. 5 with excellent reproducibility
[38, 54, 63].

3.2.1 SBSE of Urine 1. Place 0.5 ml mouse urine in a disposable 20-ml glass scintilla-
tion vial (see Fig. 5).
2. Add 2 ml of ultrapure water.
3. Pipet 5 l of internal standard 7-tridecanone (8 ng in 5 l
methanol).
4. Insert clean stir bar into the vial (see Note 8).
5. Stir for 60 min at 850 rpm.
6. Take the stir bar out from the sample, on the top of clean lint-
free paper tissue.
40 Milos V. Novotny and Helena A. Soini

7. Pipet 2 1 ml of ultrapure water on the top of the stir bar.

8. Dry the stir bar with clean dry Kimwipes sheet.
9. Place the stir bar in the desorption tube.
10. See Subheading 3.4 for thermal desorption, cryofocusing, and
the GC-MS analysis.

3.2.2 SBSE of 1. Weigh the frozen preputial gland.

Preputial Glands 2. While still frozen, cut the gland into thin slices.
3. Place the gland slices on the clean ceramic mortar vessel.
4. Pour about 50 ml of liquid nitrogen in the ceramic vessel.
5. Crush with the pestle the frozen tissue quickly into powder. If
necessary, add more liquid nitrogen.
6. Rinse the contents of the ceramic vessel into a clean 20-ml
glass scintillation vial with 2 ml of ultrapure water.
7. Add 100 mg of ammonium sulfate (see Note 9).
8. Add 50 l of 7-tridecanone (80 ng in 50 l methanol) (see
Note 10).
9. Stir for 60 min at 850 rpm.
10. Take the stir bar out from the sample vial, on the top of clean
lint-free paper tissue.
11. Pipet 2 1 ml of ultrapure water on the top of the stir bar.
12. Dry the stir bar with clean dry Kimwipes.
13. Place the stir bar in the desorption tube.
14. See Subheading 3.4 for thermal desorption, cryofocusing and
the GC-MS analysis.

3.2.3 SBSE of Serum 1. Pipet 10 ml of ultrapure water and 10 ml of 50 mM buffer into

Samples a clean 20-ml scintillation glass vial.
2. Add 200 l serum.
3. Add 2 mg of Pronase E protease.
4. Wrap vial in foil, and allow enzyme digestion to take place
overnight at room temperature (see Note 11).
5. Insert a clean stir bar in a vial and add 20 l 7-tridecanone
(40 ng in 20 l of methanol).
6. Stir for 2 h at 850 rpm.
7. Take the stir bar on the top of clean lint-free paper tissue.
8. Pipet 2 1 ml of ultrapure water on the top of the stir bar.
9. Dry the stir bar with clean dry Kimwipes.
10. Place the stir bar in the desorption tube.
11. See Subheading 3.4 for thermal desorption, cryofocusing and
the GC-MS analysis.
 Analysis of Volatile Mouse Pheromones by Gas Chromatography 41

3.2.4 SBSE of Saliva 1. Pipet 1.8 ml of 30 mM sodium phosphate buffer adjusted to

Samples pH 6.5 with o-phosphoric acid and 150 l of saliva into a 20-ml
scintillation vial.
2. Add enzymes -amylase and sialidase A (5 l of each) (see
Note 12).
3. Wrap the vial in foil and allow to digest overnight at room
temperature (see Notes 10 and 11).
4. Place a clean stir bar, 20 l of the internal standard 7-trideca-
none in methanol (16 ng in 20 l) and 300 mg of ammonium
sulfate (NH4)2SO4 in a vial.
5. Sonicate the sample vial for 60 min at room temperature.
6. Take the stir bar on the top of clean lint-free paper tissue.
7. Pipet 2 1 ml of ultrapure water on the top of the stir bar.
8. Dry the stir bar with clean dry Kimwipes.
9. Place the stir bar in the desorption tube.
10. See Subheading 3.4 below for thermal desorption, cryofocusing
and the GC-MS analysis.

3.3 Hot Injector 1. Injector temperature is set at 250 C.

Desorption-Cooled 2. Helium carrier gas flow is stopped.
Capillary Loop: GC-MS
3. Injector liner containing Tenax adsorbent is placed inside the
Analysis
injector.
4. Place the precolumn uncoated capillary loop inside the 200-ml
liquid nitrogen container in the GC oven, keep door open.
5. Resume carrier gas flow (1.0 ml/min).
6. Let helium flow thermal desorption to release the volatiles
from Tenax and cryofocus them in the deactivated precolumn
loop immersed in liquid nitrogen (150 C) for 15 min.
7. Remove the liquid nitrogen container from the GC oven and
close the oven door.
8. Start the GC oven temperature program: 40 C (hold 5 min),
an increase 2 C/min to 200 C (hold 10 min), the helium
carrier gas head pressure of 12.0 psi, flow rate of 1.0 ml/min.
9. Start manually the MS acquisition: The manifold and transfer
line temperatures are set at 220 C and 300 C, respectively,
the ion trap in the positive electron ionization mode (70 eV),
spectra scanning from 40 to 350 msu (1 scan/s).

3.4 TDSACIS/ 1. TDSA is in a splitless mode for thermal desorption: TDSA

GC-MS Analysis temperature program for desorption 20 C (hold 0.5 min), a
60 C/min ramp to 270 C (final hold 5 min), temperature of
the transfer line to the CIS, at 280 C.
2. CIS (cooled injection system) is used for cryofocusing, injec-
tion in the solvent vent mode: The CIS-4 is cooled with liquid
42 Milos V. Novotny and Helena A. Soini

nitrogen to 80 C, after desorption and cryotrapping, CIS-4

is heated at 12 C/s to 280 C (hold 10 min) for injection,
vent pressure 9.0 psi, a vent flow 50 ml/min, and a purge flow
at 50 ml/min.
3. The GC temperature program is 40 C (hold 5 min), an
increase 2 C/min to 200 C (hold 10 min), the helium carrier
gas head pressure of 9.0 psi, flow rate of 1.0 ml/min at con-
stant flow mode.
4. The MS instrument is at positive electron ionization (EI) mode
at 70 eV, the scanning rate of 2.41 scans/s over the mass range
of 40350 amu, the MSD transfer line temperature is set at
280 C, the ion source and quadrupole temperatures are set at
230 C and 150 C, respectively.

4 Notes

1. Internal standard, 7-tridecanone (dihexylketone), is stored in a

silica desiccator after opening the bottle due to hygroscopic
nature of 7-tridecanone crystals. Methanol solutions can be
stored at +4 C.
2. Helium flow (carrying volatiles) passes through an ice-water-
chilled section to minimize water excess before the volatiles are
trapped on the Tenax polymer.
3. Stir bars are cleaned and reconditioned after use in the TC-2
Tube Conditioner with AUX 163 Controller (Gerstel GmbH,
Mlheim an der Ruhr, Germany). Stir bars are placed in Gerstel
TDS desorption tubes at 300 C under the high flow of ultra-
pure helium (120 ml/min in each desorption tube).
4. Disposable glass scintillation vials are rinsed with acetone and
dried at 95 C prior to use. Vial caps with metal foil lining are
used. Caps do not need acetone wash. As the alternative, Gerstel
(Gerstel GmbH, Mlheim an der Ruhr, Germany) 10 or 20-ml
headspace glass vials can be used with screw caps provided with
blue silicone/PTFE septa (low bleed). The vials and caps are
ultraclean, not requiring acetone wash prior to use.
5. TDS tubes are (1) washed with acetone, (2) dried in the oven
at 95 C, (3) conditioned in the TC-2 Tube Conditioner with
AUX 163 Controller (Gerstel GmbH, Mlheim an der Ruhr,
Germany) at 300 C under the high flow of ultrapure helium
(120 ml/min in each desorption tube), (4) rinsed with ace-
tone, and (5) dried at 95. These steps were found necessary,
especially when tissue extractions were performed with numer-
ous samples.
6. Store liquid nitrogen in the small Dewar vessel for the use
within 12 days.
 Analysis of Volatile Mouse Pheromones by Gas Chromatography 43

7. A small container size, e.g., 50 g has been found optimal for

ammonium sulfate to avoid exogenous volatile compounds
entering inside the reagent bottle headspace over a long time.
8. Stir bar extraction can be also performed in the headspace
mode. While the sample is placed in the bottom of the vial, the
Twister stir bar is hung in air space above the sample for an
appropriate time (112 h). The sample can be kept at room
temperature or heated to 37 C.
9. Ammonium sulfate increases small-molecule recoveries from
the matrix containing proteins due to the salting-out effects
and protein denaturation [64].
10. Large amount of 7-tridecanone is necessary because of the
presence of competing hydrophobic matrix material.
11. Enzyme digest takes place in dark for 12 h at room temperature,
loosely covered with aluminum foil to allow oxygen flow into the
sample. Protease enzyme is digesting macromolecular proteins
into smaller peptide fragments, while releasing small organic vola-
tile molecules that have affinity to the protein surfaces.
12. Sialidase enzyme breaks very large mucin glycoprotein struc-
tures while reducing viscosity in the saliva sample and subse-
quently improving the extraction recoveries for small organic
target compounds.

Acknowledgments

This work was jointly sponsored by the METACyt Initiative of

Indiana University, a major grant from the Lilly Endowment, Inc.,
and the Lilly Chemistry Alumni Chair funds (to M.V.N.).

References

1. Karlson P, Lscher M (1959) Pheromones: 6. Jemiolo B, Harvey S, Novotny M (1986)

a new term for a class of biologically active sub- Promotion of the Whitten effect in female mice
stances. Nature 183:5556 by synthetic analogs of male urinary constitu-
2. Lee S, van der Boot LM (1955) Spontaneous ents. Proc Natl Acad Sci USA 83:45764579
pseudopregnancy in mice. Acta Physiol 7. Novotny M, Jemiolo B, Harvey S, Wiesler D,
Pharmacol Neerl 4:442443 Marchlewska-Koj A (1986) Adrenal-mediated
3. Lee S, van der Boot LM (1956) Spontaneous endogenous metabolites inhibit puberty in
pseudopregnancy in mice II. Acta Physiol female mice. Science 231:722725
Pharmacol Neerl 5:213214 8. Novotny M, Harvey S, Novotny M (1990)
4. Whitten WK (1956) Modification of the oes- Chemistry of male dominance in the house
trus cycle of the mouse by external stimuli mouse, Mus domesticus. Experientia 46:109113
associated with the male. J Endocrinol 9. Buck LB, Axel R (1991) A novel multigene
13:399404 family may encode odorant receptors: a molec-
5. Whitten WK, Bronson FH, Greenstein JA ular basis for odor recognition. Cell
(1968) Estrus-inducing pheromone of male 65:175187
mice: transport by movement of air. Science 10. Moss RL, Flynn RE, Shen X-M, Dudley C, Shi
161:584585 J, Novotny M (1997) Urine-derived compound
44 Milos V. Novotny and Helena A. Soini

evokes membrane responses in mouse vomero- 25. Flower DR, North ACT, Sansom CE (2000)
nasal receptor neurons. J Neurophysiol 77: The lipocalin protein family: structural and
28562862 sequence overview. Biochim Biophys Acta
11. Leinders-Zufall T, Lane AP, Puche AC, Ma W, 1482:924
Novotny MV, Shipley MT, Zufall F (2000) 26. Parfentjer JA (1932) Calcium and nitrogen
Ultrasensitive pheromone detection by mam- content in urine of normal and cancer mice.
malian vomeronasal neurons. Nature 405: Proc Soc Exp Biol Med 29:12851287
792796 27. Robertson DH, Cox KA, Gaskell SJ, Evershed
12. Holy TE, Dulac C, Meister M (2000) RP, Beynon RJ (1996) Molecular heterogeneity
Responses of vomeronasal neurons to natural in the major urinary proteins of the house mouse
stimuli. Science 289:15691572 Mus musculus. Biochem J 316:265272
13. Sam M, Vora S, Malnic B, Ma W, Novotny 28. Mechref Y, Zidek L, Ma W, Novotny MV
MV, Buck LB (2001) Odorants may arouse (2000) Glycosylated major urinary protein of
instinctive behaviours. Nature 412:142 the house mouse: characterization of its
14. Boehm U, Zou Z, Buck LB (2005) Feedback N-linked oligosaccharides. Glycobiology
loops link odor and pheromone signaling with 10:231235
reproduction. Cell 123:683695 29. Zidek L, Novotny MV, Stone MJ (1999)
15. Soucy ER, Albeanu DF, Fantana AL, Murthy Increased protein backbone conformational
VN, Meister M (2009) Precision and diversity entropy upon hydrophobic ligand binding.
in an odor map on the olfactory bulb. Nat Nat Struct Biol 6:11181121
Neurosci 12:210220 30. Timm DE, Baker LJ, Mueller H, Zidek L,
16. Bacchini A, Gaetani E, Cavaggioni A (1992) Novotny MV (2001) Structural basis of phero-
Pheromone binding proteins of the mouse, mone binding to mouse major urinary protein
Mus musculus. Experientia 48:419421 (MUP-I). Protein Sci 10:9971004
17. Hurst JL, Robertson DHL, Tolladay U, 31. Hurst JL, Payne CE, Nevison CM, Marie AD,
Beynon RJ (1998) Proteins in urine scent Humphries RE, Robertson DHL, Cavaggioni
marks of male mouse extend the longevity of A, Beynon RJ (2001) Individual recognition
olfactory signals. Anim Behav 55:12891297 in mice mediated by major urinary proteins.
18. Zidek L, Stone MJ, Lato SM, Ellington AD, Nature 414:631634
Novotny MV (1999) NMR Mapping of 32. Stopka P, Janotova K, Heyrovsky D (2007)
recombinant mouse major urinary protein I The advertisement role of major urinary pro-
binding site occupied by a pheromone 2-sec- teins in mice. Physiol Behav 91:667670
butyl-4,5-dihydrothiazole. Biochemistry 33. Jemiolo B, Alberts J, Sochinski-Wiggins S,
38:98509861 Harvey S, Novotny M (1985) Behavioural and
19. Novotny MV, Ma W, Wiesler D, Zidek L endocrine responses of female mice to syn-
(1999) Positive identification of the puberty- thetic analogues of volatile compounds in male
accelerating pheromone of the house mouse: urine. Anim Behav 33:11141118
the volatiles associating with the major urinary 34. Schwende FJ, Wiesler D, Jorgenson JW,
protein. Proc R Soc Lond B 266: Carmack M, Novotny M (1986) Urinary vola-
20172022 tile constituents of the house mouse, Mus mus-
20. Sharrow SD, Vaughn JL, Zidek L, Novotny culus, and their endocrine dependency. J Chem
MV (2002) Pheromone binding by polymor- Ecol 12:277296
phic mouse major urinary proteins. Protein Sci 35. Andreolini F, Jemiolo B, Novotny M (1987)
11:22472256 Dynamics of extraction of urinary chemosig-
21. Cavaggioni A, Mucignat-Caretta C, Zagotto nals in the house mouse (Mus musculus) dur-
G (2003) Absolute configuration of 2-sec- ing the natural estrous cycle. Experientia
butyl-4,5-dihydrothiazole in male mouse 43:9981002
urine. Chem Senses 28:791797 36. Harvey S, Jemiolo B, Novotny M (1989)
22. Armstrong SD, Robertson DHL, Cheetman Pattern of volatile compounds in dominant
SA, Hurst JL, Beynon RJ (2005) Structural and subordinate male mouse urine. J Chem
and functional differences in isoforms of Ecol 15:20612071
mouse major urinary proteins: a male-specific 37. Ma W, Miao Z, Novotny MV (1998) The role
protein that preferentially binds a male phero- of the adrenal gland and adrenal-mediated
mone. Biochem J 391:343350 chemosignals in estrus suppression in the
23. Bronson FH (1971) Rodent pheromones. Biol house mouse: the Lee-Boot effect revisited.
Reprod 4:344357 Biol Reprod 59:13171320
24. Finlayson JS, Asofsky R, Potter M, Runner CC 38. Novotny MV, Soini HA, Koyama S, Bruce KE,
(1965) Major urinary protein complex of nor- Penn D (2007) Chemical identification of
mal mice: origin. Science 149:981982 MHC-influenced volatile compounds in mouse
 Analysis of Volatile Mouse Pheromones by Gas Chromatography 45

urine. I: Quantitative proportions of major 52. Soini HA, Bruce KE, Wiesler D, David F,
chemosignals. J Chem Ecol 33:417434 Sandra P, Novotny MV (2005) Stir bar sorp-
39. Soini HA, Wiesler D, Koyama S, Fron C, tive extraction: a new quantitative and com-
Baudoin C, Novotny MV (2009) Differences prehensive sampling technique for
in urinary scents of two related mouse species, determination of chemical signal profiles from
Mus spicilegus and Mus domesticus. J Chem biological media. J Chem Ecol 31:377392
Ecol 35:580589 53. Rck F, Hadeler KP, Rammensee HG, Overath
40. Patris B, Baudoin C (2000) A comparative P (2007) Quantitative analysis of mouse urine
study of parental care between two rodent spe- volatiles: in search of MHC-dependent differ-
cies: implications for the mating system of the ences. PLoS One 5(e429):18
mound-building mouse Mus spicilegus. Behav 54. Baltussen E, David F, Sandra P, Janssen H-G,
Processes 51:3543 Cramers CA (2002) Sorptive sample prepara-
41. Poteaux B, Gouat P, Jacquot C, Christophe N, tiona review. Anal Bioanal Chem 373:322
Baudoin C (2008) Socio-genetic structure of 55. Novotny MV, Soini HA (2008) Volatile mam-
mound-building mice, Mus spicilegus, in malian chemosignals: structural and quantita-
autumn and early spring. Biol J Linn Soc Lond tive aspects. In: Hurst JL, Beynon RJ, Roberts
93:689699 SC, Wyatt T (eds) Chemical signals in verte-
42. Nishimura K, Utsumi K, Yuhara M, Fujitani Y brates 11. Springer, New York, pp 1323
(1989) Identification of puberty-accelerating 56. Soini HA, Koyama S, Karn RC, Novotny MV
pheromones in male mouse urine. J Exp Zool (2013) To be published
251:3003005 57. Kovats E (1965) Gas chromatographic charac-
43. Singer AG, Beauchamp GK, Yamazaki K terization of organic substances in the reten-
(1997) Volatile signals of the major histocom- tion index system. In: Giddings JC, Keller RA
patibility complex in male mouse urine. Proc (eds) Advances in chromatography. Marcel
Natl Acad Sci USA 94:22102214 Dekker, New York, pp 229247
44. Zhang JX, Rao XP, Sun L, Zhao CH, Qin XW 58. Novotny M, Lee ML, Bartle KD (1974) Some
(2007) Putative chemical signals about sex, analytical aspects of the chromatographic head-
individuality, and genetic background in the space concentration method using a porous
preputial gland and urine of the house mouse polymer. Chromatographia 7:333338
(Mus musculus). Chem Senses 32:293303 59. Novotny M, McConnell ML, Lee ML, Farlow
45. Lin DY, Zhang SZ, Block E, Katz LC (2005) R (1974) High resolution gas-chromato-
Encoding social signals in the mouse main graphic analysis of the volatile constituents of
olfactory bulb. Nature 434:470477 body fluids with use of glass capillary columns.
46. Willse A, Kwak J, Yamazaki K, Preti G, Wahl Clin Chem 20:11051110
JH, Beauchamp GK (2006) Individual odor- 60. Novotny M, McConnell ML, Lee ML (1974)
types: interaction of MHC and background Some aspects of high-resolution gas-chro-
genes. Immunogenetics 58:967982 matographic analysis of complex volatile sam-
47. Osada K, Tashiro T, Mori K, Izumi H (2008) ples. J Agric Food Chem 22:765769
The identification of attractive volatiles in aged 61. Baltussen E, David F, Sandra P, Janssen H-G,
male mouse urine. Chem Senses 33:815823 Cramers CA (1998) Sorption tubes packed
48. Schaefer ML, Wongravee K, Holmboe ME, with polydimethylsiloxane: a new and promis-
Heinrich NM, Dixon SJ, Zeskind JE, Kulaga ing technique for the preconcentration of vola-
HM, Brereton RG, Reed RR, Trevejo JM tiles and semivolatiles from air and gaseous
(2010) Mouse urinary biomarkers provide sig- samples. J High Resolut Chromatogr 21:
natures of maturation, diet, stress level and 332340
diurnal rhythm. Chem Senses 35:459471 62. Baltussen E, Sandra P, David F, Cramers C
49. Schwende FJ, Jorgenson JW, Novotny M (1999) Stir bar sorptive extraction (SBSE), a
(1984) A possible chemical basis for the histo- novel extraction technique for aqueous sam-
compatibility-related mating preference in ples: theory and principles. J Microcolumn Sep
mice. J Chem Ecol 10:16031615 11:737747
50. Jemiolo B, Gubernick DJ, Yoder MC, Novotny 63. Tienpont B, David F, Desmet K, Sandra P
M (1994) Chemical characterization of uri- (2002) Stir bar sorptive extraction-thermal
nary volatile compounds of Peromyscus califor- desorption-capillary GC-MS applied to bio-
nicus, a monogamous biparental rodent. J logical fluids. Anal Bioanal Chem 373:4655
Chem Ecol 20:24892500 64. Blanchard J (1981) Evaluation of the relative
51. Rck F, Mueller S, Weimar U, Rammensee efficacy of various techniques for deproteiniz-
HG, Overath P (2006) Comparative analysis ing plasma samples prior to high-performance
of volatile constituents from mice and their liquid chromatographic analysis. J Chromatogr
urine. J Chem Ecol 32:13331346 226:455460
 Chapter 4

Murine Nonvolatile Pheromones:

Isolation of Exocrine-Gland Secreting Peptide 1
Hiroko Kimoto and Kazushige Touhara

Abstract
Our search for a substance recognized by the vomeronasal neurons revealed that the extra-orbital lacrimal
gland (ELG) isolated from adult male mice produced the male-specific peptide pheromone exocrine
gland-secreting peptide 1 (ESP1). The following protocol reveals how ESP1 may be extracted from the
ELG, purified using anion-exchange and reverse-phase high-performance liquid chromatography (HPLC),
and analyzed by mass spectrometry. This protocol has been specifically designed for the purification of
ESP1, but may be modified to isolate a variety of peptides from the exocrine glands. Peptides purified in
this manner may help further define the molecular mechanisms underlying pheromone communication in
the vomeronasal system.

Key words Exocrine gland-secreting peptide 1, Extra-orbital lacrimal gland, Vomeronasal organ,
Dialysis, High-performance liquid chromatography, Mass spectrometry

1 Introduction

The vomeronasal organ (VNO) plays an essential role in the detec-

tion of pheromones by mice [1]. The vomeronasal pumping sys-
tem enables the uptake of not only volatile chemicals, but also
nonvolatile peptides and proteins into the VNO [2, 3]. Whereas
volatile chemical substances are transferred immediately, but tran-
siently, and vary in accordance with the physiological conditions,
peptides and proteins remain intact at scent marks for extended
periods of time and convey definite information throughout their
life [1]. Volatiles and nonvolatile peptides have been suggested to
mediate distinct types of social and sexual information [4].
Several chemical compounds originating in the urine and the
major urinary proteins have been reported to stimulate the vomero-
nasal sensory neurons [5, 6]. Neuronal firing in the vomeronasal
system of behaving mice has also been shown to be modulated fol-
lowing physical contact with the facial area [7], implicating the exis-
tence of additional sources of pheromone secretion other than urine.

Kazushige Touhara (ed.), Pheromone Signaling: Methods and Protocols, Methods in Molecular Biology, vol. 1068,
DOI 10.1007/978-1-62703-619-1_4, Springer Science+Business Media, LLC 2013

47
48 Hiroko Kimoto and Kazushige Touhara

Whilst searching for the source of pheromone substances

recognized by the vomeronasal neurons among extracts of exo-
crine glands, we discovered that the extra-orbital lacrimal gland
(ELG), which lies subcutaneously at the base of the ear, secreted an
active component [8]. Via a three-step purification process using
anion-exchange and reverse-phase high-performance liquid chro-
matography (HPLC), we identified a male specific 7-kDa peptide
that we termed exocrine gland-secreting peptide 1 (ESP1) [8].
ESP1 is a member of a multi-gene family that is clustered within
close proximity to the class I major histocompatibility complex
region [8, 9]. ESP1 induced c-Fos expression in the vomeronasal
sensory neurons expressing the V2Rp5 receptor in female mice, and
evoked a neural action potential [810]. In addition, ESP1 released
into the tears of males exhibits an important role as a sex phero-
mone that enhances female sexual receptive behavior [10].
Here, we describe protocols for extraction, purification, and
analysis of ESP1 from the ELG. The innate ESP1 isolated in this
manner has proved useful for further defining the molecular basis
underlying information transmission between individuals via the
vomeronasal system. Recombinant ESP1 produced in E. coli, which
has a c-Fos inducing activity practically equivalent to innate ESP1
[8], has also be utilized.

2 Materials

2.1 ELG Sampling 1. Adult male BALB/c mice.

Components 2. Forceps and fine dissection scissors.
3. Tris buffer (20 mM TrisHCl, pH 7.5).
4. 1.5-ml microcentrifuge tubes.
5. Pellet pestle for 1.5-ml microcentrifuge tube.
6. High speed refrigerated microcentrifuge.
7. Dialysis tubing (10,000 MWCO, Spectra/Por; Spectrum
Laboratories, see Note 1).
8. Closure clips for dialysis tubing (see Note 2).
9. Magnetic stirrer.
10. Magnetic stir bar.

2.2 HPLC All buffers and solvents used in the following method should be
Components HPLC-grade or filtered using a 0.22-m filter prior to use.
Degassing mobile-phase solvents by sonication under pressure or a
HPLC degasser is necessary to prevent air bubble formation in
flow lines.
1. 0.45-m filter (MILLEX-HV; Millipore).
2. Microcentrifuge tubes.
 Murine Nonvolatile Pheromones: Isolation of Exocrine-Gland Secreting Peptide 1 49

3. HPLC system (see Note 3).

4. Anion-exchange DEAE column (TSK-GEL DEAE-5PW,
7.5 75 mm or 22.5 150 mm; Tosoh, see Note 4).
5. Tris buffer (20 mM TrisHCl, pH 7.5).
6. 0.5 M NaCl in Tris buffer.
7. 50-ml conical tube (BD Falcon; BD Biosciences).
8. Liquid nitrogen.
9. Vacuum freeze drying equipment.
10. Reverse-phase C4P column (PEGASIL-300, 4.6 250 mm;
Senshu, see Note 4).
11. 10 % acetonitrile (ACN) in 0.1 % trifluoroacetic acid (TFA).
12. 60 % ACN in 0.1 % TFA.
13. Siliconized microcentrifuge tubes (see Note 5).
14. SpeedVac concentrator (GMI).
15. 30 % ACN in 0.1 % heptafluorobutyric acid (HFBA).
16. 60 % ACN in 0.1 % HFBA.

2.3 Mass 1. Matrix: 10 mg of -cyano-4-hydroxycinnamic acid (CHCA) in

Spectrometry 50 % ACN containing 0.1 % TFA (see Note 6).
Components 2. Matrix-assisted laser desorption ionization-time of flight/mass
spectrometer (Voyager-DE STR; Applied Biosystems).
3. Bovine insulin (Sigma).

3 Methods

3.1 ELG Sampling 1. Collect the intact ELGs from sacrificed adult male BALB/c
mice. Rinse the collected samples with ice-cold Tris buffer and
transfer to a 1.5-ml microcentrifuge tube (see Note 7).
2. Add ice-cold Tris buffer (10 l/mg of gland tissue) to the tube
and homogenize with a pellet pestle on ice.
3. Centrifuge at 14,000 g for 10 min at 4 C.
4. During centrifugation, prepare dialysis tubing by rinsing in
deionized water and seal one end of the tubing with a closure
clip (see Note 8).
5. Transfer the supernatant into the dialysis tubing and close the
open end using a second closure clip (see Note 2).
6. Place the dialysis tubing in a large volume of Tris buffer, and
dialyze with stirring overnight at 4 C with two to three buffer
changes (see Note 9).
7. At the completion of dialysis, remove one closure clip and
pipette the dialyzed sample into a clean microcentrifuge tube
(see Note 10).
50 Hiroko Kimoto and Kazushige Touhara

3.2 Purification Please note, that when not in use, columns and flow lines need to
of ESP1 be filled with an organic solvent such as 80 % methanol or 90 %
acetonitrile.
1. Filter the dialyzed sample through a 0.45-m filter into a fresh
microcentrifuge tube (see Note 11).
2. Prepare the mobile phase for anion-exchange HPLC by filling
the apparatus with Tris buffer with or without 0.5 M NaCl. Set
the flow rate to 0.5 ml/min.
3. Connect an anion-exchange DEAE column and rinse with Tris
buffer (see Note 12).
4. Inject the filtered sample acquired at step 1 onto the DEAE
column and elute with Tris buffer. While monitoring the pep-
tide effluent at 280 nm, collect the flow through into a 50 ml
conical tube (see Note 13).
5. Freeze the flow through sample in liquid nitrogen and lyophi-
lize using vacuum freeze drying equipment (see Note 14).
6. Set up the mobile phase for reverse-phase HPLC by changing
the solution filling the apparatus to 10 % or 60 % ACN in 0.1 %
TFA. Set the flow rate to 1 ml/min.
7. Connect a reverse-phase C4P column and wash using a linear
gradient program of 1060 % ACN in 0.1 % TFA over 10 min
at a flow rate of 1 ml/min.
8. Dissolve the lyophilized flow through sample (acquired at step 5)
in 10 % ACN in 0.1 % TFA.
9. Inject the dissolved sample onto the C4P column and run a
linear gradient program of 1060 % ACN in 0.1 % TFA over
50 min at a flow rate of 1 ml/min (Fig. 1a). Collect 1 ml frac-
tions in siliconized microcentrifuge tubes.

Fig. 1 Purification and analysis of ESP1 from the ELGs of male BALB/c mice. (a) The second step in HPLC
purification; a C4 column and TFA were used. This elution profile was obtained from a sample that contained
the equivalent of 48 ELGs. (b) The third step in the HPLC purification; a C4 column and HFBA were used. This
elution profile was obtained from a sample that contained the equivalent of 43 ELGs. Fractions A and B had
c-Fos-inducing activity. These figures were reproduced from [8]
 Murine Nonvolatile Pheromones: Isolation of Exocrine-Gland Secreting Peptide 1 51

10. Check the activity using a sample of each fraction equivalent to

five ELGs. Transfer to a 1.5-ml microcentrifuge tube and dry
in a SpeedVac concentrator. Dissolve peptide in 200 l of Tris
buffer, transfer all the peptides with 3540 mg of a cotton
swab, and repeat the drying process. Expose the cotton swab
transfused with the fraction to a BALB/c female mouse and
examine c-Fos induction in the vomeronasal sensory neurons
(see Note 15 and Chapter 19 for the detailed c-Fos induction
methodology).
11. Set up the mobile phase for the second reverse-phase HPLC
using the same C4P column. Transfer filled solution to 30 or
60 % ACN in 0.1 % HFBA (see Note 16). Set the flow rate to
1 ml/min.
12. Connect the C4P column and wash using a linear gradient
program of 3060 % ACN in 0.1 % HFBA over 10 min at a
flow rate of 1 ml/min.
13. Dry the active fractions (acquired at step 9) in a SpeedVac
concentrator and dissolve in 30 % ACN in 0.1 % HFBA.
14. Inject the dissolved active component onto the C4P column
and run a linear gradient program of 3050 % ACN in 0.1 %
HFBA over 80 min followed by 5060 % ACN in 0.1 % HFBA
over 10 min at a flow rate of 1 ml/min (Fig. 1b). Collect 1-ml
fractions in siliconized microcentrifuge tubes.
15. Check the activity using a portion of each fraction and the
c-Fos assay (see step 10 and Note 17).

3.3 Analysis by Mass 1. Mix a few microliters of the peptide fraction with 1 l of the
Spectrometry matrix.
2. Spot 1 l of the mixture onto the MALDI plate and dry.
Record the plate position numbers for each of the samples.
3. Prepare for mass spectrometry. Calibrate the mass scale by
acquiring spectra of a known peptide such as bovine insulin.
4. Measure the molecular mass of the peptides in the active frac-
tions following the manufacturers instructions (Fig. 2a).
If you would like to confirm the amino acid sequence of the
purified ESP1, we suggest using N-terminal protein sequencing
(Fig. 2b). For N-terminal amino acid analysis, transfer the HPLC
active fraction onto glass support disks pre-cycled with polybrene
and run the sequencer. MS/MS experiments may also be utilized
to identify the sequence by searching MS/MS spectra of peptides
against sequences in known DNA or protein databases.
52 Hiroko Kimoto and Kazushige Touhara

Fig. 2 (a) Mass spectra of the active fractions A and B from the final column. Fraction A exhibited a 7,106 Da
main peak and a 7,505 Da minor peak (left ). Fraction B exhibited a 7,376 Da peak (right ). (b) The amino-acid
sequence of ESP1. The N-terminal amino-acid sequence obtained from the peptides in fractions A and B is
underlined. The molecular weights determined by mass spectrometry corresponded to three different
C-terminal truncated forms of the peptide. The numbers show the theoretical molecular weights for each
peptide. These figures were reproduced from [8]

4 Notes

1. We were able to use this dialysis tubing even though the size of
ESP1 was under 10 kDa. We hypothesize that this was likely
due to oligomerization of ESP1. Consider using other MWCO
size tubing, depending on the size of a target peptide.
2. A knot tied in the tubing may also be used as a substitute for a
closure clip. However, simply forming a knot risks leakage and
alteration to pore size.
3. The HPLC system should be able to mix two mobile phases in
a gradient manner, be set to manage a flow rate precisely and
detect on a UV absorption spectrum. Although we collected
each fraction manually, a fraction collector and an auto-sampler
may be used.
4. HPLC columns produced by different manufacturers may be
used successfully. However, results may vary due to column
matrixes exhibiting slightly different characteristics.
5. We used siliconized microcentrifuge tubes to prevent sample
loss. Other tubes which have a low binding surface may also be
used for collection.
6. The matrix is used for improving crystallization. CHCA is
appropriate for measuring peptides under 10 kDa.
7. The ELG is located under the skin at the base of the ear and
should be easily isolated from the surrounding tissue. ELG
samples may be stored at 80 C until required.
8. Always follow the manufacturers instructions when using dial-
ysis tubing from different sources.
9. The volume of the dialysis buffer should be at least 200 times
the sample volume. This step drastically decreased UV
 Murine Nonvolatile Pheromones: Isolation of Exocrine-Gland Secreting Peptide 1 53

absorbance at 280 nm of the flow-through sample at step 4 in

Subheading 3.2.
10. The tube containing the dialyzed extract may be stored at
80 C until required.
11. This step protects the HPLC column and improves reproduc-
ibility of the analytical data. The sample may be stored at
80 C for further use.
12. In this case, it is sufficient to wash the column with Tris buffer.
When components absorbed to the column are to be collected,
simply run a gradient program of 00.5 M NaCl in Tris buffer.
13. ESP1 was not bound to the DEAE column, probably a result
of its high isoelectric point. Following the collection of flow-
through sample, the DEAE column should be washed using a
linear gradient of 00.5 M NaCl in Tris buffer.
14. The lyophilized sample may be stored at 4 C or 20 C.
15. The dried cotton swabs soaked with each fraction diluted in
ACN buffer did not exhibit any c-Fos-inducing activity. The
cotton swabs soaked with each fraction that was redissolved in
Tris buffer demonstrated very weak activity. Therefore, we
concluded that redissolving in Tris buffer and drying in the
SpeedVac are required to ensure optimum activity.
16. A TMS column (PEGASIL-300 TMS, 4.6 250 mm; Senshu)
did not improve the separation or reduce the ACN concentra-
tion in active fractions.
17. In our experiments, the active fractions including ESP1 were
eluted at approximately 4142 % ACN (Fig. 1b).

References
1. Touhara K (2008) Sexual communication via that promote aggressive behaviour. Nature
peptide and protein pheromones. Curr Opin 450(7171):899902
Pharmacol 8(6):759764 7. Luo M, Fee MS, Katz LC (2003) Encoding
2. Wysocki CJ, Wellington JL, Beauchamp GK pheromonal signals in the accessory olfactory
(1980) Access of urinary nonvolatiles to the bulb of behaving mice. Science 299(5610):
mammalian vomeronasal organ. Science 11961201
207(4432):781783 8. Kimoto H, Haga S, Sato K, Touhara K (2005)
3. Meredith M, OConnell RJ (1979) Efferent Sex-specific peptides from exocrine glands
control of stimulus access to the hamster vom- stimulate mouse vomeronasal sensory neurons.
eronasal organ. J Physiol 286:301316 Nature 437(7060):898901
4. Isogai Y, Si S, Pont-Lezica L, Tan T, Kapoor V, 9. Kimoto H, Sato K, Nodari F, Haga S, Holy
Murthy VN, Dulac C (2011) Molecular orga- TE, Touhara K (2007) Sex- and strain-specific
nization of vomeronasal chemoreception. expression and vomeronasal activity of mouse
Nature 478(7368):241245 ESP family peptides. Curr Biol
5. Leinders-Zufall T, Lane AP, Puche AC, Ma W, 17(21):18791884
Novotny MV, Shipley MT, Zufall F (2000) 10. Haga S, Hattori T, Sato T, Sato K, Matsuda
Ultrasensitive pheromone detection by S, Kobayakawa R, Sakano H, Yoshihara Y,
mammalian vomeronasal neurons. Nature Kikusui T, Touhara K (2010) The male
405(6788):792796 mouse pheromone ESP1 enhances female
6. Chamero P, Marton TF, Logan DW, Flanagan sexual receptive behaviour through a specific
K, Cruz JR, Saghatelian A, Cravatt BF, Stowers vomeronasal receptor. Nature 466(7302):
L (2007) Identification of protein pheromones 118122
 Chapter 5

Chemical Analysis of Aquatic Pheromones in Fish

Michael Stewart, Cindy F. Baker, and Peter W. Sorensen

Abstract
Pheromones are chemicals that pass between members of the same species that have inherent meaning. In
the case of fish, pheromones are water-soluble and found in low concentrations. As such, sensitive and
selective methods are needed to separate and analyze these pheromones from an environmental matrix that
may contain many other chemicals. This chapter describes a generic method used to concentrate and iden-
tify these chemicals and two extremely sensitive and selective methods for analysis, namely, mass spectrom-
etry and enzyme-linked immunosorbent assay.

Key words Fish pheromone, Liquid chromatography, Solid phase extraction chromatography, Mass
spectrometry, Enzyme-linked immunosorbent assay, ELISA

1 Introduction

Pheromones, or chemical signals that pass between members of the

same species, are used by fish to mediate a variety of physiological
responses and behaviors including endocrinological synchrony,
migratory orientation, sexual arousal, attraction, and aggression.
While most fish pheromones are mixtures of relatively unspecial-
ised bodily metabolites, some can function as independent entities [1].
Approximately a dozen fish pheromones have now been identified,
meaning that their release, detection and biological function have
all been elucidated and their chemical structures identified [2].
There are also many chemicals utilized by fish that are not strictly
pheromones, but these compounds still play an important role in
the life cycle of fish, such as predator detection, locating food,
shoaling, and homing. This chapter only covers techniques used
for the detection and analysis of chemicals utilized by fish and does
not address their biological importance.

1.1 Pheromone Steroids and prostaglandins are the predominant classes of com-
Classes pounds that have been identified as pheromones used by fish.
These are usually modified with polar functional groups, thereby

Kazushige Touhara (ed.), Pheromone Signaling: Methods and Protocols, Methods in Molecular Biology, vol. 1068,
DOI 10.1007/978-1-62703-619-1_5, Springer Science+Business Media, LLC 2013

55
56 Michael Stewart et al.

Fig. 1 Four important fish odorants with pheromonal function

increasing their water solubility. Four important classes of chemicals

used by fishes as odorants, and putative pheromones, are illustrated
in Fig. 1. Bile acids (e.g., taurocholic acid (TCA; 1) have been
shown to be potent olfactory stimuli for many fish species [35].
Amino sterols (e.g., petromyzonamine disulfate (PADS; 2) are
part of the sea lamprey migratory pheromone mixture, which
contains at least three related steroids [6]. 17,20,21-Trihydroxy-
4-pregnen-3-one (20-S; 3) has been identified as a maturation-
inducing hormone and pheromonal precursor in the percid fish,
Gymnocephalus cernuus [7]. F series prostaglandins (e.g., PGF2; 4)
are part of the goldfish, Carassius auratus, and common carp,
Cyprinus carpio, postovulatory hormonal pheromone mixture [2, 8].
Polar functional groups that ensure water solubilityi.e., sul-
fate, glucuronic acid, carboxylic acid, amide, hydroxylare present
in many pheromones including the four compounds depicted
(Fig. 1). These polar functional groups also reduce volatility of the
compound, so the methods used to separate them for analysis are
different from those used for airborne pheromones; namely liquid
chromatography (LC) instead of gas chromatography (GC).
Detection of very low concentrations of fish pheromones
requires highly specific and highly sensitive methods. Two tech-
niques used commonly are mass spectrometry (MS)including
hyphenated derivatives (MS/MS)and enzyme-linked immuno-
sorbent assay (ELISA). Even with these extremely sensitive tech-
niques, a pre-concentration step is usually also required, and solid
phase extraction (SPE) chromatography is the method commonly
employed.
 Chemical Analysis of Aquatic Pheromones in Fish 57

1.2 SPE Solid phase extraction (SPE) is a method used to concentrate

chemicals of interest from water for analysis. This may also involve
separating them from an interfering matrix, such as urine. SPE
methods are generically similar but are application specific, so the
chromatographic supplier literature should be consulted for
the best conditions for a specific application. This chapter will
describe a generic method developed in our laboratories to con-
centrate lamprey bile acids from water samples using Oasis HLB
SPE cartridges (Waters Corporation).

1.3 LC/MS Liquid chromatography/mass spectrometry (LC/MS) is a rapidly

evolving technique that is becoming more widespread in the analysis
of fish pheromones, due to its high sensitivity and selectivity [811].
The analysis of water soluble chemicals is primarily carried out using
reverse phase high pressure liquid chromatography (RP HPLC),
which employs columns containing small uniform sorbent particles
(35 m) to provide a highly efficient, automated separation proce-
dure that delivers chemicals in a suitable format for analysis. For
chemicals with acidic or basic functional groups, buffers (e.g.,
ammonium acetate or ammonium formate) and/or pH modifiers
(e.g., formic acid or triethylamine) are usually added to the HPLC
solvent. These buffers and modifiers give reliable retention times,
good peak shape and improved detection by mass spectrometry.

1.4 ELISA ELISA is a type of immunoassay which uses competitive binding to

antibodies to measure miniscule quantities of compounds, usually
via a fluorescent label (see below). ELISA often represents a rea-
sonable alternative to direct chemical measurement because it can
be very quick and inexpensive. It can also analyze many samples at
once (typically several dozen through the use of a 96 well assay).
Additionally, some structures either do not fragment (e.g., andro-
stenedione) or do not fragment in a distinctive manner (e.g.,
13,14-dihydro-15-keto-prostaglandin F2) in mass spectrometry,
so ELISA may be a good primary method [8]. ELISAs require
both an antibody for the structure of interest as well as synthesized
standards, both of which can be difficult (and expensive) to pro-
duce and beyond the scope of most workers (and this chapter).
Thus, almost all workers, including us, use commercially available
kits to measure fish pheromones. ELISA kits have been developed
for some pheromones that also happen to be common metabolites
in mammals (e.g., prostaglandin F2 and metabolites, androstene-
dione). Many types of ELISA exist (indirect, sandwich, competi-
tive, etc.) and we will not review them here other than to mention
that ELISAs either employ an antibody which binds to the antigen
(compound of interest), and then either a standard that is labeled
or an enzyme that is linked to standard. Either way, the amount of
antigen present in the sample is determined by measuring the
presence of label bound to an antibody which in turn reflects some
58 Michael Stewart et al.

type of competition between the quantity of antigen found in the

sample and that added as part of the kits protocol. These reactions
are carried out in plates with wells (usually 96) that are also gener-
ally treated with antibodies to capture the product. Not surpris-
ingly, different companies have produced a wide variety of kits.
These employ different suites of antibodies, enzymes, labels and
reagents which need to deployed in kit-specific manners, and
whose instructions must be followed very carefully. Generally, the
strategy behind the chemical reaction being exploited is provided
by the kit. Here, we provide some tips on issues to consider when
using these kits.

2 Materials

2.1 SPE 1. Oasis HLB solid phase extraction cartridges (60 mg).
2. Nanopure water.
3. Analytical grade methanol (see Note 1).
4. Automatic pipette (101,000 L).
5. Vacuum manifold or syringe with luer slip end and adaptor for
SPE.
6. Rotary evaporator (if available).
7. Nitrogen gas blow down apparatus.
8. 2 mL amber glass vials with lids and septa.
9. Low volume glass inserts (150 L).
10. Oxygen free nitrogen.
11. Sonicator (recommended).

2.2 LC/MS 1. Acetonitrile (LC/MS grade).

2. Nanopure water.
3. Triethylamine.
4. Automatic pipette (101,000 L).
5. LC/MS: Thermo Finnigan Surveyor LC with MS pump and
autosampler. LCQ Advantage Ion Trap mass spectrometer
with ESI probe operating in negative mode.
6. Nitrogen generator: Peak Scientific.
7. Column: Phenomenex Gemini, 5 m, 150 2 mm, 110 .
8. Standards (prepared at a concentration of 100 g/mL).
9. 2 mL amber glass vials with lids and septa.

2.3 ELISA 1. ELISA kit (typically contains 96 well antibody-coated micro-

plate, standards, enzyme conjugate, substrate, EIA buffer,
wash buffer, instructions, information on antibody specificity)
(see Note 2).
 Chemical Analysis of Aquatic Pheromones in Fish 59

2. Ultrapure water (see Note 3).

3. Automatic, calibrated pipette (101,000 L) (see Note 4).
4. Pipette tips.
5. Pheromone standards (can be prepared at a concentration of
100 g/mL).
6. Clean test tubes to mix standards.
7. Plastic film to cover plates.
8. ELISA microplate reader with appropriate wavelength as well
as associated computer and software.
9. Microplate shaker (recommended).

3 Methods

3.1 SPE The following method describes a common approach used for
concentrating moderately polar chemicals from water samples,
using Oasis HLB cartridges [10]. This method was designed for
low volumes of water (up to 20 mL) but can be scaled up accord-
ingly by increasing the size of the cartridge to allow for larger water
volumes.
1. Wash SPE cartridge (60 mg) with methanol (2 mL) (see Notes
5 and 6).
2. Wash SPE cartridge with H2O (2 mL).
3. Elute water samples (20 mL) through the SPE, discarding the
eluent (see Note 7).
4. Remove excess H2O by vacuum or several passes of air through
with positive pressure.
5. Immediately elute the material retained by the SPE cartridge
into a 2 mL amber glass vial with methanol (2 mL) (see Note 8).
6. Remove all methanol under a stream of N2 gas (see Note 9).
7. Redissolve in 1:1 methanol:H2O (100 L), cap vial and soni-
cate briefly (see Note 10).
8. Transfer to a low volume 150 L glass insert, inside the 2 mL
amber glass vial. Seal with a lid and septum.
9. Store vials in a 20 C freezer.

3.2 LC/MS An LC/MS method was developed within the Stewart laboratory
to analyze fish tissues and water for common bile acids (Fig. 2).
This particular method uses an alkaline modifier (2 mM triethyl-
amine; see Note 11) and an LC column that is stable to basic con-
ditions (Phenomenex Gemini; see Note 12). The rationale behind
this was that as all bile acids analyzed had acidic functional groups,
an alkaline mobile phase would enhance formation of negatively
charged ions and improve the MS analysis.
60 Michael Stewart et al.

Fig. 2 LC/MS traces of 10 bile acid standards. Top trace is the full scan chromatogram (m/z 200700) and
bottom trace is the SIM windows chromatogram. Standards are DHC (Dehydrocholic acid), TLS (Taurolithocholic
acid 3-sulfate disodium salt), GCA (Glycocholic acid), TCA (Taurocholic acid), CDC (Chenodeoxycholic acid),
DOC (Deoxycholic acid), TCD (Sodium taurochenodeoxycholate), TDC (Sodium taurodeoxycholate hydrate), LCA
(Lithocholic acid), and TLC (Sodium taurolithocholate). The peak labelled with an asterisk in the full scan chro-
matogram is an impurity

1. Ensure solvent reservoirs on the LC have sufficient solvent for

all the required analyses. Place H2O in line A and acetonitrile
in line C (see Note 13).
2. Prepare 20 mM triethylamine modifier by pipetting 697 L of
triethylamine into 250 mL of H2O. Place this in line D (see
Note 14).
3. Set column temperature to 30 C.
4. Use the following gradient chromatography method: Linear
gradient from 80:10:10 to 0:90:10 A:C:D over 60 min. Hold
0:90:10 for 5 min. Linear gradient from 0:90:10 to 80:10:10
ACD over 1 min. Hold 80:10:10 for 9 min. Total run time
75 min. Flow 250 L/min (see Note 15).
5. Use the following parameters for MS: Sheath gas 80 arbitrary
units (approx. 1.2 L/min), aux/sweep gas 10 arbitrary units
 Chemical Analysis of Aquatic Pheromones in Fish 61

(approx. 0.3 L/min), spray voltage 5.00 V, capillary temperature

280 C, capillary voltage 17.00 V, Tube lens offset 35.00.
6. Set following analysis windows for MS: negative mode, full
scan m/z 200700 (see Note 16).
7. Equilibrate the LC column with starting solvent (see Note 17).
8. Run at least one blank analysis to ensure column has been
properly cleaned from previous use, is equilibrated for sample
analysis and that MS is at correct operating conditions (see
Note 18).
9. Inject 10 L of each standard using the autosampler (see Note 19).
10. Set SIM windows for individual compounds being analyzed
(see Note 20).
11. Inject 10 L of each sample using the autosampler (see Note 21).
12. Obtain spectrum for each peak observed in full scan analysis
(see Note 22).
Modification of the MS SIM windows (see Note 23)but
keeping the same LC methodallowed for two subsequent appli-
cations in the analysis of fish pheromones.
The first application of this particular method was the analysis of
the sea lamprey (Petromyzon marinus) migratory pheromones pet-
romyzonol sulfate (PS), petromyzonamine disulfate (PADS; com-
pound 2; Fig. 1), and petromyzosterol disulfate (PSDS) from tank
water containing ammocetes of the southern pouched lamprey,
Geotria australis (unpublished). Migratory pheromones utilized by
G. australis had not previously been studied. Full scan, SIM (see
Fig. 3) and SRM (see Note 24) analyses were undertaken using
authentic standards to confirm or deny presence. As Fig. 3 suggests,
PS was detected in the lamprey holding water with PADS and PSDS
undetected. A subsequent follow-up study was performed using a
much more sensitive triple quadrupole MS, which confirmed that
PS was the dominant pheromone released, PADS was a very minor
component (0.33 %) and PSDS was not detected [12].
A second application involved female redfin perch, Perca flu-
viatilis. Pre-spawning females were injected with a priming hor-
mone (Ova-RH, a salmon GnRH analogue), and the maturation
inducing hormone 20-S [3] and then placed in small (2 L) aer-
ated tanks. Water was replaced every 12 h and released analytes
concentrated by SPE. Time series LC/MS analyses were under-
taken using both full scan and SIM windows for conjugates and
metabolites of 20-S (Fig. 4). 20-S was the dominant chemical
released after 12 h, but reduced substantially in the 24 and 36 h
analyses. Many conjugates were detected in the 24 and 36 h analyses,
including a putative sulfate and glucuronides (2) of 20-S (no stan-
dards are available to confirm the presence of these compounds,
however, they had similar retention times to 17,20-P sulfate and
glucuronide and correct pseudo-molecular ions). The LC method
62 Michael Stewart et al.

Fig. 3 LC/MS SIM analysis of the southern pouched lamprey (Geotria australis) ammocete holding water. The
top three traces are the standards: petromyzonol sulfate (PS), petromyzonamine disulfate (PADS), and petro-
myzosterol disulfate (PSDS). The bottom trace is pre-concentrated holding water which clearly shows pres-
ence of PS, but PADS or PSDS were not detected

was subsequently used to fractionate holding water for further

chemical analysis, electro-olfactogram (EOG) measurements, and
behavior tests (unpublished).

3.3 ELISA An ELISA method was recently developed by the Sorensen labora-
tory to analyze water containing ovulated common carp and gold-
fish for the F prostaglandins that are known to function as sex
pheromones in this species [8]. We have successfully used kits pro-
duced by both Neogen (Lansing, Michigan; kit #404710) and
Cayman Chemical (Ann Arbor, Michigan; kit #5160111) to mea-
sure this class of pheromone, as well as sex steroids. These kits
employ different approaches, so here we describe commonalities
and ask readers to defer to manufacturers for specifics. For ELISA
analysis of pheromones, we typically collect 1 L samples of fish
holding waters (20100 g fish are held for 1 h in well water) and
extract these waters using SPE cartridges, as described above. In initial
tests we also collect three types of controls; at least three samples
of well water which lack fish (blank controls to test for
 Chemical Analysis of Aquatic Pheromones in Fish 63

Fig. 4 LC/MS SIM time series analysis of female redfin perch (Perca fluviatilis) holding water. The top trace
contains four standards including sulfate and glucuronide conjugates of 17,20-P. No standards were avail-
able for 20-S-sulfate or 20-S-glucuronide. The bottom three traces contain time series analyses (PI = post
injection)

contamination and the possible presence of interfering compounds

in water), three samples of water which lack fish but to which we
add known amounts of pheromone standards (1 g) (to confirm
antibody specificity and determine extraction efficiency), and three
samples which contain fish which we know not to be releasing any
pheromone (e.g., another species, or the same species but imma-
ture, to confirm antibody specificity, which is never 100 %).
Depending on the published specificities of the antibody used in a
64 Michael Stewart et al.

kit, we may also test compounds closely related to the compound

of interest. If in doubt, results may also be confirmed by LC/MS,
as described elsewhere [8]. Although all ELISA kits differ, many of
the steps they use are same.
The following description is based on that used for acetylcho-
linesterase competitive ELISAs with some notes for other assay
types added:
1. Read kit instructions very carefully. Note that different kits use
different combinations of reagents, tracers and standards. Very
likely, there will be differences from the protocols we describe
here. Expired kits, or kits not kept under specified conditions,
should not be used.
2. Plan your experiment to use all 96 wells in an efficient manner.
Typically samples (unknowns) are run in duplicate and may
include a few blanks and positive controls. Also, at least eight
pairs of wells are also reserved for standards to create an eight-
point calibration curve. Depending upon the assay type, wells
may also be needed for blanks (reagent interference), total
activity, non-specific binding (NSB), maximum binding (B0),
B0/Bmax (see Note 25).
3. Prepare EIA buffer and wash buffers as specified by kit direc-
tions using supplied materials.
4. Prepare standards as specified by kit instructions using supplied
materials. Usually this involves serial dilution.
5. If requested by the kit, prepare enzyme conjugate (tracer), as
specified by kit instructions. Store at 4 C.
6. Prepare antiserum, as specified by the kit. Store at 4 C.
7. Transfer EIA buffer to wells, as described by kit instructions
(see Notes 26 and 27).
8. Transfer standards to wells by automatic pipette (see Note 28).
9. Prepare samples (unknowns) and add to the appropriate wells
that are not controls or standards. This is accomplished by
transferring predetermined aliquots of SPE extracted fish water
samples into clean glass vials, drying down under a stream of
N2, and then adding 1 mL of EIA buffer to these vials. Gently
shake (or vortex) and allow to sit for 510 min to assure com-
plete dissolution. As specified by the kit instructions, then add
these (diluted) samples to the wells. 50 L is a typical volume.
10. If specified by kit instructions, use an automatic pipette to add
specified quantities of tracer to all wells, except those that
might be designated to measure total activity and blanks.
11. If specified by kit instructions, add specified quantities of anti-
body to each well except those used for total activity, non-
specific binding, and the kit blank. Then very gently move the
plate in a circle (or use microplate shaker) to promote mixing,
 Chemical Analysis of Aquatic Pheromones in Fish 65

cover with plastic film, and then incubate per kit instructions
(times and conditions will vary greatly by kit) (see Note 29).
12. Wash and develop (e.g., incubate) the plate as specified by kit
instructions. This involves several steps and differs between
kits. Usually several washes are suggested using supplied wash
buffer which is added by pipette and then the plate is turned
upside down onto a paper towel. After washing the plate, assays
that measure acetylcholinesterase activity may use Ellmans
reagent, after which a tracer is added to the total activity well.
The plate is then typically covered and placed on an orbital
shaker (or simply shaken by hand). Allow the plate to develop,
following kit instructions, which in some cases may involve
adding a stop reagent after a set period of time (see Note 30).
13. After developing the plate, gently shake the plate to ensure
even mixing. Clean its bottom with a Kimwipe, remove its
plastic film, and place into the plate reader. Operate the reader
per kit instructions. Some readers will automatically plot data
and determine the necessary relationships.
14. Analyze data, following kit instructions. These typically provide
great detail. Briefly, all duplicates should be averaged. Average
NSB (substrate background) is then subtracted from the NSB
average to yield corrected maximum binding (B0). Calculate
B/B0 (standard bound/ maximum bound) for the remaining
wells after averaging. Plot %B/B0. The quantity of unknown
may then be derived by the logistic equation used to describe
the plot. Data that fall outside that of the linear portion of the
standard curve should not be used and re-run after dilution or
concentration so they fall within the specified relationship.
Great attention should be taken to ensure that this relationship
is credible; each kit should have its own criteria for fit as well as
acceptable intra- and inter-assay variation (see Note 31).
The first application of this particular method was the analysis
of prostaglandin F2 released by ovulated goldfish (unpublished)
but later we applied it to the common carp [8]. Analyses were
undertaken using a variety of authentic standards in urine and
whole waters to confirm or deny presence. A subsequent follow-up
study was performed using triple quadrupole MS, which confirmed
all results [8].

4 Notes

1. All solvents need to be of analytical grade or higher.

2. Ideally antibody specificity to compounds of likely importance
has been predetermined by the kit manufacturer.
3. Ultrapure water generally comes with the kit but you may
need extra. Water quality is extremely important for ELISA
66 Michael Stewart et al.

because small amounts of organics may interfere with the assay.

Even HPLC quality water sometimes is not adequate so this is
best purchased from the company.
4. A repeat pipette can be useful, especially if many samples are
measured.
5. Keep the SPE covered by solvent at all times, to prevent it dry-
ing out. Although polymeric phases like Oasis are not as sus-
ceptible to problems with drying as reverse phase cartridges,
this is considered best laboratory practice.
6. Apply vacuum (or positive pressure if using a syringe) to ensure
the solvent is eluting through the cartridge at a steady drip, not
a constant stream, which is too fast. This also ensures adequate
time for equilibration of analyte between liquid and solid
phases, allowing retention of the analyte.
7. If a pheromone is known to be highly polar, extraction effi-
ciency should be confirmed using standards first. If fish hold-
ing water contains a lot debris, it should be passed through
filter paper first to prevent clogging of the SPE cartridge.
8. Amber glass vials ensure any compounds that are unstable to
light are preserved. This is especially important when samples
are exposed to light during work-up and analysis.
9. Oxygen free nitrogen is essential to avoid potential issues with
oxidation of your sample. Take care to use a gentle flow of
nitrogen gas to avoid splashing and spilling. The process is
expedited by concurrent heating of the vial, however, beware
of temperature instability and use a maximum temperature of
40 C.
10. Do not place any part of the hand in the sonicator water while
in operation. Observe the inside of the vial to assess when all
material is dissolved.
11. Triethylamine has a pKb of 11.0 and a working pH range of
between 10 and 12. For LC/MS it is imperative that volatile
buffers are used to avoid build-up of unwanted buffer on the
probe and potential degradation (or loss) of signal.
12. Traditional silica based LC columns are susceptible to hydroly-
sis under alkaline conditions (pH >8), so it is important to use
a specially designed column that is stable under these condi-
tions (up to pH 12).
13. All solvents should be filtered through 0.45 mm nylon filters
prior to each analysis.
14. Water should be filtered prior to addition of the triethylamine
as it is volatile.
15. The final step of 9 min ensures the LC column has been equili-
brated before the next injection. Flow rate should be increased
 Chemical Analysis of Aquatic Pheromones in Fish 67

slowly from zero to avoid pressure shock to the LC column.

Pressure will change throughout the run and should be moni-
tored over time to assess potential blockages to the system or
column.
16. For negative mode ESI under alkaline conditions (2 mM tri-
ethylamine) most chemicals produce the pseudo-molecular
ions [MH] or [MNa].
17. HPLC columns are usually shipped in the solvent they were
tested with. Carry out small incremental (ca. 10 %) changes to
the solvent proportions to reach initial conditions. End caps
should always be in place when the column is not in use to
avoid drying.
18. This also allows the user to assess baseline data and potential
contamination on the column or in the MS prior to analysis.
19. Prior to analyzing samples, parameters need to be set for stan-
dards. Normally a full scan is used to identify pseudo-molecular
ions (see Note 16), retention time and any fragment ions.
20. SIM windows are user definable, and for this method, were set
at m/z pseudo-molecular ion 1. The method can be set to run
concurrent full scan and SIM windows. Full scan gives more
information (pseudo-molecular ion, adducts, fragment ions)
but is less sensitive. SIM analyses are more sensitive but give
limited information.
21. The autosampler usually has different modes of injection. If
quantitation is required a loop overfill procedure will ensure an
exact volume is injected each time. If sample is precious then a
partial loop, or no waste, procedure will minimize the amount
used.
22. The full scan spectrum gives much higher confidence that the
analyte is detected than SIM analysis, as it can be compared
with authentic standards. Isotope ratios, fragments and adduct
formation can be compared. This information is not available
in SIM analysis so it is imperative that both full scan and SIM
analyses are run concurrently.
23. Adjusting SIM windows to accommodate new analytes requires
method development as each chemical analyzed will exhibit
different behavior in the MS (i.e., adduct formation, fragmen-
tation, formation of doubly charged species, and formation of
clusters). This cannot necessarily be predicted so it is advisable
to run full scan to ascertain the most intense ion formed for
each analyte under the elution conditions.
24. SRM analyses can be undertaken with Ion Trap MS. These
involve capturing ions of choice and fragmenting to daughter
ions, which gives a highly selective analysis. This selectivity
however is compromised by lack of sensitivity and (in this case)
68 Michael Stewart et al.

was an inferior technique to SIM. Tandem mass spectrometers

(MS/MS) use a similar technique (MRM) which are usually
highly selective and sensitive.
25. Depending on results from pilot studies you may choose to
run samples in triplicate instead of duplicate. Some assays will
not need all of these controls.
26. Pipetting technique is absolutely critical. Make sure pipettes
are calibrated and practice repeatability if you lack experience.
Use different pipette tips for each reagent. Equilibrate pipette
tips in reagent and then withdraw slowly. Do not insert pipette
tips into fluids found in the wells.
27. Pilot studies should be performed using different dilutions of
fish extract along with blank and spiked standards to establish
the concentrations that are both present and can best be run
using the kit. Most assays are designed to measure between
0.001 and 1.0 ng of standard per mL (determined by antibody
binding). How much sample and fish water this represents will
have to be determined by pilot studies that test various dilu-
tions. For common carp, we diluted samples between 10 (for
control samples) and 1,000 times (for fully ovulated fish sam-
ples releasing large quantities of prostaglandin F2).
28. Aliquot low concentrations of standards first.
29. Some kits do not require adding antibody because it will
already be coating the wells.
30. Increasing development time can sometimes increase assay
sensitivity. Developed plates may be kept in the refrigerator
and reread at a later date, if in question.
31. Kits usually provide excellent advice on trouble shooting. Here
are a few pointers. High variance between duplicates may be a
sign of poor pipetting techniques or contamination. A high
NSB indicates poor washing. A low B0 value could indicate con-
taminants and/or inadequate development time. Low sensitiv-
ity could mean the standard is degraded. If only the total activity
well develops, then the water source is probably contaminated.

References

1. Stacey NE, Sorensen PW (2011) Hormonal namaycush) function as chemical signals.

pheromones. In: Farrell AP (eds) Encyclopedia J Comp Physiol B 171(2):161171
of fish physiology: from genome to environ- 4. Michel WC, Lubomudrov LM (1995)
ment, vol 2. Academic: San Diego Specificity and sensitivity of the olfactory organ
2. Sorensen PW, Hoye TR (2010) Pheromones in of the zebrafish, Danio rerio. J Comp Physiol A
Vertebrates. In: Lew M, Hung-Wen L (eds) Neuroethol Sens Neural Behav Physiol
Comprehensive Natural Products II. Elsevier, 177(2):191199
Oxford, pp 225262 5. Baker CF, Carton AG, Fine JM, Sorensen PW
3. Zhang C, Brown S, Hara T (2001) Biochemical (2006) Can bile acids function as a migratory
and physiological evidence that bile acids pro- pheromone in banded kokopu, Galaxias fascia-
duced and released by lake char (Salvelinus tus (Gray)? Ecol Freshw Fish 15(3):275283
 Chemical Analysis of Aquatic Pheromones in Fish 69

6. Sorensen PW, Fine JM, Dvornikovs V, Jeffrey lamprey migratory pheromone. J Chem Ecol
CS, Shao F, Wang JZ, Vrieze LA, Anderson 34(10):12591267
KR, Hoye TR (2005) Mixture of new sulfated 10. Stewart M, Baker C, Cooney T (2011) A rapid,
steroids functions as a migratory pheromone in sensitive, and selective method for quantitation
the sea lamprey. Nat Chem Biol 1(6):324328 of lamprey migratory pheromones in river
7. Sorensen PW, Murphy CA, Loomis K, Maniak water. J Chem Ecol 37(11):12031207
P, Thomas P (2004) Evidence that 4-pregnen- 11. Xi X, Johnson NS, Brant CO, Yun S-S,
17,20,21-triol-3-one functions as a Chambers KL, Jones AD, Li W (2011)
maturation-inducing hormone and phero- Quantification of a male sea lamprey phero-
monal precursor in the percid fish, mone in tributaries of Laurentian Great Lakes
Gymnocephalus cernuus. Gen Comp Endocrinol by Liquid Chromatography Tandem Mass
139(1):111 Spectrometry. Environ Sci Technol
8. Lim H, Sorensen P (2011) Polar metabolites 45(11):64376443
synergize the activity of prostaglandin F2 in a 12. Stewart M, Baker C (2012) A sensitive analyti-
species-specific hormonal sex pheromone cal method for quantifying petromyzonol sul-
released by ovulated common carp. J Chem fate in water as a potential tool for population
Ecol 37(7):695704 monitoring of the Southern Pouched Lamprey,
9. Fine JM, Sorensen PW (2008) Isolation and Geotria Australis, in New Zealand streams.
biological activity of the multi-component sea J Chem Ecol 38(2):135144
 Chapter 6

Analysis of Ascarosides from Caenorhabditis elegans

Using Mass Spectrometry and NMR Spectroscopy
Xinxing Zhang, Jaime H. Noguez, Yue Zhou, and Rebecca A. Butcher

Abstract
The nematode Caenorhabditis elegans secretes a family of water-soluble small molecules, known as the
ascarosides, into its environment and uses these ascarosides in chemical communication. The ascarosides
are derivatives of the 3,6-dideoxysugar ascarylose, modified with different fatty acid-derived side chains. C.
elegans uses specific ascarosides, which are together known as the dauer pheromone, to trigger entry into
the stress-resistant dauer larval stage. In addition, C. elegans uses specific ascarosides to control certain
behaviors, including mating attraction, aggregation, and avoidance. Although in general the concentration
of the ascarosides in the environment increases with population density, C. elegans can vary the types and
amounts of ascarosides that it secretes depending on the culture conditions under which it has been grown
and its developmental history. Here, we describe how to grow high-density worm cultures and the bacte-
rial food for those cultures, as well as how to extract the culture medium to generate a crude pheromone
extract. Then, we discuss how to analyze the types and amounts of ascarosides in that extract using mass
spectrometry and NMR spectroscopy.

Key words Dauer, Ascarosides, Pheromone, C. elegans, Mass spectrometry, NMR spectroscopy,
dqf-COSY

1 Introduction

C. elegans secretes ascarosides as chemical signals to coordinate its

development and behavior. All ascarosides are derivatives of the
3,6-dideoxysugar ascarylose, with different fatty acid-derived side
chains [19]. The ascarosides that are found in C. elegans culture
medium can be divided into several structural classes (Table 1). In
this chapter, we describe a general method for growing high-
density worm cultures and extracting the culture medium in order
to generate a crude pheromone extract that contains secreted asca-
rosides (Subheadings 3.1 and 3.2). This crude pheromone extract
can be directly used in biological assays, such as the dauer forma-
tion assay (see Chapter 20), it can be fractionated chromatographi-
cally for bioactivity-guided purification of ascarosides [14, 6], or
the ascarosides in the extract can be analyzed using analytical

Kazushige Touhara (ed.), Pheromone Signaling: Methods and Protocols, Methods in Molecular Biology, vol. 1068,
DOI 10.1007/978-1-62703-619-1_6, Springer Science+Business Media, LLC 2013

71
Table 1
Summary of ascarosides found in wild-type (N2) C. elegans culture medium that have had their structures confirmed through chemical synthesisa

Structure-based name Other name(s) Orig. ref. Mol. formula [M+Na]+ [MH] n R Structure
Ascarosides with a saturated fatty acid side chain that is attached to the sugar at the penultimate (1) carbon
asc-C4 ascr#11 [7] C10H18O6 257.1001 233.1025 1 H O

asc-C5 ascr#9 [7, 8] C11H20O6 271.1158 247.1182 2 H O OH

n
IC-asc-C5 indolecarboxyl- [6] C20H25NO7 414.1529 390.1553 2 IC
O
ascaroside C5; RO
ascaroside C5; icas#9 OH
OS-asc-C5 osas#9 [9] C23H33NO10 506.2002 482.2026 2 OS
asc-C7 daumone 1; ascr#1; [1] C13H24O6 299.1471 275.1495 4 H
ascaroside C7
IC-asc-C7 icas#1 [7, 8] C22H29NO7 442.1842 418.1866 4 IC
asc-C9 ascr#10 [7, 8] C15H28O6 327.1784 303.1808 6 H
Ascarosides with an ,-unsaturated () fatty acid side chain that is attached to the sugar at the penultimate (1) carbon
asc-C7 ascr#7 [5] C13H22O6 297.1314 273.1339 2 H
OH
IC-asc-C7 icas#7 [7, 8] C24H33NO7 440.1685 416.1709 2 IC O n
asc-C9 ascaroside C9; [2] C15H26O6 325.1627 301.1651 4 H O O
ascr#3; daumone 3 RO

IC-asc-C9 icas#3 [7, 8] C24H31NO7 468.1998 444.2022 4 IC OH

HB-asc-C9 hbas#3 [7] C22H30O8 445.1838 421.1862 4 HB
MB-asc-C9 mbas#3 [7] C20H32O7 407.2046 383.2070 4 MB
Ascarosides with a side chain that is attached to the sugar at the penultimate (1) carbon and terminates in a methyl ketone (MK)b
asc-C6-MK ascaroside C6; [2] C12H22O5 269.1365 n.a. 2 H
ascr#2; daumone 2 O
Glu-asc-C6-MK ascr#4; daumone 4 [4] C18H32O10 431.1893 n.a. 2 Glu O
O
HO

OR
 Ascarosides with a saturated fatty acid side chain that is attached to the sugar at the terminal () carbon
O
asc-C3 ascaroside C3; ascr#5; [3] C9H16O6 243.0845 219.0869 1
daumone 5 O n OH
asc-C5 oscr#9 [7] C11H20O6 271.1158 247.1182 3
O
asc-C9 oscr#10 [7] C15H28O6 327.1784 303.1808 7 HO

OH

Ascarosides with an ,-unsaturated () side chain attached at the penultimate (1) carbon and linked to p-aminobenzoic acid (PABA)
asc-C7-PABA ascr#8 [5] C20H27NO7 n.a. 392.1709 2 CO2H
O

O n N
H
O
HO

OH

R group definitions:
O OH O
O H
HO O N
O
HO
HO O O
OH HN HO HO
Glu, glucosyl IC, indole-3-carbonyl HB, 4-hydroxybenzoyl MB, 2-(E)-methyl-2-butenoyl OS, octopamine succinyl
The structure-based name given in the first column assumes that the canonical ascaroside structure is an ascaroside with a saturated fatty acid side chain that is attached to the sugar at the
penultimate (1) carbon. Deviations from the canonical structure are indicated such that each name conforms to the following:
(modifications to the sugar)-asc-(deviations in the fatty acid)C#-(modifications at the end of the fatty acid)
Modifications to the sugar and their abbreviations are listed at the bottom of the table. Deviations in the fatty acid include attachment of the fatty acid to the sugar at the terminal () carbon,
instead of penultimate (1) carbon, as well as - () unsaturation of the fatty acid. Modifications at the end of the fatty acid include the methyl ketone (MK) moiety and addition of
p-aminobenzoic acid (PABA)
a
Many other ascarosides, such as asc-C6 (ascr#12), asc-C8 (ascr#14), IC-asc-C9 (icas#10), asc-C11 (ascr#18), asc-C11 (oscr#18), asc-C12 (ascr#20), OS-asc-C9 (osas#10), and OS-asc-C6-MK
(osas#2), have also been detected in wild-type C. elegans culture medium, but their structures have not yet been fully confirmed through chemical synthesis [79]
b
Reduced versions of asc-C6-MK (ascr#6.1 and ascr#6.2) with a hydroxyl instead of a ketone have also been reported [5]
74 Xinxing Zhang et al.

techniques, such as liquid chromatography-mass spectrometry

(LC-MS) (Subheading 3.3), LC-MS/MS (Subheading 3.4), and
NMR spectroscopy (Subheading 3.5). Each of these techniques
has its own advantages and disadvantages, as described below. The
structural class of a specific ascaroside, as well as its concentration
in the crude pheromone extract, can affect whether or not it can be
detected in the extract using a specific analytical technique.
LC-MS analysis on a single quadrupole mass spectrometer
(Subheading 3.3) is a rapid technique for analyzing the concentra-
tions of specific known ascarosides in a sample [10, 11].
In LC-ESI-MS, the molecules in the extract are separated on a
column based on their polarity. As the molecules elute from the
column, they are ionized by electrospray ionization (ESI), and the
mass-to-charge (m/z) ratios of the ionized species are then
detected. Most ascarosides will ionize most strongly as positively
charged, sodium adducts. Thus, they can be detected by positive
ion mode LC-MS, in which the ionized ascaroside is detected as
[M + Na]+, where M is the mass of the ascaroside (see Table 1 for
which ascarosides ionize in positive ion mode). Many ascarosides,
but not all, will also ionize as negatively charged, deprotonated
species. Thus, they can also be detected by negative ion mode
LC-MS, in which the ionized ascaroside is detected as [MH] (see
Table 1 for which ascarosides ionize in negative ion mode).
In order to use LC-MS to identify the presence of a specific asca-
roside and to determine the concentration of that ascaroside in the
extract, a synthetic standard is needed for the ascaroside of interest.
The crude pheromone extract can contain non-ascaroside
compounds with masses and retention times that are similar to the
ascarosides of interest. For this reason, LC-MS-based identification
of ascarosides is usually confirmed with additional techniques,
including LC-MS/MS or NMR spectroscopy. LC-MS can be used
to analyze the ascarosides in (1) the crude pheromone extract from
the culture medium of a non-synchronized culture (generated by
method given in Subheading 3.2), (2) the crude pheromone
extract from the culture medium of a synchronized culture [11],
(3) the crude extract from the worms themselves [10], or (4) the
secretions from worms produced over a short period of time
(worm water) [4]. LC-MS is not useful for characterizing the
chemical structures of novel ascarosides, as this must be done using
techniques that provide additional structural information, such as
LC-MS/MS and NMR spectroscopy.
LC-MS/MS analysis on a triple quadrupole mass spectrometer
(Subheading 3.4) allows the simultaneous identification of many
of the ascarosides present in the crude pheromone extract [7].
LC-MS/MS is similar to LC-MS in that the molecules in the
extract are separated on a column based on their polarity and ion-
ized. The m/z ratios of the ionized species are then detected.
However, in LC-MS/MS, the molecules are not only ionized to
detect the ionized species (called the precursor ion), but also
 Analysis of Ascarosides 75

broken apart, resulting in product ions that provide additional

structural information about the molecules. When LC-MS/MS is
performed on a triple quadrupole mass spectrometer, a precursor
ion scan can be used to detect only molecules that give rise to a
specific product ion and report back the precursor ions that pro-
duced the specific product ion. This technique is particularly useful
for the detection of classes of molecules with similar structures,
such as the ascarosides, that all produce a common product ion.
Those ascarosides that ionize in negative ion mode can be frag-
mented in LC-MS/MS to produce a common product ion of m/z
73 [C3H5O2]. Thus, this technique, which was developed by von
Reuss and coworkers [7], can be used to detect all ascarosides in a
crude pheromone extract that ionize as negative ions and report
back the precursor ions (that is, [MH]) corresponding to those
ascarosides (Subheading 3.4). Those ascarosides that ionize only in
positive ion mode can be monitored by looking for a neutral loss
of 130 amu [C6H10O3] that corresponds to the loss of the ascary-
lose sugar [7]. However, fragmentation of the ascarosides in posi-
tive ion mode is not as efficient as that in negative ion mode,
leading to lower and less reliable signal, and will not be described
in this review.
NMR spectroscopy can be used to analyze the ascarosides in a
crude pheromone extract or chromatography fractions derived
from that extract (Subheading 3.5). One type of NMR experiment
that is particularly informative is the double quantum-filtered cor-
relation spectroscopy (dqf-COSY) experiment [5, 10]. Like a stan-
dard COSY experiment, this experiment detects correlations
between protons on neighboring carbons in a molecule, and these
correlations appear as peaks off the diagonal of the spectrum.
However, the dqf-COSY experiment is generally run in a phase-
sensitive mode and gives a much cleaner spectrum near the diago-
nal than COSY, which is important for analysis of mixtures
containing multiple components, such as crude pheromone extract,
which may have many peaks near the diagonal. Each dqf-COSY
correlation provides additional information about the protons
involved in the correlation, including the coupling constant
between those two protons (active coupling), as well as the cou-
pling constants of those protons with other neighboring protons
(passive coupling). The dqf-COSY spectrum of the crude phero-
mone extract contains peaks that are diagnostic for the ascarosides.
Furthermore, dqf-COSY spectra have been acquired for many of
the known ascarosides, and thus, the presence of a particular asca-
roside in a crude pheromone extract or chromatography fraction
can be substantiated by dqf-COSY. It should be noted, however,
that the dqf-COSY spectrum of ascarosides from the same struc-
tural class, but with different side chain lengths, will be similar.
One drawback of using NMR spectroscopy to analyze the ascaro-
sides in the crude pheromone extract is that less abundant
ascarosides cannot be detected because NMR spectroscopy is an
76 Xinxing Zhang et al.

inherently less sensitive technique than mass spectrometry.

Further fractionation of the crude pheromone extract to give a less
complex sample containing the specific ascaroside of interest is
needed. However, chromatographic purification of individual asca-
rosides is dependent on the polarity of those ascarosides and done
on an ad hoc basis, and thus, it is beyond the scope of this review.

2 Materials

2.1 25 Bacterial 1. E. coli strain OP50 or HB101, grown on Luria-Bertani

Stock (LB)-agar.
2. LB broth.
3. Falcon tube (50 mL) for bacterial mini-culture.
4. Laminar flow hood.
5. 2.8 L baffled flask for large bacterial culture.
6. Shaker for shaking bacterial culture at 37 C.
7. Refrigerated table-top centrifuge.
8. Potassium phosphate (1 M, pH 6): In 0.8 L water, dissolve
136.1 g KH2PO4. Add potassium hydroxide until the pH
reaches 6. Then add water to a final volume of 1 L. Autoclave.
9. S basal: 0.1 M NaCl, 0.05 M potassium phosphate (pH 6).
Autoclave.
10. Microfuge and 1.5 mL Eppendorf tubes.

2.2 Worm Culturing 1. C. elegans, wild-type (N2) or mutant strain.

2. Stereomicroscope and worm pick for manipulating worms.
3. Cholesterol (5 mg/mL in ethanol): Dissolve 50 mg of choles-
terol in 10 mL of ethanol (200 proof). Sonicate for approxi-
mately 10 min or shake until the cholesterol dissolves.
4. 25 bacterial stock from Subheading 3.1 for feeding worm
cultures. 25 mL of 25 stock (corresponding to 625 mL-
worth of bacterial culture) is needed for every 150 mL of
worm culture that will be grown.
5. 10 cm NGM-agar plates with bacterial lawn: In 500 mL water,
add 1.5 g NaCl, 8.5 g granulated agar, and 1.25 g peptone.
Autoclave, and then add: 0.5 mL 1 M CaCl2, 0.5 mL 1 M
MgSO4, 12.5 mL potassium phosphate (pH 6), and 0.5 mL
5 mg/mL cholesterol. Pour into 10 cm plates, and let them
dry on the benchtop overnight. Pipet 0.75 mL 25 bacterial
stock onto each plate, spread on plate (but do not allow lawn
to touch plate edge), and dry in laminar flow hood or on
benchtop. Approximately one plate is needed for every
150 mL of worm culture that will be grown.
 Analysis of Ascarosides 77

6. Potassium citrate (1 M, pH 6): In 0.8 L water, dissolve

210.14 g citric acid monohydrate. Add potassium hydroxide
until the pH reaches 6. Then add water to a final volume of
1 L. Autoclave.
7. Trace metals solution: In 1 L water, dissolve 1.86 g disodium
EDTA (5 mM), 0.69 g FeSO4 7H2O (2.5 mM), 0.20 g
MnCl2 4H2O (1 mM), 0.29 g ZnSO4 7H2O (1 mM),
0.025 g CuSO4 5H2O (0.1 mM). Autoclave. Wrap bottle in
foil, and store in the dark.
8. S medium: Autoclave 150 mL of S basal in a 500 mL
Erlenmeyer flask, covering the top with foil. Just before use,
add 1.5 mL 1 M potassium citrate (pH 6), 1.5 mL trace metals
solution, 0.45 mL 1 M CaCl2, 0.45 mL 1 M MgSO4, and
150 L cholesterol (5 mg/mL).
9. Optional: To help prevent contamination of the cultures, add
the antifungal nystatin (100 U/mL final concentration) and
the antibiotics penicillinstreptomycin (50 U/mL final con-
centration for penicillin and 50 g/mL final concentration for
streptomycin).
10. Refrigerated shaker for culturing worms in shaker flasks at
22.5 C.
11. Graduated cylinder (250 mL) and ice bath.
12. Refrigerated table-top centrifuge and polypropylene centri-
fuge bottles with screw caps.
13. Lyophilizer and lyophilization flask (600 mL).
14. Mortar and pestle.
15. Erlenmeyer flask (500 mL) and funnel.
16. Ethanol (190 proof).
17. Tabletop shaker for performing extractions.
18. Filter paper (Whatman, No. 41), Buchner funnel, neoprene
adaptor, and Erlenmeyer filtering flask (500 mL).
19. Rotary evaporator and round bottom flask (250 mL).
20. Short (5.75 in.) and long (9 in.) Pasteur pipets and bulbs.
21. Clean ball of cotton.
22. Bendable spatula.
23. Speedvac, rotor for 8 mL glass vials, and 8 mL glass vials (e.g.,
Fisher, cat. 03-340-60B) (see Note 1).

2.3 LC-MS 1. Methanol.

2. Sonicator.
3. Speedvac and rotor for 8 mL glass vials (see Note 1).
4. Microfuge and methanol-safe 1.5 mL Eppendorf tubes (e.g.,
Fisher, cat. 05-408-129).
78 Xinxing Zhang et al.

5. Mass spectrometer: Agilent 6130 single quad mass spectrom-

eter with API-ES source and operating in dual negative/posi-
tive scan mode.
6. LC: Agilent 1260 infinity binary pump with autosampler.
7. Column: Phenomenex 5 m Luna C18 2, 100 (100 4.6 mm).
8. LCMS-grade water with 0.1 % formic acid (solvent A).
9. LCMS-grade acetonitrile with 0.1 % formic acid (solvent B).
10. Autosampler vials and glass inserts with springs.

2.4 LC-MS/MS 1. Methanol.

2. Sonicator.
3. Speedvac and rotor for 8 mL glass vials (see Note 1).
4. Microfuge and methanol-safe 1.5 mL Eppendorf tubes (e.g.,
Fisher, cat. 05-408-129).
5. Mass spectrometer: Thermo TSQ quantum ultra triple quad-
rupole mass spectrometer using argon collision gas, operating
in negative ion, heated (H)-ESI, precursor scanning mode,
selecting for a product ion of 73.0.
6. LC: Accela UHPLC (max pressure of 1,000 bar) with
autosampler.
7. Column: Kinetex 2.6 m C18 100 (100 2.10 mm).
8. LCMS-grade water (solvent A).
9. LCMS-grade acetonitrile with 0.1 % acetic acid (solvent B).
10. Autosampler vials and glass inserts with springs.

2.5 NMR 1. Lyophilizer.

Spectroscopy 2. Methanol-d4.
3. Sonicator.
4. Short (5.75 in.) and long (9 in.) Pasteur pipets and bulbs.
5. Clean ball of cotton.
6. 5 mm methanol-matched Shigemi NMR tube (Shigemi, cat.
MMS-005v) (see Note 2).
7. Parafilm.
8. NMR spectrometer: Varian 500 MHz NMR spectrometer.
9. NMR probe: 5 mm Varian indirect-detection, triple resonance
probe (1H/13C/(31P-15N)PFG).
10. Pasteur pipets with extra long tips for removing samples from
NMR tubes (Fisher, cat. 13-678-6).
11. Speedvac, rotor for 8 mL glass vials, and 8 mL glass vials (e.g.,
Fisher, cat. 03-340-60B).
 Analysis of Ascarosides 79

3 Methods

3.1 Preparing 25 This method prepares the bacterial food (25 bacterial stock)
Bacterial Stock needed for growing the worm cultures in Subheading 3.2. Sterile
technique is important in the preparation of the 25 bacterial
stock. If the 25 bacterial stock is contaminated with bacteria or
fungus, this contamination may overwhelm the worm cultures that
are grown in Subheading 3.2, regardless of whether antibiotics and
antifungals are added to the worm culture medium.
1. Inoculate 5 mL LB with a colony of OP50 or HB101 bacteria.
Let the mini-culture grow for 68 h in a shaker at 250 rpm at
37 C (see Note 3).
2. Use 1 mL of this mini-culture to inoculate 1 L of LB broth in
a 2.8 L baffled flask. Grow culture overnight in a shaker at
250 rpm at 37 C.
3. Collect bacteria by centrifuging at 2,700 g for 10 min.
Discard the supernatant, and resuspend in enough S basal to
make the final volume equal to 40 mL to give a 25 bacterial
stock. Transfer to a 50 mL falcon tube, and store at 4 C (see
Notes 46).

3.2 Unsynchronized This method describes how to grow unsynchronized worm cul-
Worm Cultures and tures and how to extract the culture medium to generate crude
Generating Crude pheromone extract. Production of many of the ascarosides is maxi-
Pheromone Extract mized by feeding the worm cultures regularly with sufficient bacte-
rial food to allow reproduction and increase worm density. It is
important not to feed the worms too much on any 1 day because
they will suffocate, but it is also important not to deprive the
worms of food. The exact culture conditions used will influence
the amounts of the ascarosides produced.
1. Passage 20 N2 worms onto a 10 cm NGM-agar plate with a
bacterial lawn.
2. Let worms grow for ~4 days at room temperature until just
before the bacterial food supply runs out.
3. Wash the plate with 10 mL of S medium, and put in a 15 mL
tube. Count the number of worms in 3 10 L drops under
the stereomicroscope to gauge the density of worms. Make
sure that you invert the tube several times before taking out
each 10 L sample (see Note 7).
4. Add ~90,000 worms per 150 mL culture.
5. Feed the worms 3 mL of 25 bacterial stock (110125 mg/
mL). For our feeding schedule, this day is considered day 0.
6. Shake at 22.5 C at 225 rpm for 9 days. Feed worms 1 mL
25 bacterial stock on day 1, and 3 mL 25 bacterial stock on
days 28.
80 Xinxing Zhang et al.

7. On day 9, harvest the culture by transferring the culture to a

tall container, such as a 250 mL graduated cylinder, and place
the container in an ice-bath for 30 min to 1 h in order to settle
the worms. Transfer the supernatant to a centrifuge bottle
until you cannot pipet any more of the supernatant without
disturbing the settled worms. Transfer the remaining worms
and culture medium from the bottle to a 50 mL falcon tube,
centrifuge at 200 g for 5 min. Pipet the remaining superna-
tant from the falcon tube into the centrifuge bottle together
with the previously pipetted supernatant. Centrifuge the cen-
trifuge bottle at 2,700 g for 10 min to remove the bacteria
from the supernatant. Then carefully transfer the supernatant
to a 600 mL lyophilization flask, and freeze by placing it at
20 C overnight.
8. Lyophilize the frozen culture medium to dryness (23 days)
(see Note 8).
9. Scrape out solid material into a mortar, and grind with a pestle
to a fine powder.
10. Place the powder in a 500 mL Erlenmeyer flask using a funnel,
and extract by adding 150 mL ethanol (190 proof) and shak-
ing in a table-top shaker at 250 rpm for 2 h.
11. Filter the extraction mixture through filter paper in a Buchner
funnel.
12. Place the retentate back in the Erlenmeyer flask from step 10,
and extract a second time with 150 mL of ethanol (190 proof),
shaking overnight (see Note 9).
13. Filter the extraction mixture through filter paper in a Buchner
funnel, and combine the filtrate with the filtrate from step 11.
14. Rotovap the combined filtrates in a round bottom flask to a
small volume (~3 mL).
15. Place a small amount of cotton in a short Pasteur pipet, and
pack it down inside the Pasteur pipet using a long Pasteur
pipet, such that the packed cotton plug is about 3 mm in
length.
16. Wash the cotton in the Pasteur pipet with a small amount of
ethanol to remove any impurities from the cotton. Use a bulb
to push the ethanol through the Pasteur pipet and dry the
cotton.
17. Filter the sample through the cotton plug in the Pasteur pipet
into a 8 mL glass vial. Use a bulb to push the sample through
the Pasteur pipet.
18. Wash the round bottom with ~3 mL of ethanol (190 proof).
Scrape the sides of the round bottom with a metal spatula to
make sure that as much of the solid is exposed to the ethanol
as possible.
 Analysis of Ascarosides 81

19. Filter this ethanol wash through the cotton plug in the Pasteur
pipet into the 8 mL glass vial from step 17. Use a small amount
of ethanol to wash the plug, and add this last wash to the glass
vial as well.
20. Speedvac the sample in the glass vial to dryness. Store the sam-
ple at 20 C until use.

3.3 LC-MS This method can be used to quantify the concentrations of the asca-
Quantitation rosides in a particular sample, such as in the crude pheromone
of Ascarosides extract (Fig. 1) or in worm secretions produced over a short period
of time (worm water) [4]. The mass spectrometer must be well
maintained with low background signal. It should be noted that
this method can underestimate the amount of an ascaroside in a
sample, especially if the sample is complex, if ion suppression occurs.
1. Resuspend the sample from Subheading 3.2, step 20, by add-
ing 1 mL methanol. Sonicate for 20 min, vortex briefly to dis-
lodge any remaining solids on the sides of the vial, and sonicate
again for 20 min. Centrifuge the vial (with the cap closed) in
the speedvac for 2 min to pellet most of the particulates.
2. Pipet 100 L of the supernatant into a 1.5 mL Eppendorf
tube, and centrifuge at 17,600 g for 1 min to pellet any
remaining particulates. Pipet 50 L from the top of the sample
to an autosampler vial. Pipet the remaining 50 L of the sam-
ple from the tube back into vial (see Note 10).
3. Make 0.2, 0.4, 2, 4, 8, 20, 40, 60, 80, and 100 M stocks of
each synthetic standard in methanol. Transfer 20 L of each
stock to an autosampler vial (see Note 11).
4. For an Agilent 6130 single quadrupole mass spectrometer, use
the following parameters for the spray chamber of the ion
source: Drying gas flow of 12 L/min, nebulizer pressure of
35 psig, capillary voltage of 3,000 V, and drying gas tempera-
ture of 350 C. Acquire signals in dual positivenegative scan
mode with a 50 % cycle time between the two modes, peak
width of 0.1 min, and cycle time of 1.08 s/cycle. Use a mass
range of 1001,000, fragmentor voltage of 125 V, gain of
1.00, threshold of 150, and stepsize of 0.10 (see Note 12).
5. Set the column temperature to 25 C.
6. Set up the following chromatography method, where the sol-
vent A is water with 0.1 % formic acid and solvent B is aceto-
nitrile with 0.1 % formic acid: (1) ramp from 5 % B to 100 %
B over 20 min, (2) hold at 100 % B for 2 min, (3) ramp from
100 % B to 5 % B over 1 min, and (4) hold at 5 % B for 2 min.
Set the flow rate at 0.7 mL/min (see Note 13).
7. In order to clean the column, wash the column with 100 %
acetonitrile for 10 min at 0.7 mL/min.
82 Xinxing Zhang et al.

a asc-C9, MSD1 325, EIC=324.8:325.8

Area asc-C9+ at exp. RT: 9.497
5 MSD1 325, EIC=324.8:325.8
80000 1E6 Correlation: 0.99945
asc-C9 4 Residual Std. Dev.: 15643.54053
800000
Formula: y = ax + bx + c
60000 7.21e+005 3 a: -2.18009
600000
2
b: 3743.74712
40000 400000 c: 44334.55299
1
x: Amount
200000 y: Area
20000
205.279
0
0 200 400
0 Amount[pmol]

2.5 5 7.5 10 12.5 min

b asc-C6-MK, MSD1 269, EIC=268.8:269.8 Area asc-C6-MK+ at exp. RT: 6.952

5
MSD1 269, EIC=268.7:269.7
50000 400000 4.15e+005 Correlation: 0.99992
Residual Std. Dev. : 2643.67442
40000 asc-C6-MK 300000 Formula: y = ax + bx + c
4
a: -5.35225
30000
200000 b: 3463.91963
3
c: 8847.85835
20000
100000 2 x: Amount
1
10000 y: Area
153.710
0
0 100 200
0 Amount[pmol]

2.5 5 7.5 10 12.5 min

c asc-C3, MSD1 243, EIC=242.8:243.8

Area asc-C3+ at exp. RT: 4.318
5
MSD1 243, EIC=242.8:243.8
160000 140000
Correlation: 0.99965
140000 120000
Residual Std. Dev.: 2031.15163
120000 100000 1.08e+005
Formula: y = ax + bx + c
4
100000 80000 a: -2.63607
80000 60000 b: 1392.01335
3 c: -11618.09570
60000 40000
2 x: Amount
40000 20000
asc-C3 1 108.413
y: Area
20000 0
0 100 200
0 Amount[pmol]

2.5 5 7.5 10 12.5 min

d IC-asc-C5, MSD1 414, EIC=413.8:414.8 Area IC-asc-C5+ at exp. RT: 11.582

5
MSD1 414, EIC=413.7:414.7
800000
IC-asc-C5 Correlation: 0.99998
40000 Residual Std. Dev.: 2477.30542
600000 Formula: y = ax + bx + c
4
30000 a: -9.09933
400000 3.70e+005 b: 6613.92172
3
20000 c: 16129.27854
200000 2 x: Amount
10000 1
y: Area
58.108
0
0 100 200
0 Amount[pmol]

2.5 5 7.5 10 12.5 min

Fig. 1 LC-MS-based determination of the concentrations of asc-C9, asc-C6-MK, asc-C3, and IC-asc-C5 in
crude pheromone extract. A calibration curve of synthetic ascarosides was generated by injecting 5 L of
several concentrations of the ascarosides and fitting a quadratic curve to the data. 5 L of crude pheromone
extract was also injected, and the relevant ion for each ascaroside was extracted from the positive total ion
count trace. The area under the peak for each ascaroside was used to estimate the number of pmol in 5 L of
crude pheromone extract. This number then was used to estimate the concentration of each ascaroside in the
conditioned medium that was used to generate the crude pheromone extract. (a) The concentration of asc-C9
in the conditioned medium is ~273.7 nM. (b) The concentration of asc-C6-MK in the conditioned medium is
~204.9 nM. (c) The concentration of asc-C3 in the conditioned medium is ~144.5 nM. (d) The concentration
of IC-asc-C5 in the conditioned medium is ~77.4 nM
 Analysis of Ascarosides 83

8. In order to equilibrate the column, wash the column with 5 %

B for 10 min at 0.7 mL/min.
9. Use the autosampler to inject 5 L of methanol into the
LC-MS. This blank run will allow you to assess the back-
ground level of signal (see Note 14).
10. Use the autosampler to inject 5 L of sample into the LC-MS.
11. Extract the relevant ion for the ascaroside of interest. That is,
extract [M + Na]+ for the ascaroside of interest from the posi-
tive total ion count trace, or extract [MH] for the ascaroside
of interest from the negative total ion count trace. Obtain the
area under the peak for the extracted ion.
12. Use the autosampler to inject 5 L of each concentration of
the synthetic standards in turn into the LC-MS.
13. For each concentration of synthetic standard injected into the
LC-MS, extract the relevant ion from the total ion count trace,
and use the area under the peak to generate a standard curve.
14. Use the standard curve generated to estimate the concentra-
tion of the ascaroside of interest in the sample (see Fig. 1).
15. In order to clean and store the column, wash the column with
100 % acetonitrile for 10 min.

3.4 LC-MS/MS This method can be used to analyze the ascarosides in the crude
Analysis of pheromone extract that ionize in negative ion mode (Fig. 2). This
Ascarosides method assumes that the user has a basic knowledge of LC-MS/
MS and is assisted by an experienced mass spectrometry facility
manager. The mass spectrometer must be well maintained with low
background signal.
1. Resuspend the sample from Subheading 3.2, step 20, by add-
ing 1 mL methanol. Sonicate for 20 min, vortex briefly to dis-
lodge any remaining solids on the sides of the vial, and sonicate
again for 20 min. Centrifuge the vial (with the cap closed) in
the speedvac for 2 min to pellet most of the particulates.
2. Pipet 100 L of the supernatant into a 1.5 mL Eppendorf
tube, and centrifuge at 13,700 rpm for 1 min to pellet any
remaining particulates. Pipet 50 L from the top of the sample
to an autosampler vial with glass insert with spring. Pipet the
remaining 50 L of the sample from the tube back into vial
(see Note 10).
3. Set the mass spectrometer to operate in negative ion, heated
(H)-ESI, precursor scanning mode, selecting for a product ion
of 73.0. If you are using a Thermo TSQ quantum ultra mass
spectrometer, use the following parameters for the mass spec-
trometer: spray voltage of 3,000 V, vaporizer temperature of
300 C, sheath gas pressure of 40 arb, auxillary gas pressure of
20 arb, capillary temperature of 350 C, skimmer offset of
84 Xinxing Zhang et al.

50
OS-asc-C5

45

40
asc-C7 asc-C9 IC-asc-C5
Relative Abundance (%)

35

30

25
asc-C9
20 asc-C7

asc-C6
15
asc-C5
10
asc-C3
asc-C5
asc-C8 asc-C11
5 asc-C4 IC-asc-C9

0
5 10 15 20 25 30
Time (min)

Fig. 2 LC-MS/MS trace in precursor ion scanning mode (selecting for m/z 73.0 product ion) for crude phero-
mone extract. Asterisks indicate signals from unidentified compounds or non-ascarosides

5 V, collision pressure of 0.8 mTorr, collision energy of 25 eV,

and scan time 0.4 s. We usually set the mass range to m/v
200500 (see Notes 15 and 16).
4. Set the column temperature to 30 C.
5. Set up the following chromatography method, where solvent
A is water and solvent B is acetonitrile with 0.1 % acetic acid:
(1) hold at 2 % B for 1 min, (2) ramp from 2 % B to 50 % B
over 34 min, (3) ramp from 50 % B to 100 % B over 1 min, (4)
hold at 100 % B for 4 min, (5) ramp from 100 % B to 2 % B
over 1 min, (6) hold at 2 % B for 4 min. Set the flow rate at
0.25 mL/min (see Note 17).
6. In order to clean the column before use, wash the column
with 100 % acetonitrile for 10 min at 0.25 mL/min.
7. In order to equilibrate the column, wash the column with 2 %
solvent B for 10 min at 0.25 mL/min.
8. Use the autosampler to inject 5 L of methanol into the
LC-MS/MS. This blank run will allow you to assess the back-
ground level of signal (see Note 14).
9. Use the autosampler to inject 5 L of sample into the LC-MS/
MS.
10. After the run is complete, smooth the trace using a Boxcar or
Gaussian function, with 57 smoothing points (see Note 18).
 Analysis of Ascarosides 85

11. Obtain the areas under the peaks for each ascaroside by extract-
ing each specific precursor ion from the total ion count trace
(see Note 19).
12. In order to validate that a given peak is a given ascaroside, you
will need to acquire a synthetic standard. Dissolve the
standard(s) in ethanol (200 proof) at 270 M. Transfer some
of the sample to an autosampler vial, and use the autosampler
to inject 5 L from the autosampler vial into the LC-MS/MS.
13. After the run is complete, verify that the synthetic standard has
the same retention time as the tentatively assigned peak from
step 11.
14. In order to clean and store the column, wash the column with
100 % acetonitrile for 10 min.

3.5 NMR This method can be used to analyze the ascarosides in the crude
Spectroscopy-Based pheromone extract (Fig. 3) or chromatography fractions generated
Analysis of from the crude pheromone extract. The ascarosides must be pres-
Ascarosides Using the ent at sufficient concentrations in the crude pheromone extract for
dqf-COSY Experiment detection by NMR spectroscopy. Subheading 3.5 assumes that the
user has a basic knowledge of NMR and is assisted by an experi-
enced NMR facility manager. The NMR spectrometer must be
well maintained with an updated shim set. Although we use a
Varian NMR spectrometer in our own experiments, we employ
terminology in this method that should be interpretable by users
familiar with either Agilent (Varian) or Bruker NMR spectrome-
ters. In order to have the necessary chemical shift separation in the
complex spectrum obtained from crude pheromone extract, an
NMR spectrometer with a magnetic field of 500 MHz or higher
should be used. The number of points in both dimensions should
provide for a resolution of 12 Hz per point. The concentrations
of the ascarosides in the crude pheromone extract are relatively low
compared to the concentrations of other more abundant metabo-
lites. Therefore, it is critical to maximize the signal-to-noise ratio
for the sample, and thus, an indirect detection probe or a cold
probe is recommended.
1. Resuspend the sample from Subheading 3.2, step 20, by add-
ing 1 mL methanol-d4. Sonicate for 20 min, vortex briefly to
dislodge any remaining solids on the sides of the vial, and soni-
cate again for 20 min. Centrifuge in the speedvac (with the cap
closed) for 2 min to pellet any particulates that do not
dissolve.
2. Transfer the supernatant to a new vial, and speedvac or roto-
vap the sample to dryness (see Note 20).
3. Place the vial of dry sample, a short Pasteur pipet, and an
empty 5 mm methanol-matched Shigemi tube on the lyophi-
lizer for 12 h. This drying step will remove any trace amounts
86 Xinxing Zhang et al.

1.0

1.5
c

2.0

2.5

3.0

3.5

f1 (ppm)
4.0

4.5

5.0

5.5
b
6.0

6.5

a
7.0
7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0
f2 (ppm)

asc-C9 5.81 ppm a 15.6

Hz
f1
b 6.8
Hz
f1

H H H
OH
O 6.8 15.6
6.92 5.81
Hz Hz
O H O
HO 6.92 ppm 15.6
Hz

OH f2 5.81 f2 6.92

asc-C6-MK 2.61 ppm

c 14.8 f1
HH Hz
7.4 1.71
Hz
O 7.4
Hz
Ha Hb O 1.77
O
HO 1.77 1.71
ppm ppm

OH
f2 2.61
asc-C3 3.67 3.94
ppm ppm d 12.6
Hz
f1
Hb Ha O
6.3
Hz
O 6.3 OH 3.94
Hz
HH
O 2.56
HO ppm

OH f2 2.56

Fig. 3 dqf-COSY spectrum of crude pheromone extract, highlighting specific correlations for bolded protons for
specific ascarosides (asc-C9, asc-C6-MK, and asc-C3) that are visible in the spectrum. (a and b) Highlight
 Analysis of Ascarosides 87

of methanol and water in the sample that would otherwise

interfere with the signal of the sample in the NMR spectrum.
4. Resuspend your sample by adding 300 L of methanol-d4 and
sonicating for 5 min.
5. Place a small amount of cotton in the short Pasteur pipet, and
pack it down inside the Pasteur pipet using a long Pasteur
pipet, such that the packed cotton plug is about 3 mm in
length.
6. Wash the cotton in the Pasteur pipet with a small amount of
methanol-d4 to remove any impurities from the cotton. Use a
bulb to push the methanol-d4 through the Pasteur pipet and
dry the cotton.
7. Filter the sample through the cotton plug in the Pasteur pipet,
directly into the Shigemi tube. Use a bulb to push the sample
through the Pasteur pipet. Use a few drops of methanol-d4 to
wash the cotton plug, allowing the wash to drip directly into
the Shigemi tube (see Note 21).
8. Insert the solvent-matched Shigemi tube insert into the
Shigemi tube, taking care to avoid any bubbles between the
sample and the insert. Use a small strip of parafilm (about
0.5 2 cm in length) to wrap around the top of the tube to
prevent any evaporation of the sample while it is in the NMR
spectrometer (see Notes 22 and 23).
9. Insert your sample into the spectrometer. Setup parameters
for a 1H spectrum in methanol (see Note 24).
10. Lock the sample, and optimize the lock phase.
11. Gradient shim the sample.
12. Tune the 1H channel of the probe.
13. Turn off the spinner (see Note 25).
14. Acquire a 1H spectrum with 16 scans (and with the spinner
off) (see Notes 26 and 27).

Fig. 3 (continued) the below diagonal and above diagonal correlations between the bolded proton at 5.81 ppm
and the bolded proton at 6.92 ppm in asc-C9. The active coupling can be measured in (a) and (b) to give the
coupling constant between the two bolded protons (15.6 Hz). The passive coupling can be measured in (b) to
give the coupling constant between the bolded proton at 6.92 ppm and the other protons to which it is coupled
(6.8 Hz). (c) Highlights the above diagonal correlations between the bolded protons at 2.61 ppm and the two
nonequivalent bolded protons at 1.71 ppm and 1.77 ppm in asc-C6-MK. The active coupling between the bolded
protons can be measured in (c). The measured value of 14.8 Hz must be divided in two (to give coupling con-
stants of approximately 7.4 Hz) because the bolded proton at 2.61 ppm is a triplet and oppositely phased peaks
in the center of the triplet cancel each other. The below diagonal correlation is not highlighted as it is too faint.
(d) Highlights the below diagonal correlation between the two bolded protons at 2.56 ppm and the bolded proton
at 3.94 ppm in asc-C3. The above diagonal correlation is not highlighted as it is obscured by a vertical stripe
88 Xinxing Zhang et al.

15. Set the sweep width to the region of the 1H spectrum that you
are interested in for the 2D experiment. Usually you will select
the region where there are peaks, plus 0.5 ppm on either side.
In general, for the crude pheromone extract, the sweep width
will extend from 0.5 to 9.5 ppm.
16. Determine the 90 pulse width. Make sure that the pulse
width is equal to this new optimized value (see Note 28).
17. Acquire another 1H spectrum (see Note 29).
18. Move the 1H spectrum into a new experiment in order to
move all optimized parameters, including sweep width, 90
pulse width, and gain, into the new experiment. Setup up a
dqf-COSY experiment with the default parameters.
19. Change the default parameters in the following manner: the
number of scans to 16, the number of iterations to 1,024, the
number of points to 4,096, and the relaxation delay to 1 (see
Notes 30 and 31).
20. Start the experiment, and once finished, process the data (see
Note 32).

4 Notes

1. If a speedvac with a rotor for 8 mL glass vials is not available,

the material can be dried on a rotovap with a 24/40 glass
adaptor (e.g., Chemglass cat. CG-1318-10) and connector
(e.g., Chemglass cat. CG-1318-21) for the 8 mL glass vials.
2. This tube is magnetic susceptibility-matched to methanol.
The catalog number given is for a tube that is designed for use
with a 5 mm Varian NMR probe.
3. Because sterile technique is important, we perform as many of
the steps as possible in a laminar flow hood.
4. To gauge the sterility of the 25 bacterial stock, streak it out
on an LB-agar plate and allow colonies to grow. There should
be only one colony type.
5. The 25 bacterial stock can be stored at 4 C for up to
1 month.
6. Typically, if HB101 or OP50 is grown in LB broth as described,
110125 mg/mL of bacteria will be obtained for the 25
stock. Calculate the stock concentration by weighing an empty
1.5 mL Eppendorf tube and then adding 1 mL of 25 stock
to it. Centrifuge the tube in the microfuge at 13,700 rpm for
1 min, and remove the supernatant by pipetting. Centrifuge
again, and remove any remaining supernatant by pipetting.
Reweigh the tube with the pellet in it. The weight of the pellet
gives the stock concentration in mg/mL.
 Analysis of Ascarosides 89

7. Generally, for N2, ~90,000 worms are obtained from a single

10 cm plate, but for some mutants, especially for those that
like to dig into the agar, this number can be lower.
8. It is important that there are no chunks of frozen material
remaining, since it will interfere with the extraction process.
After 2 days, it is helpful to use a spatula to break apart any
remaining material, and place the flask back on the
lyophilizer.
9. The second extraction significantly increases the pheromone
yield.
10. If the sample will be used again within a week, store at 20 C.
If the sample will not be used again within a week, speedvac it
to dryness, and store at 20 C.
11. If a standard curve will be generated for more than one syn-
thetic ascaroside, it is convenient to put all synthetic standards
together at a given concentration for simultaneous injection.
12. These parameters are instrument dependent. The fragmentor
voltage was optimized on our instrument using a synthetic
ascaroside standard (20 M asc-C9 in ethanol). If a mass
spectrometer other than an Agilent 6130 single quadrupole
mass spectrometer will be used, these parameters may need to
be re-optimized.
13. This chromatography protocol was optimized for monitoring
the concentrations of specific ascarosides (asc-C3, asc-C6-
MK, asc-C9, IC-asc-C5, as shown in Fig. 1). It should be
noted that this gradient is a short one, and so two or more
ascarosides in the crude pheromone extract may elute at the
same time. For example, two ascarosides with similar polarities
and the same mass (e.g., asc-C5 and asc-C5) may elute from
the column at the same time, which would be problematic if
the objective was to analyze their concentrations. To prevent
co-elution of two or more ascarosides, a longer gradient may
be required such as the gradient used in Subheading 3.4.
14. If there is a high background level of signal, it most likely indi-
cates that the column is dirty and must be cleaned more thor-
oughly before use.
15. These parameters are instrument dependent. They were opti-
mized on our instrument based on direct infusion of a syn-
thetic ascaroside standard (4 M asc-C9 in ethanol). If a
mass spectrometer other than a Thermo TSQ quantum mass
spectrometer will be used, these parameters will need to be
re-optimized.
16. This mass range should be shifted if you are interested in ana-
lyzing ascarosides with larger masses, such as the long-chain
ascarosides.
90 Xinxing Zhang et al.

17. This chromatography protocol was optimized for the

separation and analysis of more polar, short-chain ascarosides,
but not for the separation and analysis of less polar, long-chain
ascarosides. If you are interested in separating and analyzing
long-chain ascarosides, you will need to extend the ramp from
50 % B to 100 % B over a longer period of time.
18. Using greater than seven smoothing points may create artifi-
cial peaks.
19. Different ascarosides ionize and fragment with different effi-
ciencies, so the area of a given peak is not a very good indica-
tion of its relative abundance.
20. This drying step will remove trace amounts of methanol and
water in your sample that would otherwise interfere with the
signal of the sample the NMR spectrum.
21. The sample should extend exactly 2 cm in the Shigemi tube.
This volume of sample will allow the sample to extend the full
length of the receiver coil when inserted into the NMR spec-
trometer. When washing the cotton filter with a few drops of
methanol-d4, take great care not to add too much volume,
since the sample will then extend longer than 2 cm in the
Shigemi tube and will be more dilute. A more dilute sample
will give lower signal in the NMR spectrum.
22. If bubbles appear, they must be removed since they may inter-
fere with optimal shimming. They can be removed by slightly
pushing the insert 12 mm further into the tube with a quick
turning motion.
23. Shigemi tubes have a smaller sample window than regular
5 mm NMR tubes because both the bottom of the tube and
the insert that fits into the tube are made of solvent-matched
glass. Thus, the volume of solvent required to dissolve the
sample and fill the tube is smaller, thereby increasing the con-
centration of the sample and the signal-to-noise ratio in the
NMR spectrum.
24. Be sure to align the Shigemi tube in the spinner using the
depth gauge so that once the tube is inserted into the NMR
spectrometer, the deuterated solvent will be centered within
the receiver coil and will extend the full length of the receiver
coil. If the deuterated solvent does not extend the full length
of the receiver coil, it will cause issues with shimming.
25. If the lock level drops significantly (>10 %) when the spinner
is turned off, it indicates poor shimming. Adjust the non-
spinning (x, y) shims accordingly.
26. The number of scans determines the signal-to-noise ratio of
the spectrum. For concentrated samples such as the crude
pheromone extract, 16 scans is generally good, but for more
 Analysis of Ascarosides 91

dilute samples, such as chromatography fractions from the

crude pheromone extract, the number of scans should be set
higher (e.g., 32). In general, if a sample is more dilute by a
factor of 2, then the number of scans must be increased by a
factor of 4 in order to obtain a spectrum with a similar signal-
to-noise ratio. The number of scans in a dqf-COSY should be
a multiple of 8 due to phase-cycling.
27. Make sure that the peak shapes are sharp and that the metha-
nol peak at 3.31 ppm is a fully defined quintet. If the peak
shape is not sharp or is skewed, it indicates poor shimming.
Spinning the tube improves peak shape, but since spinning
will be turned off during the 2D dqf-COSY experiment, it is
important to assess peak shape in a 1H spectrum in which the
spinner has been turned off.
28. Because the crude pheromone extract has a high concentration
of salt, the 90 pulse width can be much higher than for more
dilute samples. Optimizing the 90 pulse width will improve
the signal-to-noise ratio in the 2D dqf-COSY experiment.
29. Acquiring this second 1H spectrum with an optimized 90
pulse width will give you a lower gain value than that obtained
in the original 1H spectrum. Use this new lower gain value in
the dqf-COSY experiment.
30. The number of iterations determines the resolution in the
indirect (f1) dimension and should be at least 512 for a dqf-
COSY experiment performed on a complex mixture. If the
experiment time (which is proportional to the number of iter-
ations) is not an issue, then the number of iterations should be
increased to 1,024.
31. The resolution in the direct (f2) dimension depends on the
number of points. Since increasing the number of points does
not significantly increase the experiment time, there is no dis-
advantage to setting the number of points to as high as 4,096
(except that it will increase the file size). The number of points
should be equal to a power of two since the Fourier transform
uses 2n points. Given a sweep width of approximately 9 ppm,
setting the number of points to 4,096 will give an acquisition
time of approximately 0.5.
32. The desktop processing software, MestreNova, is a convenient
way to process NMR data. In the f2 dimension, use a sine-bell
window function with a 90 shift (to obtain high resolution)
or a squared sine-bell window function with a 55 shift (if high
noise is present). In the f1 dimension, use a sine-bell window
function with a 90 shift. For zero-filling, the size of the
Fourier transform should be 4,096 in both dimensions. The
spectrum will also need to be phased.
92 Xinxing Zhang et al.

Acknowledgments

Funding for this work was provided by the National Institutes of

Health (GM87533 to R.A.B.) and the Human Frontiers Science
Program (RGY0042/2010 subcontract to R.A.B.). Our work on
the triple quadrupole mass spectrometer was supported in part by
the National Institutes of Health and National Center for Research
Resources CTSA grant 1UL 1RR029890. We would like to thank
Dr. Tim Garrett (University of Florida) for helpful suggestions on
the LC-MS/MS parameters and Dr. Ion Ghiviriga (University of
Florida) for helpful suggestions on the NMR parameters. We
would like to thank Prof. Justin R. Ragains (Louisiana State
University) for providing the synthetic ascaroside standards that
are used in our work.

References
1. Jeong PY, Jung M, Yim YH et al (2005) with unusual dauer inhibitory activity at higher
Chemical structure and biological activity of concentrations. Org Lett 11:31003103
the Caenorhabditis elegans dauer-inducing 7. von Reuss SH, Bose N, Srinivasan J et al
pheromone. Nature 433:541545 (2012) Comparative metabolomics reveals
2. Butcher RA, Fujita M, Schroeder FC et al biogenesis of ascarosides, a modular library of
(2007) Small-molecule pheromones that con- small-molecule signals in C. elegans. J Am
trol dauer development in Caenorhabditis ele- Chem Soc 134:18171824
gans. Nat Chem Biol 3:420422 8. Srinivasan J, von Reuss SH, Bose N et al (2012)
3. Butcher RA, Ragains JR, Kim E et al (2008) A A modular library of small molecule signals
potent dauer pheromone component in C. regulates social behaviors in Caenorhabditis
elegans that acts synergistically with other elegans. PLoS Biol 10:e1001237
components. Proc Natl Acad Sci USA 9. Artyukhin AB, Yim JJ, Srinivasan J et al (2013)
105:1428814292 Succinylated octopamine ascarosides and a
4. Srinivasan J, Kaplan F, Ajredini R et al (2008) new pathway of biogenic amine metabolism in
A blend of small molecules regulates both mat- C. elegans. J Biol Chem 288:1877818783
ing and development in Caenorhabditis ele- 10. Butcher RA, Ragains JR, Li W et al (2009)
gans. Nature 454:11151118 Biosynthesis of the Caenorhabditis elegans
5. Pungaliya C, Srinivasan J, Fox BW et al (2009) dauer pheromone. Proc Natl Acad Sci USA
A shortcut to identifying small molecule sig- 106:18751879
nals that regulate behavior and development in 11. Kaplan F, Srinivasan J, Mahanti P et al (2011)
Caenorhabditis elegans. Proc Natl Acad Sci Ascaroside expression in Caenorhabditis
USA 106(19):77087713 elegans is strongly dependent on diet and
6. Butcher RA, Ragains JR, Clardy J (2009) An developmental stage. PLoS One 6:
indole-containing dauer pheromone component e17804
 Part II

Approaches to Characterize Gene and Function

of Pheromone Receptors
 Chapter 7

Identification of Chemosensory Receptor Genes

from Vertebrate Genomes
Yoshihito Niimura

Abstract
Chemical senses are essential for the survival of animals. In vertebrates, mainly three different types of
receptors, olfactory receptors (ORs), vomeronasal receptors type 1 (V1Rs), and vomeronasal receptors
type 2 (V2Rs), are responsible for the detection of chemicals in the environment. Mouse or rat genomes
contain >1,000 OR genes, forming the largest multigene family in vertebrates, and have >100 V1R and
V2R genes as well. Recent advancement in genome sequencing enabled us to computationally identify
nearly complete repertories of OR, V1R, and V2R genes from various organisms, revealing that the num-
bers of these genes are highly variable among different organisms depending on each species living envi-
ronment. Here I would explain bioinformatic methods to identify the entire repertoires of OR, V1R, and
V2R genes from vertebrate genome sequences.

Key words Olfactory receptor, Vomeronasal receptor, Multigene family, Bioinformatics, Vertebrate,
G protein-coupled receptor

1 Introduction

Chemical senses are essential for the survival of most animals. In

vertebrates, chemical molecules in the environment are mainly
detected by three different types chemosensory receptors, named
olfactory receptors (ORs), vomeronasal receptors type 1 (V1Rs),
and vomeronasal receptors type 2 (V2Rs) (for review, Ref. 1). All of
them are G protein-coupled receptors (GPCRs), membrane
proteins having seven transmembrane (TM) -helical regions
(Fig. 1). Each type of receptors is encoded by a multigene family.
The three gene families share almost no sequence similarity, though
their molecular structures are similar to one another.
Among the three gene families, the OR gene family is by far
the largest (for review, Ref. 2). OR genes are predominantly
expressed in the olfactory epithelium of the nasal cavity and were
first identified from rats by Linda Buck and Richard Axel in 1991 [3].
The genomes of many mammalian species harbor ~1,000 or more

Kazushige Touhara (ed.), Pheromone Signaling: Methods and Protocols, Methods in Molecular Biology, vol. 1068,
DOI 10.1007/978-1-62703-619-1_7, Springer Science+Business Media, LLC 2013

95
96 Yoshihito Niimura

OR V1R V2R
N

N N

C C C

Extracellular domain TM domain

Single exon Single exon Multiple exons

Fig. 1 Structures of ORs, V1Rs, and V2Rs. Membrane topologies and exonintron
structures of genes are shown

OR genes, which comprise 45 % of their proteomes. An OR is

~310 amino acids long on average, and typically the gene does not
have any introns in its coding region. OR genes are present in all
vertebrate species. Fishes have much smaller repertoires of OR
genes (~100) than mammals. However, despite smaller OR gene
repertoires in fishes, OR gene sequences in fishes are more diverse
than those in mammals [4, 5]. Amphibians and fishes have some
additional groups of OR genes that are not present in mammals.
V1R and V2R genes were discovered in 1995 [6] and 1997 [7
9], respectively. They are expressed in vomeronasal organs in mam-
mals and are involved in pheromone detection. Fishes do not have a
discrete vomeronasal organ, but they do have both V1R and V2R
genes, which are expressed in olfactory epithelium in fishes [10, 11].
V1R genes are intronless like OR genes, whereas V2R genes consist
of multiple exons (Fig. 1). A V2R gene is characterized by a long
N-terminal extracellular domain, which is encoded by several exons,
while the TM domain typically corresponds to a single exon.
Recently the whole genome sequences became available from
diverse organisms, which enabled us to identify nearly complete
repertoires of chemosensory receptor genes in a given species.
Comparative genomic studies revealed that the repertoires of che-
mosensory receptor genes are highly variable among different
species (Refs. 4, 5, 1214 for OR genes, Refs. 10, 1518 for V1R
genes, and Refs. 17, 19, 20 for V2R genes). In this chapter,
I describe the methods to identify OR, V1R, and V2R genes from
vertebrate genomes. OR and V1R genes do not have any introns;
therefore, detection of these genes from genome sequences is
relatively easy. However, because the numbers of the genes are
huge, development of computation methods is requisite for the
identification of the entire gene repertoires. On the other hand,
identification of V2R genes is more difficult due to their complex
gene structures.
 Identification of Chemosensory Receptor Genes from Vertebrate Genomes 97

2 Materials

1. The following computer programs need to be installed: BLAST

(http://blast.ncbi.nlm.nih.gov/; Ref. 21) for homology
searches, MAFFT (http://mafft.cbrc.jp/alignment/soft-
ware/; Ref. 22) for constructing multiple alignments, and
LINTREE (http://www.personal.psu.edu/nxm2/software.
htm; Ref. 23) for phylogenetic tree construction. Moreover,
the HMMER package (http://hmmer.janelia.org; Ref. 24)
and the Wise2 package (http://www.ebi.ac.uk/Tools/
Wise2/; Ref. 25) are used for the identification of V2R genes.
2. The genome sequences of various vertebrate species can be
downloaded from the Web sites of the University of California
Santa Cruz (http://genome.ucsc.edu), Ensembl Genome
Browser (http://ensembl.org), the Genome Sequencing
Center at Washington University School of Medicine (http://
genome.wustl.edu), the Broad Institute (http://www.
broadinstitute.org), etc.

3 Methods

Here I would explain the method for identifying OR genes in some

detail. Figure 2 illustrates the flowchart of OR gene identification.
OR genes are classified into three categories: intact genes, trun-
cated genes, and pseudogenes. An intact gene putatively encodes a
functional OR. A pseudogene is defined as a sequence containing
interrupting stop codons, frameshifts, and/or gaps within con-
served regions. A truncated gene is a partial intact sequence encod-
ing a part of OR and is located at the contig end. It is either a
functional gene or a pseudogene. Therefore, the fraction of pseu-
dogenes in a given species is overestimated if only intact genes are
considered to be functional. This effect is critical for the species
with low-coverage genome data, which contain many short contigs
[5, 12, 14]. For this reason, some caution is necessary to estimate
the fraction of pseudogenes.
The methods to identify V1R and V2R genes were described
in Refs. 15, 16 and Refs. 19, 20, 26, respectively. The methods
explained here are based on these articles with some modification.

3.1 Identification 1. Conduct TBLASTN searches [21] with a cutoff E-value of

of OR Genes 1e5 against a given genome sequence using known OR genes
as queries (see Note 1). Use the option -F F (filtering
low-complexity regions is not used) (see Note 2).
2. Because multiple sequences are used as queries, a number of
different queries may hit the same genomic region. In such a
case, choose a BLAST hit with the lowest E-value (called a
98 Yoshihito Niimura

Genome sequence

Query: Representative Query:

Intact OR genes all intact
TBLASTN E < 1e-5 ORs TBLASTN E < 1e-20

Blast best-hits Blast best-hits

1
6
<250 aa?
Intact OR?
Extension of a best-hit
Extract an intact coding sequence No No
(ATG ~ stop codon)
2 7
Longest coding Interrupting stop
sequence is <250 aa? codons or frameshifts?
Yes
Multiple alignment by MAFFT No No
3
Long (>5 aa) gap 8
within a TM? Close (<30bp) to
the contig end? No
Assignment of a proper initiation codon
Phylogenetic tree construction No - Extension of a best-hit
with non-OR genes by LINTREE - Assignment of a proper initiation
4 Yes codon for C-missing sequences
Form a clade with - Multiple alignment by MAFFT
ORs? with intact OR genes
Multiple alignment by MAFFT Yes 9
Intact N- or C-
5 terminal portion?
Gaps within TMs or No
conserved regions?
Yes Visual inspection
Visual inspection No

Intact OR genes Truncated OR genes OR pseudogenes

Fig. 2 Flowchart for the identification of intact OR genes, truncated OR genes, and OR pseudogenes from
vertebrate genome sequences

best-hit) among all BLAST hits to a given genomic region.

Extract all best-hits from the genome sequence.
3. To identify intact OR genes from the best-hit sequences, several
criteria are used. First, discard the best-hit sequences shorter
than 250 amino acids (criterion 1 in Fig. 2; see Note 3).
4. For each of the best-hit sequences remaining after step 3,
extend it to both directions along the genome sequence. Then,
extract the longest coding sequence from an ATG codon to a
stop codon (Fig. 3a). If the length of the sequence after exten-
sion is less than 250 amino acids, discard it (criterion 2 in
Fig. 2). If the extended sequence contains undetermined
nucleotides, the sequence should also be discarded.
5. Construct a multiple alignment from the remaining sequences
after step 4 by using the program MAFFT [22] (see Note 4),
and assign the positions of seven TM regions according to
Ref. 27.
Identification of Chemosensory Receptor Genes from Vertebrate Genomes 99

a
Genome sequence
TAA
ATG TAG
TGA
Best-hit

b C-missing <30 bp

ATG
Best-hit

N-missing <30 bp
TAA
TAG
TGA
Best-hit

<30 bp <30 bp
NC-missing

Best-hit

Fig. 3 (a) Extension of a best-hit along the genome sequence to take the longest
coding sequence. (b) Extension of a best-hit that is located near the end of a
contig. C-missing, N-missing, and NC-missing sequences are shown separately

6. If a sequence has a gap of five or more amino acids within TM

regions, exclude it (criterion 3 in Fig. 2).
7. Construct a multiple alignment once again from the remain-
ing sequences after step 6 by using MAFFT [22] (see Note 4).
Then choose the most proper ATG codon as the initiation
codon in case that a sequence examined contains two or more
ATG codons in the N-terminal tail region (the upstream of the
first TM region; see Note 5).
8. Construct a neighbor-joining phylogenetic tree [28] using the
remaining sequences after step 7 together with several non-
OR GPCR genes as the outgroup (see Note 6). Run the pro-
gram njboot in LINTREE [23] with the option -d28
(Poisson correction distance) and -b500 (bootstrap resam-
plings for 500 times).
9. Eliminate non-OR genes on the basis of the phylogenetic tree
constructed in step 8 (criterion 4 in Fig. 2; see Note 7). When
a given sequence is located out of the OR gene clade in the
phylogenetic tree, it should be discarded (see Note 8).
10. Construct a multiple alignment from the remaining sequences
after step 9 by MAFFT [22] (see Note 4), and eliminate the
sequences having gaps within TM regions or at other conserva-
100 Yoshihito Niimura

tive amino acid sites by visual inspection (criterion 5 in Fig. 2).

The remaining sequences are regarded as intact OR genes.
11. To identify non-intact OR genes, perform TBLASTN searches
[21] against the genome sequence using all intact OR genes
identified in step 10 as queries with the E-value below 1e20
(see Note 9).
12. Extract all best-hit sequences in the same way as step 2.
13. Exclude all of the intact OR genes obtained in step 10 (crite-
rion 6 in Fig. 2). All remaining sequences are regarded to be
truncated genes or pseudogenes.
14. Extract the best-hit sequences that meet both of the following
conditions. (1) There are no interrupting stop codons and
frameshifts (criterion 7 in Fig. 2). (2) The distance between
the end of the sequence and the end of the contig containing
the sequence is less than 30 bp (criterion 8 in Fig. 2).
15. Classify the remaining sequences after step 14 into three cat-
egory, C-missing, N-missing, and NC-missing (Fig. 3b). For a
C-missing sequence, the upstream of the best-hit is present in
the contig examined, whereas its downstream is missing.
Conversely, for an N-missing sequence, its downstream is pres-
ent in the contig, while its upstream is missing. An NC-missing
sequence lacks both upstream and downstream portions (see
Note 10).
16. For a C-missing sequence, extend it along the genome
sequence and extract a sequence from the most upstream ATG
codon to the most downstream in-frame codon without any
interrupting stop codons (Fig. 3b, top).
17. Construct a multiple alignment using the extended C-missing
sequences obtained in step 16 together with some representa-
tive intact OR genes by MAFFT [22], and choose the most
proper ATG codon as the initiation codon for each sequence in
the same manner as step 7 and Note 5.
18. For an N-missing sequence, extend it and extract a sequence
from the most upstream in-frame codon to the stop codon
(Fig. 3b, middle).
19. For an NC-missing sequence, extend it and extract the longest
sequence from the most upstream in-frame codon to the most
downstream one (Fig. 3b, bottom).
20. Construct a multiple alignment using all of the extended
C-missing, N-missing, and NC-missing sequences obtained in
steps 1719 together with some representative intact OR
genes by MAFFT [22].
21. Exclude the sequences that contain gaps within TM regions
or at other conservative amino acid sites by visual inspection
 Identification of Chemosensory Receptor Genes from Vertebrate Genomes 101

(criterion 9 in Fig. 2). The remaining sequences are regarded

to be truncated genes.
22. Exclude all truncated genes from the best-hit sequences
obtained in step 13. All remaining sequences are regarded as
OR pseudogenes.

3.2 Identification 1. Conduct TBLASTN searches [21] with a cutoff E-value of

of V1R Genes 1e5 against a given genome sequence using known V1R
genes as queries. The following options should be used: -F F
for that filtering low-complexity regions is not used and -v
1000 -b 1000 for the number of hits reported.
2. From the results obtained in step 1, extract all best-hit
sequences in the same manner as Subheading 3.1, step 2.
3. For each of the best-hit sequences, conduct BLASTP searches
[21] against the nr database of GenBank to ensure that the
best hit is a V1R gene. Discard the sequences showing higher
similarity to non-V1R genes (see Note 11).
4. Among the remaining sequences after step 3, extract sequences
that do not contain any interrupting stop codons or frame-
shifts. For each of these sequences, extend it to both directions
along the genome sequence and extract the longest coding
sequence from an ATG codon to a stop codon.
5. Construct a multiple alignment from the sequences obtained
in step 4 by MAFFT [22] and choose the most proper initia-
tion codon from each sequence.
6. Assign the location of TM regions in the multiple alignment.
If a sequence contains gaps within TM regions or other highly
conserved regions, it should be regarded as a pseudogene. The
remaining sequences are regarded to be intact V1R genes.
7. Exclude all intact V1R genes (step 6) from the sequences
obtained after step 3. The remaining sequences are regarded
as V1R pseudogenes.

3.3 Identification 1. The first step is to perform TBLASTN searches [21] against a
of V2R Genes given genome sequence using known V2R genes as queries. To
this end, construct a multiple alignment by MAFFT [22] from
the known V2R query sequences (see Note 12).
2. Trim the multiple alignment to extract the TM domain (from
the first TM region to the C-terminal end) according to Ref. 8.
The TM domain in each V2R gene is used as a query sequence
of TBLASTN searches (see Note 13).
3. Conduct TBLASTN searches [21] with a cutoff E-value of
1e5 using the TM domain of known V2R genes as queries.
The following options should be used: -F F for that filtering
low-complexity regions is not used and -v 1000 -b 1000 for
the number of hits reported.
102 Yoshihito Niimura

4. From the results obtained in step 3, extract all best-hit

sequences in the same manner as Subheading 3.1, step 2.
5. For each of the best-hit sequences, extract the genomic
sequence together with 200 kb upstream and 1 kb downstream
regions (see Note 14).
6. Construct a profile Hidden Markov Model (HMM) from the
alignment generated in step 1. (Here use the alignment includ-
ing the entire coding region rather than a TM domain). Run
the program hmmbuild in the HMMER package, version
2.3.2 [24] for the MAFFT output file with options --fast
--gapmax 0.95 -s [20]. Then run the hmmcalibrate program
subsequently.
7. Align each of the elongated best-hit sequences obtained in
step 5 with a profile HMM created in step 6 by the program
genewisedb in the Wise2 package [25]. Use the following
options: -splice flat -cut 20 -alg 623 -aalg 623 -pretty -para
-pseudo -genes -sum -cdna -trans -gff -gener [20].
8. When two neighboring elongated best-hit sequences (from
step 5) are overlapped, the same genomic region may be
detected as a result of the genewisedb searches. In such cases,
extract only one genewisedb hit showing the highest score
among all overlapping hits.
9. For each of the genewisedb hits obtained in step 8, conduct
BLASTP searches [21] against the nr database of GenBank to
ensure that the best hit is a V2R gene. Discard the sequences
showing higher similarity to non-V2R genes (see Note 15).
All remaining sequences are regarded to be V2R genes.
10. Discard the sequences (1) that contain interrupting stop
codons or frameshifts, (2) that have gaps within TM regions or
other highly conserved regions, and (3) that are shorter than
750 amino acids. The remaining sequences are regarded to be
intact V2R genes.
11. Exclude all intact V2R genes (step 10) from the sequences
obtained after step 9. The remaining sequences are regarded
as V2R pseudogenes.

4 Notes

1. As for query sequences, an OR gene set with sufficient sequence

diversity should be used to retrieve all putative OR genes from
the genome. However, currently thousands of OR genes are
available in the databases; therefore, to reduce a computational
time, highly similar sequences should be eliminated from the
queries. To this aim, classify candidate query genes into groups
using a given sequence similarity cutoff, e.g., 50 % amino acid
Identification of Chemosensory Receptor Genes from Vertebrate Genomes 103

identity, and choose one representative sequence from each

group. To examine amniote (mammalian, avian, and reptilian)
genomes, human and/or mouse OR genes can be used as que-
ries. On the other hand, to investigate amphibian or fish
genomes, OR genes from fishes (e.g., zebrafish) and/or frogs
should be used as queries, because amphibian and fish OR
genes are more diverse than amniote OR genes [5]. OR gene
sequences in mammals and other vertebrates can be obtained
from Refs. 12, 14, 29 and Ref. 5, respectively.
2. If the number of BLAST hits is expected to large, e.g., when
mammalian genome sequences are examined, -v and -b
options should also be used to change the number of hits and
alignments shown.
3. The cutoff length of 250 amino acids is sufficiently shorter
than that of any known functional OR genes.
4. When the number of sequences is large (e.g., >400), it is better
to separate them into several groups to reduce a computational
time.
5. For most of the known functional OR genes in mammals, the
length of the N-terminal tail is between 21 and 34 amino acids.
Therefore, to choose the initiation codon, the following crite-
ria can be used. If an ATG codon is present in the region
(named region A) between the positions -34 and -21 (here
the amino acid position is indicated as the relative position to
the boundary between the N-terminal tail and the first TM
region), choose the one as the initiation codon. In case that
two or more ATG codons are present within the region A,
choose the most downstream one among them. If ATG codons
are not present in the region A, choose the closest one to the
region A.
6. The following genes can be used as the outgroup sequences:
alpha-1A-adrenergic receptor isoform 1 (GenBank protein id,
NP_000671), beta-1-adrenergic receptor (NP_000675), ade-
nosine A2b receptor (NP_000667), histamine receptor H2
(NP_071640), 5-hydroxytryptamine (serotonin) receptor 1B
(NP_000854), 5-hydroxytryptamine (serotonin) receptor 1F
(NP_000857), 5-hydroxytryptamine (serotonin) receptor 6
(NP_000862), galanin receptor 1 (NP_001471), and soma-
tostatin receptor 4 (NP_001043). These genes are relatively
close to OR genes among the genes belonging to the
rhodopsin-like GPCR superfamily [30].
7. When the number of sequences is large (e.g., >200), it is better to
separate them into several groups to reduce a computational time.
8. In a phylogenetic tree, OR genes form a monophyletic clade
with a high bootstrap support [5]. Therefore, non-OR genes
are easily distinguishable from OR genes.
104 Yoshihito Niimura

9. The reason for using the cutoff E-value of 1e20 is as follows.

First, the E-value of a best-hit to the genomic region corre-
sponding to a non-OR gene is 1e18 or larger. Second, all
best-hit sequences with the E-value below 1e20 obtained by
OR gene queries are more similar to OR genes than to any
known non-OR genes. Therefore, non-intact OR genes can be
distinguished from non-OR genes by using the cutoff E-value
of 1e20.
10. Note that NC-missing sequences are found only on a contig
shorter than the length of an OR gene (~930 bp).
11. This step is necessary to exclude some other receptors (e.g.,
T2R taste receptors) that are homologous to V1Rs.
12. A V2R gene set with sufficient sequence diversity should be
used as query sequences. These sequences are also used to con-
struct a profile HMM (see step 6). To search for mammalian
V2R genes, for example, 75 intact V2R genes in mice with the
prefix mouseMay04V2R in Ref. 20 can be used as queries.
Fish V2R gene sequences are available in Ref. 19.
13. The N-terminal extracellular domain shows a higher extent of
sequence diversity [26]; therefore, if N-terminal extracellular
domain is used as a query, homology searches give a large
number of non-V2R hits. For this reason, it is better to use
only a TM domain rather than the entire sequence of a V2R
gene as a query.
14. For extracting the entire coding exons of a V2R gene, it is
necessary to examine a long upstream genomic sequence,
because the N-terminal extracellular domain is encoded by
multiple exons (Fig. 1). (Note that a query sequence for
homology searches contains only a TM domain.) The reason
for using the 200 kb limit is that it is longer than the genomic
extent of all previously described V2R genes [20].
15. This step is necessary, because some other receptors (e.g., cal-
cium-sensing receptors and T1R taste receptors) are known to
be homologous to V2Rs.

Acknowledgments

This work was supported by grant (20770192 and 23770271)

from the Ministry of Education, Culture, Sports, Science, and
Technology, Japan.
 Identification of Chemosensory Receptor Genes from Vertebrate Genomes 105

References
1. Nei M, Niimura Y, Nozawa M (2008) The Amplification in rodents, decimation in pri-
evolution of animal chemosensory receptor mates, and a surprisingly small repertoire in
gene repertoires: roles of chance and necessity. dogs. Genome Res 15:231240
Nat Rev Genet 9:951963 17. Shi P, Zhang J (2007) Comparative genomic
2. Niimura Y (2012) Olfactory receptor multi- analysis identifies an evolutionary shift of vom-
gene family in vertebrates: from the viewpoint eronasal receptor gene repertoires in the verte-
of evolutionary genomics. Curr Genomics brate transition from water to land. Genome
13:103114 Res 17:166174
3. Buck L, Axel R (1991) A novel multigene fam- 18. Young JM, Massa HF, Hsu L et al (2010)
ily may encode odorant receptors: a molecular Extreme variability among mammalian V1R
basis for odor recognition. Cell 65:175187 gene families. Genome Res 20:1018
4. Niimura Y, Nei M (2005) Evolutionary dynam- 19. Hashiguchi Y, Nishida M (2006) Evolution and
ics of olfactory receptor genes in fishes and tetra- origin of vomeronasal-type odorant receptor
pods. Proc Natl Acad Sci USA 102:60396044 gene repertoire in fishes. BMC Evol Biol 6:76
5. Niimura Y (2009) On the origin and evolution 20. Young JM, Trask BJ (2007) V2R gene families
of vertebrate olfactory receptor genes: degenerated in primates, dog and cow, but
Comparative genome analysis among 23 chor- expanded in opossum. Trends Genet
date species. Genome Biol Evol 1:3444 23:212215
6. Dulac C, Axel R (1995) A novel family of genes 21. Altschul SF, Madden TL, Schaffer AA et al
encoding putative pheromone receptors in (1997) Gapped BLAST and PSI-BLAST: a
mammals. Cell 83:195206 new generation of protein database search pro-
7. Herrada G, Dulac C (1997) A novel family of grams. Nucleic Acids Res 25:33893402
putative pheromone receptors in mammals 22. Katoh K, Kuma K, Toh H et al (2005) MAFFT
with a topographically organized and sexually version 5: improvement in accuracy of multiple
dimorphic distribution. Cell 90:763773 sequence alignment. Nucleic Acids Res
8. Matsunami H, Buck LB (1997) A multigene 33:511518
family encoding a diverse array of putative 23. Takezaki N, Rzhetsky A, Nei M (1995)
pheromone receptors in mammals. Cell Phylogenetic test of molecular clock and linear-
90:775784 ized trees. Mol Biol Evol 12:823833
9. Ryba NJ, Tirindelli R (1997) A new multigene 24. Eddy SR (2011) Accelerated Profile HMM
family of putative pheromone receptors. Searches. PLoS Comput Biol 7:e1002195
Neuron 19:371379 25. Birney E, Clamp M, Durbin R (2004)
10. Saraiva LR, Korsching SI (2007) A novel olfac- GeneWise and Genomewise. Genome Res
tory receptor gene family in teleost fish. 14:988995
Genome Res 17:14481457 26. Yang H, Shi P, Zhang YP et al (2005)
11. Hashiguchi Y, Nishida M (2005) Evolution of Composition and evolution of the V2r vom-
vomeronasal-type odorant receptor genes in eronasal receptor gene repertoire in mice and
the zebrafish genome. Gene 362:1928 rats. Genomics 86:306315
12. Niimura Y, Nei M (2007) Extensive gains and 27. Man O, Gilad Y, Lancet D (2004) Prediction
losses of olfactory receptor genes in mamma- of the odorant binding site of olfactory recep-
lian evolution. PLoS One 2:e708 tor proteins by human-mouse comparisons.
13. Hayden S, Bekaert M, Crider TA et al (2010) Protein Sci 13:240254
Ecological adaptation determines functional 28. Saitou N, Nei M (1987) The neighbor-join-
mammalian olfactory subgenomes. Genome ing method: a new method for reconstruct-
Res 20:19 ing phylogenetic trees. Mol Biol Evol 4:
14. Matsui A, Go Y, Niimura Y (2010) 406425
Degeneration of olfactory receptor gene reper- 29. Go Y, Niimura Y (2008) Similar numbers but
tories in primates: No direct link to full trichro- different repertoires of olfactory receptor genes
matic vision. Mol Biol Evol 27:11921200 in humans and chimpanzees. Mol Biol Evol
15. Grus WE, Shi P, Zhang YP et al (2005) 25:18971907
Dramatic variation of the vomeronasal phero- 30. Fredriksson R, Lagerstrm MC, Lundin LG
mone receptor gene repertoire among five et al (2003) The G-protein-coupled recep-
orders of placental and marsupial mammals. tors in the human genome form five main
Proc Natl Acad Sci USA 102:57675772 families. Phylogenetic analysis, paralogon
16. Young JM, Kambere M, Trask BJ et al (2005) groups, and fingerprints. Mol Pharmacol 63:
Divergent V1R repertoires in five species: 12561272
 Chapter 8

Functional Assays for Insect Olfactory Receptors

in Xenopus Oocytes
Tatsuro Nakagawa and Kazushige Touhara

Abstract
Reconstitution of olfactory or pheromone receptors in heterologous expression systems greatly facilitates
the functional analysis of these receptors. Xenopus laevis oocytes can be used to efficiently express insect
olfactory or pheromone receptors. In this chapter, we describe how to use Xenopus laevis oocytes for
functional assays of insect olfactory receptors. The procedure can be subdivided into the four following
steps: (1) in vitro complementary RNA (cRNA) synthesis, (2) isolation of oocytes from female Xenopus
laevis, (3) cRNA microinjection into oocytes, and (4) two-electrode voltage-clamp recording. This system
can be used to identify odor or pheromone ligands and to analyze structurefunction relationships involving
receptor proteins of interest.

Key words Xenopus laevis oocyte, In vitro cRNA transcription, Heterologous expression,
Electrophysiology

1 Introduction

To understand the function of olfactory or pheromone receptors in

insects, it is necessary to identify cognate ligands, analyze signal
transduction mechanisms, and examine structurefunction rela-
tionships within the receptor proteins. Currently, there are two
types of expression systemsin vivo and in vitrofor recording
the responses of olfactory or pheromone receptors in insects. The
in vivo recording systems, such as electroantennograms, single
sensillum recordings, and antennal lobe imaging techniques
(see Chapters 11 and 12), have been used to accumulate vast
amounts of knowledge as to physiological roles of some pheromone
compounds and some olfactory or pheromone receptors [13].
However, these in vivo systems are not suitable for exhaustive, effi-
cient, and high-throughput analysis of many insect olfactory or
pheromone receptors because it is difficult to generate a variety of
mutants quickly. Alternative in vitro recording systems would be
invaluable for overcoming such situations because in vitro systems

Kazushige Touhara (ed.), Pheromone Signaling: Methods and Protocols, Methods in Molecular Biology, vol. 1068,
DOI 10.1007/978-1-62703-619-1_8, Springer Science+Business Media, LLC 2013

107
108 Tatsuro Nakagawa and Kazushige Touhara

could make it much easier and faster to analyze the function of

olfactory or pheromone receptors.
Xenopus laevis oocyte has long been used as an excellent
expression system for functional analyses of myriad G-protein
coupled receptors (GPCRs) and ion channels. In functional assay
using Xenopus oocytes, which is called two-electrode voltage-
clamp recording (TEVC), the activity of the expressed receptor is
detected as changes in membrane current. Xenopus oocytes have
two major advantages to analyze the function of membrane pro-
teins; first, oocytes have an abundance of ribosomal RNAs and
transfer RNAs and can, therefore, express exogenously introduced
mRNA as functional receptor proteins; second, oocytes possess
many kinds of endogenous signal transduction components, such
as G-proteins (Gq, Gs, and Gi/o), adenylyl cyclase, phospho-
lipace C, and Ca2+-activated Cl channels, and therefore, the
responses of GPCRs, as well as those of ion channels, can be
detected by co-expressing some ion channels that are activated
downstream of the signal transduction cascade of the GPCRs.
Many researchers have put their efforts into reconstituting insect
olfactory or pheromone receptors in Xenopus oocytes. In 2001,
Wetzel and colleagues succeeded, for the first time, in detecting odor
responses of Drosophila Or43a expressed in Xenopus oocytes [4].
After that, Nakagawa and colleagues found that co-expression of
BmOrco (the Orco family receptor in Bombyx mori) with either
BmOr-1 (the bombykol receptor) or BmOr-3 (the bombykal recep-
tor) in oocytes greatly enhances the sensitivities to the respective
BmOr-1 or BmOr-3 cognate ligand [5]. This result has been shown
to be applicable to other combinations of Or and the Orco family
receptors, and to other heterologous expression systems as well
[611]. These findings provided invaluable methods for efficient
analysis of the function of many insect olfactory and pheromone
receptors in vitro and lead a rapid increase in our knowledge of the
function of olfactory and pheromone receptors in insects. For instance,
it was shown that insect Or and Orco proteins form heteromeric
receptors of unknown stoichiometry [12]. And more recently, several
groups have shown that the insect OrOrco complexes function as
odor- or pheromone-activated nonselective cation channels [810].
Furthermore, some site-directed mutagenesis studies using Xenopus
oocytes provided important insights into the functional domains
within insect OrOrco complexes, specifically the domains that are
involved in ligand binding and in ion permeation properties [13, 14].
Method for expressing insect olfactory or pheromone receptors
in Xenopus oocytes is now firmly established. This method can be
subdivided into the following four steps: (1) in vitro transcription of
cRNAs of a canonical Or and a Orco protein, (2) isolation of oocytes
from female Xenopus laevis, (3) microinjection of the synthesized
cRNAs into oocytes, (4) electrophysiological recording. In this sec-
tion we will describe a detailed protocol for performing functional
analysis of an OrOrco complex of interest in Xenopus oocytes.
 Functional Assays for Insect Olfactory Receptors in Xenopus Oocytes 109

2 Materials

2.1 In Vitro cRNA 1. cDNA clones encoding Ors in a Xenopus oocyte expression
Synthesis vector (see Note 1). Store these reagents at 20 C.
2. RNA synthesis solution (for 1 sample) (see Note 2). All the
reagents should be stored at 20 C.
For RNA synthesis reactions using T7 polymerase, please
use the following volumes and concentrations for each reagent
listed in each reaction mixture: 46.5 l of DEPC water, 20 l
of 5 transcription optimized buffer (40 mM TrisHCl, pH
7.9, 6 mM MgCl2, 2 mM spermidine, and 10 mM NaCl), 5 l
of ATP (10 mM), 2.5 l of GTP (10 mM), 5 l of CTP
(10 mM), 5 l of UTP (10 mM), 5 l of DTT (0.1 M), 5 l of
m7G(5)ppp(5)G RNA Capping Analog (10 mM), 3 l
of Ribonuclease inhibitor, and 3 l of T7 polymerase (1020
units/l).
For RNA synthesis reactions using SP6 polymerase, please
use the following volumes and concentrations for each reagent
listed in each reaction mixture: 41.5 l of DEPC water, 20 l
of 5 transcription optimized buffer (40 mM TrisHCl,
pH 7.9, 6 mM MgCl2, 2 mM spermidine and 10 mM NaCl),
5 l of ATP (10 mM), 2.5 l of GTP (10 mM), 5 l of CTP
(10 mM), 5 l of UTP (10 mM), 10 l of DTT (10 mM), 5 l
of m7G(5)ppp(5)G RNA Capping Analog (10 mM), 3 l of
Ribonuclease inhibitor, and 3 l of SP6 polymerase (1020
units/l).
3. RNase-Free DNase (1 unit/l), 10 reaction buffer (400 mM
TrisHCl, pH 8.0, 100 mM MgSO4 and 10 mM CaCl2). Store
at 20 C.
4. RNA electrophoresis buffer: (Percentages of total volume)
77 % DEPC water, 18 % formaldehyde, 5 % 20 MOPS (0.4
M MOPS, 160 mM NaOAc, 20 mM EDTA, pH 7.0). Store
at 4 C.
5. 1 % denaturing agarose gel: (for 100 ml volume) 1 g agarose,
77 ml DEPC water, 5 ml 20 MOPS (0.4 M MOPS, 100 mM
CH3COONa, 20 mM EDTA, pH7.0 with NaOH), 18 ml
formaldehyde (see Note 3).
6. RNase-free phenolchloroformisoamyl alcohol. Store at 4 C.
7. RNase-free chloroform.
8. Ethanol.
9. RNase-free 3 M sodium acetate (NaOAc), pH 5.2
10. RNase-free 4 M LiCl.
11. UltraPure DNase/RNase-free distilled water.
110 Tatsuro Nakagawa and Kazushige Touhara

12. RNA denaturing buffer: for 1 sample, use 5 l formamide,

1.75 l formaldehyde, 0.5 l 20 MOPS (0.4 M MOPS,
100 mM CH3COONa, 20 mM EDTA, pH 7.0 with NaOH),
10 loading buffer, 0.25 l EtBr (10 mg/ml), and 0.5 l
DEPC water.

2.2 Isolation of As Xenopus oocytes are susceptible to bacterial contamination, all

Oocytes from Female the implements, solutions, and buffersexcept for the anesthesia
Xenopus laevis (Fig. 1) must be sterilized prior to each experiment.
1. Female Xenopus laevis. Keep at 16 C.
2. Anesthesia: (in 200 ml water) 0.3 % 3-aminobenzoic acid ethyl
ester, 0.1 % NaHCO3.
3. MBSH buffer: 88 mM NaCl, 1 mM KCl, 0.33 mM
Ca(NO3)2 4H2O, 0.41 mM CaCl2 2H2O, 0.82 mM
MgSO4 7H2O, 2.4 mM NaHCO3, 10 mM HEPES, 60 g/l
penicillin, and 50 g/l streptomycin (see Note 4). Store at
16 C.

tear oocyte clusters into small pieces

by tweezers

oocyte cluster

Xenopus laevis (

Wash by OR2 buffer Gently rotate the vial

and MBSH for 1.5-2 hours at 18C

OR2 buffer containing

2 mg/ml collagenase

Fig. 1 A flow diagram for the isolation oocytes from female Xenopus laevis. Clusters of oocytes were taken
from the stomach of each anesthetized frog. Tweezers were used to tear each cluster into small pieces. The
follicle cells surrounding the oocytes were digested by treating oocyte clusters with 2 mg/ml collagenase
diluted in Ca2+-free buffer (OR2 buffer) for 12 h at 16 C. Finally, the collagenase solution was washed away
with OR2 buffer and then with Ca2+-containing MBSH. Isolated oocytes were then collected for injections
 Functional Assays for Insect Olfactory Receptors in Xenopus Oocytes 111

4. OR2 buffer: 82.5 mM NaCl, 2 mM KCl, 1 mM MgCl2 6H2O,

5 mM HEPES, pH 7.5 with NaOH. Store at 16 C.
5. Collagenase B (Roche Applied Science).
6. Surgical implements: scalpel, surgical needles, tweezers, and
sutures (see Note 5).
7. Sterile 10-cm dish.
8. Shaker for glass vials.

2.3 Microinjection 1. Standard dissecting scope equipped with a cold light source.
of cRNA into Oocytes 2. Injector mounted on a micromanipulator (for example;
(Fig. 2) NANOJECT II Variable Volume (2.369 nl) Automatic Injector
with Glass Capillaries (110 V), Drummond Scientific Company).
3. Glass capillaries (for example; 3 1/2 inch capillary (8.9 cm),
OD = 1.14 mm, ID = 0.53 mm, Drummond Scientific Company).
4. Sterile gauze.
5. Needle puller (for example; P-97 Flaming/Brown type micro-
pipette puller, Sutter Instruments).
6. Mineral oil.
7. Sterile Pasteur pipettes.

Microinjector

stereomicroscope micromanipulator control unit

aspiration filled with

cRNA solution Xenopus laevis
glass needle oocyte cRNA injection

cRNA solution

Fig. 2 A schematic drawing of the apparatus used for microinjection of cRNA into oocytes. The micropipette Puller
P-97 (SUTTER Ins.) was used to make the glass needles. A microinjector fitted with a glass needle filled with cRNA
solution was used to inject cRNA solution (50 nl) into each oocyte
112 Tatsuro Nakagawa and Kazushige Touhara

2.4 Two-Electrode 1. Standard dissecting scope equipped with a cold light source.
Voltage Clamp Method 2. Standard bath solution (see Note 6): 115 mM NaCl, 2.5 mM
(Fig. 3) KCl, 1.8 mM BaCl2, 10 mM HEPES, pH 7.2 with NaOH.
3. Oocyte Clamp Amplifier Model OC-725C (WARNER
instruments).

a Normal solution

Current recording

V command Odorants
+ Normal solution

Voltage electrode Current electrode

Ground
electrode Electric valves
the Or-Orco complex

A waste tank

b
None injected BmOr-1-BmOrco

bombykol bombykol

0.5 A

1 min 0.5 A 1 min

Fig. 3 Electrophysiological recordings from oocytes that express an insect OrOrco complex (two-electrode
voltage clamp method). (a) A schematic diagram of the apparatus used for two-electrode voltage clamp
recording. Ligands are applied via a superfusing bath solution, which is supplied via a silicon tube that is con-
nected to a computer-driven solenoid valve; alternatively, peristaltic pumps can be used to apply the ligand.
Membrane voltage is measured via the voltage electrode, and the recorded signal is compared to the com-
mand voltage and amplified through a feedback amplifier. The difference between the two voltages is then
converted into current and applied to the cell via the current electrode. (b) Representative current traces of
none-injected oocyte (left ) and BmOr-1-BmOrco expressing oocyte (right ) in response to bombykol. The
responses are measured as the changes of membrane current
 Functional Assays for Insect Olfactory Receptors in Xenopus Oocytes 113

4. Glass capillaries (For example; Capillary glass tubing with flame

polished ends Borosilicate-thin wall with filament, OD = 1.50 mm,
ID = 1.17 mm, Length = 10 cm, WARNER instruments).
5. Needle puller (for example; P-97 Flaming/Brown type micro-
pipette puller, Sutter Instruments).
6. Peristaltic pump (see Note 7).
7. Perfusion chamber (for example; Oocyte Recording Chamber,
WARNER instruments or a hand-made chamber.).
8. 3 M KCl solution.
9. A needle and a syringe for loading the KCl solution into the
glass capillaries.
10. Software for recording and analyzing electrophysiological data.

3 Methods

Carry out all procedures at room temperature unless otherwise

specified.

3.1 In Vitro cRNA Use gloves throughout this procedure.

Synthesis
1. Linearize the plasmid DNA (5 g) by digesting it for at least
2 h with a restriction enzyme specific for a site located on the
3 end of polyadenylation region of the plasmid vector (see
Note 8). The reaction volume should be 100200 l. Check
that the template is completely digested by electrophoresis
using 1 l of the digestion mix.
2. Purify the digested DNA either by phenol/chloroform/
isoamyl alcohol extraction or by using a DNA purification kit
(elute the DNA with 100 l of RNase-free water).
3. Ethanol precipitate the DNA for 25 min at 80 C using 10 l
of RNase-free 3 M NaOAc and 250 l of 100 % ethanol.
Centrifuge at 20,000 g for 25 min at 4 C, wash once with
200 l of 70 % ethanol, and remove supernatant. Spin the tube
briefly and collect the remaining ethanol from the bottom with
a pipette, and thus completely remove all ethanol from the
tube (see Note 9).
4. Resuspend the pellet with 100 l of RNA synthesis solution by
repeated pipetting. Incubate the mixture for 2 h at 37 C.
5. To digest template DNA, add 5 l of RNase-free DNase and
11 l of 10 reaction buffer, and incubate the mixture for
510 min at room temperature.
6. Purify synthesized RNA either by phenol/chloroform/iso-
amyl alcohol extraction or using a RNA purification kit (elute
with 100 l of RNase-free water).
114 Tatsuro Nakagawa and Kazushige Touhara

7. Ethanol precipitate the RNA for 25 min at 80 C in 10 l of

RNase-free 4 M LiCl and 250 l of 100 % ethanol. Centrifuge
at 20,000 g for 25 min at 4 C, wash once with 200 l of
70 % ethanol, and decant supernatant completely. Ethanol
should be removed completely by spinning the tube briefly and
collecting all the ethanol at the bottom with a pipette (see Note
10). Resuspend the pellet in 21 l of ultrapure-distilled RNase/
DNase free water.
8. Use 1 l of the RNA solution to check the amount and size of
the synthesized RNA via electrophoresis through a denaturing
agarose gel. Briefly, denature RNA by mixing 1 l of RNA
solution and 9 l of RNA denaturing buffer, incubate it at
70 C for 10 min, and put it on ice for 2 min. Load the mixture
onto a denaturing agarose gel, and perform electrophoresis
(see Note 11). Estimate RNA concentration by spectropho-
tometry, and make the final concentration of RNA 1 g/l.

3.2 Isolation of 1. Anesthetize an adult female Xenopus laevis by immersing it in

Oocytes from Female iced-anesthesia for 2030 min prior to surgery. After the frog
Xenopus laevis becomes unconscious, place the frog on its back on top of a
bed of crushed ice.
2. Make a 12 cm incision in the lower right (or left) abdominal
wall near right (or left) leg to expose an ovary.
3. Use tweezers to carefully remove about one third of one ovary;
place this tissue sample in OR2 buffer. Use tweezers or surgical
scissors to tear the oocyte clusters into small pieces (~ 0.5 cm
in diameter).
4. Suture the skin and the abdominal wall of the stomach of the
frog separately using thin silk sutures; let the frog recover from
anesthesia in shallow water (see Note 12).
5. Treat the oocytes with collagenase to remove follicular cells.
Add 5 ml of collagenase B-diluted OR2 solution (2 mg/ml
(see Note 13)) to 20 ml glass vials, and add approximately an
equal volume of ovary suspension that contains ovaries torn
into small pieces. Gently rotate the vial for 12 h at 16 C (see
Note 14). Remove the collagenase from the solution by wash-
ing four times in fresh OR2 buffer, and then by washing in
MBSH four times. Transfer the defolliculated oocytes into an
MBSH-filled 10-cm dish. Prior to microinjection, incubate the
dish at 16 C for a few hours or overnight.

3.3 Microinjection Use gloves throughout the procedure

of cRNA into Oocytes
1. Use a needle-puller (e.g., the P-97 puller) to pull microinjec-
tion needles (see Note 15).
2. Use tweezers to snap off the tips of the needles and to adjust
the size of the tips to 1030 m (see Note 16).
 Functional Assays for Insect Olfactory Receptors in Xenopus Oocytes 115

3. Backfill the needles with mineral oil before attaching the

needle to the injector by using 30 g 2 needle and a syringe.
4. Eject air out of the needle by pushing the empty button on
the injector pedal because air bubbles often affect the accuracy
of the injection volumes.
5. Place equal volumes (0.51.0 l) of canonical Or cRNA (1 g/
l) and of the Orco family cRNA (1 g/l) on a small piece of
Parafilm, and mix them by pipetting. Immerse the tip of the
needle into the drop of RNA solution and aspirate the solution
by pushing the fill button on the injector pedal (see Note
17).
6. Fill a 10-cm dish with MBSH, and immerse a small piece
(2 2 cm) of sterile gauze in the MBSH forcing it to the bot-
tom of the dish (see Note 18). Use a Pasteur pipette to transfer
2030 oocytes onto the gauze by.
7. For each injection, select a large oocyte (approximately 1 mm
diameter, the size of stage V or VI oocyte) that has lost the fol-
licle cell layer (see Notes 19 and 20). Gently inject 50 nl of the
RNA solution into each oocyte by piercing the membrane; an
oocyte will plump-up if injected correctly.
8. Incubate the injected oocytes in MBSH at 16 C for 34 days
(see Note 21).

3.4 Two-Electrode Before performing TEVC, precise setting-up of the measurement

Voltage Clamp Method system is necessary. This system is well established and has been
widely used to analyze various types of ion channels; therefore, we
will describe this method only very briefly here. Importantly, the
information provided here is insufficient for a reader who is
unfamiliar with the TEVC to successfully perform the technique;
more detailed instructions about how to set-up the system can be
found at and freely downloaded from a Web site maintained by
WARNER instruments (http://www.warneronline.com/).
1. Use a needle puller (e.g., the P-97 puller) to pull two capillar-
ies (see Notes 22 and 23); use a syringe needle and a syringe to
backfill each glass needle with 3 M KCl.
2. Install one capillary each onto the current-electrode holder
and the voltage-electrode holder.
3. Perfuse the Standard bath solution in the perfusion chamber.
4. Use the micromanipulator to dip the tip of the current-
electrode capillary and of the voltage-electrode capillary into
the Standard bath solution, and then adjust the voltage of both
electrodes to 0 mV. Also, check if the resistances of the elec-
trodes are 110 M.
5. Place an oocyte at the center of the perfusion chamber
( see Note 24).
116 Tatsuro Nakagawa and Kazushige Touhara

6. Pierce the oocyte with the current electrode capillary and with
the voltage electrode capillary (see Note 25).
7. Hold the membrane potential at 80 mV, and start recording.
8. Apply the odorant or pheromone of interest, and record the
responses as changes in membrane current (see Notes 26
and 27).

4 Notes

1. Currently available Xenopus oocyte expression vectors possess

specified features that facilitate stable and efficient translation
of cRNAs in Xenopus oocytes. First, within the upstream region
of multicloning site, the vectors contain RNA polymerase pro-
moter sites (T7 or SP6) for in vitro transcription. Second, the
vectors possess the 5 and 3 UTRs from the Xenopus laevis
-globin mRNA, one at each terminus of the multicloning site;
these UTRs greatly enhance translation efficiency, [15]. In
addition, polyadenylation (poly(A)) sequences are positioned
at the 3 end of multicloning site; these sequences enhance the
stability of transcribed RNA [16].
2. GTP inhibits the RNA capping reaction; therefore, the concen-
tration of GTP should be lower than that of the other NTPs.
3. Make the gel in a fume hood because vapor from formalde-
hyde is very harmful. Formaldehyde should be added to the
solution after the temperature of the mixture of other reagents
drops below 70 C.
4. For convenience, make and store a 25 high-salt stock
(Stock A) and a 25 divalent cation stock (Stock B) of MBSH.
Compositions of the stocks are as follows: Stock A contains 2.2
M NaCl, 25 mM KCl, 60 mM NaHCO3, 375 mM HEPES,
titrated to pH 7.6; Stock B contains 8.25 mM Ca(NO3)2 4H2O,
10.25 mM CaCl2 2H2O, 20.5 mM MgSO4 7H2O. Both solu-
tions should be filter sterilized with a 0.20 m pore diameter
filter. Store both stock solutions at 4 C. Prior to oocyte
manipulation, make the 1 MBSH solution by diluting Stock
A (4 % of total volume), Stock B (4 % of total volume), 60 mg/
ml penicillin (0.1 % of total volume), and 50 mg/ml strepto-
mycin (0.1 % of total volume) into sterilized water.
5. All the implements should be sterilized in 70 % EtOH
before use.
6. To avoid activation of endogenous Ca2+-activated Cl channels,
Ca2+ should not be included in the extracellular solution. For
example, Ba2+ is included instead of Ca2+ in the Standard bath
solution described here.
Functional Assays for Insect Olfactory Receptors in Xenopus Oocytes 117

7. Choose a peristaltic pump that has small flow pulsation that

causes a low S/N ratio.
8. If the template remains a circular plasmid, RNA polymerases
generate long and heterogeneous RNA transcripts. Linearization
of the template is necessary to avoid this problem. Also, check
that the plasmid has only one site for the restriction enzyme
that is used to linearize the plasmid.
9. Drying the pelleted DNA is not recommended because drying
the pellet may reduce the DNA solubility and the success of
the subsequent transcription. Make sure that almost all the
ethanol is removed from the DNA because ethanol will inhibit
the transcription reaction.
10. Drying the pelleted DNA completely is not recommended
because drying may drastically reduce the solubility of the
DNA. Make sure that all the ethanol is removed since it is toxic
for oocytes.
11. Use a scalpel to remove the upper region of the denaturing
agarose gel before EtBr flows out of the gel; you can see the
EtBr as orange-colored bands.
12. If the frog has already been used to harvest oocytes two or
three times, it should be euthanized and discarded; repeated
surgery degrades the condition of frogs and oocytes.
13. As collagenase is toxic to oocytes in the presence of Ca2+, you
need to use an OR2 buffer that does not contain Ca2+.
14. As efficiency of the collagenase treatment is greatly dependent
on collagenase product lot, you must adjust the incubation
time according to the actual follicle dissolution rate. If the
activity of the collagenase is low, you should perform one addi-
tional collagenase treatment. To perform an additional colla-
genase treatment, remove collagenase from the glass vial by
washing the oocytes and the vial with OR2 solution two to
three times, and then add fresh collagenase solution to the vial
and gently rotate it for another hour.
15. We use the following program to pull the appropriate needles;
Step 1: HEAT = 557, PULL = 30, VEL = 30, TIME = 200, Step
2: HEAT = 530, PULL = 85, VEL = 60, TIME = 100.
16. The size of the needle tip is critical for successful microinjec-
tion. If a tip is too thick, it will damage oocytes, which will
then die before the day of TEVC recording. Conversely, if a tip
is too thin, aspirating the RNA solution will be difficult.
17. To avoid aspirating air bubbles, adjust the location of the tip
with micromanipulator to keep the tip of needle in the drop of
RNA solution drop.
118 Tatsuro Nakagawa and Kazushige Touhara

18. Fibers of the gauze prevent the oocytes from rolling over dur-
ing the microinjection.
19. For each injection, choose a completely defolliculated oocyte
because the follicle cell layer makes it difficult to pierce the
oocyte; alternatively, you can use tweezers to remove the folli-
cle cell layer, although careful manipulation of the oocytes is
needed with this method. Also, you must choose good
(healthy) oocytes that have resilient membranes and clearly
separated animal (black side) and vegetal (white side) poles.
Do not choose bad oocytes such as those that are too small or
too large or those for which some or all of animal pole is white.
20. As a result, 25 ng of canonical Or cRNA and 25 ng of the Orco
family cRNA are injected to each oocyte.
21. Exchange the MBSH in the dish at least once a day, and remove
the weakened or dead oocytes from the dishes; the dead and
damaged oocytes degrade the condition of the culture and can
make the condition of other, healthy oocytes worse.
22. For instance, we use the following program to pull the
appropriate needles; Step 1: HEAT = 570, PULL = Blank,
VEL = 25, TIME = 200, Step 2: HEAT = 565, PULL = 30,
VEL = 25, TIME = 200, Step 3: HEAT = 580, PULL = 75,
VEL = 75, TIME = 200).
23. The size of the tips of capillaries should be thin because thick
tips cause KCl solution to flow into the oocyte.
24. Adjust the flow rate of the solution so that the oocyte in the
chamber does not move.
25. Here you can see the resting membrane potential (usually
40 mV to 20 mV) of the oocyte.
26. For hydrophobic odorants or pheromones that are difficult to
dissolve in the Standard bath solution, make high-
concentration stock solution of the ligand in DMSO, and
then dissolve this solution in the Standard bath solution prior
to each experiment. The final DMSO concentration should be
less than 1 %. Alternatively, the solubility of the ligands would
be enhanced by adding odor- or pheromone-binding proteins
into the solution [17].
27. If you want to measure currentvoltage relationship of an
Or-Orco receptor, use voltage-steps or voltage ramps to change
the membrane current from 80 mV to different currents
between +50 and +100 mV before and after the onset of odor-
ant or pheromone exposure. You can obtain the currentvoltage
relationship by subtracting the traces before the onset of ligand
exposure from the trace after the onset.
 Functional Assays for Insect Olfactory Receptors in Xenopus Oocytes 119

References
1. Larsson MC, Domingos AI, Jones WD, brane domain proteins that can signal indepen-
Chiappe ME, Amrein H, Vosshall LB (2004) dently of heterotrimeric G proteins. Insect
Or83b encodes a broadly expressed odorant Biochem Mol Biol 38:770780
receptor essential for Drosophila olfaction. 10. Wicher D, Schafer R, Bauernfeind R, Stensmyr
Neuron 43:703714 MC, Heller R, Heinemann SH, Hansson BS
2. de Bruyne M, Foster K, Carlson JR (2001) (2008) Drosophila odorant receptors are both
Odor coding in the Drosophila antenna. ligand-gated and cyclic-nucleotide-activated
Neuron 30:537552 cation channels. Nature 452:10071011
3. Wang JW, Wong AM, Flores J, Vosshall LB, 11. Wang G, Carey AF, Carlson JR, Zwiebel LJ
Axel R (2003) Two-photon calcium imaging (2010) Molecular basis of odor coding in the
reveals an odor-evoked map of activity in the malaria vector mosquito Anopheles gambiae.
fly brain. Cell 112:271282 Proc Natl Acad Sci USA 107:44184423
4. Wetzel CH, Behrendt HJ, Gisselmann G, 12. Neuhaus EM, Gisselmann G, Zhang W,
Stortkuhl KF, Hovemann B, Hatt H (2001) Dooley R, Stortkuhl K, Hatt H (2005)
Functional expression and characterization of a Odorant receptor heterodimerization in the
Drosophila odorant receptor in a heterologous olfactory system of Drosophila melanogaster.
cell system. Proc Natl Acad Sci USA Nat Neurosci 8:1517
98:93779380 13. Nichols AS, Luetje CW (2010) Transmembrane
5. Nakagawa T, Sakurai T, Nishioka T, Touhara segment 3 of Drosophila melanogaster odorant
K (2005) Insect sex-pheromone signals medi- receptor subunit 85b contributes to ligand-
ated by specific combinations of olfactory receptor interactions. J Biol Chem
receptors. Science 307:16381642 285:1185411862
6. Wanner KW, Nichols AS, Walden KK, 14. Nakagawa T, Pellegrino M, Sato K, Vosshall
Brockmann A, Luetje CW, Robertson HM LB, Touhara K (2012) Amino acid residues
(2007) A honey bee odorant receptor for the contributing to function of the heteromeric
queen substance 9-oxo-2-decenoic acid. Proc insect olfactory receptor complex. PLoS One
Natl Acad Sci USA 104:1438314388 7:e32372
7. Mitsuno H, Sakurai T, Murai M, Yasuda T, 15. Krieg PA, Melton DA (1984) Functional mes-
Kugimiya S, Ozawa R, Toyohara H, senger RNAs are produced by SP6 in vitro
Takabayashi J, Miyoshi H, Nishioka T (2008) transcription of cloned cDNAs. Nucleic Acids
Identification of receptors of main sex- Res 12:70577070
pheromone components of three Lepidopteran 16. Huez G, Marbaix G, Hubert E, Leclercq M,
species. Eur J Neurosci 28:893902 Nudel U, Soreq H, Salomon R, Lebleu B,
8. Sato K, Pellegrino M, Nakagawa T, Vosshall Revel M, Littauer UZ (1974) Proceedings:
LB, Touhara K (2008) Insect olfactory recep- Role of the polyadenylic segment in the trans-
tors are heteromeric ligand-gated ion chan- lation of globin messenger RNA. Hoppe
nels. Nature 452:10021006 Seylers Z Physiol Chem 355:1211
9. Smart R, Kiely A, Beale M, Vargas E, Carraher 17. Grosse-Wilde E, Svatos A, Krieger J (2006) A
C, Kralicek AV, Christie DL, Chen C, pheromone-binding protein mediates the
Newcomb RD, Warr CG (2008) Drosophila bombykol-induced activation of a pheromone
odorant receptors are novel seven transmem- receptor in vitro. Chem Senses 31:547555
 Chapter 9

A Protocol for Heterologous Expression and Functional

Assay for Mouse Pheromone Receptors
Sandeepa Dey, Senmiao Zhan, and Hiroaki Matsunami

Abstract
Innate social behaviors like intermale aggression, fear, and mating rituals are important for survival and
propagation of a species. In mice, these behaviors have been implicated to be mediated by peptide phero-
mones that are sensed by a class of G protein-coupled receptors, vomeronasal receptor type 2 (V2Rs),
expressed in the pheromone-detecting vomeronasal organ (VNO) (Chamero et al., Nature 450:899902,
2007; Haga et al., Nature 466:118122, 2010; Kimoto et al., Curr Biol 17:18791884, 2007; Leinders-
Zufall et al., Nat Neurosci 12:15511558, 2009; Papes et al., Cell 141:692703, 2010). Matching V2Rs
with their cognate ligands is required to understand what receptors the biologically relevant pheromones
are acting on. However, this goal has been greatly limited by the unavailability of appropriate heterologous
tools commonly used to carry out receptor deorphanization, due to the fact that this family of receptors
fails to traffic to the surface of heterologous cells. We have demonstrated that calreticulin, a housekeeping
chaperone commonly expressed in most eukaryotic cells, is sparsely expressed in the vomeronasal sensory
neurons (VSNs). Stable knock down of calreticulin in a HEK293T derived cell line (R24 cells) allows us
to functionally express V2Rs on the surface of heterologous cells. In this chapter we describe protocols for
maintenance and expansion of the R24 cell line and functional assays for V2Rs using these cells.

Key words Heterologous, Calcium, V2Rs, R24, Cells

1 Introduction

In vertebrates, pheromones are largely detected by the accessory

olfactory organ, the vomeronasal organ (VNO). In mice, the VNO
has been implicated in detecting a large number of inter and intra
specific chemosignals that mediate innate behavior like fear,
aggression, mating rituals, pregnancy block, mate recognition, etc.
[15]. The VNO contains closely packed sensory neurons that
express at least 250 putative pheromone receptors spanning three
unrelated families of G protein-coupled receptors (GPCRs): vom-
eronasal receptors classes I and II (V1Rs and V2Rs respectively)
and formyl peptide receptors (FPRs) [611]. Each sensory neuron
expresses either one V1R or an FPR or a combination of two V2Rs
in spatially distinct regions. A key question in studying pheromonal

Kazushige Touhara (ed.), Pheromone Signaling: Methods and Protocols, Methods in Molecular Biology, vol. 1068,
DOI 10.1007/978-1-62703-619-1_9, Springer Science+Business Media, LLC 2013

121
122 Sandeepa Dey et al.

olfaction is what biological ligands activate a receptor or related

receptors; however, the complex expression pattern of such a large
receptor repertoire makes it challenging to adequately address this
question. Heterologous cells provide a convenient platform to
carry out large scale functional and deorphanization studies for
GPCRs; however, a major road block in studying the pheromone
receptors using this tool was the inability to express the receptors
functionally on the surface of heterologous cells. We have recently
demonstrated that vomeronasal sensory neurons expressing V2Rs
show low expression of an otherwise ubiquitous cellular chaper-
one, calreticulin. Depletion of calreticulin in heterologous cells
(HEK293T) enabled export of V2Rs to the surface. Using this
knowledge, we established the cell line R24, from HEK293T, with
constitutive knock down of calreticulin. We then used this line to
carry out V2R ligand binding assays with calcium imaging [12].
In this chapter we describe the protocol to maintain and expand
the cells and carry out the functional assays.
Heterologous cells have been extensively exploited to study G
protein-coupled receptors. The advantages of deorphanizing recep-
tors heterologously in comparison to in vivo protocols are many.
One can selectively transfect a receptor-expressing plasmid into
cells, and examine the response of the transfected cells easily to a
ligand or set of ligands by applying a series of ligand pulses. To test
the specificity of a particular receptor-cognate ligand pair, one can
set up control experiments easily by testing either other ligands with
the receptor of choice or other receptors with the ligand of choice.
The heterologous system allows for identification of receptors for a
set of biologically relevant ligands without performing genetic
manipulations and/or in vivo assays. The heterologous cells may be
further used to study the signal transduction pathway by addition
or subtraction of cofactors in the transfection step and comparing
how the activity of the cells is affected. One can also create chimeric
receptors, receptors with point mutations, or domain switched
receptor constructs, transfect into cells, and test their response to a
ligand or set of ligands. It allows us to identify motifs or domains
and particular amino acids that are important for either ligand
binding or signaling or both. Additionally, the heterologous cells
provide a means to study various other cell biological areas, such as
mechanisms of endoplasmic reticulum targeting of receptors, sur-
face export and interactions with chaperones [13].
To assay for ligand binding, a V2R receptor of interest is co-
transfected with H2M-10.4 and B2m as well as G15, a Gq class
of G protein widely used in functional assays of GPCRs. Activation
of G15, which couples to many but not all of the GPCRs, leads
to a transient intracellular calcium release [14, 15]. Changes in
intracellular calcium can be easily detected by chemical dyes with
high affinity for calcium which change their spectral properties
upon binding calcium ions. Using this principle, we perform
 A Protocol for Heterologous Expression and Functional Assay 123

calcium dye based imaging because this provides an extremely

sensitive and efficient method. We use a combination of two cal-
cium dye indicators, Fluo-4 and Fura Red, which are both visible
light excitable (488 nm). The green-fluorescent emission
(~525 nm) of Fluo-4 increases on binding calcium while the red
fluorescence (~660 nm) of the Fura Red indicator decreases once
the indicator binds Ca2+. Using two dyes in combination enables
one to carry out ratiometric quantification for changes in cytosolic
calcium concentration.
R24 cells, like most other heterologous cell lines pose their
own set of disadvantages, which should be closely considered in
parallel with the advantages of deorphanizing V2Rs. In an attempt
to mimic the vomeronasal sensory neurons, this heterologous cell
line was established by knocking down calreticulin, a very essential
endoplasmic reticulum chaperone that plays multiple roles in sus-
tenance of cell health and calcium homeostasis. As a result, the cells
are slower growing than the parent cell line (HEK293T), an
important parameter to bear in mind while planning experiments.
The cell must be handled with extreme cautionover-trypsinization
should be avoided, cells should be triturated slowly, and all media
changes must be done gently and in a timely way. Each batch of
cells may be used for a limited number of passages.
Keeping these limitations in mind, one can design experiments
using the R24 heterologous cells and calcium imaging to study the
ligand specificities and functions of mouse V2Rs.

2 Materials

Use cell culture grade commercially available reagents. Prepare all

solutions in a sterile environment, preferably in sterile laminar flow
chambers. Diligently follow all waste disposal rules when disposing
waste materials.

2.1 R24 Cell Culture 1. Minimum essential medium (MEM) containing Earles salts
and Maintenance and L-glutamine.
2. Fetal bovine serum (FBS): heat inactivated.
3. 10 % supplemented medium (M10): Mix 45 ml MEM, with
5 ml FBS to obtain 10 % supplemented medium (M10). Store
M10 in 4 C.
4. Penicillin-Streptomycin-Amphotericin (PSF) culturing
medium: Prepare M10 as above and add Penicillin-Streptomycin
and Amphotericin to final concentrations 100 g/ml and
1.25 g/ml respectively.
5. Puromycin containing PSF, R24 maintenance medium:
Prepare PSF culturing medium as described. Prepare two R24
124 Sandeepa Dey et al.

maintenance media by adding puromycin to final concentra-

tions 5 g/ml and 20 g/ml respectively.
6. 100 mm coated sterile cell culture dishes.
7. Centrifuge tubes.

2.2 Cell Transfer 1. 1 Phosphate Buffer Saline, PBS.

and Plating for 2. Trypsin (0.05 %).
Calcium Imaging
3. 35 mm sterile cell culture dishes with 10 mm optical glass well.

2.3 Transfection 1. Transfection reagent.

2. Plasmid DNA: V2R pheromone receptor, major histocompat-
ibility factor 10.4, 2 microglobulin and mouse G15 com-
plete open reading frames cloned in mammalian expression
vector pCI.

2.4 Calcium 1. Fura Red solution: Dissolve 50 g Fura Red in 12.5 l DMSO
Imaging Assay to obtain 4 g/l Fura Red.
2. Fluo-4: Dissolve 50 g Fluo-4 in 12.5 l DMSO to obtain
4 g/l Fluo-4.
3. Pluronic acid: Weigh out 20 mg pluronic acid (amorphous
solid) in an eppendorf tube, add 100 l cell culture grade
DMSO and incubate at 37 C for 1015 min, intermittently
tapping the mixture to dissolve the pluronic acid (see Note 1).
4. Bovine serum albumin (BSA), 7.5 %.
5. Glucose, 45 %.
6. Hanks Balanced Salt Solution (HBSS) 1, with glucose: anhy-
drous calcium chloride 140 mg/L, magnesium chloride-6H2O
100 mg/L, magnesium sulfate-7H2O 100 mg/L, potassium
chloride 400 mg/L, potassium phosphate monobasic 60 mg/L,
sodium chloride 8,000 mg/L, sodium phosphate dibasic-7H2O
48 mg/L, Dextrose 1,000 mg/L; pH = 7.4.
7. 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES),
1 M.
8. Imaging buffer: HBSS, 10 mM HEPES, 0.45 % glucose. Mix
5 ml 1 M HEPES and 5 ml 45 % glucose in 500 ml HBSS to
obtain imaging buffer.
9. Loading buffer: Add 13.3 l, 7.5 % BSA to 1 ml imaging buf-
fer to obtain 0.1 % BSA in imaging buffer.
10. Dye mixture: Mix well 0.5 l 20 % pluronic acid, 0.5 l Fluo-4,
and 1 l Fura Red with vigorous pipeting, and add 500 l load-
ing buffer to obtain dye mixture (see Notes 2 and 3).
 A Protocol for Heterologous Expression and Functional Assay 125

2.5 Equipments 1. 37 C cell culture incubator with 5 % carbon dioxide.

2. Certified class II biological safety cabinet with laminar flow.
3. Centrifuge with swinging bucket rotor for 15 ml conical tubes.
4. Phase contrast microscope with 10, 20 objectives.
5. Fluorescence microscope with Fluo 4 and Fura Red filters
(excitation, 488 nm; emission, 500560 nm for Fluo-4 and
605700 nm for Fura red), 40 objective.
6. Peristalitic pump.

3 Methods

All cell culture procedures must be carried out in a sterile way, in

class II biological safety cabinet with laminar flow, at room tem-
perature unless otherwise mentioned.

3.1 R24 Cell Culture 1. Quickly thaw a frozen vial of R24 cells in a 37 C water bath.
and Maintenance 2. Immediately after thawing, transfer cells from vial to sterile
15 ml tube containing 10 ml M10.
3. Centrifuge at 200 g for 5 min.
4. Aspirate supernatant and resuspend cells in 8 ml M10 with
gentle trituration.
5. Plate cells in a 100 mm sterile cell culture dish.
6. Observe the cells closely in a phase contrast microscope, they
should appear like round particles in suspension in the media.
7. Incubate overnight in a cell culture incubator at 37 C, 5 % CO2.
8. Observe the cells again after overnight incubation, they should
have settled to the bottom of the cell culture dish and attached
to it.
9. If the cells appear to be firmly attached, aspirate M10 plating
medium and replace with 8 ml PSF culture medium (see Note 4).
10. 24 hours later, change medium was to PSF containing 5 g/
ml puromycin (see Note 5).
11. 24 hours after addition of 5 g/ml puromycin, change the
medium to 20 g/ml puromycin containing PSF (see Note 6).
12. Thereafter maintain the cells in 20 g/ml puromycin and PSF
containing medium for experiments.

3.2 Transfer R24 1. Allow cells to grow till confluent or up to desired confluency,
Cells for Transfection checking cell density periodically (see Note 7).
2. Prior to cell transfer, aspirate media and gently add 8 ml sterile
PBS for washing out media (see Note 8).
126 Sandeepa Dey et al.

3. Incubate cells in 3 ml trypsin (0.05 %) at room temperature;

when all cells are loosened from the sterile cell culture dish,
gently triturate them a few times to detach them.
4. Stop trypsinization by adding 5 ml M10; gently triturate the
mixture several times to dissociate the cells from each other
and make a homogeneous cell suspension (see Note 9).
5. At this point, the cells need to be plated for assay and mainte-
nance. Depending on anticipated frequency of future uses,
transfer a portion of the trypsinized cells into a sterile conical
tube; for the immediate assay, depending on the number of
dishes to be set up, transfer another portion into a second ster-
ile conical tube.
6. Centrifuge both aliquots at 200 g, for 5 min at room
temperature.
7. Resuspend the aliquot for maintenance in 8 ml 20 g/ml
puromycin containing PSF media by gentle trituration and
transfer to a sterile 100 mm cell culture dish. Maintain the dish
in the cell culture incubator at 37 C, 5 % CO2 till further use.
8. Resuspend the aliquot of cells for imaging in M10 (antibiotic
free), 1 ml per 10 mm dish.
9. Transfer 1 ml of media containing resuspended cells into each
glass bottomed dish.
10. Incubate dishes overnight in 37 C, 5 % CO2.

3.3 Transfection 1. Observe the cells closely using a phase contrast microscope
of Cells for Imaging 24 h after plating. If cells appear to have settled on the glass
bottomed dish, they are ready for transfection. R24 cells should
be ~40 % confluent at the time of transfection in the 35 mm
cell culture dishes with 10 mm optical glass wells.
2. Mix plasmid DNA in the following amounts (per glass bot-
tomed dish): V2R pheromone receptor 1,000 ng, H2-M10.4
1,200 ng, 2-microglobulin 200 ng and G15 400 ng (see
Note 10).
3. For transfection, follow instructions from the transfection
reagents manufacturers manual.
4. Incubate the cells for ~36 h in 37 C, 5 % CO2.

3.4 Calcium Imaging 1. Approximately 36 h after transfection, slowly aspirate the

medium out of the glass bottomed culture dish.
2. Gently add 1 ml imaging buffer to wash remaining media, tak-
ing care to avoid pipetting the buffer in the glass bottom.
3. Slowly aspirate the buffer from the culture dish.
4. Pipette 50 l dye mixture to the glass bottom.
 A Protocol for Heterologous Expression and Functional Assay 127

5. Incubate the cells in the dye mixture for 45 min in dark (see
Note 11).
6. Dilute stimulant to desired strength in imaging buffer.
7. Wash the stimulant/buffer delivery tubing thoroughly with
distilled water.
8. For imaging, expose the cells to a constant flow of imaging
buffer (see Note 12).
9. Collect the data at wavelengths appropriate for Fluo-4 and
Fura Red (excitation 488 nm, emission 500560 nm for Fluo-
4, 605700 nm for Fura-red) at regular intervals. Apply stimu-
lants for desired length of time punctuating with buffer flow to
allow the cells to recover in case they respond to a particular
pulse of test ligand (see Notes 13 and 14).
10. At the end of the ligand pulses, apply a pulse of 10 nM isopro-
terenol to use for positive control (see Note 15).

3.5 Preparation of 1. Pre cool an isopropyl alcohol bath to 4 C.

R24 Cell Frozen Stocks 2. To prepare frozen stocks of R24 cells, start culture and grow
them in PSF containing medium as discussed in Subheading 3.1.
3. When confluent, split them as discussed in Subheading 3.2.
After aspirating out the trypsin medium, add 2 ml freezing
medium and gently triturate.
4. Transfer 1 ml of resuspended cells to one freezing vial (two
vials of frozen stocks from each confluent 100 mm plate of
R24 cells).
5. Freeze cell vials in 80 C.
6. Store frozen vials of R24 cells for long term at 80 C.
7. Cells split as discussed and resuspended after spinning in freez-
ing mixture (each 100 ml confluent plate of cells resuspended
in 1 ml freezing mixture, frozen in a single vial for usage next
time) and frozen in 80 C in isopropyl alcohol bath.

3.6 Data Analysis 1. Base fluorescence intensities, changes in the Fluo-2 and Fura
Red intensities, and the ratio of the changes can be plotted
with graphing softwares like Microsoft Excel or ImageJ. Use
the ratio to analyze the data for responses (Fig. 1).
2. Determine cells that have responded to a positive control stim-
ulant applied at the end of the experimental run, for example
10 nM Isoproterenol (see Note 16).
3. Response to ligand pulse should appear within 90 s of applica-
tion of the stimulant (see Note 17).
4. Responses that appear before the stimulant front reaches the
cells should be counted as non specific.
128 Sandeepa Dey et al.

B A B A B A B A B A B A B A

Fluo4
Fura red
4.0
V2Rp1+H2-M10.4+2m
3.5
3.0
2.5

Fflou4/Ffura red
2.0
1.5
1.0
0.5
60s
0.0

Buffer ESP15 ESP36 ESP3 ESP5 ESP6 Isoproterenol

(111) (312) (504) (696) (888) (1089) (1287) (sec)

Fig. 1 Response of single R24 cell transfected with V2Rp1 to seriel application of
stimulants. Upper panel: change of Fluo-4 and Fura Red fluorescence intensities
(b, before application of stimulant indicated in X axis; a, after application of stim-
ulant indicated in X axis). Lower panel: Calcium concentration trace of the cell on
application of recombinant peptide stimulants (ESP15, 36, 3, 5, and 6, respec-
tively), 100 nM each and positive control (isoproterenol, 1 nM)

5. A cell that responds nonspecifically outside the range of time

interest should be discounted from analysis.
6. A cell must show an increase of Fluo-4 intensity and a concomi-
tant decrease in Fura Red signal intensity; ratio of Fluo-4 to Fura
Red signals must exceed 50 % of the base line (see Note 18).
7. Efficiency of transfection may be calculated by expressing the
number of isoproterenol responsive cells as a percentage of the
total number of cells in the field of view (see Note 19). Number
of cells responding to a particular ligand should be normalized by
expressing as a percentage of the total number of isoproterenol
responsive cells. Normalization enables us to compare responses
between different dishes, conditions, and days of experiments.

4 Notes

1. For best results, make fresh solutions of 20 % pluronic acid

every week.
2. After mixing pluronic acid, Fluo-4, and Fura Red thoroughly,
consider incubating in a 37 C water bath or heating block for
5 min. This increases the total uptake of the dyes by each cell.
3. Once loading buffer is prepared, keep wrapped in aluminum
foil in the dark till the end of the experiment, since the fluores-
cent dyes are light sensitive.
A Protocol for Heterologous Expression and Functional Assay 129

4. Observe the cells closely before replacing the medium. If cells

look unsettled, wait until the cells are attached to the dish
before replacing the M10 medium.
5. Some cells usually die after addition of lower concentration of
puromycin.
6. A larger number of cells usually die after adding media con-
taining higher concentration of puromycin.
7. R24 cells grow and divide approximately every 36 h.
8. It is important to wash out supplemented medium before
adding trypsin since any trace of medium will inhibit trypsin
activity.
9. R24 cells attach less robustly to cell culture dishes than the par-
ent cell line HEK293T; therefore while carrying out cell trans-
fer, maintain caution to not over-trypsinize. Stop the enzymatic
reaction as soon as the cells detach from the culture dish.
10. These amounts of plasmid DNA have been optimized for pCI
(mammalian vector) based constructs and transfections carried
out with Lipofectamine 2000 (Invitrogen) mediated transfec-
tions. For other expression and transfection systems, the rela-
tive amounts of the different constructs in the transfection
mixture may have to be optimized.
11. For best results, soak transfected cells in dye solution, one dish
at a time so that cells can be imaged promptly after loading
with dye. Letting the loaded cells wait longer than the required
incubation time results in formation of intracellular dye aggre-
gates that appear as tiny fluorescent specks while carrying out
the assay and may interfere with the read-out.
12. R24 cells attach lightly to the glass bottomed dish; delivery
must be minutely regulated such that buffer or stimulants are
delivered at a steady rate and at the same time does not dis-
lodge the cells from the dish. We optimized our assays at deliv-
ery rate of ~50 l/s, regulated by a peristaltic pump (Rainin).
13. We used Leica confocal microscope and the live imaging mode
of Leica confocal software for data acquisition. Data was col-
lected at 3 s interval.
14. We pulsed up to 100 nM of ESP family of peptides to stimulate
cells transfected with test receptors; ligand pulses were applied
for 10 s and buffer bath was applied for 2 min following ligand
pulse.
15. Isoproterenol is a cognate ligand for the beta adrenergic recep-
tor, endogenously expressed in the HEK293T cells or cells
derived from it. On binding isoproterenol, the receptor cou-
ples with transfected G15, resulting in an increase in intracel-
lular calcium.
130 Sandeepa Dey et al.

16. Applying the pulse of isoproterenol at the end of the assay

enables the experimenter to ensure that cells are healthy and
responsive throughout the experiment. Cells not responding
to isoproterenol in the end should not be counted.
17. We empirically determined responses of R24 cells transfected
with V2Rp receptors to applied concentrations of ESP ligands
appeared within 90 s time frame. This may have to be opti-
mized for different ligand-receptor pairs and instrument set up.
18. A concomitant increase in fluorescence intensity of one fluoro-
phore and decrease in the other ensures real signals are selected
for analysis over noise. We empirically determined an increase
in ratio of Fluo-4 to Fura Red by 50 % from the base line
enables us to segregate noise from real signals. This may have
to be optimized for different ligand-receptor pairs, calcium
dyes and instrument set up.
19. R24 cells are transfection efficient; typically 80100 % of cells
respond to 10 nM isoproterenol within seconds of applica-
tion. An assay that shows less efficient calcium response to
the positive control should not be considered for statistical
purposes.

Acknowledgement

This work was supported by NIH grant and Duke University

Medical Center, Department of Molecular Genetics and
Microbiology.

References

1. Chamero P, Marton TF, Logan DW, Flanagan 5. Papes F, Logan DW, Stowers L (2010) The
K, Cruz JR, Saghatelian A, Cravatt BF, Stowers vomeronasal organ mediates interspecies defen-
L (2007) Identification of protein pheromones sive behaviors through detection of protein
that promote aggressive behaviour. Nature pheromone homologs. Cell 141:692703
450:899902 6. Dulac C, Axel R (1995) A novel family of genes
2. Haga S, Hattori T, Sato T, Sato K, Matsuda S, encoding putative pheromone receptors in
Kobayakawa R, Sakano H, Yoshihara Y, Kikusui mammals. Cell 83:195206
T, Touhara K (2010) The male mouse phero- 7. Herrada G, Dulac C (1997) A novel family of
mone ESP1 enhances female sexual receptive putative pheromone receptors in mammals
behaviour through a specific vomeronasal with a topographically organized and sexually
receptor. Nature 466:118122 dimorphic distribution. Cell 90:763773
3. Kimoto H, Sato K, Nodari F, Haga S, Holy 8. Liberles SD, Horowitz LF, Kuang D, Contos
TE, Touhara K (2007) Sex- and strain-specific JJ, Wilson KL, Siltberg-Liberles J, Liberles DA,
expression and vomeronasal activity of mouse Buck LB (2009) Formyl peptide receptors are
ESP family peptides. Curr Biol 17:18791884 candidate chemosensory receptors in the vom-
4. Leinders-Zufall T, Ishii T, Mombaerts P, Zufall eronasal organ. Proc Natl Acad Sci USA
F, Boehm T (2009) Structural requirements 106:98429847
for the activation of vomeronasal sensory neu- 9. Matsunami H, Buck LB (1997) A multigene
rons by MHC peptides. Nat Neurosci 12: family encoding a diverse array of putative phero-
15511558 mone receptors in mammals. Cell 90:775784
 A Protocol for Heterologous Expression and Functional Assay 131

10. Riviere S, Challet L, Fluegge D, Spehr M, 13. Strader CD, Fong TM, Tota MR, Underwood
Rodriguez I (2009) Formyl peptide receptor- D, Dixon RA (1994) Structure and function of
like proteins are a novel family of vomeronasal G protein-coupled receptors. Annu Rev
chemosensors. Nature 459:574577 Biochem 63:101132
11. Ryba NJ, Tirindelli R (1997) A new multigene 14. Kostenis E, Waelbroeck M, Milligan G (2005)
family of putative pheromone receptors. Techniques: promiscuous Galpha proteins in
Neuron 19:371379 basic research and drug discovery. Trends
12. Dey S, Matsunami H (2011) Calreticulin Pharmacol Sci 26:595602
chaperones regulate functional expression 15. Offermanns S, Simon MI (1995) G alpha 15
of vomeronasal type 2 pheromone recep- and G alpha 16 couple a wide variety of recep-
tors. Proc Natl Acad Sci USA 108: tors to phospholipase C. J Biol Chem
1665116656 270:1517515180
 Chapter 10

Genetic Manipulation to Analyze Pheromone

Responses: Knockouts of Multiple Receptor Genes
Tomohiro Ishii

Abstract
Gene targeting in the mouse is an essential technique to study gene function in vivo. Multigene families
encoding vomeronasal receptor (VR) type 1 and type 2 consist of ~300 intact genes, which are clustered
at multiple loci in the mouse genome. To understand the function of VRs and neurons expressing a par-
ticular VR in vivo, individual endogenous receptor genes can be manipulated by conventional gene target-
ing to create loss-of-function mutations or to visualize neurons and their axons expressing the VR. Multiple
receptor genes in a cluster can also be deleted simultaneously by chromosome engineering, allowing analy-
sis of function of a particular VR subfamily. Here, we describe protocols for conventional gene targeting
and chromosome engineering for deleting a large genomic region in mouse embryonic stem (ES) cells.

Key words Embryonic stem cell, Embryonic fibroblast, Gene targeting, Knock-out, Chromosome
engineering, Vomeronasal receptor

1 Introduction

ES cells are pluripotent cells and can be maintained in an undif-

ferentiated state in vitro [1, 2]. To prevent differentiation, ES cells
are usually grown on a feeder layer of mitotically inactivated pri-
mary mouse embryonic fibroblasts (EFs). Targeting constructs can
be introduced into mouse ES cells by electroporation and inte-
grated into ES cell genome by homologous recombination [3].
Gene targeting technology has allowed the generation of mice
with desired mutations in the mouse genome. The mutations
include gene inactivation, point mutation, insertions, and small
deletions. For deletions, removal of 20 kb DNA region has been
succeeded in ES cells [4], but larger deletions are difficult to gener-
ate by conventional gene targeting. Chromosomal deletions of
DNA covering several hundred kb to a few Mb can be achieved by
consecutive targeting of loxP sites to the end points of the region
to be deleted followed by Cre-loxP mediated recombination in ES
cells [5, 6]. This chromosome engineering technique can also be

Kazushige Touhara (ed.), Pheromone Signaling: Methods and Protocols, Methods in Molecular Biology, vol. 1068,
DOI 10.1007/978-1-62703-619-1_10, Springer Science+Business Media, LLC 2013

133
134 Tomohiro Ishii

applied to create inversions and duplications of a large chromo-

somal region.
Several mice with targeted mutations in VR loci have been
reported [714]. These mutations include gene knockout by
replacement of the coding sequence with a marker gene, replace-
ment of the coding sequence with that of another receptor, inser-
tion of an internal ribosome entry site (IRES)-marker cassette into
the 3 untranslated region (UTR) of a VR gene to allow for bicis-
tronic expression of the marker gene with the intact VR gene, and
deletion of a VR cluster containing 16 intact receptor genes.
Although the genome-wide knockout programs KOMP,
NorCOMM, and EUCOMM [1517] provide useful resources of
mutant ES cell lines, most of VRs have not been the subject of
gene knockout. Furthermore, mice with targeted mutations of a
chemosensory receptor gene using fluorescent marker genes are
often useful in the field of the olfactory research, and such experi-
ments need to be done on a customized basis. Whole mouse
genome-wide sequence information [18] has enabled to design
gene targeting vectors relatively easily and quickly. Databases for
bacterial artificial chromosomes (BACs) library from 129 and
C57BL/6 strains are useful to obtain genomic DNA fragments for
targeting vector constructs [19]. Several convenient gene con-
struct cassettes are available from Addgene: IRES-taulacZ [20]
and IRES-tauGFP [8] are convenient to visualize cell bodies and
axons, and ACNf cassette [21, 22], which is neo auto-excising
sequence during germ line transmission of the targeted gene, saves
time and labor to remove the neo. Furthermore, the MICER proj-
ect provides a resource with vectors for chromosome engineering
[23]. This information and these tools facilitate greatly the genera-
tion of mutant mice.
Here we describe our methods of designing targeting vectors,
preparation of mouse EF cells, conventional ES cell culture, genetic
manipulations for an individual gene in E14 ES cells [20, 24], and
deletion of clustered multiple genes in AB2.2 ES cells [5]. It takes
approximately 2 months to obtain mutant ES cells from starting a
culture of wild-type ES cells by conventional gene targeting. It is
possible to perform two or more independent gene targeting pro-
cedures simultaneously or in a staggered way, depending on the
skill of the investigator. The methods described here have been
successful for gene targeting of many genes including chemosen-
sory receptors with high efficiency of homologous recombination
rate ranging from 10 to 60 % among drug-resistant clones. Mutant
ES cells can be injected into mouse blastocysts to produce chimeric
mice. For behavior analysis, it is advisable to use mice in an inbred
background by mating the chimeric mouse with mice of 129 strain
or backcrossing F1 mice in a heterogeneous background with mice
of inbred strain several times.
 Gene Targeting in the Mouse 135

2 Materials

Tissue culture facility is essential: laminar flow hood, humidified

37 C, 5 % CO2 incubator, inverted phase contrast microscope
with an objective lens (10 to 20), water bath, liquid nitrogen
storage tanks.

2.1 Conventional 1. neo auto-excising cassette, ACNf (Addgene, Cambridge, MA,

Gene Targeting USA; ID15510).
Vectors 2. Mouse BAC clones (see Subheading 3).
3. Restriction enzymes and PCR enzymes.

2.2 Preparation of a 1. Dulbeccos modified Eagles medium (DMEM) with

Stock of Mouse EF 4,500 mg/L Glucose and L-Glutamine (Millipore, Benford,
Cells for Conventional MA, USA; cat. no. SLM-121-B).
Gene Targeting 2. Fetal bovine serum (FBS) (Fisher Scientific, Pittsburgh, PA,
USA; cat. no. SH3007103). FBS should be heat-inactivated
before use as follows: Remove a bottle of FBS from the 80 C
freezer and place it at 4 C for >1 day to thaw. Warm the bottle
completely in a 37 C bath. Place the bottle up to its neck in a
56 C bath for 25 min, with shaking every 5 min. Equilibrate
the bottle on ice. Store at 4 C.
3. 100 penicillin-streptomycin (Millipore, cat. no. TMS-
AB2-C). Store 5 mL aliquots at 20 C.
4. 100 L-glutamine (Millipore, cat. no. TMS-AB-002-C). Store
5 mL aliquots at 20 C.
5. EF medium: For 500 mL, combine 440 mL of DMEM (item
1) with 50 mL of FBS, 5 mL of 100 penicillin-streptomycin,
and 5 mL of 100 L-glutamine. Filter-sterilize using a 0.22 m
filter. Store at 4 C.
6. Ca2+- and Mg2+-free Dulbeccos Phosphate-buffered saline (PBS).
7. 0.025 % Trypsin and 0.75 mM EDTA without Ca2+ and Mg2+
(Millipore, cat. no. SM-2004-C).
8. Freezing medium (Millipore, cat. no. S-002-D).
9. 20 Mitomycin C. Dissolve 2 mg Mitomycin C (Sigma, St.
Louis, MO, USA; cat. no. M0503) in 10 mL of EF medium.
Freshly prepare before use.
10. Embryonic day 13.5 (E13.5) embryos from G418 resistant
mouse strains. Pregnant C57BL/6 mice crossed with the
TCR KO mouse (The Jackson Laboratory, Bar Harbor, ME,
USA; stock number 002120, B6.129P2-TcrdtmMom) [25]
are used (see Note 1).
11. Cryovials.
12. Styrofoam boxes.
136 Tomohiro Ishii

13. Culture dishes (T75 and T175 flasks).

14. Petri dishes (60 and 100 mm).
15. Dissecting instruments.
16. 70 % Ethanol.

2.3 General ES 1. EmbryoMax ES DMEM (Millipore, cat. no. SLM-220-B) (see

Cell Culture Note 2).
2. FBS (see Subheading 2.2, item 2).
3. 100 penicillin-streptomycin (see Subheading 2.2, item 3).
4. 100 L-glutamine (see Subheading 2.2, item 4).
5. 100 Nonessential Amino Acids (Millipore, cat. no. TMS-
001-C). Store 5 mL aliquots at 20 C.
6. 100 Nucleosides for ES cells (Millipore, cat. no. ES-008-D).
Store 5 mL aliquots at 20 C.
7. 100 -mercaptoethanol for ES cells (Millipore, cat. no.
ES-007-E). Store 5 mL aliquots at 20 C.
8. Leukemia inhibitory factor (LIF) (Millipore, cat. no. ESG1107)
(see Note 3). Store 50 L aliquots at 80 C.
9. ES + LIF medium: For 500 mL, combine 400 mL of
EmbryoMax ES DMEM with 75 mL of FBS, 5 mL of 100
penicillin-streptomycin, 5 mL of 100 L-glutamine, 5 mL of
100 Nonessential amino acids, 5 mL of 100 Nucleosides,
5 mL of 100 -mercaptoethanol, and 50 L of LIF. Filter-
sterilize using a 0.22 m filter. Store at 4 C.
10. EF medium (see Subheading 2.2, item 5).
11. PBS (see Subheading 2.2, item 6).
12. Trypsin (see Subheading 2.2, item 7).
13. Freezing medium (see Subheading 2.2, item 8).
14. E14 ES cell line [24].
15. EF cells (see Subheading 3.2).
16. Culture dishes (T12.5 and T25 flasks).
17. Cryovials.
18. Styrofoam boxes.

2.4 Conventional In addition to materials in Subheading 2.3, prepare the followings.

Gene Targeting
1. 100 G418. G418 (Invitrogen, Carlsbad, CA, USA; cat. no.
in ES Cells
11811-031). 15 mg/mL active ingredient in PBS. Filter-
sterilize using a 0.22 m filter. Store 5 mL aliquots at 20 C
(see Note 4).
2. ES + LIF + G418 medium: Add 5 mL of 100 G418 to 500 mL
of ES + LIF medium. Filter-sterilize using a 0.22 m filter.
Store at 4 C.
 Gene Targeting in the Mouse 137

3. 2 Freezing medium: 80 % FBS, 20 % DMSO (Sigma, cat. no.

D2650). Filter-sterilize using a 0.22 m filter. Freshly prepare
before use.
4. PBS with Ca2+, Mg2+.
5. Large scale plasmid preparation kit (see Note 5).
6. Restriction enzymes.
7. ES lysis buffer: 10 mM TrisHCl, pH 7.5, 10 mM EDTA,
10 mM NaCl, 0.5 % N-Lauroyl Sarcosine Sodium Salt, 1 mg/
mL Protenase K. Protenase K is freshly added before use.
8. Precipitation buffer (for one 96-well plate): Mix 10 mL of
100 % ethanol and 150 L of 5 M NaCl.
9. TE buffer: 10 mM TrisHCl, pH 8.0, 1 mM EDTA.
10. Ethanol (70 % and 100 %).
11. Isopropanol.
12. Culture dishes (96-well plate, 48-well plate, 24-well plate,
12-well plate, 6-well plate, 60 mm dish).
13. 96-well round bottom plates.
14. Electroporator Gene Pulser II equipped with a Pulse Controller
plus module (Bio-Rad, Richmond, CA, USA).
15. Electroporation cuvettes (Bio-Rad, cat. no. 165-2088).
16. Parafilm.
17. Sealing film for 96-well plates.
18. P20- and P200-L Multi-channel Pipetmans and tips.
19. Disposable Pippette Basins for Multi-channel Pipetmans.
20. Pipetman and tip for picking colonies (see Note 6).

2.5 Deletion of a In addition to all materials above except ACNf cassette and E14
Receptor Gene Cluster ES cell line, prepare the followings.
by Chromosome
1. Chromosome engineering cassettes Hprt3 and Hprt5 [5]
Engineering
and Cre-expression vector pOG231 [26].
2. E13.5 embryos from puromycin-resistant mouse strains.
Pregnant DR4 mice (The Jackson Laboratory, stock number
003208, Tg(DR4)1Jae/J) [27] crossed with the DR4 mouse
are used (see Note 7).
3. 1,000 Puromycin. 1 mg/mL puromycin (Sigma, cat. no.
P8833) in tissue culture-grade water. Filter-sterilize using a
0.22 m filter. Store 500 L aliquots at 20 C.
4. 50 Hybri-Max hypoxanthin/aminopterin/thymidine (HAT)
Media Supplement (Sigma, cat. no. H0262).
5. ES + LIF + Puromycin medium: Add 500 L of 1,000
Puromycin to 500 mL of ES + LIF medium (see Note 8).
Filter-sterilize using a 0.22 m filter. Store at 4 C.
138 Tomohiro Ishii

6. ES + LIF + HAT medium: ES + LIF medium with 1 Hybri-Max

HAT Media Supplement. Filter-sterilize using a 0.22 m filter.
Store at 4 C.
7. AB2.2 ES cell line (see Note 9).

3 Methods

Before starting gene targeting, visit the International Knockout

Mouse Consortium Web site (http://www.knockoutmouse.org/)
and ensure that the gene of interest has not yet been mutated in a
way that could be useful for the purpose of the project. All tissue
culture procedures are carried out under sterile conditions using
disposable sterile plasticware and all culture incubations are per-
formed in a humidified 37 C, 5 % CO2 incubator. Medium and
trypsin should be warmed in a 37 C water bath before adding
them to cells unless otherwise specified. Tissue culture plates are
used without any coating procedures in our protocol (see Note 10).
We use the growth surface area of culture plate as a size of cell
number for convenience (e.g., 8 cm2 area of cells) if the area was
confluent at the time of freezing, instead of counting cell number.

3.1 Conventional Higher homologous recombination rates in ES cells can be

Gene Targeting achieved using DNA isogenic to the ES cells. ES cell line E14 and
Vectors AB2.2 are derived from 129P2/OlaHsd and of 129/SvEv strains,
respectively. Therefore, gene targeting vectors should be con-
structed using DNA of 129 origin.
1. Find the sequence of the gene of interest and research the
structure in the Web sites of European Nucleotide Archive at
EMBL-EBI (http://www.ebi.ac.uk/ena/). Choose DNA
sequences of 129 origin, if available (see Note 11).
2. Find BAC clones in the Web site of Mus musculus Clone
Finder at NCBI (http://www.ncbi.nlm.nih.gov/projects/
mapview/mvhome/mvclone.cgi?taxid=10090) (see Note 12).
Choose clones in 129S7/SvEv ES cell BAC library and
purchase the clone from Source Bioscience (http://www.life-
sciences.sourcebioscience.com/clone-products/sanger-
resources/mouse-bmq- bac-librar y/bmq-mouse-bac-
clone---search-page.aspx) (see Note 13).
3. Design and construct targeting vectors. The basic consider-
ations are the region of homologous sequences to the
chromosome integration site (see Note 14), a positive select-
able marker, the region of probes and restriction enzyme diges-
tion sites for screening by Southern blotting, and a linearization
restriction enzyme site in the backbone vector (Fig. 1a). Obtain
a DNA fragment of homologous sequences by long PCR using
 Gene Targeting in the Mouse 139

Fig. 1 Gene targeting strategy. (a) General strategy for targeted insertion of the
IRES-tauGFP sequence into V2R loci. Top horizontal line represents a genomic
structure around the last coding exon, mostly exon 6, of the V2R. The coding
region of exon 6 is shown as a closed box. The IRES-tauGFP-ACNf cassette is
inserted immediately after the stop codon of the V2R by homologous recombina-
tion in ES cells. The probe for screening clones by Southern blotting for homolo-
gous recombination must be located outside the targeting vector. The other
probe on the opposite side of the targeting vector, which is also preferably
located outside the vector, is used for confirmation of homologous recombination
in candidate clones. Restriction enzymes and location of the probes must be
chosen so that the fragment sizes detected by Southern blotting can be distin-
guished between wild type and mutated allele. The ACNf cassette is self-excised
during transmission through the male germ line, leaving a single loxP site behind
in the locus. Filled triangles represent the loxP site. The probes used to detect
homologous recombination in Southern blotting are represented as a horizontal
bar. R1 and R2 are restriction enzyme sites. The left probes and right probes are
used in combination with R1 and R2, respectively. (b) General strategy for tar-
geted deletion of a cluster of VR genes. Top horizontal line represents an unmodi-
fied VR cluster. Replacement vectors containing the Hprt (exons 1, 2)-loxP-neo
sequence and the puro-loxP-Hprt (exons 39) sequence are consecutively tar-
geted to the end points of the cluster in ES cells. Transient expression of Cre
induces recombination between the loxP sites, resulting in a deletion of the VR
cluster and the reconstitution of functional Hprt gene structure allowing the posi-
tive selection in HAT medium. The grey boxes and closed triangles represent a VR
gene and the loxP site, respectively
140 Tomohiro Ishii

BAC clones as a template or excising the fragment from BAC

clones with restriction enzymes. For knockout of a VR gene by
replacement of the coding sequence with GFP, for example,
replace the coding sequence with GFP-ACNf by conventional
recombinant PCR and DNA manipulation (see Note 15). For
bicistronic expression of tauGFP with a VR gene, insert an
IRES-tauGFP-ACNf cassette into a restriction enzyme site
introduced in the 3 UTR of the VR gene (Fig. 1a).

3.2 Preparation of a ES cells are usually cultured on monolayers of mitotically inacti-

Stock of Mouse EFs vated EF cells. We use E13.5 embryos of the TCR KO mouse as
for Conventional Gene a source of G418-resistant EF cells (see Note 16). Use sterile dis-
Targeting section instruments washed in 70 % ethanol.

3.2.1 Isolation of 1. Kill the 13.5 post-coitus pregnant female by cervical disloca-
Embryonic Fibroblasts tion or another approved method.
2. Swab the abdomen with 70 % ethanol.
3. Open the body wall with a pair of forceps and scissors, and
expose the uterine horns.
4. Using a fresh set of forceps and scissors dissect out the uterus
and place it in a 100 mm petri dish. Avoid dissection instru-
ments and uterus touching the fur.
5. Dissect the embryos away from the uterus and place them in a
60 mm dish containing PBS.
6. Transfer the dish of embryos to a laminar flow hood. At this
point start warming 10 mL of trypsin in a 15 mL conical tube
in a 37 C water bath for step 13.
7. Transfer the embryos into the second new 60 mm dish con-
taining PBS.
8. Remove heads using scissors and transfer to the third new
60 mm dish containing PBS.
9. Carefully remove all internal organs using forceps and transfer
to the fourth new 60 mm dish containing PBS.
10. Remove as many red blood cells as possible and rinse in the
fifth new 60 mm dish containing PBS.
11. Transfer the remaining embryo carcasses to a new 60 mm dish.
12. Mince the carcasses thoroughly with scissors for 5 min.
13. Transfer the minced embryos to pre-warmed trypsin in a
15 mL conical tube.
14. Incubate in a 37 C water bath for 30 min. Shake the tube
every 5 min. Pipette up and down ten times with a 10 mL
pipette during the incubation if the minced embryos are not
dissociated well.
 Gene Targeting in the Mouse 141

15. Pipette 20 times with a 10 mL pipette (see Note 17) and allow
large tissue pieces to settle to the bottom for 1 min.
16. Carefully transfer the supernatant to 20 mL of EF medium in
a 50 mL conical tube.
17. Spin at 50 g for 1 min to remove large chunks of tissue and
transfer the supernatant (~25 mL) to a new 50 mL conical
tube.
18. Plate the cell suspension (35 mL) on T75 flasks containing
20 mL of EF medium using approximately one embryo per
flask and culture overnight (see Note 18).
19. Change medium for continuous culture, or if confluent wash
cells with 10 mL of PBS, add 2 mL of trypsin per flask and
incubate in a tissue culture incubator for 5 min.
20. Tap the flasks to detach the cells.
21. Add 18 mL of EF medium and triturate with a pipette to break
cell clumps.
22. Plate cells on two T175 flasks from one T75 flask (i.e., dilution
of 1:4 to 1:5), in a total volume of 50 mL per flask, and culture
for ~2 days (see Note 19).
23. If confluent trypsinize and split cells into T175 flasks at a dilu-
tion of 1:4 or 1:5, and culture for ~2 days. This is used for
mitomycin C treatment in Subheading 3.2.2.

3.2.2 Mitotic EF cells are inactivated by mitomycin C treatment before using as

Inactivation of EF Cells a feeder layer for ES cells (see Note 20). Treat a small number of
with Mitomycin C flasks at the same time (usually three flasks). If six flasks are treated,
it should be done in a staggered way with an interval of 30 min
to 1 h.
1. Prepare T175 flasks of EF cells (80100 % confluent).
2. Remove medium and add 30 mL of EF medium and 1.5 mL
of 20 mitomycin C.
3. Incubate in a tissue culture incubator for exactly 2 h.
4. Wash flasks twice with 20 mL of PBS and trypsinize with 5 mL
of trypsin.
5. Add 10 mL of EF medium and triturate with a pipette to break
cell clumps.
6. Pool the cell suspensions, if multiple flasks are treated at the
same time, and pellet by centrifuging at 200 g for 5 min.
7. Resuspend pellet in 40 mL of EF medium and pellet by centri-
fuging at 200 g for 5 min.
8. Resuspend pellet at 100 cm2 surface area of EF cells per 1 mL
in freezing medium (see Note 21) and dispense into cryovials
in 1 mL-aliquots.
142 Tomohiro Ishii

9. Place cryovials in a styrofoam box at 80 C and leave them

overnight (see Note 22). Transfer the cryovials to liquid
nitrogen.

3.3 General ES Cell ES cells are cultured on a monolayer of EF cells.

Culture

3.3.1 Thawing Frozen EF 1. Thaw vial(s) of frozen EF cells in a 37 C water bath.

Cells for ES Cell Culture 2. Transfer EF cells to 30 mL of EF medium (see Note 23) in a
50 mL conical tube and pellet cells by centrifuging at 200 g
for 5 min.
3. Resuspend pellet in the appropriate amount of EF medium
and plate the cell suspension on a flask.
4. After 12 h, cells should form a monolayer covering almost the
whole area of the flask bottom. If the coverage is too sparse,
thaw more feeders as above and plate them on top of the exist-
ing one.

3.3.2 Thawing Frozen Prepare T12.5 flask of EF cells. ES cell colonies have an oval
ES Cells appearance with a smooth outline (Fig. 2a).
1. Thaw a vial of frozen ES cells (8 cm2) in a 37 C water bath.
2. Transfer ES cells to 10 mL of ES + LIF medium in a 15 mL
conical tube. Pellet cells by centrifuging at 200 g for 5 min.
3. Resuspend pellet in 5 mL of ES + LIF medium and plate on a
confluent T12.5 flask of EF cells (see Note 24).

3.3.3 Passaging ES Cells 1. Check cells under the microscope for 7080 % confluence.
2. Wash cells with PBS and add the appropriate volume of trypsin
(see Table 1). Incubate in a tissue culture incubator for
510 min.
3. Tap the plate to detach the cells.

Fig. 2 Phase-contrast images of E14 ES cells. ES cell colonies on EF cells have an oval appearance with a
smooth outline (a), while ES colonies appear more flattened when cultured on a layer of EF cells that have been
plated at lower density (b). Scale bars, 100 m
 Gene Targeting in the Mouse 143

Table 1
Working volumes for cell culture vessels

Dish Medium Trypsin Growth area

48 well plate 1.1 mL 100 L 0.75 cm2
24 well plate 1.5 mL 200 L 2.00 cm2
12 well plate 3 mL 300 L 3.8 cm2
6 well plate 6 mL 700 L 9.6 cm2
T12.5 flask 6 mL 700 L 12.5 cm2
T25 flask 10 mL 1 mL 25 cm2

4. Add the appropriate volume of ES + LIF medium (see Table 1)

(see Note 25) and triturate with a pipette to break cell clumps.
5. Plate on feeders at a dilution of 1:3 to 1:4.

3.3.4 Freezing ES Cells The protocol below is for a T25 flask of ES cells into three cryovi-
als (~8 cm2 aliquots). Label cryovials with cell line, passage num-
ber, date, and plating area. Place freezing medium on ice next to
the hood.
1. Trypsinize a T25 flask of ES cells.
2. Stop with 10 mL of ES + LIF medium and triturate with a
10 mL pipette to break cell clumps.
3. Pellet cells by centrifuging at 200 g for 5 min.
4. Resuspend pellet in 1.2 mL of freezing medium. Dispense into
cryovials in 400 L-aliquots, and quickly place the cryovials on
ice.
5. Place the cryovials in a styrofoam box at 80 C, and leave
them overnight. Transfer the cryovials to liquid nitrogen.

3.4 Conventional The E14 ES cell line is used in combination with the ACNf cas-
Gene Targeting sette in the targeting vector and G418-resistant EF cells in this
in ES Cells protocol.

3.4.1 Preparation 1. Isolate targeting construct DNA using a large scale plasmid
of DNA for Electroporation preparation kit.
2. One day before electroporation, digest 35 g of the DNA with
the appropriate restriction enzyme (150 L scale) to linearize,
and assess the completion of the digestion by agarose gel elec-
trophoresis using 0.5 L of the digested DNA. Store at 20 C.
3. On the day of electroporation, thaw the linearized DNA on ice.
4. Add 1.2 mL of 100 % ethanol and mix by inverting the tube
(see Note 26). White DNA clump can be observed.
144 Tomohiro Ishii

5. Remove the supernatant using a P200 Pipetman (see Note 27).

6. Wash DNA with 1.4 mL of 70 % ethanol and air-dry pellet for
1015 min (see Note 28).
7. Dissolve DNA in 800 L of PBS with Ca2+ and Mg2+ by incu-
bating at 65 C for 10 min. Mix well using a P1000 Pipetman.
8. Place on ice.

3.4.2 Electroporation DNA can be transfected into ES cells by high-voltage

of ES Cells electroporation.
1. Prepare three 60 mm dishes of EF cells and one T25 flask of
ES cells (7080 % confluent) after passaging recently thawed
ES cells.
2. Feed ES cells with fresh ES + LIF medium 4 h before
harvesting.
3. Wash ES cells twice with PBS. Add 1 mL of trypsin and incu-
bate in a tissue culture incubator for 10 min.
4. Tap the flask to detach the cells. Add 10 mL of ES + LIF
medium and triturate well by pipetting up and down five to ten
times to break cell clumps into a single-cell suspension.
5. Pellet cells by centrifuging at 200 g for 5 min.
6. Remove the supernatant and tap the pellet several times.
Resuspend pellet in 800 L of DNA/PBS solution from
Subheading 3.4.1.
7. Transfer the cell suspension into precooled electroporation
cuvette on ice (see Note 29).
8. Transfer the cuvette into the Bio-Rad electroporation chamber
and apply the electric pulse (800 V and 3 F).
9. Transfer the cell suspension to 10 mL of ES + LIF medium in a
15 mL conical tube.
10. Divide over three 60 mm dishes of EF cells. Drop the cell sus-
pension evenly onto the dish. Culture for 1520 h without
changing medium.
11. Change to ES + LIF + G418 medium to start selection.
12. Change medium when the medium color begins to turn orange
(see Note 30).
13. Pick colonies on 610th day of G418 selection when colonies
are visible (see Note 31).

3.4.3 Picking Colonies Use an inverted microscope to see ES colonies and pick them in a
laminar flow hood. Prepare the Pipetman for picking colonies,
P20- and P200-L Multi-channel Pipetmans and their tips, and
basins in the hood. For one targeting construct we usually pick 96
ES colonies per day for 3 consecutive days. Prepare EF cells in
96-well flat-bottom plates.
 Gene Targeting in the Mouse 145

1. Prepare three 96-well round bottom plates with 15 L of

trypsin per well. Store at 4 C until use (see Note 32).
2. Change ES + LIF + G418 medium in three 60 mm culture
dishes of electroporated ES cells. Replace medium in a 96-well
flat-bottom plate of EF cells with 100 L of ES + LIF + G418
medium per well.
3. Transfer one 96-well trypsin plate to the hood.
4. Put one 60 mm dish of electroporated ES cells on the micro-
scope in the hood.
5. Under the microscope dislodge the colony slightly with the
pipette tip, and suck up the colony using the Pipetman set to
1 L. Transfer the colony to a well of the 96-well plate con-
taining trypsin.
6. Repeat picking colonies for no more than 30 min (see Note 33).
7. Incubate the plate in a tissue culture incubator for 10 min.
8. Tap the 96-well plate and add 100 L of ES + LIF + G418
medium to each well by a Multi-channel Pipetman. Pipette up
and down ten times.
9. Transfer the cell suspension to the 96-well plate of EF cells,
and return the plate to the tissue culture incubator. Change
medium daily.
10. Repeat steps 59 using a new 96-well trypsin plate and another
60 mm dish of the ES cells until picking ~96 colonies per day.
11. Repeat steps 110 on the second and third days (see Note 34).
12. When the 96-well plate is confluent overall (~3 days after pick-
ing colonies), proceed to the cell freezing step in
Subheading 3.4.4 (see Note 35).

3.4.4 Freezing ES Prepare a 96-well flat-bottom plate with 65 L 2 freezing medium

Cells in 96-Well Plates per well and store at 4 C until use.
1. Feed ES cells with fresh ES + LIF + G418 medium 4 h before
harvesting.
2. Wash twice with PBS.
3. Add 35 L of trypsin and incubate in a tissue culture incubator
for 10 min.
4. Tap the plate to detach the cells. Add 65 L of ES + LIF + G418
medium and triturate well by pipetting up and down ten times
(see Note 36).
5. Transfer 65 L of the cell suspension to a well of the 96-well
flat-bottom plate containing 2 freezing medium, and pipet
up and down three times.
146 Tomohiro Ishii

6. Label the plate with clone name and date. Seal the plate with
parafilm around all the edges.
7. Place the plate in a styrofoam box and store at 80 C.
8. Add 200 L of ES + LIF + G418 medium to the remaining cell
suspension (35 L) at step 5 and incubate the plate.
9. Change medium daily.
10. If the plate is mostly confluent, proceed to genomic DNA
extraction in Subheading 3.4.5.

3.4.5 Genomic DNA Prepare 70 % ethanol at 20 C.

Extraction from 96-Well
1. Wash cells in the 96-well plate from Subheading 3.4.4. with
Plate and Restriction
PBS.
Digestion for Southern
Blotting 2. Add 50 L of ES lysis buffer per well.
3. Seal the 96-well plate completely with a sealing film. Place the
plate in a moisture chamber and incubate at 60 C for 24 h.
4. Add 100 L of precipitation buffer per well and wrap the plate
in plastic wrap (see Note 37), and leave the plate overnight at
room temperature.
5. Discard the supernatant by turning upside down on paper
towel.
6. Add 150 L of prechilled 70 % ethanol per well.
7. Discard the supernatant by turning upside down on paper
towel.
8. Place the plate at 55 C for 20 min to air-dry the DNA (see
Note 38).
9. Add the appropriate restriction enzyme mixture (30 L/well)
for Southern blotting and seal the plate with a sealing film.
Incubate the plate for restriction reaction.
10. Perform standard Southern blotting using a 32P-labeled or
DIG-labeled probe for screening homologous recombinant
clones (see Note 39).

3.4.6 Thawing and 1. Prepare a 48-well plate of EF cells with 400 L of ES + LIF
Expanding Mutant ES Cells medium (see Note 40).
from Frozen 96-Well Plates 2. Remove the 96-well plate of ES cells from 80 C and wipe it
to remove ice.
3. Locate the clone to be thawed and warm with a finger under
the well.
4. Add 100 L of ES + LIF medium to the partially thawed ES
cells. Triturate by pipetting up and down a few times and trans-
fer to the 48-well plate of EF cells.
5. Repeat step 4 until you transfer all cell suspensions. A total
volume of 500 L per well can be added.
 Gene Targeting in the Mouse 147

6. Add ES + LIF medium if a total volume of 500 L per well is not

added, and return the 48-well plate to the tissue culture incubator.
7. Proceed to the next clone and repeat steps 36 (see Note 41).
8. Rewrap the 96-well plate in parafilm and return to the 80 C
freezer.
9. Change medium after 12 h, and then change every 24 h.
10. If the plate is confluent, passage stepwise to 24-well plate,
12-well plate, 6-well plate, T25 flask, and three T25 flasks (see
Note 42).
11. Make four cryovials of 8 cm2 stock from two T25 flasks.
Continue culture using the remaining ES cells in the T25 flask
for genomic DNA extraction.
12. If the plate is confluent, trypsinize the cells, pellet by centrifu-
gation, and resuspend in 2 mL of ES lysis buffer.
13. Incubate at 55 C overnight.
14. Add 2 mL of isopropanol and mix well.
15. Fish genomic DNA, wash in 70 % ethanol and 100 % ethanol,
and briefly air-dry.
16. Dissolve the DNA in 150 L of TE buffer. This DNA is subject
to Southern blotting for confirmation of the homologous
recombination in the clones (see Note 43).

3.5 Deletion of a The AB2.2 ES cell line, which is deficient in hypoxanthine phos-
Receptor Gene phoribosyl transferase (Hprt), has been used traditionally for chro-
Cluster by mosome engineering. Two complementary but nonfunctional
Chromosome fragments of a Hprt minigene are linked to the loxP site with either
Engineering neo or puro in the targeting constructs. Cre-loxP mediated recom-
bination in AB2.2 ES cells bearing these two loxP cassettes in a
chromosome result in the reconstitution of functional Hprt gene
structure allowing the selection in HAT medium (Fig. 1b).

3.5.1 Deletion 1. Design the targeting vectors at both end points of a VR gene
of a Receptor Gene cluster as in Subheading 3.1. One targeting vector should contain
Cluster in ES Cells Hprt (exons 1, 2)-loxP-neo from the Hprt3 plasmid for G418
selection and the other should contain puro-loxP-Hprt (exons
39) from the Hprt5 plasmid for puromycin selection. Make
sure the Hprt-minigene fragments are placed in the correct order
and orientation on the targeted chromosome(s) to reconstitute
the functional Hprt after Cre-loxP mediated recombination. The
correct order of DNA fragments is as follows: Hprt (exons 1,
2)/loxP/neo/VR cluster to be removed/puro/loxP/Hprt (exons
39) (Fig. 1b) (see Note 44).
2. Prepare G418-resistant EF cells and puromycin-resistant EF
cells using embryos of TCR KO mouse and DR4 mouse,
respectively, as in Subheading 3.2.
148 Tomohiro Ishii

3. Perform gene targeting as in Subheading 3.4 using AB2.2 ES

cells and a linearized targeting vector containing Hprt (exons
1, 2)-loxP-neo for electroporation. Use the G418-resistant EF
cells as a feeder layer and ES + LIF + G418 medium for selec-
tion (see Note 45).
4. Perform gene targeting as in Subheading 3.4 using ES clones
from step 3 (see Note 46) and a linearized targeting vector
containing puro-loxP-Hprt (exons 39) for electroporation.
Use the puromycin-resistant EF cells as a feeder layer and
ES + LIF + Puromycin medium for selection (see Notes 47
and 48).
5. Perform electroporation of ES cells and pick colonies as in
Subheadings 3.4.23.4.3 using several independent double
targeted ES clones from step 4 and a Cre-expression vector,
pOG231, for electroporation for transient expression of Cre.
Preparation of pOG231 for electroporation is described in
Subheading 3.5.2. The G418-resistant EF cells can be used as
a feeder layer (see Note 49). At step 10 in Subheading 3.4.2,
plate ES cells at three different densities such as 3 106, 3 105,
and 3 104 cells per 60 mm culture dish. The normal number
of ES cells with 80 % confluency is ~1 107 per T25 flask (see
Note 50). Use ES + LIF + HAT medium for selection and pick
610 HAT-resistant colonies. Passage HAT-resistant ES cells
stepwise from 96-well plate to three T25 flasks without freez-
ing the cells (see Note 51). Perform sib selection as in
Subheading 3.5.3 during passaging the ES cells. Genomic
DNA from the ES cells is subject to Southern blotting for con-
firmation of the correct Cre-loxP mediated recombination.

3.5.2 Preparation of For transient expression of Cre, use circular pOG231 plasmid for
pOG231 for Electroporation electroporation. Do not linearize the DNA to avoid increased
probability of genomic integration.
1. Isolate pOG231 using a large scale plasmid preparation kit.
2. Put 25 g of the plasmid in 1 restriction buffer as if it is
digested (100 L volume) (see Note 52).
3. Add 150 L of 100 % ethanol and mix by inverting the tube
(see Note 53).
4. Pellet DNA by centrifuging at 10,000 g for 5 min.
5. Remove the supernatant.
6. Wash DNA with 1.4 mL of 70 % ethanol and air-dry pellet
briefly.
7. Dissolve DNA in 800 L of PBS with Ca2+ and Mg2+ by incu-
bating at 65 C for 10 min. Mix well using a P1000 Pipetman.
 Gene Targeting in the Mouse 149

3.5.3 Sib Selection Although most HAT-resistant clones should have the allele with
the desired chromosomal deletion, there could be a few different
genotypes on the other allele (Fig. 3). The genotype of ES clones
can be distinguished by sib selection [6]. We perform sib selection
during passaging HAT-resistant ES clones from 6-well plates to
T25 flasks in Subheading 3.5.1.
1. Trypsinize HAT-resistant ES cells in a 6-well plate and stop
with 10 mL of ES + LIF medium. Triturate with a 10 mL
pipette.
2. Plate 0.3 mL of cell suspension on each of three wells of a
24-well plate containing 1 mL of ES + LIF (see Note 54).
Incubate for 612 h.
3. Change medium of each of three wells to ES + LIF + HAT,
ES + LIF + G418, or ES + LIF + Puromycin medium.
4. Change medium every 24 h and observe the growth of ES cells
in each medium on the 4th day. Identify the genotype accord-
ing to Fig. 3.

4 Notes

1. This is used for preparation of G418-resistant feeder cells

expressing neo.
2. SLM-220-M can also be used as formulations of SLM-220-M
and SLM-220-B are same.
3. LIF is a cytokine that can suppress differentiation of ES cells.
4. G418 is an analog of neomycin.
5. We use QIAfilter Plasmid Midi Kit (QIAGEN, Valencia, CA,
USA; cat. no. 12243).
6. We use the Pipetman (Gray control button for 20 L pipette
tips, 0.510 L, Eppendorf, Germany; cat. no. 022470051)
and tips (epT.I.P.S. Biopur, 0.120 L volume, Eppendorf,
cat. no. 022491067).
7. Although EF cells prepared from the DR4 transgenic mouse
strain is resistant to concentrations of the drug G418,
6-thioguanine, puromycin, and hygromycin, we use the DR4
EF cells only for puromycin selection in this protocol.
8. DR4 EF cells are resistant to 1 g/mL, but sensitive to 2 g/
mL puromycin.
9. AB2.2 ES cells can be requested through the Welcome Trust
Sanger Institute Web site: http://www.sanger.ac.uk/technol-
ogy/clonerequests/resources/.
10. We do not usually gelatinize tissue culture plates. Alternatively,
gelatinization of tissue culture plates can be performed with
150 Tomohiro Ishii

Fig. 3 HAT-resistant ES clones. HAT-resistant clones should harbor a VR cluster deletion if the targeting vectors
are integrated in the correct order and orientation. Among the HAT-resistant clones there could be G418/
Puromycin-sensitive, G418/Puromycin-resistant, G418-sensitive/Puromycin-resistant, and G418-resistant/
Puromycin-sensitive clones. It depends on integration of two targeting vectors into the same chromosome (in
cis) or different chromosomes (in trans), G1 or G2 phase of the cell cycle during which Cre-loxP mediated
recombination occurs, and orientation to the centromere. Closed circle centromere, triangle loxP site, closed
box VR cluster, 5 Hprt (exons 1, 2), 3 Hprt (exons 39), N neo, P puro, Del cluster deletion, Dup cluster duplica-
tion, R resistant, S sensitive, G1/Cre, G2/Cre phases of the cell cycle in which Cre-recombination occurs. The
various DNA sequences involved in chromosome engineering are not drawn to scale
 Gene Targeting in the Mouse 151

0.1 % gelatin solution (Millipore, cat. no. ES-006-B) for

10 min at room temperature followed by aspirating off the
solution and allowing it to dry for 5 min.
11. Possible other Web sites include Mouse Genome Resources at
NCBI: http://www.ncbi.nlm.nih.gov/projects/genome/
guide/mouse/; UCSC genome browser: http://genome.
ucsc.edu/; Ensembl genome browser: http://www.ensembl.
org/Mus_musculus/Info/Index.
12. A possible other Web site is Ensembl genome browser: http://
www.ensembl.org/Mus_musculus/Info/Index.
13. Alternatively, BAC clones (RP21 and RP22) from 129S6/
SvEvTac strain can be obtained by screening high-density
microarray filters to acquire the BAC ID and purchase the
clone from Childrens Hospital Oakland Research Institute
(CHORI).
14. The length of left and right homology arms is 35 kb and the
total length of homology is 810 kb in most of our targeting
vectors.
15. ACNf is a neo expression cassette for G418 selection with self-
excising property during germ line transmission of the targeted
gene.
16. G418-resistant EF cells are also commercially available from
Millipore (Millipore, cat. no. PMEF-N) and gene targeting in
E14 ES cells using this EF cells is successful.
17. Avoid making air bubbles.
18. Although EF cells can be expanded in EF medium with G418
to select for resistant cells, we usually do not use the drug.
19. Alternatively make EF cell stocks (75 cm2 area of EF cells in
1 mL of freezing medium per vial) for future mitomycin C
treatment.
20. Work carefully with gloves and protective clothing because
mitomycin C is toxic and teratogenic.
21. Remove all of the supernatant by first sucking off all bubbles,
and then slowly inverting the tube while aspirating medium.
Tap the pellet several times and add freezing medium.
22. The cells should be frozen slowly at ~1 C per min to 80 C.
23. EF cells can also be plated in ES + LIF medium.
24. The recovery may vary from 25 to 75 %. Cells should be plated
and left alone for 1224 h as they take a while to recover.
25. Trypsin action should be stopped by adding more than 10
volume of ES + LIF medium.
26. We do not purify the digested DNA by phenol/chloroform
extraction.
152 Tomohiro Ishii

27. Avoid touching the Pipetman tip to DNA clump because DNA
is sticky.
28. Do not over-dry DNA because it might become harder to
dissolve.
29. Avoid making air bubbles. Do not touch the metal plate of the
cuvette. Using plastic wrap, avoid the surface of the cuvette
touching ice directly.
30. Watch the color of medium carefully. We change medium
every 8 h on the first and second day, and every 12 h on the
third-fifth day of G418 selection.
31. Try to pick round ES colonies with a smooth outline.
32. We usually add trypsin to 36 wells in each 96-well plate because
all wells are not used.
33. 2436 colonies can be picked for 30 min.
34. 288 ES colonies in total are sufficient to obtain homologous
recombinant clones in most cases. If you expect higher or
lower efficiency of homologous recombination, adjust the
number of colonies to be picked.
35. Cell growth will vary significantly among wells.
36. Avoid making air bubbles during pipetting.
37. Do not mix by pipetting or shaking.
38. If you perform Southern blotting later, you can seal the dried
plate with a sealing film and store at 20 C.
39. It is convenient to use high-throughput gel system allowing
multi-channel pipetting.
40. It is not necessary to use G418 for selection any more.
41. We usually thaw six to eight clones in total from three frozen
96-well plates.
42. We do not plate EF cells on these plates. ES cells cultured
without EF cells form more flattened colonies than cells cul-
tured on EF cells (Fig. 2b). When the ES cells are plated on EF
cells again, ES cell colonies revert to their original oval appear-
ance with a smooth outline.
43. Analyze the homologous recombination on the side of the
homology arm of the targeting vector different from the one
analyzed during ES clone screening.
44. A library of ready-made targeting vectors, incorporating a drug
resistance gene for gene targeting, the 5 or 3 portion of an
Hprt minigene, and loxP site for chromosome engineering, is
available and clones can be requested through the MICER
Web site [23]. http://www.sanger.ac.uk/resources/mouse/
micer/. It is worth searching for clones.
 Gene Targeting in the Mouse 153

45. It is advisable to perform karyotyping to confirm that selected

ES clones have maintained the correct number of chromo-
somes prior to step 4.
46. It is advisable to use two or more independent clones.
47. The duration of selection varies depending on the drug.
Puromycin selection is faster than G418 selection.
48. It is advisable to perform karyotyping to confirm that selected
ES clones have maintained the correct number of chromo-
somes prior to step 5.
49. G418-resistance in EF cells is not necessary for HAT
selection.
50. The frequency of Cre-loxP mediated recombination varied
from 3 106 to 5 103 in our hand, and could probably be
lower depending on the genes or distance to be deleted [5].
The frequency is lower in case of two loxP sites on chromo-
some in trans than those in cis. Therefore plate more ES cells
in case the frequency should be very low.
51. Most HAT-resistant clones should have desired chromosome
deletion if the design of targeting vector is appropriate.
52. Do not add enzymes.
53. Many small white clumps are observed.
54. The 24-well plate does not contain EF cells. The remaining
cell suspension is used to expand the clone.

Acknowledgments

I am grateful to P. Mombaerts for critical reading the manuscript,

M. Omura for photos of ES cells and valuable comments,
S. Nishimura for comments on chromosome engineering, and
members of the Mombaerts laboratory for improvements in the
methods. This work was supported in part by JSPS KAKENHI
Grant Number 24500456.

References

1. Martin GR (1981) Isolation of a pluripotent 4. Zhang H, Hasty P, Bradley A (1994) Targeting

cell line from early mouse embryos cultured frequency for deletion vectors in embryonic
in medium conditioned by teratocarcinoma stem cells. Mol Cell Biol 14:24042410
stem cells. Proc Natl Acad Sci USA 78: 5. Ramirez-Solis R, Liu P, Bradley A (1995)
76347638 Chromosome engineering in mice. Nature
2. Evans MJ, Kaufman MH (1981) Establishment 378:720724
in culture of pluripotential cells from mouse 6. Yu Y, Bradley A (2001) Engineering chromo-
embryos. Nature 292:154156 somal rearrangements in mice. Nat Rev Genet
3. Vasquez KM, Marburger K, Intody Z et al 2:780790
(2001) Manipulating the mammalian genome 7. Belluscio L, Koentges G, Axel R et al (1999) A
by homologous recombination. Proc Natl map of pheromone receptor activation in the
Acad Sci USA 98:84038410 mammalian brain. Cell 97:209220
154 Tomohiro Ishii

8. Rodriguez I, Feinstein P, Mombaerts P (1999) 18. Mouse Genome Sequencing Consortium,

Variable patterns of axonal projections of Waterston RH, Lindblad-Toh K, Birney E et al
sensory neurons in the mouse vomeronasal (2002) Initial sequencing and comparative
system. Cell 97:199208 analysis of the mouse genome. Nature 420:
9. Del Punta K, Leinders-Zufall T, Rodriguez I 520562
et al (2002) Deficient pheromone responses in 19. Adams DJ, Quail MA, Cox T et al (2005) A
mice lacking a cluster of vomeronasal receptor genome-wide, end-sequenced 129Sv BAC
genes. Nature 419:7074 library resource targeting vector construction.
10. Del Punta K, Puche A, Adams NC et al (2002) Genomics 86:753758
A divergent pattern of sensory axonal projec- 20. Mombaerts P, Wang F, Dulac C et al (1996)
tions is rendered convergent by second-order Visualizing an olfactory sensory map. Cell
neurons in the accessory olfactory bulb. 87:675686
Neuron 35:10571066 21. Bunting M, Bernstein KE, Greer JM et al
11. Ishii T, Mombaerts P (2008) Expression of (1999) Targeting genes for self-excision in the
nonclassical class I major histocompatibility germ line. Genes Dev 13:15241528
genes defines a tripartite organization of the 22. Bozza T, Feinstein P, Zheng C et al (2002)
mouse vomeronasal system. J Neurosci 28: Odorant receptor expression defines functional
23322341 units in the mouse olfactory system. J Neurosci
12. Haga S, Hattori T, Sato T et al (2010) The 22:30333043
male mouse pheromone ESP1 enhances female 23. Adams DJ, Biggs PJ, Cox T et al (2004)
sexual receptive behaviour through a specific Mutagenic insertion and chromosome engi-
vomeronasal receptor. Nature 466:118122 neering resource (MICER). Nat Genet 36:
13. Leinders-Zufall T, Ishii T, Mombaerts P et al 867871
(2009) Structural requirements for the activa- 24. Hooper M, Hardy K, Handyside A et al (1987)
tion of vomeronasal sensory neurons by MHC HPRT-deficient (Lesch-Nyhan) mouse
peptides. Nat Neurosci 12:15511558 embryos derived from germline colonization
14. Ishii T, Mombaerts P (2011) Coordinated by cultured cells. Nature 326:292295
coexpression of two vomeronasal receptor V2R 25. Itohara S, Mombaerts P, Lafaille J et al (1993)
genes per neuron in the mouse. Mol Cell T cell receptor delta gene mutant mice: inde-
Neurosci 46:397408 pendent generation of alpha beta T cells and
15. Ringwald M, Iyer V, Mason JC et al (2011) programmed rearrangements of gamma delta
The IKMC web portal: a central point of entry TCR genes. Cell 72:337348
to data and resources from the International 26. OGorman S, Dagenais NA, Qian M et al
Knockout Mouse Consortium. Nucleic Acids (1997) Protamine-Cre recombinase transgenes
Res 39:D849D855 efficiently recombine target sequences in the
16. Skarnes WC, Rosen B, West AP et al (2011) A male germ line of mice, but not in embryonic
conditional knockout resource for the genome- stem cells. Proc Natl Acad Sci USA
wide study of mouse gene function. Nature 94:1460214607
474:337342 27. Tucker KL, Wang Y, Dausman J et al (1997) A
17. Guan C, Ye C, Yang X et al (2010) A review of transgenic mouse strain expressing four drug-
current large-scale mouse knockout efforts. selectable marker genes. Nucleic Acids Res
Genesis 48:7385 25:37453746
 Part III

Methods in Measuring Neural Responses to Pheromones

 Chapter 11

Electroantennogram and Single Sensillum

Recording in Insect Antennae
Shannon B. Olsson and Bill S. Hansson

Abstract
Electrophysiology is an invaluable technique to quickly and quantitatively assess the response of the olfactory
system to odor stimuli. For measuring the response of the insect antenna, two basic techniques exist, elec-
troantennography and single sensillum recording. Here, we describe the general practice of both methods
in terms of equipment used, insect preparation, recording technique, and basic analysis.

Key words Invertebrate, Olfaction, Electrophysiology, Odor stimulation, GC-EAD, EAG, SSR

1 Introduction

For most insects, the chemical senses are indispensable for survival.
Insects rely on chemical cues to locate food, hosts, predators, kin,
and, of course, potential mates. It is perhaps, then, no surprise that
the first pheromone identified was the sex attractant of the female
silkworm moth, Bombyx mori, in 1959 [1].
Given that electrical potentials are the modus operandi for
transmitting information within the nervous system, electrophysi-
ology provides an incredibly efficient and effective means for
assessing chemical signaling. In insects, the primary sensory
structure for detecting volatile chemical cues is the antenna.
Indeed, the first physiological recordings from the antenna
occurred nearly simultaneously with the identification of the first
pheromone [2, 3]. From that time until now, electrophysiological
studies have been invaluable for detecting, identifying, and assess-
ing pheromone communication.
Insect antennae vary widely in structure, from the long pecti-
nate (feathered) antenna of the silkworm moth to the lamellate
club antenna of scarab beetles (see descriptions in [4]). Adult ptery-
gote (winged) insect antennae have three sections: the basal scapus,
attached to the head capsule, the pedicellus, housing the Johnstons

Kazushige Touhara (ed.), Pheromone Signaling: Methods and Protocols, Methods in Molecular Biology, vol. 1068,
DOI 10.1007/978-1-62703-619-1_11, Springer Science+Business Media, LLC 2013

157
158 Shannon B. Olsson and Bill S. Hansson

Fig. 1 Electrophysiology setup showing basic equipment, clockwise from bottom

center: (a) anti-vibration table, (b) grounding block with ground wires, (c) manual
micromanipulator with ground electrode, (d) microscope with long working distance
objectives and stage, (e) odor stimulus outlet with tubing, (f) motorized microma-
nipulator with recording electrode and amplifier, (g) Faraday cage, (h) motorized
micromanipulator control board

organ, and the flagellum housing the majority of the olfactory sen-
sory neurons (OSNs) [4]. The OSNs are housed in cuticular struc-
tures called sensilla, which also range from long hair-like sensillum
trichodea to plate-like sensillum chaetica. These sensilla house one
or several neurons depending on species and sensillum type (up to
50 in some basiconic sensilla of locusts [5]). It is thus the flagellum
that is the focus of most olfactory studies.
Despite the incredible morphological diversity of insect anten-
nae and the large body of electrophysiological studies over the past
50 years, the basic equipment required for electrophysiological
recording has remained relatively unchanged from Schneiders initial
experiments ([6], Fig. 1). The basic setup consists of: conductive
metal electrodes, an amplifier, an anti-vibration table, a microscope
with long working distance objectives, and a Faraday cage to
reduce external electrical noise. Likewise, there are mainly two
types of antennal recording, both of which were developed over 50
years ago: electroantennography [3], which records electrical activ-
ity across the antennal flagellum, and single sensillum recording
[7, 8], which measures the electrical activity within single sensilla.
Most current methods are variations of these two techniques.
Electroantennography (EAG) is an excellent technique for
quickly assessing the receptive range of an insects antenna. EAGs
Electroantennogram and Single Sensillum Recording in Insect Antennae 159

are performed by measuring the change in electrical potential

between distal and proximal sections of the antennal flagellum
caused by olfactory stimulation. EAGs are generally believed to
measure the sum of electrical potentials created by activated OSNs
on the insect antenna (see review in [9]). Kaissling [10] suggested
that the EAG is likely the summed potential of several OSNs lying
in series, and indeed the amplitude of the response potential to
pheromones in Ostrinia nubilialis was directly proportional to
antennal length [11]. Nagai further surmised that the EAG signal
is a result of the initial fast negative drop in potential caused from
OSNs situated between the recording and ground electrodes com-
bined with the slower electrotonic potential spreading from neigh-
boring regions of the recording electrode [11]. The electrotonic
spread also occurs more effectively in proximal, rather than distal
directions of the antenna [12]. Nevertheless, the nature of the EAG
potential has been questioned in antennal flagella whose OSNs do
not lie in series (e.g., flies [6, 13]). In addition, EAG amplitude is
subject to change depending on connection strength, insect vitality,
and even position of the electrode (e.g., [13, 14]). Although EAGs
show a concentrationresponse relationship with stimulus concen-
tration [9], the EAG response should in general be treated as a qual-
itative, rather than quantitative indicator of olfactory reception.
For more quantitative measurement of insect olfactory
response, single sensillum recording (SSR) is suggested. SSR is a
form of extracellular electrophysiology, where action potentials
generated by OSNs within single sensilla on the insect antenna can
be measured by an electrode in contact with the extracellular recep-
tor lymph. Individual OSNs can often be separated by action
potential amplitude, which is believed to result from the diameter
of the nerve cell generating the signal [15]. In this manner, detailed
assessment of the sensitivity and selectivity of individual OSNs can
be obtained. As OSNs expressing the same receptor type converge
onto the same region of the insect brain [16, 17], single sensillum
electrophysiology can be a very effective technique for mapping
the receptive range of the olfactory receptor proteins expressed
within the OSNs [18, 19].
From these two basic techniques, a variety of additional analy-
ses can be performed. For example, both electroantennography
(reviewed in [20]) and single-sensillum recording (reviewed in
[21]) can be combined with gas chromatography to measure
antennal responses to natural headspace extracts. The two tech-
niques can also be combined for simultaneous recording [14].
Sensillum incision allows manipulation of the internal sensillum
environment during SSR [2224]. In the absence of tissue inci-
sion, diffusion [25, 26] or the recently established microinjection
technique [27] can allow the study of pharmacological effects on
OSN response kinetics.
160 Shannon B. Olsson and Bill S. Hansson

In this chapter, we outline the basic methodology for

establishing EAG or SSR. As electrode placement and insect prep-
aration vary widely with chosen species (see included figures), we
attempt a general overview of important considerations rather than
a description of specialized techniques. As with all electrophysiol-
ogy, effective recording requires practice and experience with the
chosen organism. We consider this chapter a primer to introduce
the reader to the technique. However, researchers wishing to
perform these analyses for the first time are highly encouraged to
practice under supervision of an experienced electrophysiologist
who can give additional tips specific to the species being assessed.

2 Materials

2.1 Insect 1. Animals: Use healthy, young insects with undamaged antenna.
Preparation Generally, insects older than 1 day and less than 1 week gener-
ate the best electrophysiological signals.
2. Mounting equipment: For excised antennae: electrode gel or
glass capillary electrodes filled with hemolymph ringer (see
Subheading 2.2). For whole animal preparations: microscope
slide, soft dental wax or modeling clay, parafilm, double-sided
tape, glass capillaries, magnetic stand, test tube clamp.
3. Other materials: For long-term preparations (particularly for
single-sensillum recording), moistened cotton or paper towels
can be used on the body of the preparation to maintain
humidity.

2.2 Electrodes 1. Electrodes: conductive metal (usually tungsten wire, silver wire
and/or two-pronged fork electrode; choice depends on pro-
cedure, see Subheading 3), glass capillaries with filament,
laboratory tissue.
2. Electrode sharpening/etching device: For glass: micropipette
puller (e.g., Sutter Instrument Co). For tungsten: approxi-
mately 10 V voltage source, electrolytic solution (e.g., satu-
rated potassium nitrite, KNO2), cathode (carbon electrode),
manual micromanipulator.
3. Hemolymph Ringer: 6.4 mM KCl, 20 mM KH2PO4, 12 mM
MgCl2, 1 mM CaCl2, 9.6 mM KOH, 354 mM glucose, 12 mM
NaCl, Osmotic pressure 450, pH 6.5 after [28, 29].
4. Sensillum lymph ringer: 171.9 mM KCl, 9.2 mM KH2PO4,
10.8 mM K2HPO4, 3 mM MgCl2, 1 mM CaCl2, 1.5 mM HCl,
22.5 mM glucose, 25.0 mM NaCl, Osmotic pressure 475,
pH 6.5 after [28].
5. Electrode holders (e.g., from Ockenfels Syntech GmbH; see
Note 1).
 Electroantennogram and Single Sensillum Recording in Insect Antennae 161

2.3 Electro- 1. Air/anti-vibration table (recommended but not necessary for

physiological well-isolated environments or EAG in most cases).
Recording Equipment 2. Faraday cage (recommended but not necessary for well-isolated
environments or EAG in most cases).
3. Upright compound binocular microscope with up to 1,000
magnification and as long working distance as possible (e.g.,
Olympus BX-51 with 4, 10, and 50 objectives, Fig. 1).
4. Adjustable light source (preferably both regular and epi-
luminescence sources).
5. Manual (e.g., Narshige) and/or motorized micromanipulators
(e.g., Mrzhuser-Wetzlar PM-10 piezo micromanipulator; see
Note 2).
6. Magnetic or other appropriate mounting stands for manipulators.
7. Insulated wire with alligator clips and/or clamps for grounding.
8. Extracellular 10100 amplifier with high impedance (1012
[9]; e.g., Syntech universal AC/DC 10 probe will perform
both EAG and single sensillum recording; see Note 3).
9. Signal acquisition and analog to digital converter with computer
inputs (e.g., Syntech IDAC systems).
10. Audio output (for single sensillum recording only).
11. Desktop or laptop computer and peripherals.
12. Signal acquisition software (e.g., Autospike by Syntech).

2.4 Odor Stimulation 1. Filtered air source (e.g., compressed air with pressure regulator,
see Note 4).
2. Humidifier (e.g., gas washing bottle filled with distilled water).
3. 3-way directional-control valve (e.g., Lee solenoid control
valves) with controller.
4. Silicon tubing, between 6 and 10 mm I.D. (see Note 5).
5. Y or T-shaped silicon connectors.
6. 05 L/min adjustable gas flow meter.
7. Stimulus outlet (stainless steel or glass tube roughly 10 cm
long, 610 mm I.D., with small hole roughly 1 cm from end
large enough for Pasteur pipette tip to pass through).
8. Clamps, stand, or other holders to support outlet within 1 cm
or preferably closer to antenna.
9. Disposable glass Pasteur pipettes.
10. Filter paper cut into 1 cm diameter disks.
11. Appropriate volatiles dissolved preferably in mineral oil (see
Note 6).
162 Shannon B. Olsson and Bill S. Hansson

3 Methods

3.1 Insect 1. For many insects (e.g., with large and/or filiform antenna), an
Preparation excised antenna can be used for recording (see Notes 7 and 8).
3.1.1 For Excised 2. Remove the antenna at the base, making sure not to pinch or
Antennae close the antennal shaft.
3. Immediately attach antenna to each side of a two-pronged
electrode fork recording probe (base of antenna at the
ground position) using a small amount of electrode gel (see
Note 9, Fig. 2c). Alternatively, two glass micropipettes with
Ag/AgCl electrodes can be used (see Subheading 3.5).
4. Make sure that the antenna is straight, but not stretched, between
the two prongs/electrodes. If necessary, the antennal tip can
be cut or opened if insufficient electrical contact is obtained.

3.1.2 For Whole Animal 1. Immobilize the insects abdomen and thorax with parafilm.
Preparations Smaller insects can be placed headfirst into a disposable pipette
tip cut so that the antennae and eyes can protrude (see Notes
10 and 11).
2. Option 1 (for single sensillum or small working distances): Fix
the prostrate insect onto a microscope slide with a small
amount of dental wax or modeling clay. For many insects, place-
ment with the ventral side exposed provides the best access to
antennal sensilla.
3. Option 2 (for EAG or large working distances): Fix the insect
upright directly under the microscope objective via magnetic
stand and test tube clamp. Make sure both antennae and eyes
are accessible.
4. Immobilize antenna either with a small amount of wax or clay
or with a glass capillary placed gently over the antennal shaft.
Do not cover antenna completely or restrict airflow to antennal
surface.

Fig. 2 Sample electrodes. (a) Sharpened tungsten microelectrode, (b) sharpened glass micropipette with
filament, (c) two-pronged fork electrode for EAG of excised antennae, prepared for recording with a Heliothis
virescens antenna
 Electroantennogram and Single Sensillum Recording in Insect Antennae 163

3.2 Electrode 1. For Tungsten Electrodes: Cut tungsten wire of appropriate

Preparation length for electrode holder. Assemble an electrolytic sharpener
by obtaining a 10 V electrical source and connecting the posi-
tive via insulated wire to a carbon cathode. Insert the cathode
into a saturated potassium nitrite (KNO2) solution and place the
solution under a dissecting microscope (see Note 12). Connect
the negative to the tungsten electrode placed in an appropriate
holder and turn on the electrical source at lowest voltage setting.
Slowly dip the electrode into the solution. Increase voltage as
needed to produce bubbles within the solution when contact
with tungsten is made, indicating that the nitrite is reacting
with the tungsten metal. Repeated movement of the electrode
into and out of the solution will produce a fine tip (approxi-
mately 1 m tip; Fig. 2; see Note 13). Unless using a high-
power objective, electrode tip size will not be apparent. Rinse
electrodes with water before use (the nitrite is toxic). Make sure
electrodes are completely dry before use to avoid electrical
shorts. For immediate use, rinse electrode with ethanol and
acetone for quick drying. Be careful to avoid touching the tip.
2. For Silver Electrodes: Silver wire of appropriate size for inser-
tion into the glass micropipette shaft should be cleaned and cut
to appropriate length. In most cases, the tip of the silver elec-
trode should also be chloridized prior to use. Chloridization
(i.e., electrochemical reaction with HCl) produces a silversil-
ver chloride (Ag/AgCl) electrode that is non-polarizable and
exhibits reduced low-frequency noise. This process can be per-
formed electrolytically in a manner similar to tungsten sharp-
ening (see step 1). In this case, an AA battery provides enough
electricity for chloridization. Connect the positive end of the
battery to a carbon cathode in 0.1 N HCl, and the negative
end to the silver wire itself. Leave the silver wire in the solution
with the cathode for 0.52 h until wire turns black indicating
proper chloridization (see Note 14). This process should be
repeated periodically to maintain proper chloridization. Ag/
AgCl electrodes require a salt bridge such as a saline solution
(see Subheading 2.2), and should therefore be placed in glass
micropipettes filled with saline.
3. For Glass Micropipettes: Using an appropriate Flaming-Brown
or laser pipette puller, follow manufacturers instructions to
pull short tapered electrodes of approximately 23 M resis-
tance (see Note 15; Fig. 2b). For EAG, gently break electrode
tips by lightly tapping electrode tip against a laboratory tissue.

3.3 Electro- 1. Fix microscope in center of anti-vibration table (if used) or on

physiological a surface in a relatively electrically isolated area away from large
System Setup electrical equipment (e.g., freezers, refrigerators). If possible,
use an isolated circuit for all setup equipment.
2. Setup Faraday cage (if needed) around table and connect to
central ground with grounding wire (see Step 5).
164 Shannon B. Olsson and Bill S. Hansson

3. Place two micromanipulators (one for ground and one for

recording) on either side of microscope. Place recording
manipulator on same side that signal acquisition equipment
will be placed.
4. Attach recording amplifier/probe/holder to recording manip-
ulator and ground holder to ground manipulator.
5. Connect an insulated grounding wire (with clips or hooks as
needed) from ground holder to a central ground plate that is
earthed (see Note 16).
6. Connect the recording electrode probe/amplifier to the signal
acquisition system and/or analog-digital converter.
7. Connect signal acquisition system to computer.
8. Install appropriate signal acquisition software for your chosen
amplifier/system (e.g., Autospike or EAG software for Syntech
systems).
9. Connect audio output to audio port on signal acquisition sys-
tem or computer as needed.
10. Confirm that all signal acquisition hardware and software is
functional and installed correctly. An oscilloscope can alterna-
tively be used for system check.
11. To test the system, connect appropriate electrodes (e.g., Ag/
AgCl wire in sharpened glass micropipettes filled with saline)
to both ground and recording holders. Prepare a small Petri
dish of saline and carefully insert both probes into the liq-
uid. Check that a signal is obtained when both probes are in
the liquid.
12. Observe the noise level. Sources of noise can be systematically
isolated and reduced by connecting an insulated ground wire to
various locations (e.g., microscope, Faraday cage, light source)
until the noise is sufficiently reduced. Make sure to connect all
ground wires to a central ground terminal and use as few
grounding wires as possible. Too many wires can create ground
loops and actually increase noise (see Notes 17 and 18).

3.4 Odor 1. Securely connect pressure-regulated, filtered air source to Y or

Stimulation Setup T shaped connector.
2. Connect silicon tubing of appropriate size between one outlet
of the connector and valve inlet.
3. Connect silicon tubing (at least 610 mm ID) from other out-
let of connector to humidifier filled with distilled water, and
then to flow meter regulated to 0.51.0 L/min.
4. Mount stainless steel or glass stimulus outlet so that end is as
close to the recording site as possible (see Figs. 1, 3 and 4).
Make sure that small hole used for odor presentation is easily
accessible.
 Humidifier Flow meter

Constant
Flow

Auxiliary
Flow
Filtered Air
Source

3-way
motorized Stimulus
directional Flow
control
valve
Antenna

Fig. 3 Overview of odor stimulation equipment and setup. Arrows indicate direction of airflow

Fig. 4 Electroantennography at different scales. Electroantennography with Manduca (a, b, Hawk moth) and
Drosophila (c, d, fruit fly). G and R indicate ground and recording electrode placement, respectively. Both
preparations are undergoing gas chromatography coupled with electroantennographic detection (GC-EAD).
Images kindly provided by Dr. Andreas Reinecke and Jeanine Linz
166 Shannon B. Olsson and Bill S. Hansson

5. Connect a Y-shaped connector to opposite end of stimulus

outlet, with base of Y attached to outlet (see Notes 19 and 20,
Fig. 3).
6. Connect silicon tubing from flow meter to one arm of Y
connector at stimulus outlet.
7. Connect silicon tubing of equal length (at least 610 mm ID)
to both valve outlets and two flow meters. Adjust both flows
equally to 0.5 L/min.
8. Connect tubing of equal length from both flow meters. Make
sure tubing is long enough to reach stimulus outlet placed in
step 4.
9. Connect tube with auxiliary flow (i.e., has airflow when valve is
off ) to either the other branch of the Y-tube connector at out-
let or stationary, empty Pasteur pipette (see Note 20).
10. Connect tube with stimulus flow (i.e., has airflow when valve is
on) to stimulus pipette (see Note 21).
11. Adjust valve opening time according to manufacturers instruc-
tions (i.e., manually or via software). 0.51.0 s is a standard
stimulus presentation time.
12. Valve activation (and consequently stimulus presentation)
should be recorded along with the electrophysiological signal.
Most signal acquisition systems allow for digital input from a
trigger such a TTL pulse. Subsequent valve operation can
then trigger recording of electrophysiological responses.
Alternatively, the valve itself could be triggered by the signal
acquisition software. The choice depends on the type of
hardware used.
13. Prepare stimuli in solvent (preferably mineral oil) at behavior-
ally and/or ecologically appropriate dilutions.
14. Pipette 10 l onto filter paper circles inserted into clean, dis-
posable Pasteur pipettes (see Note 22).
15. Always prepare blank stimuli with solvent only.
16. Keep pipettes refrigerated or sealed with Parafilm or plastic tips
in a ventilated area when not in use to avoid contamination or
loss of stimuli.
17. Refill stimuli frequently after use. See [30] for details.
18. Be aware that odor concentration and timing is highly depen-
dent on stimulus presentation [31].
19. For most accurate physiological analysis, stimulus delay should
be measured between the valve and the preparation. This can
most quickly be determined by portable photoionization
detection of a volatile standard (e.g., [31, 32]). Stimulus delay
should then be subtracted from all responses if measuring
response latency.
 Electroantennogram and Single Sensillum Recording in Insect Antennae 167

3.5 Electro- 1. For two-pronged electrode: Make sure base of antenna is con-
antennography tacting ground arm of electrode (Fig. 2c). Connect electrode
to appropriate holder and amplifier. In this method, a separate
3.5.1 For Excised
ground manipulator is not necessary and should be discon-
Antennae
nected and removed to avoid additional noise. Instead, the
ground wiring should connect to the ground supplied on the
fork electrode holder.
2. For glass electrodes: Prepare two sharpened glass micropipettes
(Fig. 2b) filled with hemolymph ringer, each containing an
Ag/AgCl wire extending as far into the glass micropipettes as
possible. Break tips of micropipettes with laboratory tissue
until they are just large enough to envelop the antenna. Place
base of antenna into ground electrode. Connect electrodes to
the holders (one ground and one recording), making sure that
wire has sufficient contact with holder base. Move ground
electrode with antenna so that it is in field of view under the
microscope. Move recording electrode until it is one antennal
length from the ground electrode. Using fine forceps, gently
insert tip of antenna into recording electrode, making sure to
keep base inside recording electrode and leaving antenna
taught, but not stretched, between both points.

3.5.2 For Whole Animal 1. Mount insect preparation, either prostrate (Option 1) or
Preparations upright (Option 2) under microscope field of view.
2. Prepare and mount microelectrodes as in step 2 for excised
antenna.
3. Insert ground electrode near recording site (usually eye, Fig. 4).
4. Connect recording electrode with antennal tip (see Note 23,
Fig. 4).

3.5.3 Signal Acquisition 1. Confirm contact is made between ground, antenna, and
recording electrodes. A deflection should appear in the signal
trace when contact is made or broken.
2. Make sure the selected amplification of the signal in the acqui-
sition software matches the amplifier chosen (i.e., usually
10100).
3. Set the high pass filter to <0.5 Hz.
4. Observe the signal baseline. An unstable baseline indicates
poor electrical contact with the preparation and/or excessive
external noise. After confirming proper contact is made with
the antenna and ground (e.g., through replacement of the
electrodes), systematically test for noise as described in
Subheading 3.3, step 12.
5. A low pass filter is not generally needed for low-frequency
EAG signals.
168 Shannon B. Olsson and Bill S. Hansson

Fig. 5 Electroantennographic analysis. EAG response of an E-strain European corn borer (Ostrinia nubilalis)
excised antenna to its two major pheromone components, E-11-tetradecenyl acetate (black) and Z-11-
tetracenyl acetate (gray). Five-second EAG response traces are shown in series for four concentrations of the
two components. Line graph depicts maximum change in potential in traces displayed, dashed line indicates
blank response level

6. Present a known physiologically active stimulus to the antenna.

If no specific compounds are known, an extract/headspace
collection made of the insects food or host make good first
candidates.
7. An odor response is indicated by a sharp drop in baseline fol-
lowed by a slow decay (Fig. 5). A rise in baseline may indicate
that ground and recording electrodes are reversed. However,
some compounds are known to elicit such responses.
8. A weaker (or ideally) absent drop in baseline should be
observed for the blank stimulus.
9. Preferably, responses should produce >1 mV change in base-
line potential. However, signal strength varies with preparation
and species. For this reason, results should always be reported
in comparison to a standard response so that inter-preparation
differences due to age, health, and mounting can be avoided
(see Note 24).
10. Record stimulus responses for at least 12 s before and after
response (at least 5 s total) to allow for accurate measurement
of response.
11. Responses can be measured as change in potential between
baseline and maximum deflection. Time to maximum response
 Electroantennogram and Single Sensillum Recording in Insect Antennae 169

and/or return to baseline are also good parameters to

measure.
12. Responses should be reported as % difference from blank or
standard to avoid inter-individual differences. However, raw
mV change from baseline can also be included with standard
deviations.
13. Cut antennal preparations can generally be assessed for <2 h.
Whole animal preparations can survive for several hours if
humidity is maintained and antenna is relatively undamaged.

3.6 Single-Sensillum 1. Make sure that antenna is well stabilized and immobilize with
Electrophysiology wax, tape, or microcapillary (see Subheading 2.1).
3.6.1 For Sensillum 2. Mount insect preparation on microscope stage so that antenna
Penetration Recording is roughly parallel to and facing the recording electrode. This
positioning generally provides the best resistance for electrode
penetration of the sensillum cuticle.
3. Insert ground electrode as close to recording site as possible
(often the eye; see Note 25, Fig. 6).
4. Prepare an electrolytically sharpened tungsten electrode or
glass microelectrode (with Ag/AgCl wire) with approximately
1 m tip. Before recording, clarify under high magnification
that a sufficiently sharp tip has been created to penetrate the
sensillum cuticle (see Subheading 3.2, Note 26, and Fig. 2a).
5. Mount recording electrode on holder.
6. Check that signal acquisition equipment is on and make sure
that audio output is functional (see Note 27).
7. Slowly lower electrode to antenna. To avoid damaging the
electrode, it is recommended to first move the electrode at
least 610 mm directly above the antenna and slowly lower the
electrode to the preparation, rather than moving it horizon-
tally at the level of the preparation.
8. Move electrode a few m in front of targeted sensillum. Slowly
and carefully place electrode in base and/or shaft of sensillum
(see Note 28, Fig. 6). Careful and slow movement is required
to avoid damaging the electrode tip. Some thick or large sen-
silla may require extra force (such as via piezoelectric step
action) to allow cuticular penetration.
9. Generally, initial contact with cuticle will cause a significant
increase in signal noise, which also indicates that preparation is
both alive and properly grounded.

3.6.2 For Cut Sensillum 1. Make sure that the antenna is stabilized and immobilized with
Recording wax, tape, or microcapillary (see Subheading 2.1).
2. Prepare a glass microelectrode with Ag/AgCl wire in sensillum
lymph ringer. Tip should be <1015 m diameter.
170 Shannon B. Olsson and Bill S. Hansson

Fig. 6 Single sensillum electrophysiology across insect taxa. Recording sites indicated with white arrows. (a)
Ground and recording electrode placement in Rhagoletis pomonella (Tephritid fruit fly), (b) SSR in an ab2
Drosophila melanogaster basiconic sensillum, (c) SSR in a Tetropium fuscum basiconic sensillum (brown
spruce longhorn beetle, courtesy of Colin MacKay), (d) SSR in a male Ostrinia nubilalis trichoid pheromone
sensillum

3. Mount recording electrode on holder (see Note 29).

4. Carefully cut a few targeted sensilla using fine forceps [33],
glass knives [34], or vibrating glass stylets [35, 36].
5. To prevent sensillum from sealing closed, Vaseline can be used
on forceps, knives, or stylet. Polyvinylpyrrolidone can also be
added to the electrode ringer to prevent leakage (see ref. 37 for
details).
6. Immediately move recording electrode so that it sheathes a single
cut sensillum tip. Do not cover sensillum completely.
7. If no electrical contact is apparent, make sure sensillum and
electrode are not sealed and in contact. In addition, be careful
that electrode tip is not too large.
 Electroantennogram and Single Sensillum Recording in Insect Antennae 171

3.6.3 Signal Acquisition 1. Proper contact with the extracellular OSN space will reveal
spontaneous action potential activity. In some preparations
(particularly pheromone sensilla), very low spontaneous activ-
ity will occur.
2. If significant noise is present, first confirm that appropriate
contact is made with internal sensillum lymph. If necessary,
reposition electrode. Additional noise may indicate improper
grounding (see Subheading 3.3).
3. Set the high pass filter between 30100 Hz and the low pass at
1,0003,000 Hz. Filters can be adjusted to reduce noise, and
an additional 50/60 Hz filter can be used to reduce noise from
electronic equipment (see Note 30).
4. Action potentials should be easily distinguished from back-
ground noise (Fig. 7). Electrode replacement and/or contact-
ing other sensilla can improve signal strength.
5. Many insects house more than one sensory neuron in a single
sensillum. These neurons can often be distinguished by spon-
taneous top-to-top action potential amplitude, which should
remain relatively consistent in a stable recording (Fig. 7).
Nevertheless, some sensilla housing many neurons (e.g., ants,
locusts) or sensory neurons of similar size may be indistin-
guishable. In such cases, only sensillum response, rather
than single cell response, should be reported.
6. Present a known physiologically active stimulus to the antenna.
If no specific compounds are known, an extract/headspace
collection made of the insects food or host make good first
candidates.
7. A response is indicated by a sharp increase (or decrease) in
action potential activity.
8. A weaker (or ideally) absent change in spontaneous activity
should be observed for the blank stimulus.
9. Record stimulus responses for at least 12 s before and after
response (at least 5 s total) to allow for accurate measurement
of change in spike activity.
10. Use appropriate spike sorting software to separate sensory
neurons by amplitude (Fig. 7).
11. Responses can be measured in several ways. Responses can be
recorded as raw spike count (in Hz) during stimulation, change
in spike count with spontaneous activity, or through a peristimulus
time histogram (instantaneous spike activity over time).
12. Whole animal preparations can survive for several hours (or
even days) if humidity is maintained and antenna is relatively
undamaged. Single sensillum contacts can also be maintained
for several hours.
172 Shannon B. Olsson and Bill S. Hansson

Fig. 7 Single sensillum electrophysiology analysis. Action potential (a) and receptor potential (b) activity in a
single trichoid pheromone sensillum of a male silkworm moth, Bombyx mori. Traces were measured in AC
mode (a, 30/3,000 Hz filter settings, 2 mV display) and DC mode (b, DC/3,000 Hz filter settings, 5 mV display)
using an Ag/AgCl glass electrode penetrating the base of the sensillum near the cell bodies. (Left) Five-second
traces with stimulus window as gray bar. Top traces in each panel show the response to the main pheromone
component, bombykol, and bottom traces to the minor component, bombykal. (Right, top) Separation of the
large, bombykol-responsive OSN and smaller, bombykal-responsive OSN by amplitude. (Right, bottom)
measurement of receptor potential from beginning of stimulus onset to maximum change in potential (mV)

4 Notes

1. Holders should be chosen to interface properly with probe and

amplifier used. Syntech (http://www.syntech.nl) provides a
selection of compatible holders and probes/amplifiers.
2. For EAG and ground electrode placement, manual microma-
nipulators are sufficient. For single sensillum recording, a
motorized micromanipulator is strongly recommended for
positioning the recording electrode.
3. For best signal to noise ratio, amplifier should be placed as
close to recording site as possible.
Electroantennogram and Single Sensillum Recording in Insect Antennae 173

4. An all-in-one stimulus system can be purchased that gener-

ates, regulates, and controls airflow (e.g., Syntech stimulus
controller). Such a system can be advantageous for setups that
do not have access to compressed air.
5. Make sure to use tubing made of a relatively inert material,
such as silicon. Nevertheless, all tubing will require occasional
solvent cleaning (e.g., ethanol, water) and heat drying at
80100 C for 4 h (as allowed by tubing material) to prevent
contamination.
6. Odor cartridge preparation and maintenance is essential to
avoid nonspecific responses due to contamination, and/or
inaccurate concentrationresponse measurement due to
depleted odor sources.
7. Excised antennae provide rapid and efficient screening for elec-
troantennography. However, the preparation is short-lived
(<2 h), and is not generally practical for long-term single-
sensillum recording (but see, e.g., [22]).
8. Many insects such as moths are amenable to excised antennal
recording. If quick qualitative results are needed on a species
new to electrophysiology, excised antenna can be tested before
attempting whole animal mounting.
9. Be careful not to add too much electrode gel to an excised
antennal prep, which can reduce contact with the electrode.
Add only enough gel to ensure proper adhesion to the elec-
trode surface.
10. Ensure that all appendages are secure. Leg, wing, and obvi-
ously antennal movement can cause significant noise during
recording.
11. If using disposable pipettes to immobilize the animal, make
sure to cut the pipette tip only enough to allow the head to
protrude and not the prothorax. In many cases, only part of
the head needs be exposed to allow placement of both record-
ing and ground electrodes. Partial head exposure also helps
prevent movement.
12. Generally, it is easier to place the nitrite solution horizontally
under the microscope in, for example, a plastic syringe. 10 %
sodium nitrite (NaNO2) or 1 M potassium hydroxide (KOH)
can also be substituted for potassium nitrite as the electrolytic
solution.
13. Electrode sharpening takes practice to acquire the most effi-
cient technique. The most favorable shape for an electrode is
pencil-likea broad electrode coming to a sharp and fine point
at the end (Fig. 2). This creates a strong and sturdy electrode
for penetration into the insect cuticle. This shape can be made
by slow and deep movements at the beginning of sharpening
174 Shannon B. Olsson and Bill S. Hansson

to begin to narrow the tip, followed by small and quick move-

ments just at the tip to create a short and fine point.
14. Silver wires chloridized for too long will become brittle.
15. In general, short electrode tips are more sturdy for both single
sensillum and EAG analyses. However, be careful not to make
electrodes too short as their tips will quickly become too large
following inevitable breakage from cuticular contact.
16. Anti-vibration tables are generally grounded internally, so cen-
tral connection to the tabletop is sufficient. If using a bench
top or table, connect the grounding wire to a stainless steel
grounding plate that is then earthed.
17. Essentially, only one grounding wire is neededfrom the
ground electrode to the earth. All other wires are simply to
reduce electrical noise created by electronic and metal equip-
ment. Be very frugal with the use of grounding wires and in
general, a working rig should not need re-grounding unless
new equipment is added.
18. Despite best efforts, some preparations do not produce suffi-
cient signal-to-noise ratio due to unforeseen physiological or
environmental conditions. If no signal is obtained, first con-
firm that a complete circuit from the recording electrode
through the animal to the ground is present. Then, clean and
dry all electrode holders to remove potential salt, water or
debris buildup. After retesting setup with saline (see
Subheading 3.3, step 11), confirm that animal vitality is main-
tained. If there is still no signal, establish another animal
preparation.
19. To avoid backflow, it is best to orient the Y connector so that
the two airflows proceed in the same direction (as in both short
arms of the Y-connector). Also, the main airflow through the
humidifier should be much higher than the auxiliary flow to
avoid backflow during valve operation.
20. As an alternative to the Y connector, an additional hole can be
drilled in the stimulus outlet itself next to the stimulus inlet
hole for the auxiliary airflow to pass through an empty Pasteur
pipette fixed permanently at the outlet. This, in fact, produces
a better auxiliary flow as it replicates the stimulus flow more
precisely, and is closer to the stimulus inlet. However, as in
Note 19, make sure to angle the pipette so that its airflow
proceeds in the same direction as the main flow, and is at a
lower flow rate (generally 0.5 L/min).
21. Be aware that the line proceeding to the stimulus cartridges
can quickly become contaminated. To avoid this, disposable
connectors can be made from 1,000 L disposable plastic
pipette tips cut slightly at tip with base connected to the stimulus
Electroantennogram and Single Sensillum Recording in Insect Antennae 175

pipette base. In addition, be careful not to allow any chemicals

to contact the stimulus pipette base. See also Note 5.
22. Pipette preparation and maintenance is very important for
accurate physiological assessment.
23. For many flies and moths, the antennal tip can be placed into
the recording electrode with no additional tissue incision.
However, for insects with particularly thick cuticle (such as
beetles), a small incision can be made at the antennal tip to
allow sufficient contact with internal lymph for recording
electrical potentials.
24. There is no rule for what is a sufficient EAG signal. The
major determining factor is signal-to-noise ratio. If your stimu-
lus response is clearly distinguishable from both noise and the
blank, then you have a sufficient signal.
25. Generally, it is best to use the same material for both ground
and recording electrodes. Thus, if recording with tungsten,
use a tungsten electrode as the ground.
26. Electrode shape and sharpness depend heavily on the prepara-
tion. Animals with large and thick sensilla may require much
larger and thicker electrode tips. Generally, proper electrode
shape is determined through experience with the preparation.
27. Many researchers agree that it is very beneficial to hear the
sensory neuron activity during single sensillum recording.
Audio output from signal acquisition allows the researcher to
focus on the preparation during electrode placement rather
than a computer screen. Proper contact will produce the famil-
iar popping sound caused by spontaneous neural activity.
28. Obtaining electrical contact with the extracellular sensory neu-
ron space initially requires tremendous patience and practice.
Prepare for many bent and damaged electrodes until you
obtain the fine motor skills required for accurate and precise
electrode placement. In time, you will be able to obtain many
contacts without requiring electrode sharpening.
29. Cut tip and sensillum penetration require amplifiers with dif-
ferent impedances. In the authors experience, Syntech univer-
sal probes are not suitable for cut sensillum recording.
30. As an alternative to measuring action potential activity, the
receptor potential (i.e., the negative deflection of the tran-
sepithelial potential of the sensillum; [38]) can be measured
using an Ag/AgCl electrode in a sharpened glass micropi-
pette to contact the internal sensillum lymph (Fig. 7b). To
measure receptor potentials, the high pass filter should be set
to DC mode.
176 Shannon B. Olsson and Bill S. Hansson

Acknowledgments

This work was supported by the Max Planck Society. Special

thanks to Dr. Andreas Reinecke, Jeanine Linz, and Colin MacKay
for images.

References
1. Butenandt A, Beckmann R, Stamm D, Hecker 12. Nagai T (1983) Spread of local electroanten-
E (1959) ber den sexual-lockstoff des seiden- nogram response of the European corn borer,
spinners Bombyx mori. Z Naturforshg Ostrinia nubilalis. Pestic Biochem Physiol
14b:283284 19:291298
2. Schneider D (1957) Electrophysiological 13. Crnjar R, Scalera G, Liscia A, Angioy A-M,
investigation on the antennal receptors of the Bigiani A, Pietra P, Barbarossa IT (1989)
silk moth during chemical and mechanical Morphology and EAG mapping of the anten-
stimulation. Experientia 13:8991 nal olfactory receptors in Dacus oleae. Entomol
3. Schneider D (1957) Elektrophysiologische Exp Appl 51:7785
untersuchungen von chemorezeptoren und 14. Nagai T (1983) On the relationship between
mechanorezeptoren der antenne des seidens- the electroantennogram and simultaneously
pinners Bombyx mori-L. Z Vergl Physiol recorded single sensillum response of the
40:841 European corn borer Ostrinia nubilalis. Arch
4. Keil TA (1999) Morphology and development Insect Biochem Physiol 1:8591
of the peripheral olfactory organs. In: Hansson 15. Hansson B, Hallberg E, Lfstedt C,
BS (ed) Insect olfaction. Springer-Verlag, Steinbrecht R (1994) Correlation between
Berlin, p 457 dendrite diameter and action potential ampli-
5. Ochieng SA, Hallberg E, Hansson BS (1998) tude in sex pheromone specific receptor neu-
Fine structure and distribution of antennal rons in male Ostrinia nubilalis (Lepidoptera:
sensilla of the desert locust, Schistocerca Pyralidae). Tissue Cell 26:503512
gregaria (Orthoptera: Acrididae). Cell Tissue 16. Vosshall LB, Wong AM, Axel R (2000) An
Res 291:525536 olfactory sensory map in the fly brain. Cell
6. Ignell R, Hansson B (2004) Insect olfactory 102:147159
neuroethology: an electrophysiological per- 17. Gao Q, Yuan B, Chess A (2000) Convergent
spective. In: Christensen TA (ed) Methods in projections of Drosophila olfactory neurons to
insect sensory neuroscience (frontiers in neu- specific glomeruli in the antennal lobe. Nat
roscience). CRC Press, Boca Raton, p 435 Neurosci 3:780785
7. Schneider D, Boeckh J (1962) 18. Hallem E, Ho M, Carlson J (2004) The
Rezeptorpotential und nervenimpulse einzel- molecular basis of odor coding in the
ner olfaktorischer sensillen der insektenan- Drosophila antenna. Cell 117:965979
tenne. Z Vergl Physiol 45:405412 19. Hallem E, Carlson J (2006) Coding of odors
8. Boeckh J (1962) Elektrophysiologische unter- by a receptor repertoire. Cell 125:143160
suchungen an einzelnen geruchsrezeptoren auf 20. Struble DL, Arn H (1984) Combined gas chro-
den antennen des Totengrabers (Necrophorus, matography and electroantennogram recording
Coleoptera). Z Vergl Physiol 46:212248 of insect olfactory responses. In: Hummel HE,
9. Roelofs WL (1984) Electroantennogram Miller TA (eds) Techniques in pheromone
assays: rapid and convenient screening proce- research. Springer-Verlag, Berlin, p 464
dures for pheromones. In: Hummel HE, 21. Wadhams LJ (1984) The coupled gas
Miller TA (eds) Techniques in pheromone chromatography-single cell recording tech-
research. Springer-Verlag, Berlin, p 464 nique. In: Hummel HE, Miller TA (eds)
10. Kaissling K-E (1971) Insect olfaction. In: Techniques in pheromone research. Springer-
Beidler LM (ed) Handbook of sensory physi- Verlag, Berlin, p 464
ology, vol IV, Chemical senses 1, olfaction. 22. Kaissling K, Keil T, Williams J (1991)
Springer-Verlag, Berlin, p 518 Pheromone stimulation in perfused sensory
11. Nagai T (1981) Electroantennogram response hairs of the moth Antheraea polyphemus. J
gradient on the antenna of the European corn Insect Physiol 37:7175
borer, Ostrinia nubilalis. J Insect Physiol 23. van den Berg M, Ziegelberger G (1991) On
27:889894 the function of the pheromone binding protein
 Electroantennogram and Single Sensillum Recording in Insect Antennae 177

in the olfactory hairs of Antheraea polyphemus. 30. Andersson MN, Schlyter F, Hill SR, Dekker T
J Insect Physiol 37:7985 (2012) What reaches the antenna? How to
24. Pophof B (2004) Pheromone-binding pro- calibrate odor flux and ligand-receptor affini-
teins contribute to the activation of olfactory ties. Chem Senses 37:403420
receptor neurons in the silkmoths Antheraea 31. Vetter RS, Sage AE, Justus KA, Card RT,
polyphemus and Bombyx mori. Chem Senses Galizia CG (2006) Temporal integrity of an
29:117125 airborne odor stimulus is greatly affected by
25. Laughlin J, Ha T, Jones D, Smith D (2008) physical aspects of the odor delivery system.
Activation of pheromone-sensitive neurons is Chem Senses 31:359369
mediated by conformational activation of 32. Olsson SB, Kuebler LS, Veit D, Steck K,
pheromone-binding protein. Cell 133: Schmidt A, Knaden M, Hansson BS (2011) A
12551265 novel multicomponent stimulus device for use
26. Flecke C, Dolzer J, Krannich S, Stengl M in olfactory experiments. J Neurosci Methods
(2006) Perfusion with cGMP analogue adapts 195:19
the action potential response of pheromone- 33. Kaissling K-E (1974) Sensory transduction in
sensitive sensilla trichoidea of the Hawkmoth insect olfactory receptors. In: Jaenicke L (ed)
Manduca sexta in a daytime-dependent man- Biochemistry of sensory functions. Springer,
ner. J Exp Biol 209:38983912 Berlin, p 641
27. Olsson SB, Getahun MN, Wicher D, Hansson 34. van der Pers JNC, den Otter CJ (1978) Single
BS (2011) Piezo controlled microinjection: an cell responses from olfactory receptors of small
in vivo complement for in vitro sensory studies ermine moths to sex-attractants. J Insect
in insects. J Neurosci Methods 201:385389 Physiol 24:337343
28. Kaissling K-E, Thorson J (1980) Insect olfac- 35. Gdde J (1989) Vibrating glass stylets: Tools
tory sensilla: structural, chemical and electrical for precise microsurgery on cuticular struc-
aspects of the functional organization. In: tures. J Neurosci Methods 29:7783
Sattelle DB, Hall LM, Hildebrand JG (eds) 36. Hillier NK, Kleineidam C, Vickers NJ (2006)
Receptors for neurotransmitters, hormones, Physiology and glomerular projections of
and pheromones in insects: Proceedings of the olfactory receptor neurons on the antenna of
workshop in neurotransmitter and hormone female Heliothis virescens (Lepidoptera:
receptors in insects held in Cambridge, 1012 Noctuidae) responsive to behaviorally relevant
September, 1979. Elsevier/North Holland odors. J Comp Physiol A 192:199219
Biomedical Press, Amsterdam, p 310 37. Kaissling K, Meng L, Bestmann H (1989)
29. Kaissling K-E (1995) Single unit and electro- Responses of bombykol receptor-cells to (Z,
antennogram recordings in insect olfactory E)-4,6-hexadecadiene and linalool. J Comp
organs. In: Spielman AI, Brand JG (eds) Physiol A 165:147154
Experimental cell biology of taste and olfac- 38. Kaissling K (1986) Chemo-electrical transduc-
tion: Current techniques and protocols. CRC tion in insect olfactory receptors. Annu Rev
Press, Boca Raton, p 464 Neurosci 9:121145
 Chapter 12

Calcium Imaging of Pheromone Responses in the Insect

Antennal Lobe
Susy M. Kim and Jing W. Wang

Abstract
Calcium imaging is a powerful technique that permits the visual monitoring of neural responses to
pheromones and other odors in large ensembles of neurons. Here, we describe a method that permits the
monitoring of Drosophila antennal lobe responses to odors using the genetically encoded calcium monitor
GCaMP.

Key words Drosophila melanogaster, Antennal lobe, cis-Vaccenyl acetate (cVA), GCaMP, Olfactory
system, Calcium imaging, Pheromone

1 Introduction

Elucidating how neural circuits process pheromone information is

crucial towards understanding the role of chemical communication
in the social interactions between conspecific insects. While elec-
trophysiological methods permit monitoring of neural activity with
unparalleled resolution, they are labor intensive and not amenable
to monitoring large groups of specific neuronal types. Continuous
technical improvements of genetically expressed calcium sensors
and other activity reporters in recent years [13], however, bypass
some of these problems. A genetic approach allows one to easily
target calcium sensors to specific populations of neurons [4, 5],
thus making calcium imaging a useful approach for monitoring
pheromone responses at different levels of the fly brain.
Two-photon calcium imaging of odor responses in the anten-
nal lobe is a useful technique in the study of how odor information
is processed at this second relay in the insect nervous system [4].
In calcium imaging studies of antennal lobe odor processing in
Drosophila melanogaster, a genetically encoded calcium indicator
such as GCaMP [2] is typically expressed in either the odorant
receptor neurons or the projection neurons of this circuit [6].

Kazushige Touhara (ed.), Pheromone Signaling: Methods and Protocols, Methods in Molecular Biology, vol. 1068,
DOI 10.1007/978-1-62703-619-1_12, Springer Science+Business Media, LLC 2013

179
180 Susy M. Kim and Jing W. Wang

a Two-photon microscope b F1 d
Air to Teflon odor puffer to odor manifold

pin Tubing Tubing

8 1/8 OD x 1/16 ID 1/8 OD x 0.08 ID
7
1 Flangeless Nut
6
Flangeless Ferrule
2
AL AN 5
3
cVA 4

F2 F1 + F2 = 1000 ml/min

Air
c
Fig. 1 Schematic diagram depicting Drosophila antennal lobe preparation and odor delivery system. (a)
Close-up view of antennal lobe prep. Note that the antennae are exposed to the open air while the rest of the
brain is submerged in agarose. A rectangular coverslip is placed on top of this prep and lies between the objec-
tive and the brain. (b) Teflon odor puffer carrying humidified airstream. The total airflow is kept constant at
1,000 mL/min. (c) Truncated glass Pasteur pipette containing a small slip of filter paper with 1 L of cVA. The
solenoid valve controller diverts a small percentage of the total airflow through the glass pipette containing cVA
upon receipt of a signal from the stimulator box. (d) Vacuum fittings for odor delivery system consist of inert
tubing, a flangeless ferrule, and a flangeless nut. As the nut is screwed into threaded fittings (-28) in the
bottle cap, the ferrule seals the vacuum line at the juncture

Other methods for monitoring neural activity such as synaptopHluorin

have also been successfully used in this preparation [3, 7]. In this
technique, the fly brain is dissected out from the rest of the head
capsule with the antennae and the antennal nerve connections to
the antennal lobe still intact. Highly volatile and relatively inexpen-
sive chemical odorants are typically delivered to the intact antennae
via odor streams redirected through bottles containing odorant
[69] (Fig. 1). However, many important insect pheromones such
as 11-cis vaccenyl acetate (cVA) exhibit lower volatility than many
chemical odorants due to relatively high molecular weights
(>300 g/mol). In addition, these lipids must be purified and there-
fore can be expensive reagents. Standard odor delivery techniques,
therefore, must be modified to accommodate these limitations. We
describe here a simple system for delivering odor streams of phero-
mone using minute quantities of the pheromone cVA.

2 Materials

2.1 Antennal Lobe 1. Adult hemolymph-like (AHL) saline: 108 mM NaCl, 5 mM

Preparation Materials KCl, 2 mM CaCl2, 8.2 mM MgCl2, 4 mM NaHCO3, 1 mM
NaH2PO4, 5 mM trehalose, 10 mM sucrose, 5 mM HEPES,
pH 7.5, 265 mOsm. Weigh and transfer to a 1-L graduated flask:
 Calcium Imaging of Pheromone Responses in the Insect Antennal Lobe 181

6.312 g NaCl.
0.373 g KCl.
0.337 g NaHCO3.
0.120 g NaH2PO4.
1.892 g Trehalose2H2O.
3.423 g Sucrose.
1.192 g HEPES.
Add 8.2 mL of 1 M MgCl2. Add deionized and ultrafil-
tered (D.I.U.F.) water (Fisher W2-20) to a volume of 990 mL.
Mix and adjust pH with 1 M NaOH. Transfer to a 1 L volu-
metric flask and make up to 1 L with water. Filter sterilize
500 mL to get calcium-free AHL saline. Add 1 mL of 1 M
CaCl2 to the remaining solution and filter sterilize to get AHL
saline with calcium. Store at 4 C (see Note 1).
2. 2.5 % Agarose solution: Weigh out 0.5 g low melting point
agarose (Type VII-A; Sigma A0701) and dilute with 20 mL
AHL saline containing 2 mM Ca2+. Heat agarose solution in
microwave just until agarose is completely dissolved. Keep the
solution warm and in a liquid state on a hot plate set to 37 C.
3. Two #55 fine forceps (Dumont).
4. Scalpel.
5. Tungsten wire pins: California Fine Wire Company, Grover
Beach, CA 93433.
Tungsten 99.95 %, diameter = 0.001 in.
Tungsten 99.95 %, diameter = 0.0005 in.
6. Dissection dish with Sylgard layer: Prepare Sylgard according
to instructions. Fill small petri dish (Falcon, 35 10 mm) with
approximately 35 mm layer of Sylgard and let it cure for
2448 h until hardened.
7. Glass slide (25 75 1 mm) with small square (~5 mm
5 mm 2 mm) of Sylgard placed near one of the edges.
8. Coverslip 11 22 mm, 0.130.17 mm thick (Thomas
Scientific).

2.2 Odor The odor delivery system described here permits quantitative con-
Delivery System trol of odor stimulation by diluting saturated vapor concentration
up to a 1:1,000 ratio (Fig. 1). Two independent humidified air-
streams are generated. One airstream enters the solenoid valve sys-
tem, which can be switched from an empty vial (air) to the
pheromone odor cartridge (Fig. 2). The constant airflow before,
during, and after the 2-s odor stimulation ensures minimum
mechanical change for odor stimulation. The other airstream is
used to dilute the saturated vapor. The sum of flow rates for the
182 Susy M. Kim and Jing W. Wang

Va
b
lve
co
ntr
8 oll
er
F1 + F2 = 1000 ml/min 7
6
F1 5
4
Air 3
2
1

d
8
7
ID = 0.5 in
1 a
6
2
5
3
4 F2
40 mesh
3 2 1
4 8
5 7 Air
6

Fig. 2 Schematic diagram depicts the flow of air from the (b) manifold/solenoid valve controller, (c) glass vials
containing odor samples, and (d) the odor puffer. A small fraction of the total airstream flows into the manifold
through a side port (a). The bulk of the total airstream flows through the central port in the odor puffer (d). Upon
receipt of a signal from the stimulator box, the solenoid valve controller (b) switches off the flow of air through
valve 1/bottle 1 which is empty and switches airflow into selected valves (28) or through the glass pipette
containing cVA. Air flows from selected open valves in the manifold (b) into bottles containing odor (c). Air
containing odor is then sent into the odor puffer and delivered to the antennal lobe prep

two airstreams is kept constant at 1,000 mL/min. When a new

odor is used in the system all tubing and odor vials downstream of
the solenoid valves must be replaced to eliminate any potential
contamination from previously used odors.
1. 2 port bottle cap (SV-435, Western Analytical, CA).
2. Flangeless Fitting for 1/8" OD tubing, P-300 ETFE ferrule
(Western Analytical).
3. Delrin Nut, P-305 flangeless nut (Western Analytical).
4. 1/8 OD 1/16 ID Teflon Tubing from bottlecap to odor
puffer (PTFE2062, Western Analytical).
5. 1/8 OD 0.08 ID Teflon Tubing from bottlecap to mani-
fold (FEP2080, Western Analytical).
6. Custom manifold (see Note 2).
7. Odor puffer: 7-Port PTFE Manifold with 6 28 ports and 1
NPT port (MA-014, Western Analytical; see Note 3).
 Calcium Imaging of Pheromone Responses in the Insect Antennal Lobe 183

8. Pipe Hex Nipple, MNPT MNPT (PN-008, Western

Analytical) with steel mesh placed at both openings (Fig. 2).
9. Stimulator box (Grass Instruments).
10. Custom eight-channel solenoid valve controller system.
11. cis-vaccenyl acetate (neat, Cayman).
12. Whatman filter paper #8.
13. Pasteur glass pipette (5 ) truncated at both ends and fire
polished such that ~1 cm of the tip and 1 of the barrel
remain.

3 Methods

Carry out all procedures at room temperature unless otherwise

specified.

3.1 Antennal Lobe 1. Anesthetize fly on ice or with CO2 (see Note 4).
Dissection 2. Decapitate fly with scalpel.
3. Transfer head with paintbrush or forceps and place with ante-
rior side facing upwards on the dissection dish with antennae
positioned in front.
4. Pin down proboscis with large tungsten pin (dia. 0.001 in.).
5. Cover head with cold calcium-free saline (see Note 5).
6. Remove air bubbles by gently coaxing them away from the
surface of the head with the sides of forceps. This prevents the
brain from floating upwards towards the surface of the saline
during the dissection.
7. With a pair of forceps in either hand, begin by inserting the tips
of both forceps under one corner of the cuticle plate between
the eyes. Hold the cuticle firmly with one set of forceps and use
the other pair to tear away pieces of the cuticle and air sacs away
from the brain (see Note 6). Leaving the proboscis and anten-
nae intact, continue peeling off the outer cuticle and compound
eyes. The brain itself can be clearly distinguished from the out-
side cuticle by its opaque and white appearance.
8. Flip the head and reattach the large pin through the proboscis
such that its posterior aspect now faces upwards. Remove the
back panel of cuticle attached to posterior side of the brain.
9. Position the dissection dish so that the proboscis points away
from you. Gently pull the brain back towards yourself. Observe
the white connective tissue and nerves lying in between the
brain and proboscis. Begin gently clearing away this material
by pulling it away with your forceps. Utmost caution, however,
must be taken to avoid severing the nerves connecting the
antennae to the brain.
184 Susy M. Kim and Jing W. Wang

10. Stop once the nerves connecting the antennae to the brain
become visible. These should be located within a short dis-
tance on either side of the brains midline and must be intact
in order for information about odors detected by the antennae
to reach the antennal lobe in your experiments.
11. A slender pair of motor nerves may also be visible even closer
to the brains midline than the antennal nerves. These may be
severed at this time with your forceps to reduce spontaneous
proboscis muscle contractions.
12. Remove the pin holding down the proboscis on the Sylgard
plate. Gently suction up and transfer the brain with a glass
pipette to a slide with square piece of Sylgard. The brain should
rest on the Sylgard in a drop of saline (see Note 7).
13. Arrange the brain such that its anterior aspect is facing upwards
and the antennae are pointing out towards the edge of glass
slide (see Note 8).
14. Replace the calcium-free saline with regular saline containing
2 mM calcium.
15. Draw up agarose solution in glass pipette. Make a thin ring of
agarose around brain prep to create a well. Then cover brain
with agarose. Mix the agarose and saline by gently pipetting
the solution in and out of the glass pipette until it is uniform
(see Note 9).
16. Place the slide in a humidified chamber [a petri plate with a wet
kimwipe] to prevent dehydration. Allow the agarose approxi-
mately 20 min or more to solidify (see Note 10).
17. When ready, use a scalpel to cut away chunks of agarose from
areas immediately adjacent to the antennae and proboscis while
leaving the brain intact and submerged in agarose. Next using
a scalpel, remove the layer of agarose lying on top of the pro-
boscis (Fig. 1). Next, insert forceps into agarose above anten-
nae and gently peel agarose layer away from the antennae. Do
not disturb either the antennal nerves or the brain itself (see
Note 11). Use a kimwipe to dry off the antennae and rub off
any bits of agarose that may still be clinging to the cuticle.
18. Drop coverslip on top of the prep.

3.2 Antennal Lobe 1. Pipette 1 L of cVA on a small slip of filter paper and place in
Stimulation and glass pipette. Push filter paper towards the tip and then add
Imaging small wisps of cotton to prevent it from falling back out. Secure
the pipette to the pheromone tubing system.
2. Position slide in the imaging microscope setup. Position the
tip of the glass pipette delivering the pheromone such that it is
resting on the slide or the Sylgard slab approximately 5 mm
Calcium Imaging of Pheromone Responses in the Insect Antennal Lobe 185

away from the antennae. Position the odor puffer as close to

the antennae as possible without disturbing the glass pipette.
3. After locating the prep with a low magnification objective
(e.g., 4), add a drop of water on top of the coverslip. Add a
second drop to the tip of a higher magnification water immer-
sion objective (e.g., 40). Position the high magnification
objective until the water droplets touch before adjusting the
focus to locate the brain.
4. Starting at a zero setting, increase the laser power slowly just
until brain structures expressing GCaMP are visible. Focus and
zoom in on the antennal lobe (see Note 12). To maximize the
dynamic range of the signal and to minimize phototoxicity,
adjust the laser power to set basal fluorescence to about
1015 % of the full dynamic range of the light detector.
5. Begin stimulation and image acquisition. Stimulate for 2 s with
at least 2-min interval between odor presentations to allow any
desensitized odor receptors to recover and to permit odors to
clear from the flys headspace. For antennal lobe imaging, we
typically acquire images at ~4 frames per second (256 ms per
frame) with a resolution of 128 128 pixels (0.75 0.75 m
for each pixel; Fig. 3). After all odor stimulations are finished,
acquire a high-resolution image stack of the antennal lobe at
512 512 pixels (12 m step in z-axis) for glomerular
identification.

Fig. 3 Anterior view of antennal lobe from a female fly bearing the transgenes
GH146-Gal4 and UAS-GCaMP3. Grayscale (left) and pseudocolored (right) images
depict odor-evoked activity in DA1 projection neurons in response to delivery of
cVA at 2 % SV (saturated vapor pressure)
186 Susy M. Kim and Jing W. Wang

4 Notes

1. For best results, saline should be made fresh every week.

A freezing-point osmometer can be used to measure the osmo-
larity of the AHL saline, which should be about 265 mOsm for
adult Drosophila melanogaster [10].
2. The custom manifold with eight odor ports is under the con-
trol of the custom solenoid valve controller. Switches on the
solenoid valve controller determine which of the eight odor
ports is open at the time of odor stimulation.
3. The custom-made odor puffer in our lab contains nine ports.
A 7-port version is available commercially from Western
Analytical.
4. Antennal lobe preps from younger flies are more robust and
easier to dissect out than older flies. Flies, however, should be
old enough to permit sufficient levels of GCaMP to accumu-
late in the glomeruli. Therefore, flies approximately 23 days
old are ideal.
5. Dissection with saline containing calcium may lead to reduced
odor responses. Cold calcium-free saline serves to preserve the
integrity of the fly brain and minimize the loss of synaptic
strength.
6. Using two sets of forceps in this mannerone to hold the
cuticle, and the other to tear away pieces of itminimizes
damage to the brain by reducing the exertion of tensile forces
on the tissue.
7. Brain tissue is highly charged and tends to stick to the side of a
glass pipette. Prevent the loss of your tissue by first suctioning
up a small volume of saline into your glass pipette. Expel a tiny
volume of the saline before suctioning up the brain. The brain
should hang just at the tip of the pipette in a drop of saline.
8. Orient the fly brain such that the antennae are fairly close to
the edge of the slide. This later ensures that the antennae may
be positioned as close to the odor puffer as possible.
9. Be sure that the saline around the brain and the agarose are
thoroughly mixed. Otherwise, the agarose may fail to solidify
around the antennae, making its complete removal messy and
difficult in a later step.
10. To determine whether agarose has sufficiently dried, one can
make a small incision into the agarose with the scalpel. If it is
properly set, saline should not appear to seep out from the gel.
11. Use a Kimwipe to dry off the antennae and rub off any bits of
agarose that may still be clinging to the cuticle.
 Calcium Imaging of Pheromone Responses in the Insect Antennal Lobe 187

12. If the image of the antennal lobe is not clear, there may be too
much agarose in between the brain and the objective. Strive to
make the layer of agarose a little thinner to improve image
quality.

Acknowledgement

This work was supported by NIH (R01DK092640) and NSF

(0920668) grants to J.W.W.

References
1. Tian L et al (2009) Imaging neural activity in 6. Root CM et al (2007) Propagation of olfactory
worms, flies and mice with improved GCaMP information in Drosophila. Proc Natl Acad Sci
calcium indicators. Nat Methods 6:875881 U S A 104:1182611831
2. Nakai J, Ohkura M, Imoto K (2001) A high 7. Root CM et al (2008) A presynaptic gain con-
signal-to-noise Ca(2+) probe composed of a trol mechanism fine-tunes olfactory behavior.
single green fluorescent protein. Nat Biotechnol Neuron 59:311321
19:137141 8. Root CM et al (2011) Presynaptic facilitation
3. Ng M et al (2002) Transmission of olfactory by neuropeptide signaling mediates odor-
information between three populations of neu- driven food search. Cell 145:133144
rons in the antennal lobe of the fly. Neuron 9. Semmelhack JL, Wang JW (2009) Select
36:463474 Drosophila glomeruli mediate innate olfac-
4. Wang JW et al (2003) Two-photon calcium tory attraction and aversion. Nature 459:
imaging reveals an odor-evoked map of activity 218223
in the fly brain. Cell 112:271282 10. Singleton K, Woodruff RI (1994) The osmo-
5. Ruta V et al (2010) A dimorphic pheromone larity of adult Drosophila hemolymph and its
circuit in Drosophila from sensory input to effect on oocyte-nurse cell electrical polarity.
descending output. Nature 468:686690 Dev Biol 161:154167
 Chapter 13

Live Cell Calcium Imaging of Dissociated

Vomeronasal Neurons
Angeldeep Kaur, Sandeepa Dey, and Lisa Stowers

Abstract
Sensory neurons in the vomeronasal organ (VNO) are thought to mediate a specialized olfactory response.
Currently, very little is known about the identity of stimulating ligands or their cognate receptors that
initiate neural activation. Each sensory neuron is thought to express 1 of approximately 250 variants of
Vmn1Rs, Vmn2Rs (A, B, or D), or FPRs which enables it to be tuned to a subset of ligands (Touhara and
Vosshall, Annu Rev Physiol 71:307332, 2009). The logic of how different sources of native odors or
purified ligands are detected by this complex sensory repertoire remains mostly unknown. Here, we
describe a method to compare and analyze the response of VNO sensory neurons to multiple stimuli using
conventional calcium imaging. This method differs from other olfactory imaging approaches in that we
dissociate the tightly packed sensory epithelium into individual single cells. The advantages of this approach
include (1) the use of a relatively simple approach and inexpensive microscopy, (2) comparative analysis of
several hundreds of neurons to multiple stimuli with single-cell resolution, and (3) the possibility of isolat-
ing single cells of interest to further analyze by molecular biology techniques including in situ RNA
hybridization, immunofluorescence, or creating single-cell cDNA libraries (Malnic et al., Cell 96:713
723, 1999).

Key words Vomeronasal, Calcium imaging, Neurons, Dissociated

1 Introduction

In mice, the vomeronasal organ (VNO) has been implicated in

mediating reception of a variety of ligands informing both interspecies
and intraspecies interactions [1]. Sensory neurons within the VNO
can each express 1 of approximately 250 receptors. With such a
complex receptor family of interest, ligand identification and receptor
deorphanization have become important in order to study the
mechanisms of sensory detection and function. Calcium imaging of
dissociated sensory neurons is a relatively simple technique to
identify and analyze ligands detected by the VNO [2]. Moreover,
because the closely packed neurons are dissociated into individual
sensory neurons, single cells can be easily isolated following calcium
imaging for analysis by conventional molecular biology [3].

Kazushige Touhara (ed.), Pheromone Signaling: Methods and Protocols, Methods in Molecular Biology, vol. 1068,
DOI 10.1007/978-1-62703-619-1_13, Springer Science+Business Media, LLC 2013

189
190 Angeldeep Kaur et al.

Activation of the vomeronasal sensory neurons (VSNs) occurs

upon binding of ligands by the cognate receptors, followed by a
signaling cascade that leads to opening of a Trpc2-associated cal-
cium channel [4]. This in turn results in a high influx of extracel-
lular calcium ions [5], a property that has been exploited to identify
activated cells using chemical dyes or genetically encoded marker
proteins that detect changes in intracellular calcium (reviewed by
Grienberger, 2012 [6]). This principle can be combined with sim-
ple microscopy to study the changes in calcium concentration
within individual VSNs. The densely packed VSNs must first be
separated by physical dissociation followed by mild protease diges-
tion. The separated neurons are plated on coverslips at a density
that enables analysis of individual neurons. This process is referred
to as a dissociated preparation of VSNs. The attached VSNs are
incubated in a solution containing a calcium-sensing chemical dye
to load the neurons with the fluorescent indicator. A dissociated
preparation allows each neuron in the field of view to be indepen-
dently selected as regions of interest or areas on the coverslip for
commercially available software to monitor changes in calcium
concentration. The recorded change in fluorescence of the dye
reflects the changes in intracellular calcium within the selected
VSNs. Rapid changes in calcium can be recorded and the timing of
each rise in calcium can be correlated to specific ligands being
introduced in the cell environment. The collected data can be
studied for an in-depth comparative analysis of ligands that may
broadly or specifically activate populations of VSNs.
Dissociation of the VSNs allows for a relatively rapid and even
uptake of chemical dyes within each cell. Consequently, this prepa-
ration results in a highly reliable resolution of ligand responses by
sensory neurons normally present in both superficial and deep lay-
ers. The preparation is highly efficient; from three mice, up to four
coverslips can be prepared, each containing between 150 and 500
cells that can be imaged. Since the timing of each stimulant deliv-
ery is known, the window of time during which a neural response
may occur can be precisely defined. This allows us to parse out
spontaneous changes in intracellular calcium from real responses to
presented ligands. Additionally, the preparation is easy to learn and
does not require highly specialized equipment for imaging or
analysis of gathered data. Following imaging of a dissociated prep-
aration, it is possible to physically isolate single cells of interest
from the coverslip for molecular analysis, for instance RNA in situ
hybridizations, immunofluorescence, and creation of single-cell
cDNA libraries. The ability to do subsequent analysis on individual
activated cells is a distinct advantage of imaging dissociated neurons
over whole epithelium or slice preparations.
The most prominent advantage of using chemical dyes over
genetic markers is their high sensitivity in that they undergo a large
change in fluorescence in response to single-action potentials [7].
Live Cell Calcium Imaging of Dissociated Vomeronasal Neurons 191

Without this added sensitivity, a subset of relevant and functional

neuronal activities would go undocumented, further impairing the
ability to fully understand vomeronasal function. We use UV
excitable ratiometric dye Fura-2-AM. Once bound to Ca2+, the
absorption of Fura-2-AM shifts, which can be observed by scan-
ning the excitation spectrum from 300 to 400 nm, whereas the
emission spectrum stays at ~510 nm (product information
Invitrogen). In the calcium-free state, fluorescence at 380 nm is
higher than at 340 nm whereas in the bound state, the fluorescence
at 380 nm decreases and that at 340 nm increases. The simultaneous
change in absorption spectra results in a more reliable measure-
ment of calcium concentration. By measuring fluorescence at both
340 and 380 nm a relative quantification of calcium concentrations
can be made based on the 340/380 fluorescence values (prod-
uct informationInvitrogen). The use of the ratio also aids in
reducing artifactual information by accounting for differences in
cell thickness, uneven dye loading, loss of dye, and photobleaching
[8]. Several other commercially available fluorescent dyes may also
be used; however an advantage of Fura-2-AM over other similar
dyes is that it can be used in combination with and does not inter-
fere with the fluorescence spectrum of GFP. Addition of chemical
modifications, such as the acetoxymethyl (AM) ester [9, 10], to
calcium indicators enhances their permeability and facilitates more
uniform uptake by cells. Greater uptake of the dye reduces the
number of dye micelles observed, which can produce a background
signal as flecks of fluorescence during imaging. Noise can be fur-
ther reduced by using pluronic acid, which aids in dispersion of
Fura-2-AM, increasing dye uptake efficiency and breaking up fluo-
rescent aggregates.
Despite the simplicity and sensitivity of using a dissociated
preparation and calcium-sensing dye-based fluorescence imaging,
the technique is not without disadvantages. Protease digestion,
albeit mild, followed by dissociation of the closely packed neurons
reduces overall cell health compared to vibratome slices or whole-
epithelium preparations where cell connections are not entirely
disrupted. A limited number of sequential ligand pulses can be
perfused over the neurons before their reduced cell health affects
reliable cell responses. Therefore, each step of the protocol must
be carried out within a minimal amount of time so as to preserve
cell health. UV phototoxicity may also reduce the longevity of the
neurons during imaging; thus designing shorter experiments yields
more reliable results. For each experiment, the time of stimulant or
buffer delivery must be precisely noted so that data analysis can be
carried out with reliable knowledge of which stimuli are responsi-
ble for cell depolarization events. Most calcium dye indicators
when used in large concentrations have a calcium buffering effect
whereby the peak amplitude of calcium transients is significantly
reduced [11], so indicator loading must be carried out carefully to
192 Angeldeep Kaur et al.

maintain a good signal-to-noise ratio. Lastly, since the topological

structure of the VNO is lost during dissociation, all information on
the spatial arrangement of the neurons is irretrievable, though
molecular characterization can inform on whether imaged cells are
from the apical or the basal layer of the VNO. By designing experi-
ments while keeping these caveats in mind, a vast amount of infor-
mation can be gathered about the neurons of interest.
The combined use of chemical dye Fura-2-AM and a dissoci-
ated cell preparation enables characterization of the response
profile of a single neuron when presented with particular ligands.
With these tools, stimuli relevant to the VNO can be identified and
the specificity and tuning of vomeronasal neurons can be more
fully understood.

2 Materials

2.1 Dissection 1. 1 Phosphate buffer saline (PBS): pH 7.4 (see Note 1).
of VNO 2. Forceps and surgical scissors.

2.2 Dissociation 1. DL-Cysteine-HCl (1 M): Dissolve 15.7 mg DL-cysteine-HCl in

of Neurons 100 L distilled water (see Note 2).
2. Ethylene diamine tetra acetate (EDTA, 100 mM).
3. Papain: Dissolve papain in water to 0.22 U/L concentration,
sterile filter, and divide into 11 L aliquots for single use (see
Note 3).
4. Dissociation buffer (in 1 PBS): 1.1 mM EDTA; 5.5 mM DL-
cysteine-HCL; papain 2.2 U/mL. Mix 5.5 L DL-cysteine-HCl
solution (1 M), 11 L EDTA solution (100 mM), and 10 L
papain (0.22 U/L) in 1 mL 1 PBS (see Note 4).
5. DNAse solution: 10 DNAse buffer; DNase 1; 1 PBS. Mix
200 L of 10 DNAse buffer and 800 L 1 PBS, add 50 U
DNase 1, and mix well (see Note 5).

2.3 Plating 1. Plating media: 10 % supplemented Dulbeccos modified

of Neurons Eagle medium (D-MEM). D-MEM High Glucose (1) +
GlutaMAX + Sod.Pyr, heat-inactivated fetal bovine serum
(FBS). Mix 90 mL D-MEM with 10 mL heat-inactivated FBS
and divide into 12 mL single-use aliquots. Pre-warm aliquots
to 37 C before use (see Note 6).
2. Concanavalin A (0.5 g/L)-coated coverslips: Solid
Concanavalin A (type V), distilled water. Weigh out 10 mg
Concanavalin A and dissolve in 20 mL distilled water to obtain
a 0.5 g/L solution. Divide into 12 mL single-use aliquots.
Incubate coverslips overnight in Concanavalin A solution at
 Live Cell Calcium Imaging of Dissociated Vomeronasal Neurons 193

4 C, with continuous shaking. Place them in a coverslip rack

to dry in a laminar flow chamber for 27 h and store at 4 C
(see Note 7).
3. Stacks (optional): Prepare stacks by cutting off a center por-
tion of P-200 tips (Fig. 2). Autoclave the stacks before use.

2.4 Calcium Imaging 1. Hanks Balanced Salt Solution (HBSS) 10, anhydrous calcium
chloride 1,400 mg/L, magnesium chloride-6H2O 1,000 mg/L,
magnesium sulfate-7H2O 1,000 mg/L, potassium chloride
4,000 mg/L, potassium phosphate monobasic 600 mg/L,
sodium chloride 80,000 mg/L, sodium phosphate dibasic-
7H2O 900 mg/L, dextrose 10,000 mg/L (see Note 8).
2. 4-(2-Hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES)
buffer, 1 M (see Note 9).
3. Imaging buffer: Mix 50 mL HBSS (10), 5 mL HEPES (1 M),
and 445 mL of distilled water for a final volume of 500 mL (see
Note 10).
4. Pluronic acid 20 %: Add 20 mg pluronic acid to 100 L DMSO
in an Eppendorf tube, warm in 37 C water bath for 1015 min
with intermittent vortexing to dissolve (see Note 11).
5. Fura-2-AM dye (1 g/L): Dissolve 50 g Fura-2-AM in
50 L DMSO by pipetting up and down several times (see
Note 12).
6. Loading solution: Mix 5 L Fura-2-AM solution with 2.5 L
pluronic acid solution by pipetting up and down several times.
Add 1,000 L imaging buffer (see Note 13).

2.5 Equipment 1. 37 C cell culture incubator with 5 % carbon dioxide.

2. Certified class II biological safety cabinet with laminar flow.
3. Centrifuge machine with swinging bucket rotor for 15 mL
conical tubes.
4. Dissection microscope.
5. Phase contrast microscope with 10 and 20 objective lenses.
6. Inverted deconvolution microscope with a 20 fluar 0.75 objec-
tive or an upright microscope with water immersion objective.

3 Methods

All reagents and chemicals should be of cell culture grade. For best
results, all solutions should be prepared using sterile technique and
stored at the appropriate temperature for future use. Each step
must be carried out within minimal time to ensure optimal health
of the neurons.
194 Angeldeep Kaur et al.

3.1 VNO Dissection 1. Transfer 1 mL 1 PBS to a well in a 4-well plate and set on ice.
Transfer 20 mL 1 PBS each into two petri dishes and chill on ice.
2. Prepare papain-containing dissociation buffer immediately
prior to starting dissections. Incubate on ice.
3. Isolate both VNO lobes from a total of three mice transferring
each upon removal to a petri dish containing chilled PBS on
ice from step 1. For best results, this step should be completed
within 1015 min.
4. Move VNOs to the second petri dish containing chilled PBS
one at a time. Remove cartilage from each VNO lobe and trans-
fer tissue to chilled PBS in the 4-well plate prepared in step 1.

3.2 Dissociation of 1. Once all VNOs have been removed from their cartilage, care-
Vomeronasal Neurons fully aspirate PBS in the well and replace with 1 mL ice-cold
dissociation solution prepared in Subheading 3.1, step 2.
2. Dissociate each lobe of the VNO by tearing with fine forceps
into minute pieces (Fig. 1). For best results, complete within
10 min.
3. Transfer solution containing dissociated VNO tissue to a
15 mL Falcon tube and incubate at 37 C for 1520 min with
continuous shaking (at approximately 225 rpm).
4. Prepare the DNAse solution.
5. Add DNAse solution to dissociated VNO tissue and gently
triturate until all aggregates are dispersed (see Note 14).

3.3 Plating 1. Add 10 mL of pre-warmed D-MEMFBS media. Tap several

Vomeronasal Neurons times to mix and spin at 1,000 rpm (200g) for 5 min, at room
temperature.

Fig. 1 Dissociating VNO tissue in protease solution (a) before and (b) after. Three
whole VNOs are depicted before and after
 Live Cell Calcium Imaging of Dissociated Vomeronasal Neurons 195

Fig. 2 (a and b) Preparing a stack. (a) Cut a 200 L pipette tip at the positions indicated by white arrows. (b)
A stack obtained from the pipette tip. (c) Stack placed on a Concanavalin A-coated glass coverslip in a well. (d)
Pipetting cell suspension in media inside the stack. (e) Media containing dissociated neurons pipetted in
stacks, ready to be incubated at 37 C

2. Place one Concanavalin A-coated coverslip per well in a new

4-well plate. Place a stack in the center of each coverslip to
concentrate application of cells (Fig. 2).
3. Aspirate supernatant without disturbing the pellet. Gently
resuspend in 100 L of D-MEMFBS media.
4. Transfer 25 L of resuspended VSNs onto each coverslip inside
the stack.
5. Incubate at 37 C for 4560 min. Check to see if cells are
attached to the coverslip using a phase contrast microscope
(density similar to Fig. 3).

3.4 Loading 1. Thaw Fura-2-AM at room temperature.

Vomeronasal Sensory 2. Prepare dye loading solution, cover in aluminum foil, and store
Neurons with Calcium- in the dark through experiment.
Sensing Dye
3. Remove stack from the coverslip, transfer the coverslip to a
well, and gently layer with 250 L dye loading solution.
Incubate the coverslip for 1530 min at room temperature,
covered in aluminum foil.
4. Remove coverslip from dye solution and assemble it on the stage.
Gently cover the cells with 250 L of imaging buffer to prevent
drying (see Note 15). Coverslip is now ready for imaging.

3.5 Calcium Imaging 1. Prior to imaging, wash stimulant/buffer delivery system and
accessory tubes thoroughly with distilled water. Other wash solu-
tions may also be used based on the solubility of stimulants used.
196 Angeldeep Kaur et al.

Fig. 3 (a) A field of view of dissociated cells under 380 nm. (b) Individual cells selected, shown in colored
squares as regions of interest

2. Dilute stimulants in imaging buffer to desired strength.

3. Set up dye-loaded coverslip on microscope and set perfusion
system up to wash cells with imaging solution.
4. Using a data collection software such as MetaFluor, select each
dissociated cell as an individual region of interest (Fig. 3).
Select an empty region of the coverslip for background sub-
traction during data collection (see Note 16).
5. Apply stimulant(s) for desired duration followed by imaging
buffer for an intervening duration to wash cells before perfus-
ing the next stimulant (see Note 17). Following the last test
stimulant, perfuse a pulse of a known activator of the cells of
interest (for a positive control). If such a positive control ligand
is not defined for cells of interest, pool individual stimulants to
pulse at the end (see Notes 1820).

3.6 Data Analysis 1. Identify the cells that show an increase of the 340/380 ratio
measuring more than 1.5 times the baseline signal during the
time window(s) of positive control or pooled stimulant appli-
cation. Graphing software such as Microsoft Excel may be used
to do this (see Note 21). Plot the 340/380 fluorescence ratio
for each region of interest that shows the increase in calcium
coinciding with the delivery of the positive control or pooled
stimulant.
2. Precisely define the time during the experiment when stimulus
was perfused over the cells. In case the stimulants are delivered
through a tube, determine the length of time required for the
stimulant front to reach the coverslip so that the time of deliv-
ery of stimuli is defined precisely (see Note 22). If analyzing by
a trace plot, annotate the plot from the time the stimulus
reaches the coverslip to the time the subsequent wash reaches
the coverslip.
 Live Cell Calcium Imaging of Dissociated Vomeronasal Neurons 197

Fig. 4 Example of output data. Each colored line represents the calcium trace of a single neuron, the black
rectangles represent time windows for application of stimulus. (a) Responsive cells. Purple: Cell showing rise
in intracellular calcium on application of test stimulus and positive control. Blue: Cell showing rise in intracel-
lular calcium on application of positive control only. (b) Unresponsive and noisy cells. Red: Unresponsive cell.
Orange: Cell showing intracellular calcium increase before application of test stimulus and positive control.
Green: Cell showing intracellular calcium increase randomly on application of test stimulus but not to positive
control. (X-axis: Ratio of fluorescence change measured at 340 nm and that measured at 380 nm, Y-axis: Time
(in min))

3. To define a cell of interest as responsive, we use the following

criteria (Fig. 4):
Cells must respond to the positive control stimulus pulse at
the end.
An increase in fluorescence ratio measuring at least 1.5 times
the baseline during the time of interest is counted as a
response.
Cells must only respond during the time that stimuli are
present on the coverslip (see Note 23).
Cells must not show an increase in calcium greater than 1.5
times the baseline signal outside of the stimulus windows.

4 Notes

1. PBS used for dissection may be kept chilled on ice as VNOs

are dissected. Chilled PBS is helpful in keeping the neurons
alive if dissection takes longer than 15 min.
2. Prepare fresh each day.
3. Papain aliquots dissolved in distilled water may be stored at
20 C up to several months. To avoid repeated freezethaw
198 Angeldeep Kaur et al.

cycles, they are best divided into single-use aliquots. Thaw

before preparing dissociation buffer.
4. Prepare fresh each day. Once prepared, keep dissociation buf-
fer on ice until use to avoid losing enzyme activity.
5. Prepare DNase solution right before use while neurons are
incubating in 37 C shaker. If DNAse buffer is not available,
this may be substituted by other buffers in which DNAse is
fully active, for instance 10 PCR buffer or 10 endonuclease
buffer.
6. Prepare 10 % (by volume) FBS-supplemented DMEM in the
laminar flow chamber to keep it contamination free and store
aliqouts at 4 C. Pre-warm each aliquot at 37 C at the start of
the experiment to avoid adding cold media to cells.
7. Of the available reagents, we have found that the use of
Concanavalin A to coat coverslips allows for the maximum number
of cells to adhere after plating. Aliquots of 0.5 g/L
Concanavalin A may be stored at 20 C and thawed before use.
8. Store 10 HBSS at 4 C after opening.
9. Store 1 M HEPES at 4 C.
10. Make fresh each day. Keep at room temperature after preparation.
11. Make fresh every 710 days. Store at room temperature.
12. Fura-2-AM is light sensitive. Once dissolved, keep Fura-2-AM
solution wrapped in aluminum foil at 20 C. Thaw before use
each day.
13. Prepare loading solution fresh each day. Wrap loading dye in
aluminum foil and store at room temperature while experi-
ment is in progress.
14. At the end of protease digestion, the pieces of tissue appear to
aggregate together. After addition of DNAse, the pieces of tis-
sue appear to be more disintegrated: fewer and smaller pieces
of tissue should be seen in suspension. The solution should
appear more turbid than that in step 2 (Subheading 3.2).
15. Do not pipet the solution directly on the coverslip. Instead
pipet the solution on the side of the stage and tip the stage to
cover the cells. This will prevent cells from being forcibly dis-
lodged from the coverslip.
16. For urinary proteins and total urine, we have empirically deter-
mined that setting the 340380 gain controls at a ratio of 3:1
obtains the optimal signals to differentiate neuronal responses
from noise in our setup. This may have to be adjusted for dif-
ferent ligands and instrumental setups.
17. We apply stimuli such as mouse urinary proteins or total urine
for 1 min alternating with buffer for 2 min. These durations of
Live Cell Calcium Imaging of Dissociated Vomeronasal Neurons 199

stimulus and buffer pulses may have to be varied according to

the nature and concentrations of different ligands.
18. A positive control pulsed at the end of the experiment helps to
identify cells that survived through the entire experiment. Only
these cells can be investigated for their response to all the ligands
presented during the course of the experiment. Another possible
positive control for neurons is potassium chloride [12].
19. In our setup, a perfusion rate of approximately 5 mL per min-
ute is optimal to obtain laminar flow of buffer over the neu-
rons. This may have to be optimized differently for each setup.
20. To obtain the most accurate results, it is necessary to image
each coverslip quickly. Depending on the quality of the prepa-
ration, and the nature and concentration of ligands, up to five
total pulses may be applied, spanning approximately 15 min of
imaging time. To maximize speed, subsequent coverslips
should be incubated in dye solution while the previous is being
imaged such that one can complete imaging four coverslips in
90 min or less. If the cells are imaged for longer than 15 min,
the constant exposure to pulses of ligands and UV light may
adversely affect cell viability, leading to less accurate output. As
previously noted, one preparation yields four coverslips. If each
coverslip is imaged within 1520 min, there is no substantial
loss of cell health between neurons from the first and last cov-
erslip. Cell health may not be optimal if neurons are not imaged
within 3 h of plating.
21. We have empirically determined, for mouse urine and urinary
proteins, that a 1.5-fold increase in the 340380 nm ratio is a
good indicator of neuronal response. This may have to be opti-
mized for various ligands and instrument setup. We use this
cutoff to sort our data, separating cells that show an increase in
calcium during presentation of the positive control from the
rest of the imaged cells. This way the analysis can be focused
on responsive cells instead of every single imaged cell, many of
which show no changes in calcium during the experiment.
22. Delivery time may be determined by introducing a bubble or a
colored dye in the delivery tube and then monitoring the time
taken by the bubble or the dye front to reach the end of the
tubing with a stopwatch or a timer.
23. There are spontaneous changes in calcium in cells, a natural
phenomenon noticed in many types of preparations [12].
Being able to differentiate spontaneous activity and ligand-
based activity is important in order to use this technique to
follow ligand activity. A precise notation of the window of
stimulus perfusion and repetitive pulses of the same stimulus
enables correlation between stimulus perfusion and calcium
changes within the cell.
200 Angeldeep Kaur et al.

References

1. Touhara K, Vosshall LB (2009) Sensing odor- lations with calcium imaging. Methods
ants and pheromones with chemosensory 18(2):215221
receptors. Annu Rev Physiol 71:307332 8. Becker PL, Fay FS (1987) Photobleaching of
2. Malnic B, Hirono J, Sato T, Buck LB (1999) fura-2 and its effect on determination of cal-
Combinatorial receptor codes for odors. Cell cium concentrations. Am J Physiol 253(4 Pt 1):
96(5):713723 C613C618
3. Munger SD, Leinders-Zufall T, Zufall F (2009) 9. Tsien RY (1988) Fluorescence measurement
Subsystem organization of the mammalian and photochemical manipulation of cytosolic
sense of smell. Annu Rev Physiol 71:115140 free calcium. Trends Neurosci 11(10):419424
4. Chamero P, Marton TF, Logan DW, Flanagan K, 10. Tsien RY (1981) A non-disruptive technique
Cruz JR, Saghatelian A et al (2007) Identification for loading calcium buffers and indicators into
of protein pheromones that promote aggressive cells. Nature 290(5806):527528
behaviour. Nature 450(7171):899902 11. Saftenku EE, Teslenko VI (ed) (1995) Effect
5. Kaupp UB (2010) Olfactory signalling in ver- of fura-2 on calcium transients and its depen-
tebrates and insects: differences and common- denece on kinetics and location of endogenous
alities. Nat Rev Neurosci 11(3):188200 buffers (a model study). Kluwer Academic-
6. Grienberger C, Konnerth A (2012) Imaging Plenum, New York
calcium in neurons. Neuron 73(5):862885 12. Holy TE, Dulac C, Meister M (2000)
7. Smetters D, Majewska A, Yuste R (1999) Responses of vomeronasal neurons to natural
Detecting action potentials in neuronal popu- stimuli. Science 289(5484):15691572
 Chapter 14

Whole-Mount Imaging of Responses in Mouse

Vomeronasal Neurons
Pei Sabrina Xu and Timothy E. Holy

Abstract
Imaging permits the visualization of neural activity from the whole-mount vomeronasal sensory epithelium
with single-cell resolution. The preparation preserves an intact tissue environment, enabling the robust
detection of cellular responses upon chemical stimulation and study of the precise 3D mapping of vomero-
nasal sensory neuron (VSN) functional types within the epithelium. Using objective-coupled planar illumi-
nation (OCPI) microscopy to perform fast volumetric imaging, we routinely record the responses of
thousands of VSNs for hours from a single intact vomeronasal organ preparation. Here we document the
preparation of the whole-mounted vomeronasal epithelium, multichannel stimulus delivery, and three-
dimensional calcium imaging by OCPI microscopy.

Key words In situ, Calcium imaging, Three-dimensional, Whole-mount VNO, Neural population,
OCPI microscopy

1 Introduction

Mouse vomeronasal sensory neurons (VSNs) are situated in the

sensory epithelium of the vomeronasal organ (VNO). Within the
sensory epithelium, each bipolar VSN extends a single apical den-
drite through the supporting cells to the surface, where the dendrite
terminates as a knob covered by a tuft of microvilli [1]. It is the
vomeronasal receptor proteins, localized on the dendritic tips in
the microvilli layer, that bind ligands (often considered phero-
mones) and initiate activation of particular VSNs. To maintain the
microvilli layer, an intact whole-mount preparation of the VNO was
used, at first for electrical recording of VSN responses [2]. A similar
preparation, mounting the neuroepithelium on a nitrocellulose
membrane, was later used for imaging [3].
Calcium imaging is widely used for imaging the neural activity,
including dissociated VSNs [46] and VNO acute slices [7].
However, for the whole-mount VNO preparation, some form of

Kazushige Touhara (ed.), Pheromone Signaling: Methods and Protocols, Methods in Molecular Biology, vol. 1068,
DOI 10.1007/978-1-62703-619-1_14, Springer Science+Business Media, LLC 2013

201
202 Pei Sabrina Xu and Timothy E. Holy

optical sectioning (collecting only in-focus light) is necessary.

However, for sampling many hundreds or thousands of neurons,
conventional techniques such as two-photon microscopy are limited
in speed. We therefore developed a form of light sheet-based
microscopy called objective-coupled planar illumination (OCPI)
microscopy, which enables the fast three-dimensional fluorescence
imaging of activity in thousands of neurons [3]. With OCPI
microscopy, we have been able to image from at least one-fifth of
the whole VNO epithelium. Because OCPI microscopy is also less
phototoxic than most alternatives, a single VNO preparation can
be repetitively scanned for at least 1.5 h without significant loss of
responsiveness. This enabled us recently to test up to 13 different
stimuli on a single VNO preparation with four or more repeats of
each stimulus [8]. Consequently, responsivity of VSNs can be mea-
sured quantitatively and analyzed by systematic statistical criteria.
Given that VSN physiology is often perceived to be a needle in a
haystack problem, and VSNs show abundant spontaneous activity
with characteristics disturbingly close to those of stimulus-driven
responses [9], such systematic approaches are among the few tools
available for reducing the risk of false positives.

2 Materials

2.1 Reagents 1. Ringers solution: 115 mM NaCl, 5 mM KCl, 2 mM CaCl2,

2 mM MgCl2, 25 mM NaHCO3, 10 mM HEPES, and 10 mM
D-(+)-glucose. Solution is freshly prepared on the experiment
day from 20 Ringers A, 20 Ringers B, and glucose. Prepare
20 Ringers A: Weigh 67.17 g NaCl, 3.73 g KCl, 2.94 g
CaCl2 (dehydrate), and 4.07 g MgCl2 (hexahydrate); add
water; mix; and make up to a volume of 500 mL. Prepare 20
Ringers B: Weigh 21.00 g NaHCO3 and 23.86 g HEPES, add
water to dissolve, and make up to 500 mL. Water is prepared
by purifying deionized water to attain a sensitivity of 18 M
cm at 25 C. 20 Ringers A and B can be stored at 4 C for
3 months.
2. Stimulus stocks: Will be diluted to desired concentration by
freshly prepared carboxygenated Ringers solution on the
experiment day (see step 2 of Subheading 3.1). For water-
insoluble stimuli, they need to be dissolved in appropriate
solvent (e.g., methanol, DMSO) as high-concentration stock
(see Note 1).

2.2 Animals and VNO 1. Mice with calcium indicator-labeled VSNs, e.g., OMP-
Dissection Tools GCaMP2 mice [7, 8], mice with Oregon-Green BAPTA-
labeled VNO by surgical injection [3].
2. VNO dissection tools: Decapitation scissors, Ring forceps, scalpel,
#3 forceps, #5 forceps, Vannas spring scissors (3 mm cutting edge),
 VNO Whole Mount Imaging 203

25 mm Petri dish, Sylgard-coated 25 mm Petri dish, ice bucket

with cover, and dissection stereomicroscope.
3. VNO mounting tools: Nitrocellulose membrane (Millipore),
custom glass dropping pipet (cut out the rear ~5 cm part of a
disposable borosilicate glass Pasteur pipets, and cap the cutting
end with a rubber head) (see Note 2), custom imaging cham-
ber (can hold the nitrocellulose membrane at the bottom; is
big enough to immerse the objective and light sheet optics
inside), kimwipes tissue.

2.3 Equipment 1. Microscopy: Custom OCPI microscopy [3] (US patent appli-
and Software cation 20090174937 pending) controlled by custom imaging
acquisition software Imagine (http://holylab.wustl.edu/soft-
ware.htm).
2. Superfusion equipment: 16-channel superfusion system
(AutoMate Scientific, Inc. CA, USA) pressurized with gas cyl-
inder (40 % O2, 3 % CO2, balance helium). We use 35 mL plas-
tic syringe reservoirs. The switchable pinch valves are computer
controlled by Imagine to synchronize with the imaging
system.
3. Liquid heating equipment: PH01 heatable perfusion probe
and TC02 temperature controller (Multi Channel Systems
MCS GmbH, Reutlingen, Germany).

3 Methods

3.1 Prepare Solution 1. Mix 950 mL water, 50 mL 20 Ringers buffer stock, 50 mL

20 Ringers salt stock, and 1.8 g D-(+)-glucose in a 1 L bottle
to prepare fresh Ringers solution. Carboxygenate the Ringers
solution by bubbling with 95 % O2/5 % CO2 in 37 C water
bath for at least 30 min.
2. Dilute the stimulus stock solutions into carboxygenated
Ringers solution in 35 mL falcon tubes to desired concentra-
tion. Dilute the same amount of solvent as the Ringers con-
trol. Prepare at least two tubes of Ringers controls (see Note
3). Mix the stimuli solution gently by inverting the tubes back
and forth several times (see Note 4).
3. Load the stimulus solution into the perfusion system. First,
switch on the gas to a pressure level allowing flow rate of
2.2 mL/min. Then, for each stimulus, pour the stimulus solu-
tion into the syringe reservoir, and briefly blow the unfilled
space of the syringe with pressurized gas before sealing with
the rubber cap. Keep the stimulus solution pressurized
throughout the experiment.
204 Pei Sabrina Xu and Timothy E. Holy

4. Test each stimulus line by briefly switching on the valves one

by one. Chase the air bubbles and make sure that flow rate is
consistent among all lines.
5. Save ~200 mL carboxygenated Ringers solution on ice for
VNO dissection. Transfer the remaining Ringers solution into
a 2 L bottle reservoir in the superfusion system and check its
flow as in the previous step. It will provide Ringers flush to the
VNO tissue following each stimulus solution.
6. Before moving to the next step, make sure that the following
list is done:
Ensure that the optical table is floating.
Launch the imaging acquisition software Imagine.
Power on the laser and laser shutter controller (but keep the
shutter closed).
Empty the container for liquid waste collected by suction.
If the camera is equipped with water-cooling system, switch
on the water cooling and shut down the internal fan (to
minimize vibration). If necessary, set the optimized working
temperature for the camera (we set 55 C for the Andor
ixon 885 EMCCD camera).

3.2 Prepare 1. Prior to performing the VNO dissection, get all the following
VNO Sample items ready around the dissection microscope, including all dis-
section and mounting tools (see items 2 and 3 of Subheading 2.2).
In addition, prepare the following three Petri dishes: a 25 mm
Petri dish filled with ice-cold carboxygenated Ringers solution
(dish #1), a Sylgard-coated Petri dish in the same way (dish #2),
an empty and dry Sylgard-coated Petri dish (dish #3), a piece of
kimwipes tissue that is rolled up into a compact stick, and a
piece of nitrocellulose membrane (see Note 5). If fixing the
nitrocellulose membrane into the imaging chamber involves
assembly, get assembly tools ready.
2. Euthanize the mouse by CO2 and decapitate. Remove the VNO
quickly and place into dish #1. Under the stereomicroscope,
using #3 forceps carefully remove the bony capsule from one
VNO (see Note 6), and transfer the VNO into dish #2. With #5
forceps or spring scissors, separate the VNO neural epithelium
from the blood vessel by cutting along the edge of the VNO.
More detail about VNO dissection is described in [10].
3. Gently suck up the dissected VNO together with small amount
of liquid into the tip of glass pipet, and then place down VNO
with a drop of liquid at the center of the precooled dish #3.
Under the microscope, adjust the VNO position that (1) it is
in the center of the liquid drop and (2) the basal lamina side
faces up.
 VNO Whole Mount Imaging 205

4. Use the tip of the kimwipes tissue stick to carefully dip around
the VNO in order to get rid of the liquid drop. To best pre-
serve the VNO activity, it is recommended to complete the
following three actions within 5 s: (1) obtain a well-flat VNO
tissue (see Note 7) with basal lamina side up in minimal liquid
volume (see Note 8); (2) hold the nitrocellulose membrane
with a #3 forceps, and carefully put down the membrane in a
way that the center of the membrane touches the VNO; and
(3) add a drop of fresh ice-old oxygenated Ringers solution to
the back of the nitrocellulose membrane.
5. Add more drops of fresh Ringers solution until the nitrocel-
lulose membrane floats. The VNO should be firmly adhered.
Transfer the membrane into the imaging chamber. Fill the
imaging chamber with fresh Ringers solution as quickly as
possible to preserve the tissue health.

3.3 Recording 1. Place the imaging chamber under the OCPI microscope.
Position the heater probe above the VNO tissue. Superfuse the
tissue with heated Ringers solution at 35 C (see Note 9).
Position the suction probe on the liquid surface at a distant
corner of the imaging chamber.
2. Lower the OCPI microscope objective, until the objective
immerses into the chamber. Turn on the laser shutter and start
preview in Imagine. Adjust the chamber position by translat-
ing stage control knob while further lowering down the objec-
tive until the VNO images appear in the preview on the
computer screen (see Note 10).
3. Adjust the translating stage to position VNO in the center of
the filed of view. Sweep the piezo to scan the whole VNO and
check if the tissue is intact. Once done, close the laser shutter
and preview.
4. Since the imaging chamber has been repositioned, adjust the loca-
tion of heater probe and suction probe accordingly (see Note 11).
5. Turn off the room light. Leave the VNO tissue being super-
fused for at least 30 min to acclimate (see Note 12).
6. Set up the recording parameters in Imagine, which includes
the following: (1) generate the randomly ordered stimulus
file by custom software and load into Imagine (see Note 13);
(2) set the following parameters (values from our typical exper-
iment are listed as reference):
Piezo travel range: 200 m.
Exposure time: 50 ms/frame.
Frame number in a stack: 40.
Number of stacks to acquire: Set according to the loaded stim-
ulus file. For a typical experiment with 15 stimuli and 4 repeats,
206 Pei Sabrina Xu and Timothy E. Holy

duration of each stimulus is 5 stacks, duration of interval flush

is 10 stacks, and pre-leading flush is 10 stacks. The total
number of stacks for such a recording is 910. We acquire one
stack every 5 s, so the typical experiment with 910 stacks lasts
for ~75 min.
7. After 30 min, dump the first 20-s stimulus solution for each line
one by one before recording (see Note 14). Flush the tissue for
3 min.
8. Adjust the laser power to obtain the appropriate fluorescence level
by monitoring the intensity profile in Imagine (see Note 15).
9. Begin recording. Imaging data are saved to disk. Keep the
room dark and uninterrupted during recording.

3.4 Data Analysis 1. Analyze the imaging data offline. The resulting data sets can be
used to address many different questions. For example, from
the raw trace intensity (F) of each cell (see Fig. 1), one can
measure the F/F of recorded VSNs in response to particular
stimuli, and also get the 3D position of the VSNs. Because of
large 3D volume as well as long hours of recording, a data set
is large in size. Analyzing such data sets requires programming
skills (C, Matlab, R, Python, or similar). In addition, during
long hours of recording, VNO tissue undergoes detectable
deformity, such as swelling, shrinking, and wrapping. Thus,
3D image registration is required in the data analysis.

4 Notes

1. Be aware that solvents like methanol and DMSO may cause

tissue damage or artificial signal during recording. Adding
equivalent solvent in control groups is recommended. In our
experiments, solvent concentration never exceeds 0.01 % in the
final stimulus solutions.
2. The resulting glass dropper has a wide opening with smooth
glass walls, which generates minimal damage to the tissue during
VNO transfer.
3. Because fluorescence intensity can change even for small
differences in carboxygenation, one Ringers control will be
used to subtract any artifact response. The other Ringers
control will serve as the negative control for tested stimuli (and
is treated exactly as if it is one of the stimulus solutions).
4. To preserve the health of the VNO tissue in the experiment,
care should be taken to preserve oxygen inside the solution:
avoid rigorous shaking, minimize the time of preparing stimulus
solution before loading into the pressurized superfusion system,
etc. Handle each stimulus tube consistently to ensure consistent
carboxygenation level.
 VNO Whole Mount Imaging 207

297 295
346
305
312

267

b
Stimulus GDMNC F H EOK B J L A I EB DOCG F A NMK L H I J L J HCNA B K EG I MF DO I MGK NA F H L B EDC J O

Cell #267

Cell #295

Cell #297

Cell #305

Cell #312

Cell #346
Intensity
2000 ct
5 min

Fig. 1 Raw images and representative raw traces of individual VSNs labeled by GCaMP2. (a) Shows one of the
frames in a three-dimensional stack image of a single VNO preparation. Six example VSNs are noted in white
circles and labeled by numbers. Scale bar = 50 m. (b) Shows the fluorescent intensity traces of example VSNs
throughout a 75-min recording. The top color bars and letters represent 15 different stimuli (BN: 13 different
sulfated steroids [8], A and O: Ringers negative controls). Each stimulus was tested four times, in randomized
order. Each example VSN responds to particular sulfated steroids with high reproducibility in all four trials.
Example neurons were chosen nearly at random from regions of interest within this optical section, and are
typical of the majority of neurons in the recording

5. If the nitrocellulose membrane breaks easily when mounting in

the imaging chamber, one can attach a piece of black electrical
tape to the back of the nitrocellulose membrane in advance.
6. The other VNO tissue from the same mouse can be dissected
and mounted in the same way. Keep it (after adhering it to the
nitrocellulose membrane) in the carboxygenated Ringers on
ice. Cover the ice bucket to prevent the bleaching of fluores-
cent calcium indicator. Changing the top half of the medium
with fresh carboxygenated Ringers 12 times per hour is
recommended. The VNO tissue can last for at least 4 h in the
ice bucket without obvious compromised responsiveness.
208 Pei Sabrina Xu and Timothy E. Holy

7. It is important to make the VNO flat. Normally when the

whole dissection procedure is performed quickly (within
1218 min), the VNO is relatively flat; and it automatically
flattens as the liquid drop is wicked away. However, when
the dissection takes too long, the VNO starts to fold. In that
case, try to flatten the VNO using #5 forceps under the
stereomicroscope.
8. If too much liquid remains, VNO tissue will not attach to the
nitrocellulose membrane firmly. Tissue may float up during
later recording.
9. At room temperature, VNO tissue may die in minutes without
carboxygenated Ringers solution, so be certain to keep the
Ringers flush flowing throughout the experiment. If you
temporarily stop the superfusion, remember to switch off the
heater, and keep in mind that the solution sitting inside the
heater probe will be substantially overheatedyou will likely
need to dump this solution without flowing it onto living
tissue.
10. Never bump the objective onto the tissue or any part of the
imaging chamber. The most straightforward approach is to find
the tissue and/or bottom of the chamber using reflected light,
which, because it can be performed with very-low-intensity
illumination, has the further advantage of limiting any photo-
bleaching while finding the tissue. Set the illumination to very
low power (you may need to reduce the laser intensity by
adding an extra neutral density filter in the light path) and tem-
porarily take out the emission filter. Lower the objective until
you see a line of light corresponding to reflection of the light
sheet off the bottom of the chamber. Move the stage until the
tissue (which will not show cellular detail due to laser speckle)
is in the field of view. You can then restore the emission filter,
adjust the laser intensity, and check the fluorescent image. Keep
the light power low when the emission filter is out of the imag-
ing pathway, to avoid any risk of damaging the camera.
11. With little space between the objective and light sheet optics,
heater probes can easily hit the objective or the light sheet
optics when one lowers the objective. To avoid this, initially
position the heater probe higher up and off to the side of the
VNO itself. Once the tissue has been positioned in the field of
view, one can safely reposition the heater probe directly above
the VNO. After that, one should avoid changing the position
of the heater probe, especially after the tissue has situated
under the constant superfusion pressure. Otherwise, the tissue
may change shape due to the changed superfusion direction,
pressure, etc.
12. Be patient when waiting. A good VNO tissue will well maintain
responsiveness to stimuli in the superfused chamber for over 4 h.
 VNO Whole Mount Imaging 209

On the other hand, even a perfectly mounted VNO tissue will

shift by at least 100 m in the first 30 min (probably because of
the new superfusion environment). Such large shifts are difficult
to correct at the analysis step. Normally, after 30 min, VNO tis-
sue will keep relatively stable geometry if the heater probe
remains in the same position. The small amount of remaining
warping may be addressed by image registration during data
analysis.
13. Stimuli are randomly ordered, and repeated 45 times each
over the duration of the experiment.
14. Solutions sitting in the Teflon tubing and the silicon pinch
tubing over prolonged periods may lose oxygen and CO2,
causing artificial fluorescence increases in the whole VNO tissue.
Thus, one should dump solutions in all lines right before the
recording.
15. One should carefully choose the laser power. There are two
factors to consider: first, to achieve a decent signal-to-noise
ratio, one needs to collect enough photons from each cell; sec-
ondly, one should minimize the laser power to avoid photo-
bleaching and phototoxicity of the tissue. To balance the two
factors, calculate how many photons (N) are needed to achieve
your target signal-to-noise ratio. For example, suppose you
need to be sensitive to F/F of 1 % for each cell. When the
dominant source of noise is shot noise, the noise is N. So for
N/N < =1 %, one requires N > =10,000 photons/cell. If the
camera gain is 15 (one photon gives a signal of 15 counts),
then counts required for each cell are 10,000 15 = 150,000. If
a cell occupies 100 pixels, each pixel needs 1,500 counts. Thus,
one should adjust the laser power to get an average intensity
no less than 1,500 counts above the camera bias.

Acknowledgements

This work was supported by NIH/NIDCD 005964, NIH/NINDS

068409, and NIH/NDPA 006437 (T.E.H.).

References
1. Dving KB, Trotier D (1998) Structure and coupled planar illumination microscopy.
function of the vomeronasal organ. J Exp Biol Neuron 57:661672
201:29132925 4. Boschat C, Plofi C, Randin O et al (2002)
2. Holy TE, Dulac C, Meister M (2000) Responses Pheromone detection mediated by a V1r vom-
of vomeronasal neurons to natural stimuli. eronasal receptor. Nat Neurosci 5:12611262
Science 289:15691572 5. Leinders-Zufall T, Brennan P, Widmayer P et al
3. Holekamp TF, Turaga D, Holy TE (2008) (2004) MHC class I peptides as chemosensory
Fast three-dimensional fluorescence imaging signals in the vomeronasal organ. Science
of activity in neural populations by objective- 306:10331037
210 Pei Sabrina Xu and Timothy E. Holy

6. Chamero P, Marton TF, Logan DW et al volumetric calcium imaging. J Neurosci 32:

(2007) Identification of protein pheromones 16121621
that promote aggressive behaviour. Nature 9. Arnson HA, Holy TE (2011) Chemosensory
450:899902 burst coding by mouse vomeronasal sensory
7. He J, Ma L, Kim S et al (2008) Encoding gen- neurons. J Neurophysiol 106:409420
der and individual information in the mouse 10. Arnson HA, Fu X, Holy TE (2010)
vomeronasal organ. Science 320:535538 Multielectrode array recordings of the vom-
8. Turaga D, Holy TE (2012) Organization of eronasal epithelium. J Vis Exp. doi:
vomeronasal sensory coding revealed by fast 10.3791/1845
 Chapter 15

Calcium Imaging of Vomeronasal Organ Response

Using Slice Preparations from Transgenic Mice
Expressing G-CaMP2
C. Ron Yu

Abstract
The vomeronasal organ (VNO) in vertebrate animals detects pheromones and interspecies chemical signals.
We describe in this chapter a Ca2+ imaging approach using transgenic mice that express the genetically
encoded Ca2+ sensor G-CaMP2 in VNO tissue. This approach allows us to analyze the complex patterns
of the vomeronasal neuron response to large number of chemosensory stimuli.

Key words Vomeronasal organ, Pheromone, Urine, Slice preparation, G-CaMP2, Calcium imaging

1 Introduction

The vomeronasal organ (VNO) detects pheromones; chemical

cues that carry information about the social, sexual, and reproduc-
tive status of the individuals within the same species [1, 2]; as well
as signals emitted by animals and other species [3]. These signals
activate the vomeronasal sensory neurons (VSNs) with high levels
of specificity and sensitivity [4]. At least three distinct families of
G-protein-coupled receptors, V1R, V2R, and FPR [514], are
expressed in VNO neurons to mediate the detection of the chemo-
sensory cues. To understand how pheromone information is
encoded by the VNO, it is critical to analyze the response profiles
of individual VSNs to various stimuli and identify the specific
receptors that mediate these responses.
The neuroepithelia of VNO are enclosed in a semi-blind tubu-
lar structure, with one open end (the vomeronasal duct) connecting
to the nasal cavity. VSNs extend their dendrites to the lumen part of
the VNO, where the dendritic knobs and microvilli interact with
pheromone cues to generate electrical currents. The currents trans-
mit pheromone activation to the cell body to evoke action poten-
tials. Various approaches have been developed to detect responses

Kazushige Touhara (ed.), Pheromone Signaling: Methods and Protocols, Methods in Molecular Biology, vol. 1068,
DOI 10.1007/978-1-62703-619-1_15, Springer Science+Business Media, LLC 2013

211
212 C. Ron Yu

of VSNs to sensory stimuli, including electrophysiological recordings

and calcium imaging [4, 12, 1519]. Sensory stimulation elicits
strong Ca2+ influx in VSNs that is indicative of receptor activation
[4, 20]. Imaging calcium signals, therefore, provides a means to
monitor large number of neurons.
Here I describe a method using acute slice preparation from
transgenic mice expressing the genetically encoded calcium sensor
G-CaMP2 to conduct calcium imaging. Traditional calcium imag-
ing utilizes synthetic calcium dyes. The dye loading processes are
usually invasive and often cause damage to tissues. The detection
of the calcium signals can also be affected by the uneven loading
and the limited penetration of the dye in the tissue. The use of
genetically encoded sensor provides a sensitive readout of the neu-
ronal responses. The specific expression of the sensor in targeted
tissue and cell type make it feasible to monitor response from spe-
cific population of neurons. These sensors also make it possible for
chronic imaging in live animals. The G-CaMP proteins are calcium
sensors that have been used in a variety of animal systems. They
provide some of the highest sensitivity and temporal response to
calcium transients [2124].
We have developed transgenic mice that express G-CaMP2 in
the olfactory sensory neurons, including the VSNs [15, 25]. The
sensitivity and the genetic nature of the probe greatly facilitate Ca2+
imaging experiments by eliminating the dye loading process [4, 20].
We also employ a ligand delivery system that enables application of
various stimuli to the VNO slices. The combination of the two
techniques allows us to monitor multiple neurons simultaneously
in response to large number of stimuli. These procedures are
described in this chapter.

2 Materials

2.1 Solutions, 1. Prepare three stock 10 stock solutions (R1, R2, and R3)
Stimuli, and according to the table. They will be used to prepare the mouse
Embedding Materials artificial cerebrospinal fluid (mACSF) and Ringers solutions.
The solutions should be kept at 4 C.
10 R1 stock solution: In 0.8 L water, dissolve 73.05 g NaCl
1.86 g KCl, 2.94 g CaCl22H2O, and 1.72 g NaH2PO4H2O.
Add 10 mL of 1 M stock MgCl2 solution. Add water to a final
volume of 1 L.
10 R2 stock solution: In 0.8 L water, dissolve 21 g NaHCO3.
Add water to a final volume of 1 L.
10 R3 stock solution: In 0.8 L water, dissolve 73.05 g NaCl
1.86 g KCl, 2.94 g CaCl22H2O, and 1.72 g NaH2PO4H2O.
Add 20 mL of 1 M stock MgCl2 solution. Add 50 mL of 1 M
stock HEPES solution. Add water to a final volume of 1 L.
 Calcium Imaging of Vomeronasal Organ Response Using Slice Preparations 213

2. On the day of the experiment make fresh mACSF: In 0.7 L

double-distilled water, add 100 mL of 10 R1 and 100 mL of
10 R2. Add 1.8 g of dextrose. Add water to a final volume of
1 L. The mACSF should have osmolarity of 310315 mOsm/L.
Adjust the osmolarity with dextrose or water if necessary.
Aerate the solution with carboxygen gas, containing 95 % oxy-
gen and 5 % carbon dioxide, for at least 30 min before adjust-
ing the pH of the solution to 7.27.4. This method prevents
calcium carbonate precipitation at high pH.
3. On the day of the experiment make 1 L of Ringers solution:
In 0.7 L water, add 100 mL of 10 R2 and 100 mL of 10 R3.
Add 1.8 g of dextrose. Add water to final volume of 1 L. Adjust
if necessary to the same osmolarity and pH as the mACSF.
Aerate the solution with carboxygen gas.
4. Prepare 4 % low-melting agarose (LMA) in Ringers solution:
Weight 0.4 g LMA and add to 10 mL Ringers solution. Melt
the LMA and aliquot the agarose in Eppendorf tubes and store
at 4 C till use.
5. Prepare pheromone stimuli in Ringers solution. For mouse
urine, 1:100 dilution gives robust response. For chemical com-
pounds, peptides, and proteins, dissolve them at the desired
concentration.

2.2 Mice Compound heterozygotic mice containing both the OMP-IRES-

tTA allele and the tetO-G-CaMP2 allele are used. The two lines
have been deposited at the Jackson laboratories. Stock numbers are
no. 017754 (OMP-IRES-tTA) and no. 017755 (tetO-G-CaMP2),
respectively.

2.3 Vibratome 1. Any vibratome for live brain tissues can be used. Our labora-
and Microscope tory uses the VF300 tissue slicer (Precisionary Instruments,
Greenville, NC) to make sections.
2. We use the Zeiss AxioSkope FS2 microscope with a 10 or a
20 water-dipping lens for time-lapse imaging. Standard GFP
band-pass filter (450490 nm) is used for G-CAMP2 signals.
The epifluorescent images are acquired by a CCD camera
(Zeiss HRM) with 1 1 or 2 2 binning depending on the
expression levels of G-CaMP2.

2.4 Perfusion 1. Set up a perfusion system on the microscope stage. The perfu-
System sion system includes three main parts: a delivery system provid-
ing constant flow of oxygenated mACSF, a stimulus delivery
system, and a suction system to remove excessive liquid
(Fig. 1a).
2. Place a perfusion chamber (Siskiyou, Grants Pass, OR) on the
stage (Fig. 1b).
214 C. Ron Yu

a Perfusion Tip b

Suction
Outlet
Superfusion Suction
Inlet Outlet

Perfusion Chamber
Anchor

c Syringe pump

Sample 300-600 l/min

Loop
Pump

Sample Wash
Loading To VNO Loading
Port

Waste
Load Inject
Fig. 1 Illustration of perfusion system setup. (a) A typical perfusion chamber with inlet and outlet. The perfusion
chamber is placed on microscope stage under the dipping objective. The mACSF inlet, suction outlet, and
perfusion tip are indicated. (b) Top view of the perfusion chamber. A VNO slice is positioned in the center of the
chamber and pressed down with a tissue anchor. (c) Schematic illustration of the flow directions at the load
and inject positions. The single-barrel syringe pump provides continuous flow of Ringers through the perfu-
sion tip. Light gray contour illustrates the ports in the HPLC injection port. The dark gray line illustrates the
injection loop. Black lines illustrate the flow of liquid within the port at different positions. Arrowheads indicate
the directions of liquid flow

3. Place two perfusion port holders (MPIOH-S ALA scientific,

Inc.) on both sides of the chamber, one holding the mACSF
superfusion inlet and the other holding the suction needle.
4. Position a micromanipulator alongside the superfusion inlet
port to hold the perfusion tip such that it can be position to
deliver the stimuli.
5. Place a reservoir of oxygenated mACSF above the imaging
setup and connect the outflow tubing to the superfusion inlet.
A gravity-driven continuous flow is fed into the imaging cham-
ber at a speed of ~1 mL/min.
6. Set up stimulus delivery system. We use an HPLC injection
loop (V-451, Chromtech, Apple Valley, MN). The injection
loop (PEEK sample loop, 1803, Chromtech) has two flow
routes controlled by a switch (Fig. 1c). At either position, a
constant flow of Ringers solution is injected by a syringe pump
(NE-300, New Era Pump Systems, Farmingdale, NY). At the
load position, the flow bypasses the sample loop and goes
directly to the outlet connected to the perfusion tip. Stimulus
solution can be injected into the sample loop using a Hamilton
 Calcium Imaging of Vomeronasal Organ Response Using Slice Preparations 215

precision syringe (80630, Chromtech). The liquid contained in

the sample loop will be the volume delivered. Excess sample
exits the sample loop through waste outlet. The volume being
injected can be controlled by the size of the loop. We use a
20 L loop in our experiments. When switched to the injec-
tion position, the pump solution flows through the sample
loop and pushes the stimuli into the outlet. Other commercial
or custom-made injection systems can also be implemented for
stimulus delivery.

3 Methods

3.1 Slice Preparation 1. Before sacrificing the animal, melt two tubes of LMA on a heat
block at >60 C. As soon as the gel liquefies, transfer the tubes
to a 37 C heat block.
2. Decapitate a G-CAMP2 mouse following CO2 euthanasia
(see Note 1). Cut the mandible bones with scissors and remove
the lower jaw. Peel off the ridged upper palate tissue to expose
the nasal cavity (Fig. 2a). Insert a surgical blade between the two
front incisors to expose VNO. Lift the whole VNO from the
nasal cavity by holding on the tail bone (Fig. 2b). Immediately
transfer the VNO to oxygenated mACSF solution placed on ice.
3. Under dissection scope, separate the two VNOs by sliding the
tip of a pair of #5 forceps gently along the wall of septal bone
(Fig. 2c). Peel away the vomer bone encasing the VNO. Gently
lift the VNO from the bone cavity (see Note 2).
4. Hold the posterior end of the VNO with a pair of forceps and
gently submerge it into the melted agarose (Fig. 2d).
Immediately place the tube on ice to solidify the agarose
(see Note 3).
5. Proceed to sectioning immediately after the LMA solidifies.
Supply cold oxygenated mACSF into the sectioning chamber
and start cutting at 150200 m thickness per slice (Fig. 2d;
see Note 4).
6. Collect and transfer the sectioned slices to mACSF incubation
chamber. The slices are viable for 68 h in oxygenated mACSF
at room temperature.

3.2 Imaging 1. Place the VNO slice in the middle of the perfusion chamber
Chamber Setup and hold the slice down with a slice anchor (Warner Instruments,
Hamden, CT). Oxygenated mACSF is delivered to the perfu-
sion chamber through inlet port at ~1 mL/min and the liquid
is drained through the suction needle at the opposite side of
the perfusion chamber (see Note 5).
2. Fill a 30 mL syringe with Ringers solution and clamp it to the
syringe pump. Set the pump speed to 300600 l/min to
216 C. Ron Yu

a b
soft tissue Septum Vomer
Bone

Tail Bone VNO

B.V.
B.V.
Incisors Tail VNO
Molars Bone

P A

c Forceps d
Blade
Septum
VNO
Barrel
LMA
Holder
Vomer
Bone
B.V.
N.E.
Embed Cut

Fig. 2 Schematic illustration of VNO dissection process. (a) The anatomical location of VNO in the mouse head.
The drawing illustrates the head of a mouse laid upside down, with the jaw removed and the palate peeled to
expose the VNO. (b) A side view of the isolated VNO that is enclosed in the vomer bone. (c) A coronal view of
the VNO and dissection process. One VNO is separated from the septum and the vomer bone can then be
removed to extricate the neuroepithelium. (d) VNO is embedded into LMA. The embedded block is glued to the
tissue holder for sectioning. The tissue holder is advanced at 180200 m per slice pushing the agarose block
out of the metal barrel for sectioning. The cutting blade is positioned closely to the metal barrel. B.V. blood
vessel. N.E. neuroepithelium

provide a continuous flow of Ringers solution over the slice

(Fig. 1c; see Note 6).
3. Connect the outlet of the Ringers to the HPLC injection loop.
4. Adjust the perfusion tip under a 5 or a 10 lens so that the tip
is about 1 mm away from the VNO slice.
5. Once the perfusion system is set up, test the device by measur-
ing the sample delay time and duration using a solution con-
taining a fluorescent dye. Load 0.1 % rhodamine 6G dissolved
in Ringers solution into the sample loop, switch the valve to
inject position, and perform time-lapse imaging as would for
pheromone stimuli (see Note 7).
 Calcium Imaging of Vomeronasal Organ Response Using Slice Preparations 217

3.3 Time-Lapse 1. Set the acquisition speed to one frame per second. Adjust the
Imaging intensity of the light to minimize bleaching of the G-CaMP2
signals and photo damage to the cells.
2. Set the injection loop to load position. Load the injection
loop with stimulus solution.
3. Start image acquisition.
4. Switch the injection valve from load to injection position at a
specific time point (e.g., 5 s after start) for one set of experiment
to obtain consistent time delay in all trials (see Notes 6 and 8).
5. End acquisition at desired time. Typically we acquire a 60-frame
image stack (~6070 s).
6. Wait for 410 min for the VSNs to recover before applying the
next stimulus (see Note 9).

3.4 Data Analysis 1. Perform image registration of all the images acquired from one
slice. We use a custom-written VBA script in AxioVision (Carl
Zeiss, North America) to automate this process (see Note 10).
All image frames within the same experiment are registered
against a common chosen reference frame with elastic registra-
tion (see Note 11). This reference frame is chosen arbitrarily
from the image stacks.
2. Perform image subtraction to identify the responding cells. We
use custom-written macros in ImageJ v1.42 (http://rsb.info.
nih.gov/ij/, NIH, Bethesda, MD) to automate this process.
A minimal projection image is generated for each stack.
Responding cells emerge after the minimal projection is sub-
tracted from the raw stacks.
3. Identify the regions of interest (ROIs) from the subtracted
stacks. Obtain the ROI coordinates using Multi-Measure
PlugIn from ImageJ. Process all stacks for one experiment and
save all ROI coordinates in an ROI master list.
4. Use the master list of ROI to measure cell responses from raw
image stacks with custom-written macro and Multi-Measure
PlugIn from ImageJ.
5. Data produced by Multi-Measure are exported into a spread-
sheet for further analysis.

3.5 Alternative In addition to the mice discussed above, another mouse line could
Approach offer similar convenience and signal sensitivity for recording Ca2+
signals from VNO slices. Recently, a Cre-dependent reporter line
that expresses a new version of G-CaMP, G-CaMP3, has been
developed [26]. G-CaMP3 has been shown to have lower basal
level of fluorescence, and therefore afford a better signal [24].
One could cross this reporter line with a line that expressed the
Cre recombinase in the mouse olfactory systems [27, 28]. The
218 C. Ron Yu

resulting line is expected to express G-CaMP3 in the VNO.

However, this alternative approach has not been tested. The
G-CaMP3 expression is driven by the actin promoter. Whether it
can drive G-CaMP3 expression to a level that allows slice imaging
has yet to be confirmed.

4 Notes

1. Please follow the procedures approved by the Institutional

Animal Care and Use Committee.
2. Extreme care should be taken in this step not to damage the
neuroepithelium. Small bone fragments left on the tissue sur-
face must be removed completely before embedding.
Fragments left on the tissue may be caught by the cutting
blade to pull the tissue out of the agarose block.
3. The embedding and cooling process should take less than
2 min to ensure the health of the tissue.
4. Adjust the advancing speed and vibration frequency so that the
tissue is not compressed during sectioning and VNO does not
fall off from the agarose.
5. The threads of the anchor should only press against the LMA
part of the slice but not the VNO tissue.
6. Prior to the imaging experiment, air bubbles should be chased
out to ensure smooth flow of the perfusion fluid.
7. Perform a test run using Ringers solution as the stimulus.
Readjust the perfusion setup if movement artifact is introduced
during sample injection.
8. Perform a positive control run with mouse urine diluted at
1:100 in Ringers solution. The typical maximal F/F value of
G-CaMP2 response to mouse urine is around 2040 %. If
needed, one can confirm the viability of the slice by delivering
10 mM KCl in Ringers to stimulate the slices at the end of the
experiments.
9. Wash the Hamilton syringe in Ringers solution at least three
times after loading one stimulus. Wash the sample loop with
Ringers solution at least three times between different stimuli.
These steps prevent cross contamination among different
samples.
10. Image registration can also be done in ImageJ.
11. Elastic registration, also known as nonlinear registration, is a
category of image registration technique emphasizing the
transformation of a target image non-rigidly to a reference
image. Here we used the implementation from AxioVision.
 Calcium Imaging of Vomeronasal Organ Response Using Slice Preparations 219

Acknowledgments

This work is supported by funding from Stowers Institute and the

NIH (NIDCD 008003) to C.R.Y.

References

1. Birch MC (1974) Pheromones. American 14. Rodriguez I, Del Punta K, Rothman A, Ishii T,
Elsevier Pub. Co., New York Mombaerts P (2002) Multiple new and isolated
2. Wyatt TD (2003) Pheromones and animal families within the mouse superfamily of V1r
behaviour: communication by smell and taste. vomeronasal receptors. Nat Neurosci 5:
Cambridge University Press, Cambridge, UK 134140
3. Papes F, Logan DW, Stowers L (2010) The 15. He J, Ma L, Kim S, Nakai J, Yu CR (2008)
vomeronasal organ mediates interspecies defen- Encoding gender and individual information in
sive behaviors through detection of protein the mouse vomeronasal organ. Science 320:
pheromone homologs. Cell 141:692703 535538
4. Leinders-Zufall T, Lane AP, Puche AC, Ma W, 16. Holy TE, Dulac C, Meister M (2000)
Novotny MV, Shipley MT, Zufall F (2000) Responses of vomeronasal neurons to natural
Ultrasensitive pheromone detection by mam- stimuli. Science 289:15691572
malian vomeronasal neurons. Nature 405: 17. Chamero P, Marton TF, Logan DW, Flanagan
792796 K, Cruz JR, Saghatelian A, Cravatt BF, Stowers
5. Dulac C, Axel R (1995) A novel family of genes L (2007) Identification of protein pheromones
encoding putative pheromone receptors in that promote aggressive behaviour. Nature
mammals. Cell 83:195206 450:899902
6. Herrada G, Dulac C (1997) A novel family of 18. Leinders-Zufall T, Ishii T, Mombaerts P, Zufall
putative pheromone receptors in mammals F, Boehm T (2009) Structural requirements
with a topographically organized and sexually for the activation of vomeronasal sensory
dimorphic distribution. Cell 90:763773 neurons by MHC peptides. Nat Neurosci 12:
7. Matsunami H, Buck LB (1997) A multigene 15511558
family encoding a diverse array of putative 19. Kimoto H, Haga S, Sato K, Touhara K (2005)
pheromone receptors in mammals. Cell 90: Sex-specific peptides from exocrine glands
775784 stimulate mouse vomeronasal sensory neurons.
8. Ryba NJ, Tirindelli R (1997) A new multigene Nature 437:898901
family of putative pheromone receptors. 20. Leinders-Zufall T, Brennan P, Widmayer P, S
Neuron 19:371379 PC, Maul-Pavicic A, Jager M, Li XH, Breer H,
9. Pantages E, Dulac C (2000) A novel family of Zufall F, Boehm T (2004) MHC class I pep-
candidate pheromone receptors in mammals. tides as chemosensory signals in the vomerona-
Neuron 28:835845 sal organ. Science 306:10331037
10. Zhang X, Rodriguez I, Mombaerts P, Firestein 21. Hendel T, Mank M, Schnell B, Griesbeck O,
S (2004) Odorant and vomeronasal receptor Borst A, Reiff DF (2008) Fluorescence changes
genes in two mouse genome assemblies. of genetic calcium indicators and OGB-1 cor-
Genomics 83:802811 related with neural activity and calcium in vivo
and in vitro. J Neurosci 28:73997411
11. Liberles SD, Horowitz LF, Kuang D, Contos
JJ, Wilson KL, Siltberg-Liberles J, Liberles DA, 22. Pologruto TA, Yasuda R, Svoboda K (2004)
Buck LB (2009) Formyl peptide receptors are Monitoring neural activity and [Ca2+] with
candidate chemosensory receptors in the genetically encoded Ca2+ indicators. J Neurosci
vomeronasal organ. Proc Natl Acad Sci U S A 24:95729579
106:98429847 23. Jayaraman V, Laurent G (2007) Evaluating a
12. Riviere S, Challet L, Fluegge D, Spehr M, genetically encoded optical sensor of neural
Rodriguez I (2009) Formyl peptide receptor- activity using electrophysiology in intact adult
like proteins are a novel family of vomeronasal fruit flies. Front Neural Circuits 1:3
chemosensors. Nature 459:574577 24. Tian L, Hires SA, Mao T, Huber D, Chiappe
13. Yang H, Shi P, Zhang YP, Zhang J (2005) ME, Chalasani SH, Petreanu L, Akerboom J,
Composition and evolution of the V2r vom- McKinney SA, Schreiter ER, Bargmann CI,
eronasal receptor gene repertoire in mice and Jayaraman V, Svoboda K, Looger LL (2009)
rats. Genomics 86:306315 Imaging neural activity in worms, flies and
220 C. Ron Yu

mice with improved GCaMP calcium indica- dependent GCaMP3 reporter mouse for neuro-
tors. Nat Methods 6:875881 nal imaging in vivo. J Neurosci 32:31313141
25. He J, Ma L, Kim S, Schwartz J, Santilli M, 27. Eggan K, Baldwin K, Tackett M, Osborne J,
Wood C, Durnin MH, Yu CR (2010) Distinct Gogos J, Chess A, Axel R, Jaenisch R (2004)
signals conveyed by pheromone concentrations Mice cloned from olfactory sensory neurons.
to the mouse vomeronasal organ. J Neurosci Nature 428:4449
30:74737483 28. Li J, Ishii T, Feinstein P, Mombaerts P (2004)
26. Zariwala HA, Borghuis BG, Hoogland TM, Odorant receptor gene choice is reset by
Madisen L, Tian L, De Zeeuw CI, Zeng H, nuclear transfer from mouse olfactory sensory
Looger LL, Svoboda K, Chen TW (2012) A Cre- neurons. Nature 428:393399
 Chapter 16

The Electrovomeronasogram: Field Potential

Recordings in the Mouse Vomeronasal Organ
Trese Leinders-Zufall and Frank Zufall

Abstract
Mammalian vomeronasal neurons (VSNs) located in the sensory epithelium of the vomeronasal organ
(VNO) detect and transduce molecular cues emitted by other individuals and send this information to the
olfactory forebrain. The initial steps in the detection of pheromones and other chemosignals by VSNs
involve interaction of a ligand with a G protein-coupled receptor and downstream activation of the
primary signal transduction cascade, which includes activation of ion channels located in microvilli and the
dendritic tip of a VSN. The electrovomeronasogram (EVG) recording technique provides a sensitive
means through which ligand-induced activation of populations of VSNs can be recorded from the epithe-
lial surface using an intact, ex vivo preparation of the mouse VNO. We describe methodological aspects of
this preparation and the EVG recording technique which, together with single-cell recordings, contrib-
uted significantly to our understanding of mammalian vomeronasal function, the identification of phero-
monal ligands, and the analysis of mice with targeted deletions in specific signal transduction molecules
such as Trpc2, Go, V1R, or V2R receptors.

Key words Field potential, Generator potential, Receptor potential, EOG, EVG, Multibarrel pipette,
Agar bridge, Nonmetallic syringe needle, VNO dissection, TRP channel

1 Introduction

An important technique for measuring electrical activity of nasal

sensory neuron populations is the local field potential recorded
extracellularly from the surface of the nasal sensory epithelium in
response to chemical stimulation. In the main olfactory epithelium,
this technique is referred to as electroolfactogram (EOG) [1, 2]
whereas it is called electrovomeronasogram (EVG) when signals
from the sensory epithelium of the vomeronasal organ (VNO) in
the accessory olfactory system are measured [35]. In both cases,
these field potentials represent summated primary responses of
sensory neuron populations induced by molecular stimuli. The size
of the field potential was shown to be associated with a change in
action potential firing frequency [6] as predicted for a generator
(receptor) potential [7].

Kazushige Touhara (ed.), Pheromone Signaling: Methods and Protocols, Methods in Molecular Biology, vol. 1068,
DOI 10.1007/978-1-62703-619-1_16, Springer Science+Business Media, LLC 2013

221
222 Trese Leinders-Zufall and Frank Zufall

In this chapter we focus on EVG recordings from mouse VNO.

Classical EOG and EVG recordings initially used vapor-phase
stimulation methods, meaning that the odor stimulus was applied
via an air puff. However, the mucous layer of the sensory epithe-
lium is prone to dehydration. Therefore, a source of artifacts in
these recordings can be caused by epithelial drying and thinning of
the mucus [8] or a change in humidity of the airstream associated
with an odorant stimulus [9]. We therefore developed submerged
preparations for both mouse main olfactory epithelium [1013]
and VNO [12, 1421], comparable to the underwater EOG
first reported in snake, turtle, and fish [4, 22, 23]. These prepara-
tions are advantageous as they are independent of the volatility of
a given chemosensory stimulus and allow for application of phar-
macological agents without significantly altering the thickness of
the aqueous layer covering the epithelium. Furthermore, liquid-
phase stimulation mimics the natural environment for ligand access
in the vomeronasal lumen. Under our conditions, the EVG time
course (activation and deactivation), including its adaptational
properties, can be precisely measured in a repeated manner [15, 21].
The EVG recording technique is ideal as a high-throughput
method for identifying potential pheromonal ligands acting on
populations of vomeronasal neurons (VSNs) by using precisely
timed stimuli of known concentrations. Evaluation of complex
mixtures together with vomeronasal axotomy established that
EVG recordings reflect the summed neural activity of VSNs [4].
The first pheromonal ligands activating mouse VSNs were discov-
ered using the submerged EVG technique [21] described in this
protocol. Subsequently, VSN activity to these newly identified
stimuli was confirmed using alternative recording methods with
single-cell resolution, such as calcium imaging in acute vomerona-
sal tissue slices and electrophysiological action potential recordings
from individual VSNs [21]. The EVG recording method proved
useful in identifying vast ligand families, including peptide and
protein pheromones, that stimulate mouse VSNs [14, 17, 24, 25].
The EVG recording method has also been instrumental in analyz-
ing vomeronasal phenotypes of mice with targeted deletions in
specific signaling molecules including seven transmembrane recep-
tors, G proteins, and ion channels [14, 18, 20, 21, 24, 26]. For
instance, mice harboring a null mutation of Trpc2, which encodes
the canonical transient receptor potential channel type 2, show
severely diminished EVG responses, a finding that indicated a cen-
tral role of this cation channel in the detection of pheromones and
other chemosignals in the VNO [20]. Using a similar approach,
Go was shown to be essential for sensing peptide and protein
pheromones by a subtype of VSNs [14]. Furthermore, the first
experimental evidence that V1R receptors function as pheromone
receptors was shown by EVG recordings [18]. Thus, although
quantitative analysis of different mouse strains with specific
 The Electrovomeronasogram: Field Potential Recordings in the Mouse... 223

vomeronasal gene deletions can be demanding, use of the EVG

method has provided critical phenotypic information related to the
function of such genes. However, in many cases subsequent analy-
sis using methods that allow single-cell resolution, such as patch-
clamp recording or calcium imaging, is required to verify results
from EVG recordings. The same ex vivo mouse VNO preparation
that we initially developed for EVG recordings has also been used
successfully for confocal imaging of Ca2+ responses in single den-
dritic endings of VSNs [27].

2 Materials

2.1 Preparation For the preparation of all physiological solutions we use ultrapure
of Solutions water (>18.2 M-cm resistivity at 25 C, low ppt in divalent cations).
There are various systems available in laboratories which filter
pretreated water (reverse osmosis or demineralized water) to produce
ultrapure water with <5 ppb total organic carbon (TOC) to reduce
organic contaminants.
1. Prepare extracellular solution (S1: 120 mM NaCl, 25 mM
NaHCO3, 5 mM KCl, 5 mM N,N-Bis(2-hydroxyethyl)-2-
aminoethanesulfonic acid (BES), 1 mM MgSO4, 1 mM CaCl2,
10 mM glucose) according to the table with ultrapure water.
The pH will be ~7.3 after 10-min aeration with carbogen
(95 % O2/5 % CO2), and the osmolarity 300 mOsm. If a higher
osmolarity is required, it can be adjusted by adding more glu-
cose (1 mM equals 1 mOsm). The solution is filtered twice
using a 0.2 m membrane filter to eliminate dust particles and
possible bacterial contaminations. The solution is stored at
4 C. Make sure that the solution is at room temperature and
aerated for 10 min with carbogen (95 % O2/5 % CO2) before
use.
2. Prepare extracellular 4-(2-hydroxyethyl)-1-piperazine-
ethanesulfonic acid (HEPES)-based solution (S2: 145 mM
NaCl, 5 mM KCl, 10 mM HEPES, 1 mM MgCl2, 1 mM
CaCl2) with ultrapure water. The pH is adjusted to 7.3 using
NaOH and the osmolarity to 300 mOsm using glucose (1 mM
equals 1 mOsm; see also item 1). The solution is filtered twice
using a 0.2 m membrane filter to eliminate dust particles and
possible bacterial contaminations. The solution is stored at
4 C. Make sure that it is at room temperature before use in an
experiment.

2.2 Preparation The following items are needed to produce agar bridges for the
of Agar Bridges EVG recordings: agar-agar (CAS # 9002-18-0), S2 solution (see
Subheading 2.1), a heat source with small flame (Bunsen burner
with pilot light tube or alcohol burner), polyethylene tubing
224 Trese Leinders-Zufall and Frank Zufall

(1.193 1.574 0.215 mm), 1 cc syringe, pipette tip (yellow,

200 l), and glass scintillation vial.
1. Dissolve 4 % (w/v) agar-agar in S2 solution by heating and
stirring the solution to approximately 60 C.
2. Cut the polyethylene tubing in approximately 3.5 cm long
pieces. Every piece of tubing will become one agar bridge and
is exchanged daily. Therefore, we prepare enough for at least
2 weeks.
3. The polyethylene tubing will be bent near a small flame of a
heat source (Bunsen burner with pilot flame) into an angle of
90 (Fig. 1a). Note that experienced agar bridge preparers do
not precut the polyethylene tubing. They take the whole roll of
tubing, bend the end of the tubing over the flame, make a cut
for the total length (approximately 2.5 cm), and restart the
procedure.
4. Take the syringe and fill it with the heated dissolved agar
solution.
5. Take the pipette tip and cut as indicated in the figure (Fig. 1a).
Now the pipette tip fits onto the syringe and can be used to fill
the heated dissolved agar solution into the bent polyethylene
tubing.
6. Fill the prepared polyethylene tubings with the agar solution
and leave the tubing on a piece of paper to cool down. These
filled tubings are the agar bridges.
7. Fill a glass scintillation vial with cool S2 solution and drop the
agar bridges into this solution (Fig. 1a). This kind of agar
bridge storage prevents the loss of water (evaporation) and
keeps the amount of salt solution in the agar bridge constant.
8. Store the agar bridges at 4 C until use.

2.3 Head Mounting To enable secure positioning of the mouse head and later EVG
Dissecting Dish recordings from the VNO, a special head mounting dish should be
prepared. The following items are needed: a plastic Petri dish
(square, 9120 15 mm), Sylgard 184 silicone elastomer kit, and
insect pins (needles).
1. Prepare the Sylgard elastomer in a 1:10 ratio of the two
components (curing agent and elastomer base, respectively) by
mixing the components thoroughly in the plastic Petri dish.
Use plastic syringes to extract the two components from their
containers and add them directly into the Petri dish. A volume
of approximately 50 ml base and 5 ml curing agent fills a
90 mm square Petri dish perfectly, leaving enough latitude to
later move and manipulate the preparation and the overflow of
the perfusion solution.
 The Electrovomeronasogram: Field Potential Recordings in the Mouse... 225

Fig. 1 (a) Tools for preparing and storing agar bridges: 1, bent polyethylene tubing; 2, yellow pipette tip
(200 l); 3, syringe with pipette tip for filling the polyethylene tubing; 4, scintillation vial for storing the agar
bridges in S2 solution. To enable the yellow pipette tip to fit on the syringe, it has to be cut at the indicated
position (see red line and a small scissor). (b) Head mounting dissection dish with three insect pins. The dis-
section dish contains hard Sylgard 184 elastomer enabling the secure positioning of the mouse head for the
EVG recording. (c) Electrode storage jar containing S2 solution and three electrodes filled with colored agar
solution (green). An arrow points out one of the electrodes. (d) To produce nonmetallic syringe needles a yellow
pipette tip (200 l) is heated over a pilot flame of a Bunsen burner and stretched to fabricate thin plastic tub-
ings of preferred diameters. A single-edged razor blade cuts the needle at the desired length and the broad
end of the yellow tip is cut similar as in (a) to fit the nonmetallic needle on a syringe. (e) Multibarrel stimulation
pipettes are made from thin glass capillaries held together by shrinkage tubing (1, black). The glass capillaries
are heated, turned, and pulled to produce a small-diameter tip using an automated puller (2). On the other end,
the individual glass capillaries are heated and separated to enable tubing to connect to the individual capillar-
ies (3). These multibarrel stimulation pipettes are stored in a big Petri dish until use

2. Let the liquid mixture cure to a flexible, transparent elastomer

overnight.
3. Place three insect pins in the dish for later use (Fig. 1b; see
Subheading 3.2, Dissection of Mouse Head and VNO). Insect
pins are black-enameled stainless steel pins which offer extra
protection against corrosion and also reduce glare.

2.4 Preparation of The following items are needed: patch-pipette puller, borosilicated
EVG Recording glass, two electrode storage jars, S2 solution (see Subheading 2.1),
Electrodes agar-agar (CAS # 9002-18-0), food color, 1 cc syringe, and non-
metallic syringe needles (see Subheading 2.5).
226 Trese Leinders-Zufall and Frank Zufall

1. Pull glass pipette electrodes using a programmable puller. We

use the Narishige PC-10 vertical puller. The electrode is pulled
in two stages from borosilicated thin-walled glass capillaries
without filament (1.5 mm/1.12 mm OD/ID) and has a resis-
tance of 13 M and a tip diameter of approximately 1.5 m.
Store the electrodes in a electrode storage jar (World Precision
Instruments).
2. Prepare 2 % (w/v) agar-agar in S2 solution by heating it to
approximately 60 C. When the agar is dissolved, add food
color (McCormick Green). The tip of a colored agar-filled
electrode can be easily visualized under the microscope. Be
aware that not all types of food color can be used, since they
may contain sugar which will change the osmolarity of the
solution. It is essential at this step to have a agar solution that
is not too hard, since it can rip off microvili or cilia from the
sensory epithelium, or too soft, since it will leak solution onto
the sensory epithelium preventing a solid contact for recording
changes in potential (see Notes 1 and 2).
3. Prepare a syringe with a nonmetallic syringe needle for filling
the electrode with heated agar solution (see Subheading 2.5).
4. Fill the tip of the electrode to a height of 0.5 cm with the
heated agar solution (Fig. 1c). In this step it is essential to
work fast and consistent, since the heated agar will cool and
solidify relatively fast in the thin nonmetallic syringe needle.
During this step it is very helpful to have another lab member
hand and regain the electrodes after filling. The filled elec-
trodes are put back into the electrode storage jar to cool down
to room temperature.
5. Fill a second electrode storage jar with S2 solution (see sub-
heading 2.1) to store the electrodes after visual inspection of
the filling procedure.
6. Check visually under a microscope that the agar solution has
reached the tip of the electrode without any air bubbles and
that the agar solution did not creep up the outer glass wall.
This latter point may cause false measurements of potential
changes (see Note 3).
7. Store the agar-filled electrodes in the S2 solution-containing
storage jar (Fig. 1c). Make sure that the electrode tips contact
the solution to prevent dehydrating the agar in the electrodes.
If the consistency of the agar is incorrect (low gel strength),
the colored agar will leak out of the pipettes. In this case the
weight/volume percentage of the agar solution has to be
increased.
8. The electrodes are stored at 4 C until use.
 The Electrovomeronasogram: Field Potential Recordings in the Mouse... 227

2.5 Preparation of To fill a micropipette or glass-electrode without air bubbles is

Nonmetallic Syringe challenging if long, stainless steel hypodermic needles are used.
Needles Nonmetallic syringe needles can bend (therefore do not damage
the glass capillaries) and can be bought from various companies.
However, we produce our own nonmetallic needles from yellow
pipette tips to reduce the cost of these items (Fig. 1d). We regu-
larly exchange the needles due to the use of different filling solu-
tions (see Subheading 2.6) or salt crystals clogging the needle. The
following items are needed: pipette tips (yellow, 200 l), a heat
source with small flame (Bunsen burner with pilot light tube or
alcohol burner), and single-edge razor blade.
1. Heat up a plastic pipette tip approximately in the middle over
a small flame of a heat source (pilot flame of Bunsen burner)
holding both ends. Turn the plastic pipette tip slowly and
notice the change in plastic coloring. Take care that the plastic
does not catch fire. It is recommended to start turning the
plastic pipette tip some distance from the flame and to slowly
reduce the distance until one gets a sense of the optimal distance
between the flame and pipette tip.
2. Move the pipette away from the heat source, the moment the
plastic is changing its color and looks more liquified. Turn the
pipette and hands 90 and move the hand containing the tip
slowly in a constant velocity downward away from the hand
holding the broad end of the yellow pipette tip. A thin plastic
capillary is forming. The inner diameter of the plastic capillary
can be controlled by the speed of the pull.
3. Blow air onto the plastic to harden it at the desired thickness
and let it cool further down on a table.
4. Cut the plastic capillary in a 45 angle with a razor blade to the
desired length. Make sure that the cut is not in a 90 angle
since the plastic tubing will be squeezed together blocking the
solution outflow.
5. To finalize the nonmetallic syringe needle, cut a small piece
from the broad end of the plastic tip so that it fits on a 1 cc
syringe (Fig. 1a, d).
6. Repeat the procedure to acquire the desired amount of non-
metallic syringe needles. The long and fine tip of this needle
allows the filling of the glass capillaries (electrode and multi-
barrel stimulation pipettes; see Subheadings 2.4. and 2.6) very
close to their narrow tips eliminating air bubble formation.

2.6 Preparation and Micropipette puller exists that can consistently and automatically
Filling of Multibarrel produce multibarrel stimulation pipettes up to 7-barrels (PMP-
Stimulation Pipette 107 puller, MicroData Instrument). Multibarrel pipettes (up to
15-barrels) can also be manufactured using a Narishige PE-22
puller. However, in this case the production of the pipettes depends
228 Trese Leinders-Zufall and Frank Zufall

on the aptitude of the preparer/investigator since the rotation of

the pipette clamp (drill chuck) will rely on his/her skill. The flow
of the stimulation solution from these pipettes can be precisely
controlled by the amount of rotation in the glass pipette and the
tip opening (see step 6). The following items are needed: puller,
borosilicated glass, heat-shrink tubing (precut, 1 cm long), food
color, 1 cc syringes, nonmetallic syringe needles (see
Subheading 2.5), and a heat source with small flame (Bunsen
burner with pilot light tube or alcohol burner). Here we describe
the production and filling of 7-barrel stimulation pipettes.
1. Take seven borosilicated thin-walled glass capillaries without
filament (1 mm/0.75 mm OD/ID) and fit onto either side
heat-shrink tubing (Fig. 1e).
2. Heat the tubing by holding the glass capillaries on both ends
and turning it over a small flame of a heat source (pilot flame
of Bunsen burner). The tubing shrinks, providing a snug fit
over the seven capillaries, and helps to clamp the barrels
securely into the drill chuck of the puller without damaging
the glass.
3. Fasten the seven capillaries into the pipette clamps of the puller
(MicroData Instrument, PMP-107 puller) and use the pullers
control and functions to pull the pipette in two stages. In case
the Narishige Puller is used, the rotation of the drill chuck is
being turned in the first heating/pulling step by hand. After
the rotation, the power is stopped and air is blown onto the
capillaries to cool the glass. The second pull is initialized and
finalized by the puller without help.
4. Heat the multibarrel stimulation pipette (approximately
between the end of the capillary and the shrinkage tubing)
over a small flame of a heat source (pilot flame of Bunsen
burner). Gently push down the thin glass capillaries using old
forceps to separate the glass capillaries of the 7-barrel pipette
and to enable plastic tubing to connect to these ends (Fig. 1e).
5. Store the pipettes in a dry space and cover them with a lid to
prevent dust accumulation. We use big Petri dishes in combi-
nation with a strip of modelling clay (available from any chil-
drens store or supermarket) to hold the pipette securely in
place (Fig. 1e).
6. Before filling the 7-barrel stimulation pipette, bevel its tip
using a micropipette grinder (Narishige EG-400). The diam-
eter and the angle of the stimulation pipette tip can be
controlled by the microgrinder. The tip diameter can produce
either (a) leakage of the stimulus solution causing adaptation
of the sensory epithelium or (b)if too narrowa reduced
flow rate of the stimulus, inevitably affecting the ligand con-
centration. In addition, the angle of the cut and the placement
 The Electrovomeronasogram: Field Potential Recordings in the Mouse... 229

of the stimulation pipette into the solution flow can cause

backpropagation of the bath solution into the pipette also
affecting the stimulus concentration. To prevent the accumula-
tion of shavings in the tip of the stimulation pipette, it is essen-
tial that the grinder is equipped with a water-dripping device.
7. Secure the bevelled 7-barrel stimulation pipette on the edge of
a glass beaker using modelling clay. The tip of the pipette
should be covered by ultrapure water. Do not worry about the
hygroscopic/capillary forces causing water to creep up into the
glass pipettes. The ultrapure water boundary will be separated
from the physiological backfilling solution (stimulus solution)
by an air bubble. Later, the air bubble will be expelled by the
air pressure from the Picospritzer (Parker Instruments) indi-
cating when the stimulation solution arrived at the tip of the
stimulation pipette.
8. Prepare stimulation solution and your control solution based
on the extracellular S1 solution (see Subheading 2.1).
9. Fill a syringe with the solution and use for every syringe a sepa-
rate nonmetallic syringe needle.
10. Test the output flow of the multibarrel stimulation pipette in
the EVG setup containing a microscope and picospritzer
(Parker) using a small Petri dish filled with S1 solution. The
picospritzer is designed to rapidly and reproducibly eject small
volumes from the stimulation pipette using air pressure. Air
pressure is relayed via valves and tubings to the various ends of
the 7-barrel stimulation pipette. The output flow of the indi-
vidual barrels should be in the same direction and contain
approximately the same amount of solution. The flow can be
better visualized if the control and stimulation solution contain
a small, but equal, amount of food color (1:1,000). The stimu-
lus length is being controlled via the computer sending a TTL
signal to the picospritzer.
11. Keep the completely prepared stimulation pipette in either a
beaker or a Petri dish with S1 solution (see Subheading 2.1), until
the tissue is prepared, and the stimulation pipette needs to be
placed near the sensory epithelium for stimulation (see Note 6).

3 Methods

3.1 The Recording The recording setup consists of equipment for acquisition and
Setup recording of the electrical EVG potential changes and for visualizing
and handling the specimen and the stimulation (see Subheading 2.6)
and recording pipette (see Subheading 2.4) via micromanipulators
which are placed on a vibration-free table. The table is enclosed by
a Faraday cage to prevent electromagnetic disturbances. For the
230 Trese Leinders-Zufall and Frank Zufall

electrical recording of the signal, a recording electrode containing

the glass pipette with agar in the tip and backfilled with S1 solution
(see Subheading 2.1) is connected via an Ag/AgCl wire to a
differential amplifier (DP-301, Warner Instruments). The second
Ag/AgCl wire serving as the indifferent electrode is connected to
an agar bridge (see Subheading 2.2) and placed near the preparation
(see Subheading 3.2) using modelling clay. The amplifier output
signals are directly transmitted to an analog-to-digital converter
(Heka/Instrutech, ITC-16) that is connected to a computer and
to an oscilloscope for real-time monitoring of the signal. Software
such as Hekas Pulse/Pulsefit can be used to control the stimula-
tion protocol and to record the potential changes. The stimula-
tion/pulse protocol triggers the picospritzer (Parker) which rapidly
ejects fluid from a particular glass capillary of the stimulation
pipette using air pressure. The stereo microscope (SZX12,
Olympus) should have a great zoom range and high magnification
capabilities enabling to visualize not only the olfactory organs but
also the electrode and stimulation pipette tip. Options like fluores-
cence illumination and detection could be incorporated to help
visualize particular organ regions.

3.2 Dissection of Please make sure that all animal experimental procedures are
Mouse Head and VNO performed in accordance with the guidelines established by the
animal welfare committees of the respective institutions.
1. Euthanize a mouse with a CO2 overdose followed by
decapitation.
2. Take the mouse head and eliminate all skin and fur to expose
the skull and nasal bones. Start with a caudal to rostral cut
along the sagittal suture to the tip of the nose using sharp
Metzenbaum scissors. Remove the skin sideways and cut it off.
3. Detach the underjaw (mandible) from the head bone by cut-
ting on both sides through the temporomandibular joint with
sharp blunt scissors.
4. Cut off the top of the mouse incisors (front teeth) to ensure a
stable position of the ventral part of the head on the surface of
a cutting board (see Fig. 2a).
5. Using a single-edge razor blade, make a cut through the left
nasal cavity and the whole mouse head. More precisely, when
placing the head on the cutting board, the razor should be
positioned with one of its sharp points next to the tip of the
nose (Fig. 2a). To cut through the nasal cavity, the sharp
surface of the razor blade is carefully placed in the same angle
as the dorsal nose/head bones, parallel to the internasal suture
(midline) and approximately 1/3 the distance of the nasal
bone from the suture. Extreme care should be taken to move
the blade in a 90 angle to the cutting board to prevent damage
 The Electrovomeronasogram: Field Potential Recordings in the Mouse... 231

Fig. 2 (a) Photograph of a mouse head showing the placement of a single-edge razor blade for cutting through
the nasal cavity and brain in a single stroke in order to prevent damage to the nasal cavity. Arrow indicates the
cutting direction. (b) Sagittal view of the vomeronasal organ (VNO). The red dashed box indicates the outline of
vomer bone that should be removed using forceps. Thin black dashed line, outline of the VNO. (c) View onto the
ciliary surface of the VNO showing the placement of the perfusion input, the multibarrel stimulation pipette,
and the green agar-filled electrode. Red arrows indicate the flow of the bath perfusion. d dorsal, v ventral,
r rostral, c caudal. (d, e) Typical examples of EVG recordings from mouse VNO. The recorded potentials undergo
time-dependent adaptation, evident as a decline of the response during sustained stimulation or as a peak
response reduction during brief, repetitive stimulation. Stimulus, diluted urine (DU, 1:1,000). (f, g) Examples of
stimulus-induced EVG recordings that were scaled to compare the adaptation time course. Responses were
evoked by repetitive 6-s identical pulses of 2-heptanone (10 nM) or the major histocompatibility complex
peptide ligand SYFPEITHI (0.3 pM). Interpulse interval was 10 s. (dg, modified with permission from [15])

to the VNO which protrudes into the nasal cavity. Put pressure
onto the angled top part of the razor blade and cut through
the bones and brain tissue in one stroke (see arrow in Fig. 2a).
6. Pin the sagittally cut mouse head onto the premade mounting
dissection dish (see Subheading 2.3) using three insect pins.
Place the pins in three random brain areas.
7. Place the dish onto the recording stage under the stereo micro-
scope (SZX12, Olympus) to perform more detailed dissection.
8. Take away any obscuring tissue overlaying the VNO so that
there is a direct view onto the vomer bone. In case the cut was
too shallow, the preparer has to cautiously cut away the left-
over incisor. Its root can cover most of the VNO and may rip
off the VNO with septum.
9. Using standard #5 Dumont forceps (0.10 0.06 mm) nibble a
small rectangular window into the vomer bone (Fig. 2b, red
dashed box). The window should be located in the middle of
the vomer and not too far rostral or caudal. Care has to be
taken not to extend the window too much dorsally: VNO axon
bundles protrude underneath the dorsal part of the vomer
bone along the septum and could be damaged. The cavenous
232 Trese Leinders-Zufall and Frank Zufall

tissue of the VNO with its distinct blood vessel is now clearly
visible through the window.
10. Put one of the legs of a Biologie #5 Dumont forceps
(0.05 0.02 mm) from dorsal to ventral through the VNO
lumen at the most caudal end of the window, pinch gently the
forceps, and slowly tear the cavenous tissue from caudal to ros-
tral, away from the sensory epithelial surface. Be meticulous in
preventing the ciliary surfaces of the cavenous tissue and the
sensory epithelium to touch each other. Cilia from both sen-
sory and nonsensory epithelium could be lost due to their
entanglement (velcro hook-and-loop closure effect) and there-
fore cause the absence of chemosensory responses.
11. Carefully clear the rest of the rostral and caudal vomer bones
and cavenous tissue. Also check if the big blood vessel was
eliminated which may block access to record from the sensory
epithelium. Do not discard the ventral and dorsal vomer bone
boundaries. They will service as a wall for the input and output
perfusion system.
12. Perfuse, as quickly as possible, the sensory epithelium with
oxygenated S1 solution to prevent dehydration and oxygen
deprivation.

3.3 Perfusion of the 1. Control the flow of the bath perfusion solution (S1 solution,
Sensory Epithelium see Subheading 2.1), e.g., with a pump or a gravity feed. The
solution speed in a gravity feed system should be strictly
controlled using a combination of solution volume (container)
and gravity flow controllers [28]. Keep track of the perfusion
solution volume to maintain relatively equal pressure on the
tubing lines.
2. Install a 1-way stopcock directly at the source of the perfusate.
This enables instantaneous stoppage of the solution flow,
which is advantageous if more than one perfusate will be used
during an experiment (e.g., in the presence of pharmaca or
blockers). In addition, backpropagation of the flowing bath
solution into the tubing of the other perfusion solution will be
prevented.
3. Install a manifold, if more than one perfusate (S1 bath solu-
tion, see Subheading 2.1) will be used. The manifold output
should be located within a very short distance from the VNO
perfusion input to reduce solution exchange times.
4. Adjust the flow rate of the perfusate to approximately 100 l/s.
Always count the output flow and do not depend on the writ-
ten information on the flow controllers.
5. Replace the insect pins in a way that ensures that the tip of the
nose is located approximately 1015 higher than the brain.
This will help to modulate the perfusion height and speed in
 The Electrovomeronasogram: Field Potential Recordings in the Mouse... 233

the natural chamber composed by the boundaries of the nasal

cavity and vomer bone.
6. Use an 18 G needle as the VNO perfusion input. Place the
opening of the needle near the rostral VNO entry onto the
sagittally cut mouse head so that the solution is flowing in an
angle onto the tissue.
7. Make a small incision into the caudal region of the soft palate
near the uvula. The excess of perfusion solution will exit the
chamber constructed from the borders of the palate and
nasal/brain bones. Sometimes it helps to put a small piece of
tissue paper (Kimberly-Clark) at the height of the uvula which
connects the chamber and the bottom of the head mounting
dissection dish. The excess solution on the bottom of the dis-
section dish is also necessary to have electrical contact with the
reference electrode.
8. Align the mounting dissection dish so that the sensory epithe-
lium is centered under the microscope and secure the dish with
modelling clay.
9. Place the vacuum suction (output of perfusate) near the wall of
the Petri dish to prevent overflow. When using vacuum suc-
tion, it is necessary to precisely control the force of the suction
to be able to fine-tune the fluid height in the Petri dish [28].

3.4 Recording and 1. Place the agar bridge with reference electrode near the dorsal
Analyzing EVG Data mouse head (see Note 4).
2. Place the two micromanipulators containing the multibarrel
stimulation pipette and the recording electrode near the stage.
3. Turn on the differential amplifier (DC mode, low-pass filter at
0.1 kHz, gain 100).
4. Carefully lower the recording electrode near the surface of the
sensory epithelium. The electrode should contact the bath
solution. When the electrode comes in contact with the solu-
tion a straight baseline will appear in the oscilloscope window.
5. Gently lower the stimulation pipette so that its tip is near the
tip of the recording electrode (Fig. 2c).
6. Check the amplifier gain and operation by using the 1 mV cali-
bration signal of the amplifier by recording this signal into the
computer using your acquisition software (Heka Pulse/
Pulsefit).
7. Lower the recording electrode onto the surface of the sensory
epithelium. This step has to be done very gently. At the same
time the change in signal should be followed on the oscilloscope.
A change in the baseline indicates that the agar-filled tip of the
electrode connects to the tissue. The trick is to only connect to
the surface containing the microvilli and the knobs. A strong dip
234 Trese Leinders-Zufall and Frank Zufall

in the baseline indicates that the electrode went past the surface
and is now at the level of the somata or worse at the vomer bone
(glass electrode can break!). With practice, the investigator will
notice the small changes in potential indicating that the electrode
is near the epithelial surface (see Notes 1 and 2).
8. Lower the stimulation pipette near the electrode. Make sure
that its tip is a little elevated and placed at a small distance from
the electrode touching the sensory epithelium. Keep in mind
that the solution will be pushed out of the glass capillary, but,
due to the angle of the stimulation pipette, the solution will be
directed either over the area of interest (the recording site) or
before the area of interest which may cause a deflection from
the epithelium surface over the area of interest. In both cases,
the recording site may not be stimulated. To test the direction
of the stimulation fluid, use control S1 solution (see
Subheading 2.1) with food color by initiating a stimulation
protocol via the computer via the external trigger of the pico-
spritzer. The stimulus length of the test should not be more
than 0.3 s (see Notes 5 and 6).
9. Start the experiment on the sensory epithelium by activating the
stimulation protocols via the computer. Make sure to wait at
least 4 min between stimulations to prevent adaptation of the
sensory neurons (Fig. 2dg) [11, 15, 21] (see Notes 5 and 6).
10. Software from Heka (Pulse/Pulsefit) and Wavemetrics (Igor)
provide tools for measuring and analyzing the EVG signals.
Parameters include response amplitude, latency, time-to-peak,
and time constants of activation, adaptation/desensitization,
and termination [11, 15, 21]. It is desirable to digitally filter
the traces (low-pass 8-pole Bessel filter with corner frequency
1 kHz; digital filter frequency 60 Hz) (see Note 7).

4 Notes

These notes are provided to help identify some common problems

in EVG recordings.
1. Check the tip of the recording electrode if a spontaneous loss
of EVG responses occurred. Agar may have been lost or bro-
ken off from the electrode tip preventing electrical contact
with the tissue. In some cases the agar is boiled too long
increasing the gel strength and increasing the risk of losing
agar on the surface of the sensory epithelium.
2. Check also the tip of the recording electrode, if no stable base-
line can be obtained or no responses are measured. Colored
agar solution may be leaking from the recording electrode.
The gel strength was in this case too low.
 The Electrovomeronasogram: Field Potential Recordings in the Mouse... 235

3. Another obstacle in recording authentic EVGs is caused by a

very thin film of agar on the outside of the recording electrode.
In this case extreme hot and liquid agar solution leaked out of
the tip of the glass capillary during the preparation of the
electrode.
4. Chlorate the silver wires of the recording and reference elec-
trodes regularly. The differential amplifier has a high tolerance
to DC offsets, but cannot be adjusted indefinitely. Furthermore,
a wrong placement of the reference electrode can cause a dra-
matic shift in this offset value.
5. Start to record and stimulate at the most caudal area of the
sensory epithelium. In this way the more rostral sensory epi-
thelium has not yet detected any stimulants due to the flow of
the bath solution and should not be adapted to the ligand
stimulation.
6. Extreme care should be taken with the multibarrel stimulation
pipette. Any leakage from these pipettes can cause adaptation
of the tissue and prevent the recording of meaningful data.
7. A chemosensory stimulus should be included in the experi-
mental series that evokes broadly responses for a particular sen-
sory neuron population to enable the detection of sudden
changes in responsiveness at a particular electrode site.

Acknowledgements

This work was supported by grants from the Deutsche

Forschungsgemeinschaft (to T.L.-Z. and F.Z.) and the Volkswagen
Foundation (to T.L.-Z.). T.L.-Z. is a Lichtenberg professor of the
Volkswagen Foundation.

References

1. Hosoya Y, Yoshida H (1937) ber die bioelek- von amphibien (Rana), reptilien (Lacerta) und
trischen Erscheinungen an der Riechschleimhaut. sugetieren (Mus). Z vergl Physiol 72:
Jpn J Med Sci III Biophys 5:2223 370385
2. Ottoson D (1955) Analysis of the electrical 6. Getchell TV, Shepherd GM (1978) Responses
activity of the olfactory epithelium. Acta of olfactory receptor cells to step pulses of
Physiol Scand Suppl 35:183 odour at different concentrations in the sala-
3. Inouchi J, Wang D, Jiang XC, Kubie J, mander. J Physiol 282:521540
Halpern M (1993) Electrophysiological 7. Davis H (1961) Some principles of sensory
analysis of the nasal chemical senses in garter receptor action. Physiol Rev 41:391416
snakes. Brain Behav Evol 41:171182 8. Gesteland RC, Lettvin JY, Pitts WH (1965)
4. Taniguchi M, Wang D, Halpern M (1998) Chemical transmission in the nose of the frog.
The characteristics of the electrovomeronaso- J Physiol 181:525559
gram: its loss following vomeronasal axotomy 9. Scott JW, Scott-Johnson PE (2002) The elec-
in the garter snake. Chem Senses 23:653659 troolfactogram: a review of its history and
5. Mller W (1971) Vergleichende elektrophysi- uses. Microsc Res Tech 58:152160
ologische untersuchungen an den sinnesepi- 10. Chen S, Lane AP, Bock R, Leinders-Zufall T,
thelien des Jacobsonschen organs und der nase Zufall F (2000) Blocking adenylyl cyclase inhibits
236 Trese Leinders-Zufall and Frank Zufall

olfactory generator currents induced by IP(3)- vomeronasal neurons. Microsc Res Tech
odors. J Neurophysiol 84:575580 58:251260
11. Munger SD, Lane AP, Zhong H, Leinders- 20. Leypold BG, Yu CR, Leinders-Zufall T, Kim
Zufall T, Yau KW, Zufall F, Reed RR (2001) MM, Zufall F, Axel R (2002) Altered sexual
Central role of the CNGA4 channel subunit in and social behaviors in trp2 mutant mice. Proc
Ca2+-calmodulin-dependent odor adaptation. Natl Acad Sci U S A 99:63766381
Science 294:21722175 21. Leinders-Zufall T, Lane AP, Puche AC, Ma W,
12. Spehr M, Spehr J, Ukhanov K, Kelliher KR, Novotny MV, Shipley MT, Zufall F (2000)
Leinders-Zufall T, Zufall F (2006) Parallel Ultrasensitive pheromone detection by mam-
processing of social signals by the mammalian malian vomeronasal neurons. Nature
main and accessory olfactory systems. Cell Mol 405:792796
Life Sci 63:14761484 22. Tucker D, Shibuya T (1965) A physiologic
13. Spehr M, Kelliher KR, Li XH, Boehm T, and pharmacologic study of olfactory recep-
Leinders-Zufall T, Zufall F (2006) Essential tors. Cold Spring Harb Symp Quant Biol
role of the main olfactory system in social rec- 30:207215
ognition of major histocompatibility complex 23. Silver WL, Caprio J, Blackwell JF, Tucker D
peptide ligands. J Neurosci 26:19611970 (1976) The underwater electro-olfactogram: a
14. Chamero P, Katsoulidou V, Hendrix P, Bufe B, tool for the study of the sense of smell of
Roberts R, Matsunami H, Abramowitz J, marine fishes. Experientia 32:12161217
Birnbaumer L, Zufall F, Leinders-Zufall T 24. Haga S, Hattori T, Sato T, Sato K, Matsuda S,
(2011) G protein Go is essential for vomero- Kobayakawa R, Sakano H, Yoshihara Y,
nasal function and aggressive behavior in mice. Kikusui T, Touhara K (2010) The male mouse
Proc Natl Acad Sci U S A 108:1289812903 pheromone ESP1 enhances female sexual
15. Spehr J, Hagendorf S, Weiss J, Spehr M, receptive behaviour through a specific vom-
Leinders-Zufall T, Zufall F (2009) Ca2+- eronasal receptor. Nature 466:118122
calmodulin feedback mediates sensory adapta- 25. Kimoto H, Sato K, Nodari F, Haga S, Holy
tion and inhibits pheromone-sensitive ion TE, Touhara K (2007) Sex- and strain-specific
channels in the vomeronasal organ. J Neurosci expression and vomeronasal activity of mouse
29:21252135 ESP family peptides. Curr Biol
16. Zufall F, Ukhanov K, Lucas P, Liman ER, 17:18791884
Leinders-Zufall T (2005) Neurobiology of 26. Leinders-Zufall T, Ishii T, Mombaerts P,
TRPC2: from gene to behavior. Pflugers Arch Zufall F, Boehm T (2009) Structural require-
451:6171 ments for the activation of vomeronasal sen-
17. Leinders-Zufall T, Brennan P, Widmayer P, sory neurons by MHC peptides. Nat Neurosci
Chandramani SP, Maul-Pavicic A, Jager M, Li 12:15511558
XH, Breer H, Zufall F, Boehm T (2004) MHC 27. Riviere S, Challet L, Fluegge D, Spehr M,
class I peptides as chemosensory signals in the Rodriguez I (2009) Formyl peptide receptor-
vomeronasal organ. Science 306:10331037 like proteins are a novel family of vomeronasal
18. Del Punta K, Leinders-Zufall T, Rodriguez I, chemosensors. Nature 459:574577
Jukam D, Wysocki CJ, Ogawa S, Zufall F, 28. Leinders-Zufall T (1998) Technique for
Mombaerts P (2002) Deficient pheromone setting up the perfusion of a recording
responses in mice lacking a cluster of vomero- chamber for physiological studies with cul-
nasal receptor genes. Nature 419:7074 tured and acutely dissociated cells. http://
19. Zufall F, Kelliher KR, Leinders-Zufall T wwwwarneronlinecom/pdf/whitepapers/
(2002) Pheromone detection by mammalian perfusion_strategiespdf
 Chapter 17

Electrical Recordings from the Accessory Olfactory

Bulb in VNO-AOB Ex Vivo Preparations
Julian P. Meeks and Timothy E. Holy

Abstract
Electrical recordings from individual accessory olfactory bulb neurons allow exploration of the functional
properties of this important pheromonal processing circuit. Several approaches to performing such record-
ings have been used. Here, we describe ex vivo methods that we have found useful for recording from
accessory olfactory bulb neurons using simple extracellular glass electrodes.

Key words Accessory olfactory bulb, Vomeronasal organ, Vomeronasal nerve, Ex vivo, Microelectrode,
Extracellular recording, Tissue perfusion, Electrophysiology

1 Introduction

Electrical recordings from individual accessory olfactory bulb

(AOB) neurons have been made in awake mice [1], in anesthetized
mice [2, 3], and in an ex vivo experimental preparation that main-
tains functional inputs from the vomeronasal organ [4]. The ex
vivo preparation takes advantage of the fact that the axons of vom-
eronasal sensory neurons project medially through the ipsilateral
septal tissue, and do not cross the midline before ramifying in the
AOB glomerular layer. Although the experimental dissection is
quite involved, this procedure has many advantages in experimen-
tal contexts in which active odor sampling and/or centrifugal con-
nectivity are not required.
Because the vomeronasal synaptic inputs remain functional,
one can perform experiments linking peripheral sensory stimulation
to AOB neural activity. Because the AOB is self-contained inside an
active tissue perfusion chamber, one can use pharmacology to help
elucidate mechanisms of AOB circuit function. We have used
single-glass electrode recordings to monitor the activity of individ-
ual neurons at various depths in the AOB, but this configuration is
also compatible with more advanced electrophysiological

Kazushige Touhara (ed.), Pheromone Signaling: Methods and Protocols, Methods in Molecular Biology, vol. 1068,
DOI 10.1007/978-1-62703-619-1_17, Springer Science+Business Media, LLC 2013

237
238 Julian P. Meeks and Timothy E. Holy

techniques, including guided or blind patch-clamping and multi-

electrode array recording.
With practice, one should expect to produce successful ex vivo
preparations in approximately 80 % of attempts, and 2 targeted
extracellular glass electrode recordings from each preparation per
hour. The major cell types in the first 300 m along the depth of
the AOB remain mostly devoid of cell degradation markers for up
to 8 h, but a potential limitation of this preparation is that the
deepest granule cell layers begin to show signs of cellular distress at
4 h post dissection.
The major steps necessary to perform targeted extracellular
recordings from ex vivo preparations are.
1. Initial fast dissection to isolate one hemisphere of the mouse
snout and rostral cranium.
2. Secondary detailed dissection to expose vomeronasal axons to
the superfusion solution.
3. Insertion of a thin polyimide cannula into the vomeronasal
organ lumen.
4. Insertion of microelectrodes into the AOB.

2 Materials

2.1 Reagents 1. Chilled dissection artificial cerebrospinal fluid (aCSF):

125 mM NaCl, 2.5 mM KCl, 2 mM CaCl2, 10 mM MgCl2,
25 mM NaHCO3, 1.25 mM NaH2PO4, 0.4 mM Na-ascorbate,
2 mM Na-pyruvate, 3 mM myo-inositol, 25 mM glucose.
2. Superfusion aCSF: 125 mM NaCl, 2.5 mM KCl, 2 mM CaCl2,
1 mM MgCl2, 25 mM NaHCO3, 1.25 mM NaH2PO4, 0.4 mM
Na-ascorbate, 2 mM Na-pyruvate, 3 mM myo-inositol, 25 mM
glucose.
3. Stimulating Ringers saline: 115 mM NaCl, 5 mM KCl, 2 mM
CaCl2, 2 mM MgCl2, 25 mM NaHCO3, 10 mM HEPES,
10 mM glucose.

2.2 Dissecting Tools Decapitation scissors, sharp curved fine dissecting scissors, #11
scalpel blade and handle, two pairs of Adson Forceps (Fine Science
Tools, Foster City, CA), standard carbon steel safety razor blade,
two pairs of #5 Dumont forceps (Fine Science Tools), two pairs of
#3 Dumont forceps (Fine Science Tools).

2.3 Equipment 1. 60 mm plastic petri dishes.

and Supplies 2. Small plastic plank, approximately 1 cm 0.6 cm 1 mm.
3. Dow Corning high vacuum grease.
4. Perfusion dish or chamber (see Fig. 1).
 Electrical Recordings in the Ex Vivo AOB 239

Fig. 1 Schematic of ex vivo tissue chamber. Warmed, oxygenated aCSF flowing

at 58 mL/min enters a secondary oxygenation area prior to reaching the tissue.
Aiming the main chamber inlet stream directly at the surface of the AOB improves
tissue health over time. Note that both inlet and suction may require electrical
grounding

5. Peristaltic or gravity-fed perfusion apparatus and suction

(Gilson Minipuls 3 or an equivalent).
6. 3 M Vetbond tissue glue.
7. Stereomicroscope.
8. Long (34 in.), thin polyimide cannula (0.0056 ID, 0.001
wall thickness) attached to a pressurized fast perfusion device
(Automate Scientific ValveLink or similar).
9. Borosilicate capillary glass, pulled to diameter and impedance
specified below.
10. Fine polyimide electrode-filling syringe (World Precision
Instruments MicroFil or an equivalent).
11. 0.22 m syringe filter (EMD Millipore Millex or an
equivalent).
12. Electrode holder, headstage, extracellular amplifier, and ana-
log/digital recording and storage devices.
13. Oscilloscope attached to the voltage output of the microelec-
trode amplifier.
14. Optional audio device attached to voltage output of the micro-
electrode amplifier.
15. 4-D microelectrode positioning device, with the penetration
axis controlled with fine motion, preferably motorized (e.g.,
Siskiyou Corporation MX1641).
240 Julian P. Meeks and Timothy E. Holy

3 Methods

3.1 Fast Initial 1. Deeply anesthetize animal using inhaled isoflurane in an

Dissection induction chamber.
(See Note 1) 2. Decapitate, and immediately place the tissue into a plastic petri
dish filled with chilled dissection aCSF. Fully immerse the
tissue for 1020 s (see Fig. 2A).
3. Grasp the head by the caudal/dorsal scalp. Remove lower jaw
and tongue using curved scissors.
4. Using Adson forceps, grasp the scalp at the caudal edge, and
then peel by pulling rostrally. At nostrils, remove the scalp tis-
sue with a single cut (see Fig. 2B).
5. Using Adson forceps, carefully remove the skull, starting at the
caudal edge overlying the cerebellum or brain stem. Stop at the
rhinal sinus (the dense bone anterior to bregma; see Fig. 2C).
6. Make a single bilateral coronal cut approximately 5 mm caudal
to the rostral sinus (through the frontal pole of neocortex).
Remove the brain caudal to the cut using a forceps or a curved
scoop.
7. Using sharp scissors, cut through the remaining cranial bones
at the same location of the previous caudal cut (see Fig. 2D1).
Place the tissue into a 60 mm petri dish filled with chilled
dissection aCSF for 510 s.

Fig. 2 Overview of the ex vivo dissection. Images taken immediately after Step 1.2 (A), Step 1.4 (B), Step 1.5
(C), Step 1.7 (D1), Step 1.8 (D2), Step 1.15 (E), Step 1.16 (F), and Step 3.5 (G1G3). (G1) Ex vivo preparation
with cannula inserted into VNO. (G2) Magnified view of the accessory olfactory bulb (red boxed region in G1).
(G3) Magnified view of the polyimide cannula (orange/brown line entering from the left) placed inside the
vomeronasal organ (blue boxed region in G1)
 Electrical Recordings in the Ex Vivo AOB 241

8. Grasping the remaining tissue from the anterior side near the
eye sockets, turn the tissue ventral side up (see Note 2). Then,
using the #11 scalpel, detach the soft palate at the attachment
point to the incisors. Grasp the cut end of the palate with
Adson forceps, and peel it away with a caudal pull (see Fig. 2D2).
9. Place the sharp tip of the #11 scalpel blade into the small notch
between the molars on the left hemisphere (caudal to the vom-
eronasal organ and the palatine foramen). Crack the bone by
rotating the blade handle.
10. Make a careful caudal-rostral cut through the anterior-most
portion of the palatine foramen, guiding the blade between
the nasal cavity rostral to the left VNO to the area between the
two incisors.
11. Grasping the tissue from the ventral side around the maxillary
bones, flip the tissue dorsal side up. Using sharp scissors cut
the right zygomatic bones and thin portion of the maxilla near
its intersection with the lateral snout. Remove any attached
soft tissue with Adson forceps.
12. Grasp the tissue firmly with fingers at the anterior portion of
the tissue. Orient the safety razor blade parallel to the midline
behind the caudal end of the tissue. Place the safety razor blade
just left of the midline (<1 mm) and touching the frontal corti-
cal lobe and both dorsal and ventral bones.
13. Gently rock the razor blade along the dorsal/ventral axis,
using the rocking motion to slowly cut through the bones just
left of the midline. Continue using this rocking motion to cut
through the bone until approximately 5 mm rostral to the crib-
riform plate.
14. Grasp both hemispheres near the caudal-most region of tissue
with two pairs of Adson forceps or the thumb and index finger
of both hands. Gently separate the hemispheres. At the ante-
rior attachment point, make a small cut using sharp scissors to
separate the two hemispheres.
15. Retain the right hemisphere, which will have both vomerona-
sal organs but just one hemisphere of the olfactory bulb, plac-
ing it immediately into chilled dissection aCSF (see Fig. 2E).
16. Place the plastic plank on a flat piece of paper towel or filter
paper. Apply a small amount (~510 L) of Vetbond tissue
glue on one half of the plank. Grasp the right hemisphere with
Adson forceps via the dorsal and ventral bones. Transfer the
tissue to the paper towel or the filter paper a few centimeters
from the plank to wick away excess fluid, and then place the
tissue on the plank with the rostral tissue facing the short end
of the plank (see Fig. 2F).
242 Julian P. Meeks and Timothy E. Holy

3.2 Secondary 1. Coat a portion of the perfusion dish or chamber by a thin

Dissection (1 mm) layer of Dow Corning silicone vacuum grease. Press
(See Note 3) the flat end of the plastic plank and tissue firmly down into the
vacuum grease.
2. Initiate superfusion of the preparation. We use a peristaltic
pump from Gilson with a measured flow rate of 58 mL/min.
We additionally target the perfusion outlet towards the AOB
surface to increase convective exchange between the tissue and
perfusate.
3. While using a pair of steel #3 forceps to hold the plank in place,
use the other pair of #5 forceps, with the tips brought together,
to carefully separate the frontal cortical tissue away from the
olfactory bulb, but leaving it attached at the ventral edge
(including the lateral olfactory tract). A dense net of blood
vessels may be visible at higher magnification. It is helpful for
electrical recordings to remove these vessels, but it is a delicate
task. To accomplish this, one can, at this stage, use a sharp pair
of #5 forceps to grasp the net of vessels at the anterior/lateral
edge of the AOB, and then carefully pull the net of vessels
caudally away from the AOB, and then ventrally and medially
to peel it away. If you have already removed the frontal cortical
tissue, this is often quite difficult. After peeling the vessels from
the AOB, remove the frontal cortex from the preparation using
forceps.
4. Again while using a pair of #3 forceps to hold the plank in
place, detach the contralateral VNO with #5 forceps. This can
be accomplished by stabbing through the vomeronasal bone
and iteratively peeling it away in chunks. Alternatively, one can
remove the entire contralateral VNO by placing a single tip of
#5 forceps between the two vomeronasal organs at the caudal/
dorsal edge, and then gently running the forceps anteriorly
until they begin separating. At this point the entire VNO,
including the bone, can be removed by grasping the dorsal/
caudal edge with #5 or #3 forceps.
5. The septal cartilage will appear opaque white, and covers the
vomeronasal nerves. This must be removed in order to expose
the vomeronasal nerves to superfusion. Using a pair of #3 for-
ceps, grasp the caudal-most portion of the septal cartilage and
gently pull it slightly away from the preparation. Do not
attempt to remove the septal cartilage quickly. Instead, use a
second pair of #3 forceps (in the other hand) to gently separate
the cartilage from the septal tissue along the anterior and dor-
sal edges until reaching the rostral connective tissue. Here,
carefully sever the connections using #5 or #3 forceps. If this
process is not done carefully, the septal tissue may become
strained, and can detach from the VNO, severing the vomero-
nasal nerves.
 Electrical Recordings in the Ex Vivo AOB 243

6. The septal bone will now be visible just rostral to the cribriform
plate. It is a transparent bone with blood vessels visible in its
interior. To fully expose the vomeronasal nerves, this must also
be removed. To do so, use a pair of #3 forceps in the left hand
and a pair of #5 forceps in the right. Very carefully, approaching
from an angle nearly parallel to the septal tissue, grasp the septal
bone near the caudal/dorsal edge. While gently pulling the cau-
dal/dorsal edge away from the tissue, take the #5 forceps, with
tips together, and perforate the septal bone just rostral to the
cribriform plate. This will assist in forming a linear crack in
the bone. Once the bone cracks, it can be removed easily using
the forceps in the left hand (see Note 4).

3.3 Insertion 1. Before inserting the polyimide cannula into the VNO, it is
of Polyimide important to inspect the VNO at its most rostral end. Note
Cannula into that the VNO is connected to the nasal cavity through a very
Vomeronasal Organ thin cartilaginous tube. It is best to remove the cartilage tube
just as it touches the rostral-most VNO. Using #5 forceps,
sever the cartilage at the VNO junction by pinching the two
forceps points together. Keep the hand forceps pinched at this
junction, and then use the forceps in the other hand to pull the
cartilage tube away from the VNO.
2. Bring the polyimide cannula within 1 cm of the ex vivo prepa-
ration. Position the cannula so that it will have at least 1 cm of
extra slack (i.e., such that it can reach many areas around the
VNO, and will not be stretched to do so). It is helpful to
increase magnification at this stage if possible to focus on the
anterior VNO.
3. Initiate Ringers flow through the cannula (with a flow rate
between 0.1 and 0.4 mL/min). Submerge the cannula using a
pair of #3 forceps. Gently grasp the cannula approximately
2 mm from its tip, and bring it near the opening to the VNO
at the rostral end of the snout. It is common for the effective
flow rate out of the cannula to decrease while the tube is being
held in the forceps.
4. Bring the cannula opening near the VNO opening. Use the
Ringers flow out of the cannula to help align the cannula to
the VNO opening. When the cannula flow is parallel to the
VNO opening, you will observe the VNO inflating, which
causes the blood in the blood vessels supplying the VNO to
empty. Upon observing this, slowly move the cannula into the
VNO, stopping and retrying if your VNO deflates due to
movement of the cannula.
5. Stop advancing the cannula when it is half way into the VNO
lumen. When the cannula is released, it should relax to the
lateral portion of the VNO, containing the pump organ.
244 Julian P. Meeks and Timothy E. Holy

This may disrupt the pump organ, but minimally disrupts

the VNO epithelium. Check to ensure that the flow rate
through the cannula is between 0.1 and 0.4 mL/min and
that it is stable.

3.4 Single-Glass 1. Prepare 410 borosilicate glass electrodes using a microelec-

Electrode Recordings trode puller. Tip diameter should be no less than 1 m, and
impedance should be 38 M. Taper and smoothness of the
microelectrode tip are matters of personal preference.
2. At the time of recording, backfill the electrode with syringe-
filtered aCSF, removing any bubbles with gentle finger flicks.
3. Mount the electrode on the holder, and then use the micropo-
sitioner to bring the electrode within 100 m of the AOB sur-
face, using the dissection microscope.
4. Using slow advances of the fine (penetrating) direction of the
micropositioner (no more than 510 m per step) advance the
electrode into the AOB. Upon encountering the AOB surface,
the pitch (when listening via an acoustic monitor) will sud-
denly shift from high to low. Use this audible transition to
mark your positioner to zero depth.
5. Advance the electrode into the AOB tissue, to the desired
depth, going no faster than 1 m/s. Steps of 5 m are conve-
nient, but best results require one to wait at least several sec-
onds after each step.
6. The following recording environments are typically encoun-
tered when advancing electrodes along the AOB transverse
axis (see Note 5).
(a) In the initial 80 m of depth, small-amplitude, negative-
polarity spikes are often encountered, which may
correspond to action potentials in periglomerular den-
drites or to backpropagating action potentials in mitral
cell dendrites. Also, it is common when using smooth-
tipped microelectrodes to occasionally encounter a
large transition from high-frequency, low-amplitude
noise to extremely large-amplitude (millivolts RMS),
low-frequency oscillations. These are likely the result of
contacting blood vessels, which can clog the tip and
cause large changes in tip impedance. Retracting the tip
2050 m will often reverse this change. In some cases,
a new electrode penetration origin will be necessary to
bypass the blood vessel.
(b) From 80 m of depth to 250 m in depth, larger ampli-
tude spikes can be isolated, corresponding to the proximal
dendrites and somata of cell types in the external cellular
layer. Extremely large amplitude spikes (sometimes greater
than 1 mV) that are positive in polarity can be encountered,
 Electrical Recordings in the Ex Vivo AOB 245

likely reflecting a switch in recording microenvironment

to one in which the large capacitive currents of mitral cell
membranes dominate ionic currents. These positive-
polarity waveforms are often multi-peaked, reflecting the
propagation of action potentials through the axonal,
somatic, and dendritic regions of these larger cells. This is
especially the case when using higher impedance and
smoothed electrode tips.
(c) Recordings from depths deeper than 250 m in depth
require extra patience. The time necessary to properly
penetrate to this depth, along with the practical likelihood
that at this depth the thicker portion of the microelec-
trode taper may be entering tissue and exerting some
downward forces, means that advancement must be done
quite slowly. Because the granule cells located deep to the
lateral olfactory tract are small in diameter and often do
not respond strongly to stimuli, encountering spikes at
these depths is rare. An alternative strategy for recording
from these neurons is to advance the electrode from a
more extreme medial or dorsal-medial angle. This can
allow some granule cells to be encountered within 250 m
from the entry point.
7. Post hoc spike sorting is sometimes necessary, as even smooth,
48 M pipettes are capable of detecting spikes from multiple
nearby cells. Principal component analysis of the spike wave-
forms, along with manual, k means, or Gaussian mixture model
segmentation, is typically able to perform this function with
high confidence.

4 Notes

1. The goal for completing the initial dissection should be

approximately 58 min. Expect to practice this dissection doz-
ens of times before being able to complete the initial dissection
in this amount of time.
2. Recall that when the tissue is facing ventral-up the left hemi-
sphere will be on the right side of the midline.
3. The design of the tissue chamber is an important consider-
ation for any researcher intending to perform AOB record-
ings. The ex vivo tissue chamber described in Meeks and
Holy [4] was custom-built to allow various electrode pene-
tration angles while keeping the chamber volume minimized.
Keep in mind that this chamber also contained an aCSF oxy-
genation region just before the tissue, which can generate
electrical artifacts.
246 Julian P. Meeks and Timothy E. Holy

4. If a clear break of the septal bone does not occur on the initial
perforation, repeat one time. If the bone cracks in an unin-
tended place, one must remove the bone by peeling a small
piece at a time. There is a large risk of puncturing the septal
tissue at this stage, so proceed with great caution.
5. During electrode penetrations, it is helpful to periodically
stimulate the VNO with a positive control stimulus in place of
the standard Ringers solution. For example, a 50 mM KCl
Ringers solution (substituting KCl for NaCl on an equimolar
basis) will depolarize vomeronasal neurons and activate wide-
spread AOB activity. This has the effect of increasing spontane-
ous firing rates in the AOB, which can improve the rate of
identifying single neural units.

References
1. Luo M, Fee MS, Katz LC (2003) Encoding reveals sensory encoding of conspecific and
pheromonal signals in the accessory olfactory allospecific cues by the mouse accessory
bulb of behaving mice. Science 299:1196 olfactory bulb. Proc Natl Acad Sci U S A
2. Hendrickson RC, Krauthamer S, Essenberg 107:5172
JM, Holy TE (2008) Inhibition shapes sex 4. Meeks JP, Holy TE (2009) An ex vivo prepara-
selectivity in the mouse accessory olfactory bulb. tion of the intact mouse vomeronasal organ and
J Neurosci 28:12523 accessory olfactory bulb. J Neurosci Methods
3. Ben-Shaul Y, Katz LC, Mooney R, Dulac C 177:440
(2010) In vivo vomeronasal stimulation
 Chapter 18

Pheromone-Induced Expression of Immediate Early

Genes in the Mouse Vomeronasal Sensory System
Sachiko Haga-Yamanaka and Kazushige Touhara

Abstract
Immediate early genes (IEGs) are powerful tools for visualizing activated neurons and extended circuits
that are stimulated by sensory input. Several kinds of IEGs (e.g., c-fos, egr-1) have been utilized for detect-
ing activated receptor neurons in the pheromone sensory organ called the vomeronasal organ (VNO), as
well as for mapping the neurons within the central nervous system (CNS) excited by pheromones.
In this chapter, we describe the procedure for the detection of pheromone-induced neural activation
in the VNO and CNS using the c-Fos immunostaining technique.

Key words Vomeronasal organ (VNO), Accessory olfactory bulb (AOB), Bed nucleus of the stria
terminalis (BST), Medial amygdala (Me), Posteromedial cortical amygdaloid nucleus (PMCo),
Medial preoptic area (MPA), Ventromedial hypothalamic nucleus (VMH), Immediate early genes
(IEGs), c-Fos

1 Introduction

In the vast majority of terrestrial animals, pheromone signals are

detected and processed in the vomeronasal system. This sensory
system consists of a sensory organ called the vomeronasal organ
(VNO) where the pheromones are received by their corresponding
receptor neurons, the accessory olfactory bulb (AOB) where the
axons of the VNO neurons project, and a set of nuclei in the central
nervous system (CNS) where the information is transferred by the
AOB and is further processed and integrated. The nuclei in the
CNS involved in pheromone signal processing include those
projected by the AOB, such as the bed nucleus of the stria terminalis
(BST), the medial amygdala (Me), and the posteromedial cortical
amygdaloid nucleus (PMCo), whereas these nuclei further project
to the higher brain areas in the hypothalamus such as the medial
preoptic area (MPA) and the ventromedial hypothalamic nucleus
(VMH) [13].

Kazushige Touhara (ed.), Pheromone Signaling: Methods and Protocols, Methods in Molecular Biology, vol. 1068,
DOI 10.1007/978-1-62703-619-1_18, Springer Science+Business Media, LLC 2013

247
248 Sachiko Haga-Yamanaka and Kazushige Touhara

To those researchers who seek to understand the modes of

action of pheromones, the immediate early gene (IEG)-based
activity mapping strategy has provided valuable information about
the receptor neurons, as well as the neural pathways through which
pheromone signals are processed. Among various kinds of IEGs,
c-Fos has become the most widely used functional marker to reveal
activated neurons for several reasons: (1) it is expressed at low lev-
els in the VNO, AOB, and CNS under basal conditions; (2) it is
stereotypically induced in response to several extracellular signals
including pheromones and neurotransmitters; (3) the response is
transient; (4) its expression can be detected relatively easily; and
(5) its immunostaining can be readily combined with the detection
of various other molecules including sensory receptor proteins,
neuroanatomical marker proteins, retrograde tracers, and other
activity markers.
Indeed, by monitoring the c-Fos-inducing activity during
purification steps, a male mouse pheromone, the Exocrine gland-
Secreting Peptide 1 (ESP1), was identified from the lachrymal
grand [4, 5]. By analyzing the co-localization of the pheromone-
inducing c-Fos and the pheromone receptors, the receptor of
ESP1, V2Rp5, was identified out of more than 300 types of phero-
mone receptors [6]. By using a similar technique, it was also
revealed that pheromone receptors detect not only intraspecific
cues but also interspecific signals such as predators [7]. Moreover,
by investigating c-Fos expression patterns, neural circuitries
involved in processing distinct pheromone signals have been
successfully investigated [810].
Here we describe the basic procedures for detecting c-Fos-
immunoreactive neurons in different areas of the mouse nervous
system (VNO, AOB, and CNS) stimulated by a pheromone input.
As for the detailed protocols for double-staining with other mark-
ers, readers are directed to other recent publications [6, 7, 10].

2 Materials

2.1 Pheromone 1. Mice: 10-week-old female mice single-housed for a week prior
Stimulation to the experiment without bedding change.
2. Pheromone: 10 g ESP1 in 20 mg cotton swab (ESP1 was
dissolved in 100 L phosphate-buffered saline (PBS), absorbed
by cotton swab and dried).
3. Control stimulation: 20 mg cotton swab soaked with 100 L
PBS and dried.
4. Anesthesia: Sodium pentobarbital (150 g/g mouse), 1 mL
syringe, 27G injection needle.
 Pheromone-Induced Expression of Immediate Early Genes 249

5. Dissection tools: 70 % ethanol (EtOH), decapitation scissors,

surgical scissors, curved dissecting forceps, Dumont #7 for-
ceps, petri dish, chilled 4 % paraformaldehyde (PFA) in PBS.
6. Perfusion and fixation: 10 mL syringe with chilled PBS, 20 mL
syringe with chilled 4 % PFA/PBS, 26G injection needle.

2.2 Specimen 1. Postfixation: 4 % PFA in PBS, glass vials (20 mL, with lid).
Preparation 2. Decalcification: 500 mM ethylenediaminetetraacetic acid
(EDTA).
3. Freezing protection: 15 and 30 % sucrose in PBS.
4. Embedding: Tissue-Tek O.C.T. (Optimal Cutting
Temperature) Compound (Sakura), Tissue-Tek Cryomold
(Sakura), curved dissecting forceps.
5. Slicing the VNO and AOB: Research cryostat (Leica,
CM3050S), MAS coated glass slide (Matsunami), hair drier,
slide glass case.
6. Slicing the brain: Sliding microtome (Yamato, REM-710),
electro-freezing component (Yamato, MC-802A), 24-well cell
culture plate (Corning), ink brush, PBS.

2.3 Immunostaining 1. Staining jars, racks for the slide glass.

2. Agglutination plate (TOMY, T-2).
3. Ink brush.
4. Super Pap pen liquid blocker.
5. Fixation: 4 % PFA in PBS.
6. Rinse: Tris-buffered saline (TBS; 100 mM TrisHCl, 150 mM
NaCl, pH 7.5) or Tris-buffered saline with Triton X-100
(TBST; 100 mM TrisHCl, 150 mM NaCl, 0.1 % Triton
X-100, pH 7.5).
7. Peroxidase inactivation: 1, 0.1, or 0.3 % hydrogen peroxide
(H2O2) in PBS.
8. Moisture chamber.
9. Blocking solution: 3 % bovine serum albumin (BSA) in TBST.
10. Diluent solution: 3 % BSA in TBST.
11. Primary antibody: Rabbit anti-c-Fos antibody (Calbiochem,
PC38, Ab-5).
12. Secondary antibody: Biotinylated goat anti-rabbit IgG anti-
body (Vector Laboratories, BA-1000).
13. ABC solution: Avidin and biotin (Vector Laboratories,
VECTASTAIN ABC system, PK-6100) in PBS.
14. Detection: Diaminobenzidine (DAB; SIGMAFAST
3,3-diaminobenzidine tablets, D4293), 0.6 % (w/v) ammo-
nium nickel (II) sulfate.
250 Sachiko Haga-Yamanaka and Kazushige Touhara

15. Dehydration: H2O, EtOH, xylene.

16. Mounting: Mounting medium (Mount-Quick), cover glass
(Matsunami glass, Neo).

2.4 Image 1. Upright microscope (Olympus, BX 60).

Acquisition 2. Digital camera (Olympus, DP10).

3 Methods

3.1 Pheromone 1. Subject a mouse to either the pheromone (ESP1) or the control
Stimulation and stimulation between 22:00 and 24:00 (light period: 9:00
Intracardial Perfusion 21:00). Introduce a piece of cotton containing either ESP1 or
PBS into the mouse cage. Stimulate the mouse for 90 min
(see Note 1).
2. Anesthetize the mouse with intraperitoneal injection of pen-
tobarbital using 1 mL syringe with 27G injection needle
(see Note 2).
3. Place the mouse on the surface of a styrofoam plate with abdo-
men facing up. Using the small needles, secure the four paws
to the surface spreading them as wide as possible.
4. Wet the abdomen with 70 % EtOH.
5. Grab skin with curved dissecting forceps at the level of the
diaphragm, and cut to expose the liver. Cut laterally and then
up, cutting through the ribs. Lift flap and continue cutting
until the heart is easy to access.
6. Place 26G injection needle connected to 10 mL syringe with
PBS into the left ventricle and immediately cut the right atrium.
7. Take 5 min to slowly inject 5 mL PBS.
8. Change the 10 mL syringe to a 20 mL syringe containing 4 %
PFA/PBS.
9. Spend 15 min to slowly inject 15 mL 4 % PFA/PBS
(see Note 3).
10. Remove the small needles securing the mouse paws.
11. Decapitate and cut the skin from the base of head to the tip of
nose.
12. Peel the skin, expose the entire skull, and remove eyes using
curved dissecting forceps.
13. Remove lower jaw and tongue using surgical scissors.
14. Grasp the lateral side of nose and peel the dorsal and lateral
skull using Dumont #7 forceps starting at the caudal edge
of the brain stem to the rostral tip of the olfactory bulbs
(see Note 4).
 Pheromone-Induced Expression of Immediate Early Genes 251

15. Hold the lateral side of the nose with dorsal side down and
carefully take out the brain with the olfactory bulbs from the
skull using curved dissecting forceps. Carefully drop the brain
with the olfactory bulb into a petri dish filled with chilled 4 %
PFA/PBS.
16. Cut the bone where the tip of the olfactory bulb existed and
place the nasal tissue into a petri dish filled with chilled 4 %
PFA/PBS.

3.2 Specimen 1. Incubate the nose in a vial with 4 % PFA/PBS for 3 h at 4 C

Preparation (Nose) (see Note 5).
2. Incubate the nose in a vial with 500 mM EDTA/PBS for two
nights at 4 C (see Note 6).
3. Incubate the nose in a vial with 15 % sucrose/PBS for 3 h
at 4 C.
4. Incubate the nose in a vial with 30 % sucrose/PBS overnight
at 4 C.
5. Embed the nose in O.C.T. compound in a cryomold (see Note
7) and freeze. Embedded samples can be stored at 80 C.
6. Cut 14 m thick coronal sections using the cryostat, collect
sections containing the VNO on slides, and dry the sections
with hair drier (see Note 8). Sections can be stored in a slide
glass storage case at 80 C.

3.3 Specimen 1. Incubate the brain in a vial with 4 % PFA/PBS for 3 h at 4 C.

Preparation (Brain) 2. Incubate the brain in a vial with 15 % sucrose/PBS for 3 h at 4 C.
3. Incubate the brain in a vial with 30 % sucrose/PBS overnight
at 4 C.
4. Cut the brain at about 2.8 mm anterior from the bregma in
order to separate the olfactory bulb from the brain.
5. Embed the olfactory bulb in the O.C.T. compound in a cryo-
mold (see Note 7) and freeze. Embedded samples can be
stored at 80 C.
6. Cut 25 m thick parasagittal sections of the olfactory bulb
using a cryostat, collect sections containing AOB on slides,
and dry the sections with hair drier (see Note 8). Sections can
be stored in a slide glass storage case at 80 C.
7. Cut 50 m thick coronal sections of the brain using a sliding
microtome. Sections are collected in PBS in a 24-well plate
using an ink brush.

3.4 Immunostaining 1. Seal the edge of slides with the liquid blocker and place them
(for VNO and AOB in a staining rack (Fig. 1a).
Specimens) 2. Put the staining rack in a staining jar containing TBST and
rinse the slides in TBST for 5 min (Fig. 1a).
252 Sachiko Haga-Yamanaka and Kazushige Touhara

a
step 1

step 2 step 3 step 4

staining rack

staining jar

TBST 4 % PFA / PBS TBST

b

Fig. 1 Schematic illustration of the immunostaining procedures for the VNO and AOB. (a) Seal the edge of
slides with the liquid blocker (dashed line) and place them in a staining rack (see step 1). Transfer the staining
rack between staining jars containing different solutions. (b) Place the slides with the sections on the top in a
moisture chamber flatly and gently apply 200 L of each solution

3. Put the staining rack in a staining jar containing 4 % PFA/PBS

and incubate the slides at 4 C for 20 min (Fig. 1a).
4. Put the staining rack in a staining jar containing TBST and
rinse the slides in TBST for 5 min.
5. Put the staining rack in a staining jar containing H2O2/PBS
(1 % for nose, 0.1 % for AOB) and incubate the slides at room
temperature (RT) for 30 min (see Note 9).
6. Put the staining rack in a staining jar containing TBST and
rinse the slides in TBST for 5 min with agitation (repeat three
times).
7. Place the slides with the sections on the top in a moisture
chamber flatly and apply 200 L blocking solution to each
slide. Incubate the slides at 25 C for 45 min (Fig. 1b).
8. Drain off the blocking solution, place the slides back into the
chamber, and apply 200 L diluent solution containing the
primary antibody (1/5,000). Incubate the slides in a moisture
chamber at 4 C for three nights (Fig. 1b).
9. Drain off the antibody solution and place the slides in a staining
rack. Put the staining rack in a staining jar containing TBST
 Pheromone-Induced Expression of Immediate Early Genes 253

and rinse the slides in TBST for 5 min with agitation (repeat
three times).
10. Place the slides with the sections on the top in a moisture
chamber flatly and apply 200 L TBST containing the second-
ary antibody (1/200). Incubate the slides at 25 C for 1 h
(Fig. 1b).
11. Prepare the ABC solution of PBS containing avidin (1/100
dilution) and biotin (1/100 dilution) at 15 min before step
12 (see Note 10).
12. Drain off the antibody solution and place the slides in a stain-
ing rack. Put the staining rack in a staining jar containing TBS
and rinse the slides in TBS for 5 min with agitation (repeat
three times).
13. Place the slides with the sections on the top in a moisture
chamber flatly and apply 200 L ABC solution. Incubate the
slides at 25 C for 30 min (Fig. 1b).
14. Drain off the ABC solution and place the slides in a staining rack.
Put the staining rack in a staining jar containing TBS and rinse
the slides in TBS for 5 min with agitation (repeat three times).
15. Place the slides with the sections on the top in a moisture
chamber flatly and apply DAB solution. Incubate the slides for
510 min (until brown signals appear) (Fig. 1b).
16. Drain off the DAB solution and place the slides in a staining
rack. Put the staining rack in a staining jar containing TBS and
rinse the slides in TBS for 5 min with agitation (repeat three
times).
17. Put the staining rack in a staining jar containing EtOH/H2O.
Dehydrate the samples in a series of 80, 95, and 100 % EtOH
(two times each) and 100 % xylene (two times) at RT for
2 min each.
18. Mount the slides with 50 L of mounting medium and put a
cover glass on the top.
19. Observe the specimens under the microscope and take images
with a digital camera (Fig. 2).

3.5 Immunostaining 1. Transfer brain sections with the BST, MEA, PMCo, MPA, and
(for Brain) VMH (Fig. 3) to the wells of the agglutination plate contain-
ing 500 L PBS, at most two sections per well, using ink brush
and rinse the sections for 10 min (Fig. 4).
2. Transfer the sections to the wells containing 500 L 0.3 %
H2O2/PBS and incubate them at RT for 30 min (Fig. 4 and
see Note 11).
3. Transfer the sections to the wells containing PBS and rinse
them for 10 min (two times).
254 Sachiko Haga-Yamanaka and Kazushige Touhara

Fig. 2 ESP1 induces c-Fos expression in the VNO and AOB. (a) Representative
images of anti-c-Fos immunostaining in the VNO stimulated by PBS (left ) or
ESP1 (right ). (b) Representative images of anti-c-Fos immunostaining in the
AOB stimulated by PBS (left ) or ESP1 (right ). Lower panels show the enlarged
view of the rectangles in the upper panels. The scale bars, 200 m (upper panel )
and 100 m (lower panel )

4. Transfer the sections to the wells containing 200 L blocking

solution and incubate them at RT for 30 min.
5. Transfer the sections to the wells containing 150 L diluent
solution with the primary antibody (1/30,000). Seal the wells
with the tape and incubate the sections at 4 C overnight
(see Note 12).
6. Transfer the sections to the wells containing 500 L PBS and
rinse them for 10 min (three times) (see Note 12).
 Pheromone-Induced Expression of Immediate Early Genes 255

a Bregma 0.14 mm
b Bregma -1.46 mm
c Bregma -2.70 mm

BST

VMH

MPA PMCo
Me

Fig. 3 Schematic illustration of the nuclei involved in pheromone information processing. Representative brain
sections containing the BST and MPA (a), VMH and Me (b), and PMCo (c). Relative distance from the bregma
is shown at the top of each image. The illustrations are modified from [11]

step 1

PBS step 1

0.3% H2O2 / PBS step 2

PBS step 3

blocking soln. step 4

primary antibody step 5

Fig. 4 Schematic illustration of the immunostaining procedures for brain. Put the brain sections into the wells
of the agglutination plate, at most two sections per well, using an ink brush. Transfer the sections to the wells
containing each solution
256 Sachiko Haga-Yamanaka and Kazushige Touhara

7. Transfer the sections to the wells containing 150 L TBST

with the secondary antibody (1/2,000) and incubate them at
25 C for 1 h (see Note 12).
8. Prepare the ABC solution of PBS containing avidin (1/200
dilution) and biotin (1/200 dilution) just before step 9
(see Note 10).
9. Transfer the sections to the wells containing 500 L PBS and
rinse them for 10 min (three times) (see Note 12).
10. Transfer the sections to the wells containing 200 L ABC solu-
tion and incubate them at 25 C for 30 min (see Note 12).
11. Transfer the sections to the wells containing 500 L PBS and
rinse them for 10 min (three times) (see Note 12).
12. Prepare DABNi solution during step 11.
13. Transfer the sections to the wells containing 500 L DABNi
solution and incubate them for 510 min (until black signals
appear) (see Notes 12 and 13).
14. Transfer the sections to the wells containing 500 L PBS and
wash out DAB solution (see Note 12).
15. Transfer the sections to the wells containing 500 L PBS and
rinse them for 10 min (two times).
16. Affix the sections to slide glasses using an ink brush and let
them dry. Place slides in a staining rack.
17. Put the staining rack in a staining jar containing EtOH/H2O.
Dehydrate the samples in a series of 80, 95, and 100 % EtOH
(two times each) and 100 % xylene (two times) at RT for
2 min each.
18. Mount the slides with 50 L of mounting medium and put a
cover glass on the top.
19. Observe the specimens under the microscope and take images
with a digital camera (Fig. 5).

Fig. 5 ESP1-induced c-Fos expression in the brain. Representative images of anti-c-Fos immunostaining in the
brain stimulated by ESP1. Each section corresponds to Fig. 3
 Pheromone-Induced Expression of Immediate Early Genes 257

4 Notes

1. Mice are stimulated in the dark or under the red light in a quiet
environment. In order to reduce the neural activation for back-
ground environment, we house mice individually in mouse iso-
lators and stimulate them in the isolator without moving the
cages before stimulation.
2. To ensure that the mouse is properly sedated, pinch the toes to
judge its level of response to painful stimulus.
3. Mouse should be quite stiff at the end of this step.
4. Be careful not to damage the brain tissue with forceps. After
removing the skull, it may also be necessary to remove dura
mater using Dumont #7 forceps.
5. In order to remove the air in the nostril cavities, fill the nostril
cavities with 4 % PFA/PBS using a pipette.
6. After incubation for two nights, the bone becomes soft; or if
not, refresh the EDTA solution and keep incubating the nose
for one more night.
7. Briefly absorb the moisture of the tissue surface with a piece of
Kimwipe and soak the tissue into the O.C.T. compound in a
petri dish. Fill the nostril cavities with O.C.T. compound using
a pipette in the petri dish, and then transfer the sample into a
cryomold filled with O.C.T. compound.
8. Use cold air mode to dry slides.
9. Remove bubbles generated by the peroxidase reaction from
the surface of specimens by moving the slide rack up and down
slowly several times.
10. Prepare ABC solution 30 min prior to the application (follow
the user instruction of the kit).
11. Remove bubbles generated by the peroxidase reaction from
the surface of specimens by gently touching with an ink brush.
12. Rinse the ink brush thoroughly to avoid contamination in later
steps.
13. Keep the reaction time consistent among all the specimens
using a timer.

References
1. Wyatt TD (2003) Pheromones and animal 3. Dulac C, Wagner S (2006) Genetic analysis of
behaviour. Cambridge University Press, brain circuits underlying pheromone signaling.
Cambridge Annu Rev Genet 40:449467
2. Dulac C, Torello AT (2003) Molecular detec- 4. Kimoto H, Touhara K (2005) Induction of
tion of pheromone signals in mammals: from c-Fos expression in mouse vomeronasal neu-
genes to behaviour. Nat Rev Neurosci 4: rons by sex-specific non-volatile pheromone(s).
551562 Chem Senses 30(Suppl 1):i146i147
258 Sachiko Haga-Yamanaka and Kazushige Touhara

5. Kimoto H, Haga S, Sato K, Touhara K (2005) Fos responses to male pheromones in male and
Sex-specific peptides from exocrine glands female mice. J Neurobiol 39:249263
stimulate mouse vomeronasal sensory neu- 9. Halem HA, Baum MJ, Cherry JA (2001) Sex
rons. Nature 437:898901 difference and steroid modulation of
6. Haga S, Hattori T, Sato T, Sato K, Matsuda S, pheromone-induced immediate early genes in
Kobayakawa R, Sakano H, Yoshihara Y, the two zones of the mouse accessory olfactory
Kikusui T, Touhara K (2010) The male mouse system. J Neurosci 21:24742480
pheromone ESP1 enhances female sexual 10. Choi GB, Dong HW, Murphy AJ, Valenzuela
receptive behaviour through a specific vom- DM, Yancopoulos GD, Swanson LW,
eronasal receptor. Nature 466:118122 Anderson DJ (2005) Lhx6 delineates a path-
7. Isogai Y, Si S, Pont-Lezica L, Tan T, Kapoor V, way mediating innate reproductive behaviors
Murthy VN, Dulac C (2011) Molecular orga- from the amygdala to the hypothalamus.
nization of vomeronasal chemoreception. Neuron 46:647660
Nature 478:241245 11. Franklin KBJ, Paxinos G (2007) The mouse
8. Halem HA, Cherry JA, Baum MJ (1999) brain in stereotaxic coordinates, 3rd edn.
Vomeronasal neuroepithelium and forebrain Academic, San Diego
 Part IV

Protocols in Analyzing Behavioral and Endocrine

Changes Elicited by Pheromones
 Chapter 19

Insect Pheromone Behavior: Fruit Fly

Daisuke Yamamoto, Soh Kohatsu, and Masayuki Koganezawa

Abstract
The amenability to genetics of Drosophila melanogaster has made this organism one of the best-suited
models for studying the neurobiology of pheromone-guided behavior. Single-male assays use the minigene
encoding the thermosensitive channel dTrpA1 to activate neurons expressing fruitless (fru), a major court-
ship regulator gene, and thereby induce most of the elementary courtship acts in a solitary male exposed
to temperature increase. Tethered male assays allow Ca2+-imaging of neuronal activities of a male fly dis-
playing courtship behavior on a treadmill when stimulated with a female or pheromones. Here we describe
technical details of these assays.

Key words Drosophila, Thermogenetics, Ca2+-imaging, MARCM, Courtship, fruitless

1 Introduction

The explosive development of optogenetic and thermogenetic

tools has opened a new era in behavioral neurobiology [13].
These new tools are most powerful in resolving the neural circuitry
for behavior at the single-cell level, when combined with other
sophisticated genetic techniques available in the model organism
Drosophila melanogaster [4]. What is crucial in these studies is our
ability to rigorously identify the neurons whose activities are moni-
tored and manipulated in an effort to capture neural correlates of
specific behavior [513]. Refined versions of the GAL4-UAS sys-
tem, for example, allow us to restrict the target of study to even a
single neuron, which can be selectively labeled, activated, inacti-
vated, or activity-monitored in an animal on the move [14]. The
Mosaic Analysis with a Repressible Cell Marker (MARCM) tech-
nique [15] has been shown to be particularly useful in specifying
central neurons for the generation of male courtship behavior [16, 17].
Two critical components in MARCM, in addition to GAL4- and
UAS-fused transgenes, are the GAL4 repressor GAL80 and the
Flp-FRT system that enables one to flip out GAL80 sporadically
from the chromosome in a dividing cell [15]. As a result, one of

Kazushige Touhara (ed.), Pheromone Signaling: Methods and Protocols, Methods in Molecular Biology, vol. 1068,
DOI 10.1007/978-1-62703-619-1_19, Springer Science+Business Media, LLC 2013

261
262 Daisuke Yamamoto et al.

the two daughter cells loses GAL80, subsequently forming a

neuronal clone with GAL4 activity when flipping occurs in a neu-
roblast that remains proliferative, or forming a single neuron with
GAL4 activity when it happens in a ganglion mother cell that
undergoes the final division. In the entire brain, only these GAL4-
expressing neurons can be labeled, manipulated, or monitored for
activity recording, since the rest of the brain is not affected by
GAL4-responsive transgenes. For labeling neurons, a variety of
markers are available [4], including the widely used mCD8::GFP
[15]. The protein most favored as a neural activator is dTrpA1, a
warmth-sensitive ion channel that excites the expressing cell when
the ambient temperature is elevated beyond a certain level, e.g.,
from the permissive temperature 20 C to the restrictive tempera-
ture 30 C [18]. The other option often employed to activate neu-
rons is the use of channelrhodopsin, a light-sensitive channel that
makes the expressing neuron excitable upon illumination [19].
In inactivating neurons, reagents that reduce electrical excitability
or block synaptic outputs are typically used [4]. Halorhodopsin is
a light-sensitive Cl-transporter that hyperpolarizes the membrane
potential of the expressing neuron and reduces excitability [20].
Conditional and reversible synaptic blockage is attained with a
temperature-sensitive dominant-negative form of the Dynamin-
coding gene shibire (shits) by raising the temperature beyond 29 C
[1] while the Tetanus toxin light chain (TNT) constitutively blocks
the transmission [21]. To visualize neural activities, changes in the
intracellular concentration of various signal molecules associated
with excitation are measured based on the Fluorescence Resonance
Energy Transfer (FRET; [4]). Among those, the FRET-based Ca2+
sensor yellow cameleon [22] is widely used for the detection of
neural activities in animals displaying behavior under conditions in
which the movement of the recording site (e.g., the brain) is mini-
mized. In this chapter, we describe a few technical tips that are
critical for the effective application of these neurogenetic tools to
the analysis of cellular mechanisms of behavior, based on our recent
success in elucidating the neural circuitry for male courtship [17].

2 Materials

2.1 Generation of Depending on the behavior and cellular types to be analyzed, dif-
Flies with Desired ferent GAL4 lines are required. Here we use an enhancer-trap line,
Genotypes fruNP21, which expresses GAL4 in approximately 80 % of anti-Fru
antibody-positive neurons, which are central to the production of
male courtship behavior [16, 23].
1. Obtain fly lines harboring either the GAL4 driver fruNP21,
UAS-dTrpA1, UAS-TNT, UAS-IMP-TNT, UAS-yellow came-
leon 2.1, UAS-Channelrhodopsin, UAS-Halorhodopsin, and
 Insect Pheromone Behavior: Fruit Fly 263

UAS-shits or any other UAS-driven transgenes of interest from

stock centers or relevant laboratories (see Note 1).
2. For MARCM, obtain an additional fly line carrying hs-flp, FRT
G13 tub-Gal80/FRT G13 UAS-mCD8::GFP from a stock
center.
3. Construct flies carrying all the genetic elements necessary for
addressing specific problems, e.g., (1) hs-flp; FRTG13 tub-
Gal80/FRTG13 UAS-mCD8::GFP; fruNP21/UAS-dTrpA1 for
activating a fru-expressing neuronal clone, (2) hs-flp; FRTG13
tub-Gal80/FRTG13 UAS-mCD8::GFP; fruNP21/UAS-shits for
inactivating a fru-expressing neuronal clone, or (3) hs-flp;
FRTG13 tub-Gal80/FRTG13 UAS-mCD8::GFP; fruNP21UAS-
Yellow Cameleon 2.1/UAS-Yellow Cameleon for Ca2+-imaging
of a fru-expressing neuronal clone.
4. Isolate newly emerged female and male flies and keep singly in
small vials with cornmeal-yeast medium for several days before
use. Transfer the flies to new vials at least every 3 days.
5. For MARCM analysis to determine the brain foci for elemen-
tary courtship acts, induce clones expressing dTrpA1 (or shits)
in the brain during development. For this purpose, collect
embryos within 24 h of egg laying and subject them to heat
shock at 37 C for 60 min.

2.2 Single-Male 1. The observation chambers are constructed by cutting circular

Assays to Induce holes of 10 mm in diameter and 3 mm in height through a
Courtship Behavior copper plate, and covering them on either side with a glass slip.
in a Solitary Male 2. A thermal cycler (e.g., Gene Amp PCR system 9700; Perkin
Elmer) is used to control the temperature in the observation
chamber; the chamber plate is directly mounted on the alumi-
num block of the thermal cycler.
3. A dual LED illuminator, type W-E (Micronet), is used as a
light source to illuminate the chamber.
4. A WAT-221S (Watec) CCD camera equipped with a Micro-
NIKKOR lens (35 mm, 1:28; Nikon) is used to take movies of
fly behavior.
5. ArcSoft ShowBiz DVD software is used to control the movie
recordings.

2.3 Tethered Male 1. Fly holder. Punch a hole of approximately 8 mm in diameter in

Preparation a plastic coverslip. Seal it on one side with a piece of polyethyl-
ene film using UV glue (Fig. 1a) [24].
2.3.1 Fly Preparation
2. UV curing glue and UV pen light.
3. Syringe needle (2427 G: Terumo).
4. Tungsten needle of 0.15 mm in diameter, which is attached to
a holder.
264 Daisuke Yamamoto et al.

Fig. 1 The experimental setup for tethered male preparation. (a) The coverslip with wrap film to cover the head open-
ing for recording. (b) Lateral view of a fly attached to the wrap film. (c) Arrangement of instruments on the micro-
scopic stage. (d) Yellow cameleon-derived YFP fluorescence detected in the mushroom body (MB), lateral
protocerebrum (lpr), and optic tubercle (optu). A anterior, M medial. (e) Diagram of the entire tethered mate system

5. Silicone elastomer (Kwik-Sil; World Precision Instruments).

6. Fly saline: 130 mM NaCl, 5 mM KCl, 2 mM MgCl2, 2 mM
CaCl2, 36 mM sucrose, 5 mM HEPESNaOH (pH 7.3) [25].
7. Cool anesthetization stage made of a brass plate of
50 50 5 mm.
8. Pulled glass capillaries.

2.3.2 Recording Setup The recording system is a composite of two separable elements.
The first element is used to acquire fluorescent signals from the
brain. The second element serves to monitor behavior by video-
recording lateral views of the subject male (Fig. 1c, e). An electrical
trigger signal synchronizes the operation of these two elements.
The subject fly is stimulated by touching his foreleg with a stimulant
flys body attached to a holder mounted on the arm of a manual
micromanipulator.
1. Upright fluorescent microscopy.
2. W Plan-apochromat, 40, NA = 1.0 (Carl Zeiss), a water
immersion objective lens with high NA values and a long work-
ing distance (see Note 2).
3. C7780-20 and Aquacosmos (Hamamatsu Photonics), a CCD
camera and the software to control it.
4. HBO-100 (Carl Zeiss), a high-pressure mercury lamp.
 Insect Pheromone Behavior: Fruit Fly 265

5. ND Filters. For the preparation of excitation light and separa-

tion of emission light, a 440/20 band-path filter, a 455 dichroic
filter, and a 460 nm long-path filter are used. Emission light is
further divided by a beam-splitting prism embedded in the
CCD camera (460490 nm for CFP and 490570 nm for YFP
signals), and then projected upon separate CCD chips (see
Note 3).
6. STC-H400 (Sensor Technology Co., Ltd.), an infrared CCD
camera.
7. 12 zoom system (Navitar), a macro zoom lens.
8. VTG-55D (FOR-A), an analog video timer.
9. DVMC-DA2 (Sony), an A/D video converter.
10. Styrofoam balls of 45 mm in diameter.
11. Styrofoam ball holder. A brass base equipped with a hemi-
spherical pocket of 6 mm in diameter and an associated air inlet
of 1 mm in diameter at the bottom of the pocket. The holder
is attached to the microscope stage with appropriate adaptors
for fine adjustment of the position.
12. Air source. A compressor produces an airstream, which, after
being decompressed, passes through cotton, activated char-
coal, and water in sequence for the filtration of contaminants,
and flows into the ball holder through a needle valve. Adjust
the airflow rate so as to float the ball stably (~1 ml/s).
13. Manual micromanipulator for handling the stimulus flys body.

2.3.3 Data Analysis 1. ImageJ [17] (http://rsbweb.nih.gov/ij/ and Excel, Microsoft) as

a software environment for image processing (see Note 4) [26].

2.4 Immuno- 1. Purchase reagents for immunostaining: Rabbit polyclonal anti-

histochemistry GFP antibody (Molecular Probes, Eugene, OR) and mouse
of Brain Neurons monoclonal nc82 as primary antibodies, and Alexa548-
conjugated goat anti-mouse IgG (Invitrogen, Carlsbad, CA)
and Alexa488-conjugated goat anti-mouse IgG (Invitrogen)
as secondary antibodies.
2. Prepare solutions for dissection of tissues (PBS), fixation (4 %
paraformaldehyde in PBS), washing (PBT = 0.4 % TritonX in
PBS), blocking (PBTN = 10 % normal goat serum in PBT), and
staining (anti-GFP at 1:1,000 dilution, nc82 at 1:20, anti-
rabbit IgG at 1:200, and anti-mouse IgG at 1:200, all in
PBTN). Purchase VECTORSHIELD (Vector Laboratories) as
a clearing and mounting reagent.
3. The LSM510META confocal microscopy (Carl Zeiss) yields a
series of optical sections which are used to obtain reconstituted
images with the aid of the ImageJ software.
266 Daisuke Yamamoto et al.

3 Methods

3.1 Single- 1. Introduce a single male with dTrpA1 (or shits)-expression by

Male Assays aspiration into a chamber placed on a thermal cycler.
2. Start video-recordings of the flys behavior.
3. Maintain the temperature in the chamber at 22 C for the ini-
tial 5 min and then increase up to 35 C by a series of 1 C
temperature steps of 5-min duration. The fly commences to
exhibit elements of courtship behavior when the temperature
exceeds certain threshold levels.
4. Estimate the locomotor speed of the fly based on automatic
measurement using the program Move-tr/2D version 7.2
(Library Co., Tokyo, Japan).
5. Calculate the event frequency as the proportion of time the
male spent performing each courtship action, e.g., tapping,
wing extension, licking, or attempted copulation.
6. Construct perievent histograms in which the midpoint of every
tapping event is taken as reference time zero. For a 6-s period
starting 3 s before and ending 3 s after the midpoint of every
tapping event, fly behavior is analyzed every 0.1 s. This analysis
yields 61 data points associated with a single tapping event that
serves as the time-zero reference, defining a single data set.
The perievent histogram is regarded as the probability distri-
bution of each courtship element with time. For example, if
licking was observed at the 0.1-s period between 2.0 and 2.1 s
before time zero in 50 out of 100 data sets, then the probabil-
ity of observing licking at this time point is estimated to be
50/100 = 0.5.
7. In MARCM analysis of the courtship circuitry, classify the
assayed flies into two groups: courters, who display unilateral
wing extension or any other appropriate act, and noncourters,
who do not. Both groups are subject to immunohistochemis-
try to identify the neurons expressing dTrpA1 (or shits).
8. For the observation of neurons, fix the dissected brain with 4 %
paraformaldehyde for 60 min and incubate it in 10 % normal
goat serum for 12 h at room temperature (25 C). Thereafter,
the tissues are reacted with primary antibodies, i.e., the anti-
GFP antibody (rabbit polyclonal: A6455; Molecular Probes,
Eugene, OR) and nc82 (mouse monoclonal; Developmental
Studies Hybridoma Bank) at dilutions of 1:1,000 and 1:10,
respectively, for 7296 h at 4 C. After washing, the brain is
treated with Alexa Fluor 488 goat anti-rabbit IgG (Invitrogen
A11034) or Alexa Fluor 546 goat anti-mouse IgG (Invitrogen
A11030) at a 1:500 dilution.
9. Observe the stained brain with the LSM 510 META confocal
microscope.
 Insect Pheromone Behavior: Fruit Fly 267

3.2 The Tethered 1. Anesthetize a male fly by placing it for ~20 s in a glass tube that
Male Preparation has been cooled by ice (see Note 5 and 6).
2. After mounting the anesthetized fly on an ice-cooled prepara-
tion stage, fill the gap between the head and thorax of the fly
with UV glue to restrict head movements (see Note 7).
3. Put a small amount of UV glue on the notum, which is then
fixed to a polyethylene film placed on the fly holder. The head
cuticle flanked by the base of the antennae and ocelli must be
rigidly attached to the film, yet the aristae on the antennae
should not be in contact with the film.
4. Hold the fly ventral side down in a humid chamber and let him
grasp a Styrofoam ball that prevents him from flying. Prepare a
dozen flies at a time according to steps 13 above.
5. Keep the flies in the humid chamber for approximately 2 h to
let them recover from manipulation.
6. Make a square opening in the wrap film covering the head by
cutting it with the edge of a syringe needle, and cover the open-
ing with a drop of saline (~100 l). Be sure that saline does not
spill over the film opening onto the reverse side. At this step, it
is important that the edge of the cut film is not sealed against
the head, but the head is in tight contact with the film.
7. Open a window on the head capsule by cutting it with a syringe
needle so as to make the brain observable in situ by micros-
copy. Remove the fat bodies with fine forceps and gently set
the air sacs aside to expose the brain, particularly its lateral
protocerebral region (see Note 8). Special attention must be
paid not to damage the air sacs.
8. Mix the two components of silicone elastomer, which will be
used to seal the head opening. Drain the saline covering the
dorsal surface of the brain using a micropipette and twisted
Kimwipe, and immediately apply a small amount of the mixed
silicone elastomer onto the cuticular opening to fill the gap
between the head and the edge of the wrap film at the window.
The silicone elastomer not only seals the cuticular opening but
also helps restrict the movement of the brain during the record-
ing of neural activities. Allow the fly to rest for ~5 min, letting
it recover from the surgery.
9. Prepare a stimulant fly. Cut the head and thorax from a
cold-anesthetized fly with fine scissors. Insert a pulled glass
capillary into the abdomen from the cut end, and then seal it
with UV glue.

3.3 Recording of 1. Search for raw YFP fluorescence originating from the lateral
Neural Activities protocerebrum in the brain of the fly preparation mounted on
the microscope stage. To minimize photo-bleaching of YFP in
268 Daisuke Yamamoto et al.

this process, the strength of excitation light should be as weak

as possible.
2. Set the position of the treadmill as guided by lateral images of
the fly obtained by an infrared camera. Make sure that the fly
lands gently on the Styrofoam ball and that the legs of both
sides evenly contribute to the walking motion on the ball (see
Note 9).
3. Mount the stimulus fly on a holder. The stimulus fly will be
brought into close proximity to a foreleg of the test fly using a
micromanipulator under dim red light illumination.
4. Choose the recording site (the region of interest: ROI) from
which real-time monitoring of the fluorescent signals will be
done. Set the aperture of the excitation light in an appropriate
range, with which the cuticular region around the head open-
ing will not be illuminated (see Note 10). This helps improve
the contrast by reducing the scattered light. Choose an appro-
priate ND filter and binning and gain values for the CCD cam-
era such that exposure of the fly preparation for ~200 ms will
provide sufficient fluorescent signals in the ROI (i.e., over
1,000 signal values using a 12-bit camera) for both the YFP
and CFP channels.
5. As a tester male, choose one that is not hyperactive. In the case
that the tester male candidates persist in running or grooming,
wait ~5 min to allow them to calm down (see Note 11).
6. Start the recordings. Each recording session takes ~20 s when
the image acquisition rate is 510 Hz. Briefly touch one of the
tester fly forelegs with the stimulant fly by advancing the holder
mounting the latter and then quickly withdrawing it by manual
operation of the micromanipulator while monitoring the prep-
aration in real time with the infrared camera (see Note 12).
7. A transient increase in the YFP signal and decrease in the CFP
signal will be recorded upon touching the foreleg of the tester
fly with a stimulant fly in successful trials. A good preparation
will tolerate up to ten stimulation trials that together require
~1 h to complete.

3.4 Data Processing 1. Define the time of stimulus onset by detecting the moment of
the first contact of the stimulant fly with a foreleg of the tester
fly on the video record captured by the infrared camera. The
timestamp on the video frame in which the first contact is
recorded defines the time of stimulus onset (see Note 13).
2. Register the stack of images, using ImageJ, to cancel out any
undesirable drifts in the position of the recording site due to
inevitable brain movements.
3. Filter the images to reduce noise with a Gaussian filter (radius:
12).
 Insect Pheromone Behavior: Fruit Fly 269

4. Specify the ROI with YC2.1 expression as well as the

background brain area without YC2.1 expression. Both are
defined by circles of 10 pixels in diameter.
5. Subtract the fluorescent value at the background area from the
value at the ROI for YFP or CFP in the respective channel to
yield the crude signal value.
6. Calculate the ratio YFP/CFP frame by frame.
7. Calculate the average of the signal values for 5 frames prior to
the stimulus onset for YFP, CFP, and YFP/CFP to determine
the baseline values (F0 and R0 for the fluorescent outputs and
their ratio, respectively). Determine the signal changes (dF or
dR) by subtracting F0 or R0 from the raw signal values and
calculate dF/F0 or dR/R0 (%) for use in data quantification.
8. In the data quantification, determine the peak response value
(dR/R0) by averaging the maximal value and the values at its
two flanking frames. The peak value corresponds to the largest
dR/d0 within 3 min following the stimulus onset.

4 Notes

1. Increasing the copy number of transgenes encoding yellow

cameleon or GAL4 would improve the signal-to-noise ratio in
Ca2+-imaging of neural activities, as it reduces the baseline flu-
orescence. Aged flies may have accumulated a larger amount of
Ca2+ indicator proteins, which could enhance the signals.
2. Motorizing the diachronic filter exchange and the shutter con-
trol for excitation light will secure the preparation by reducing
the need for manual operation under dark conditions. A cus-
tomized microscopic stage with a larger working space under
the objective lens facilitates functional arrangements of the fly
preparation and instruments. A vibration-free table will be
of benefit for reducing artifacts due to movements.
3. A CCD camera having higher sensitivity and more pixels with
binning functionality is preferable. The dynamic range should
exceed 12 bits.
4. Image processing requires a function to collect signals from
the ROI in multichannel image stacks. Spatial filtrations to
reduce noises or movement artifacts as well as image registra-
tion further assist in the analysis. ImageJ, for example, offers a
free and expandable environment for batch image processing.
5. The success of the experiments will depend heavily on the state
of the flies. Discard flies whose legs are insufficiently clean,
because such flies tend to stick to grooming and show poor
responses to sexual stimuli. Tester males need to be reared sin-
gly, but the food substrate of a vial containing a single male
270 Daisuke Yamamoto et al.

easily becomes sticky due to the breeding of bacteria. This can

be a source of deposits on fly legs. Frequent transfer of flies to
new vials (e.g., every 12 days) makes them happier. Addition
of antibiotics to food is a possible solution when the sticky leg
problem is serious. Another factor that increases grooming is
the loss of leg contact with the substratethe Styrofoam ball
in this casewhich results in increased defecation and excre-
ment deposits on the legs. It is therefore important to make
the fly hold to the ball throughout the experiment.
6. Minimize the time used to place the fly under cold anesthesia
(~2 min).
7. In making the tethered male preparation, the fly needs to be
immobilized with UV glue. Reduce the amount of glue as
much as possible. A thin tungsten wire is a convenient tool for
handling a small amount of glue. The UV flash should be short
(~5 s). After glue application, adjust the fly position quickly
and cure the glue. Keep the polyethylene film tense while glu-
ing it to the coverslip hole so that an opening in the film can
be made easily.
8. In exposing the brain, special caution must be paid not to
damage the air sacs. Impaired air supply can be vital.
9. Make sure that the flies in the store chamber are keeping their
legs on the Styrofoam ball. If any fly has dropped the ball, pick
it up and give it back to him with forceps.
10. To avoid photo-bleaching of the Ca2+ sensor, do not illuminate
the brain-exposed preparation except when necessary.
11. After handling each fly, clean the scissors and forceps with a
Kimwipe moistened with ethanol, which helps avoid any
possible contamination of cuticular hydrocarbons in subse-
quent experiments.
12. In touching a leg of the tester fly for stimulation, we prefer to
move the stimulant fly on the manipulator by manual rather
than motorized operation in order to attain delicate control of
the movement.
13. In analyzing the data, choose appropriate parameters of
response (e.g., the peak amplitude, time course, and/or inte-
grals) depending on the purpose of the experiment.

Acknowledgements

This work was supported by MEXT grants (23220007, 1802012,

and 24113502 to D.Y.; 21770074 and 24570082 to M.K.; and
24770065 to S.K.), grant from the Strategic Japanese-French
 Insect Pheromone Behavior: Fruit Fly 271

Cooperative Program by JST to D.Y., a grant from the Tohoku

Neuroscience GCOE program to D.Y., and a Life Science Grant
from the Takeda Science Foundation to D.Y. We thank Hiromi
Sato for secretarial assistance.

References
1. Kitamoto T (2001) Conditional modification 13. Waddell S, Armstrong JD, Kitamoto T, Kaiser
of behavior in Drosophila by targeted expres- K, Quinn WG (2000) The amnesiac gene
sion of a temperature-sensitive shibire allele in product is expressed in two neurons in the
defined neurons. J Neurobiol 47:8192 Drosophila brain that are critical for memory.
2. Lima SQ, Miesenbck G (2005) Remote con- Cell 103:805813
trol of behavior through genetically targeted 14. Ruta V, Datta SR, Vasconcelos ML, Freeland J,
photostimulation of neurons. Cell Looger LL, Axel R (2010) A dimorphic phero-
121:141152 mone circuit in Drosophila from sensory input
3. Luo L, Callaway EM, Svoboda K (2008) to descending output. Nature 468:686690
Genetic dissection of neural circuits. Neuron 15. Lee T, Luo L (1999) Mosaic analysis with a
57:634660 repressible cell marker for studies of gene func-
4. Venken KJT, Simpson JH, Bellen HJ (2011) tion in neuronal morphogenesis. Neuron
Genetic manipulation of genes and cells in the 22:451461
Nervous System of the Fruit fly. Neuron 16. Kimura K, Hachiya T, Koganezawa M, Tazawa
72:202230 T, Yamamoto D (2008) Fruitless and
5. Fotowat H, Fayyazuddin A, Bellen HJ, Doublesex coordinate to generate male-specific
Gabbiani F (2009) A novel neuronal pathway neurons that can initiate courtship. Neuron
for visually guided escape in Drosophila melano- 59:759769
gaster. J Neurophysiol 102:875885 17. Kohatsu S, Koganezawa M, Yamamoto D
6. Gong Z, Liu J, Guo C, Zhou Y, Teng Y, Liu L (2011) Female contact activates male-specific
(2010) Two pairs of neurons in the central interneurons that trigger stereotypic courtship
brain control Drosophila innate light prefer- behavior in Drosophila. Neuron 69:498508
ence. Science 330:499502 18. Rosenzweig M, Brennan KM, Tayler TD,
7. Chiappe ME, Seelig JD, Reiser MB, Jayaraman Phelps PO, Patapoutian A, Garrity PA (2005)
V (2010) Walking modulates speed sensitivity The Drosophila ortholog of vertebrate TRPA1
in Drosophila motion vision. Curr Biol regulates thermotaxis. Genes Dev 19:419424
20:14701475 19. Boyden ES, Zhang F, Bamberg E, Nagel G,
8. Reiff DF, Plett J, Mank M, Griesbeck O, Borst Deisseroth K (2005) Millisecond-timescale,
A (2010) Visualizing retinotopic half-wave rec- genetically targeted optical control of neural
tified input to the motion detection circuitry of activity. Nat Neurosci 8:12631268
Drosophila. Nat Neurosci 13:973978 20. Chow BY, Han X, Dobry AS, Qian X, Chuong
9. Cheng LE, Song W, Looger LL, Jan LY, Jan AS, Li M, Henninger MA, Belfort GM, Lin Y,
YN (2010) The role of the TRP channel Monahan PE, Boyden ES (2010) High-
NompC in Drosophila larval and adult locomo- performance genetically targetable optical neu-
tion. Neuron 67:373380 ral silencing by light-driven proton pumps.
10. Maimon G, Straw AD, Dickinson MH (2010) Nature 463:98102
Active flight increases the gain of visual motion 21. Sweeney ST, Broadie K, Keane J, Niemann H,
processing in Drosophila. Nat Neurosci OKane CJ (1995) Targeted expression of teta-
13:393399 nus toxin light chain in Drosophila specifically
11. Claridge-Chang A, Roorda RD, Vrontou E, eliminates synaptic transmission and causes
Sjulson L, Li H, Hirsh J, Miesenbck G (2009) behavioral defects. Neuron 14:341351
Writing memories with light-addressable rein- 22. Fan GY, Fujisaki H, Miyawaki A, Tsay RK,
forcement circuitry. Cell 139:405415 Tsien RY, Ellisman MH (1999) Video-rate
12. Aso Y, Siwanowicz I, Brcker L, Ito K, scanning two-photon excitation fluorescence
Kitamoto T, Tanimoto H (2010) Specific microscopy and ratio imaging with cameleons.
dopaminergic neurons for the formation of Biophys J 76:24122420
labile aversive memory. Curr Biol 20: 23. Kimura K, Ote M, Tazawa T, Yamamoto D
14451451 (2005) Fruitless specifies sexually dimorphic
272 Daisuke Yamamoto et al.

neural circuitry in the Drosophila brain. Nature range of fluorescent indicators for Ca2+ by
438:229233 circularly permuted yellow fluorescent
24. Fiala A, Spall T (2003) In vivo calcium imaging proteins. Proc Natl Acad Sci U S A 101:
of brain activity in Drosophila by transgenic 1055410559
cameleon expression. Sci STKE 2003:PL6 26. Abramoff MD, Magelhaes PJ, Ram SJ (2004)
25. Nagai T, Yamada S, Tominaga T, Ichikawa Image processing with ImageJ. Biophoton Int
M, Miyawaki A (2004) Expanded dynamic 11:3642
 Chapter 20

Quantitative Assessment of Pheromone-Induced

Dauer Formation in Caenorhabditis elegans
Scott J. Neal, Kyuhyung Kim, and Piali Sengupta

Abstract
Environmental conditions experienced during early larval stages dictate the developmental trajectory of
the nematode C. elegans. Favorable conditions such as low population density, abundant food, and lower
temperatures allow reproductive growth, while stressful conditions promote entry of second-stage (L2)
larvae into the alternate dauer developmental stage. Population density is signaled by the concentration
and composition of a complex mixture of small molecules that is produced by all stages of animals, and is
collectively referred to as dauer pheromone; pheromone concentration is a major trigger for dauer forma-
tion. Here, we describe a quantitative dauer formation assay that provides a measure of the potency of
single or mixtures of pheromone components in regulating this critical developmental decision.

Key words Dauer larva, Ascaroside, C. elegans, Pheromone

1 Introduction

C. elegans can develop via one of the two alternative developmental

pathways [1]. This developmental decision is regulated by environ-
mental cues experienced during early larval stages. Exposure of L1
larvae to overcrowding, high temperatures, and low food abun-
dance causes them to develop as pre-dauer L2d larvae; continued
stressful conditions result in irreversible commitment to the long-
lived and stress-resistant alternative third larval stage called the
dauer stage [25]. When conditions improve, dauer larvae molt
and resume reproductive growth [6]. Conversely, favorable condi-
tions promote continuous reproductive growth. Genetic analyses
of mutants that fail to form dauers under stressful conditions, or
enter into the dauer stage regardless of environmental cues, have
identified the signaling pathways underlying the dauer develop-
mental decision [713]. In brief, environmental cues sensed by
sensory neurons regulate TGF- and insulin signaling. These sig-
nals in turn regulate the production of steroid hormones and sig-
naling via the DAF-12 nuclear hormone receptor. Hormone-bound

Kazushige Touhara (ed.), Pheromone Signaling: Methods and Protocols, Methods in Molecular Biology, vol. 1068,
DOI 10.1007/978-1-62703-619-1_20, Springer Science+Business Media, LLC 2013

273
274 Scott J. Neal et al.

DAF-12 promotes reproductive growth, whereas unliganded

DAF-12 promotes dauer arrest (reviewed in [11, 13]).
A major trigger for dauer formation is a complex mixture of
small molecules called ascarosides that are produced by all stages of
animals, and whose concentration serves as a measure of local pop-
ulation density [2, 4, 1420]. The role of secreted/excreted mol-
ecules in regulating dauer formation was demonstrated by showing
that conditioned medium from wild-type worms was sufficient to
induce dauer formation in well-fed larvae [2]. This chapter
describes an assay in which the ability of partially purified condi-
tioned medium, called crude pheromone (see Chapter 6), or chem-
ically synthesized pure ascarosides, to induce dauer formation is
assessed. Since food levels and ambient temperature significantly
modulate the efficacy of pheromone-induced dauer formation [3,
4, 21], this assay is designed to carefully control for environmental
variables so as to provide a reproducible quantification of the effi-
ciency of single or mixtures of ascarosides in inducing dauer entry.
A variation of this assay has been described [22].

2 Materials

Prepare all solutions using Milli-Q or equivalent (18 M) ultra-

pure water. Use reagent-grade ethanol (99.5 %) for reconstitut-
ing and diluting synthetic pheromones. All other materials should
be of molecular biology grade or better. This protocol assumes
basic knowledge of standard C. elegans husbandry.
1. Dissecting microscope, heat block, 20 C incubator, 25 C
incubator, water bath.
2. E. coli strain OP50, grown on Luria-Bertani (LB) medium.
3. 1 M potassium phosphate buffer pH 6.0: ~1:3 1 M K2HPO4:1 M
KH2PO4, titrate with 1 M K2HPO4 to desired pH.
4. S-basal buffer: 0.1 M NaCl, 50 mM potassium phosphate
buffer pH 6.0.
5. 5 mg/mL cholesterol in 95 % ethanol.
6. Dauer formation assay agar: 50 mM NaCl, 1.7 % Noble agar
(see Note 1), 1 mM CaCl2, 1 mM MgSO4, 25 mM potassium
phosphate buffer pH 6.0, 5 g/mL cholesterol.
7. Crude and/or synthetic pheromone (see Chapter 6).

3 Methods

3.1 Overview of This protocol involves multiple steps that are carried out on several
Assay Timeline independent days. In this protocol, it is important to be considerate
of sources of biological variation which might compromise assay
 C. elegans Dauer Formation Assay 275

results. For instance, there is increasing evidence of transgenera-

tional inheritance of epigenetic information which may influence
the results of behavioral or developmental assays [2326]. It is thus
important to control the developmental history of the parent ani-
mals that will contribute the embryos for the assay. Typically,
worms should be maintained under well-fed conditions at 20 C
(see Note 2) for at least three generations prior to the assay. An
overall summary of the protocol is shown in Fig. 1.
1. On day 1, pick five mid-L4-stage larvae to a fresh OP50-seeded
nematode growth medium (NGM) plate around midday.
Culture the worms at 20 C. This parental plate should have at
least 100 fairly synchronous well-fed L4 larvae on day 4.
2. On day 4, pick the required number of mid- to late-L4 larvae
to fresh 60 mm OP50-seeded NGM plates; on day 5, five
young adults will be required for each assay plate. Care must
be taken not to induce crowding; typically no more than 70 L4
larvae should be subcultured to the same plate in order to min-
imize their exposure to endogenous pheromone. Culture these
animals overnight at 20 C in preparation for the assay.
3. Also on day 4, prepare the dauer agar assay plates and inoculate
the OP50 culture for the heat-killed food source
(Subheadings 3.3 and 3.4, respectively).
4. On day 5, apply the heat-killed OP50 cultures to the assay
plates. Transfer five young adult animals to each assay plate and
collect eggs (Subheading 3.5).
5. Incubate the assay plates at 25 C and quantify dauer forma-
tion on day 8 or 9 (Subheading 3.6).

3.2 Preparation The methodology for preparation of crude pheromone is addressed

of Crude or Synthetic elsewhere in this volume (see Chapter 6). Purification and synthesis
Pheromone of individual ascarosides have been described [1420]. Crude or
synthetic pheromones can be stably stored dry at 20 C with
some exceptions. For instance, PABA-C7 (ascr#8) [18] is less sta-
ble and is, therefore, stored at 80 C. For use, dilutions of crude
pheromone are made in water or 50:50 water:ethanol and may
either be mixed with the molten assay agar or spread on the surface
of the assay plate. Synthetic pheromones may be reconstituted at
low millimolar concentrations in either dimethyl sulfoxide or
ethanol. These solutions are stored at 20 C and should be diluted
in the original solvent when testing multiple concentrations in a
dauer formation assay so as to keep the concentration of solvent
constant in all assay plates. Our standard practice is to use ethanol
as the solvent though in limited testing we have not observed any
differences between the two solvents in this assay.
276 Scott J. Neal et al.

3X

Preparation
Maintain well-fed worms for at least 3 generations

60 mm NGM plate
Live OP50
20C

Day 1
Transfer 5 L4 larvae to a fresh NGM plate

60 mm NGM plate
Live OP50
20C

(maximum
70 worms)
Day 4
Transfer 5 L4 larvae per assay to a fresh NGM plate
Dilute pheromone (3.2)
Make Dauer Agar Assay Plates (3.3)
Inoculate OP50 culture (3.4)
60 mm NGM plate
Live OP50
20C

Day 5
35 mm Dauer Agar Assay Plate
Heat-Killed OP50
Room Temperature
Apply heat-killed OP50 to assay plates (3.5)
3-6 hours Transfer 5 young adults to each assay plate (3.5)
Remove egg-layers and count eggs (3.5)
Incubate at 25C for 72-84 hrs (3.5)

Day 8-9
Count dauers and non-dauers (3.6)

Fig. 1 Overview of the timeline and individual steps in the described protocol. Numbers in parentheses refer
to the respective sections in Subheading 3 as described in the text. Green and red bacterial spots indicate live
and heat-killed OP50 bacteria, respectively
 C. elegans Dauer Formation Assay 277

3.3 Preparation of Use sterile technique throughout this procedure. Be sure to prepare
Dauer Agar Assay sufficient agar for all assay plates while allowing for some loss due
Plates to evaporation and pipetting error. 3 mL of dauer assay agar is
required per 35 mm assay plate (see Note 3).
1. In a clean, autoclave-safe bottle add (per 100 mL) 0.3 g NaCl,
1.7 g Noble agar, and an autoclave-safe stir bar. Add 100 mL
of MilliQ H2O and autoclave for 20 min on a liquid cycle.
Tighten the cap and cool the sterilized solution to 60 C in a
water bath.
2. Place the cooled media on a stir plate and add (per 100 mL)
100 L 1 M CaCl2, 100 L 1 M MgSO4, 2.5 mL 1 M potas-
sium phosphate buffer pH 6.0, and 100 L 5 mg/mL choles-
terol in 95 % ethanol. Stirring should be sufficient to mix in the
cholesterol while avoiding bubbles or frothing.
3. Pheromone solutions should be prepared immediately before
plate pouring, and for consistency should be prepared as a mas-
ter mix. For each plate combine 6 L of appropriately diluted
pheromone in solvent (or solvent alone) with 94 L of sterile
MilliQ water. Array the empty 35 mm assay plates on a flat
surface (see Note 4) and transfer 100 L of the above phero-
mone master mix to the center of the empty plate.
4. Dispense 3 mL of assay agar into the pheromone-containing
plates (see Note 5), and allow them to set overnight on the
bench at room temperature (see Note 6). Ensure that there are
no bubbles in the agar.

3.4 Preparation of Although pheromone-induced dauer formation can be signifi-

Heat-Killed OP50 cantly suppressed by the presence of bacterial food, food is neces-
sary to bypass the starvation-induced arrest at the L1 larval stage
[27]. The typical food source used in most C. elegans laboratories
is the E. coli strain OP50. Fresh cultures of OP50 may continue to
grow on standard NGM plates, and this self-renewing food source
suppresses dauer formation; therefore, either heat-killed or
streptomycin-killed bacteria are used. For best results, OP50
should be freshly streaked onto an LB plate from a glycerol stock
of the strain.
1. Inoculate a single OP50 colony into 35 mL of LB broth in a
50 mL conical tube. With the lid slightly loosened, grow the
culture to stationary phase overnight at 37 C with 225 rpm
shaking.
2. Pre-weigh an appropriate number (generally 24) of 1.5 mL
tubes to establish their tare weights. Transfer 1.45 mL of over-
night culture into each tube and spin for 1 min at 8,000 g.
Discard the supernatant and transfer an additional 1.45 mL of
overnight culture to the tube. Spin for 1 min at 8,000 g and
278 Scott J. Neal et al.

aspirate the supernatant carefully. Spin the pellet for 1 min at

maximum speed (13,000 g) and carefully aspirate any super-
natant with a pipette. Spin the pellet for 2 min at maximum
speed and remove any final traces of supernatant with a P10 or
a P20 pipetman.
3. Weight the tube and determine the pellet weight; the expected
mass is between 8 and 10 mg. Fully resuspend the pellet to a
concentration of 8 mg/mL in S-basal + 5 g/mL cholesterol.
4. Heat-kill the bacteria by incubating in a heat block at 95 C for
30 min with vortexing at 10-min intervals. Cool by placing at
4 C for at least 30 min before use.
5. This reagent is best prepared fresh on the day of or the day
before the assay (store overnight at 4 C), but may also be kept
for at most 1 week at 4 C. Always vortex the bacterial suspen-
sion before use.

3.5 Setting Up 1. On the day of the assay, pipette 20 L of fresh 8 mg/mL heat-killed
the Assay OP50 (Subheading 3.4) onto the center of the assay agar. Care
should be taken at all stages of this protocol not to pierce the
surface of the assay agar to prevent animals from burrowing
(see Note 7).
2. Briefly dry the food spot by incubating the plates in the fume
hood; do not offset the lids, even if they are not vented (see
Note 8). The plates are ready once the spot has completely
dried and when there is no condensation on the lid. It generally
takes 2030 min for the food spot to dry fully. If a fume hood
is not available the plates will dry on the bench but will require
more time to do so.
3. Carefully transfer (see Note 9) five young adult hermaphro-
dites to each assay plate to begin the egg laying process (see
Note 10). It is essential that only minimal live bacteria are
transferred to the assay plates (see Note 11).
4. Allow the worms to lay approximately 75 eggs on each plate at
room temperature. About 6585 eggs are sufficient, although
more consistent results are achieved with 75 eggs. Wild-type
worms that have been correctly staged should lay sufficient
eggs in about 3 h. It is useful to check the plates after the first
hour and to return worms that have moved off the food back
to the bacterial spot. When sufficient eggs are present on the
plate, remove the adult worms and discard them. Be certain
that all five adults are accounted for and removed, even those
that may have begun to desiccate on the plate walls and those
that may have burrowed (see Note 7).
 C. elegans Dauer Formation Assay 279

5. Count the number of eggs on the plate and remove any eggs
in excess of 85 (see Note 12); do not remove any of the heat-
killed OP50 in this process.
6. Invert the plates in a loosely sealed plastic container and place
them in a 25 C incubator (see Note 13) for 7284 h.

3.6 Scoring The scoring of dauers can be subjective to untrained observers.

the Assay Many physical hallmarks of true dauer larvae are not readily visible
under a standard dissection microscope, and the use of a com-
pound microscope is not compatible with large-scale experiments.
Since true dauer larvae are resistant to 1 % sodium dodecyl sulfate
(SDS) [1, 28], SDS selection can be used to identify dauers.
However, it is difficult to accurately count all dauer larvae present
on SDS-flooded plates, and we do not routinely perform this pro-
cedure. With some practice, the size, movement, body posture,
and gut refraction phenotypes of dauer larvae [1, 10, 12, 29] can
be readily used to discriminate them from non-dauer animals. In
reporting the results from any assay it is important to state whether
worms exhibiting pre-dauer arrest or characteristics of partial dau-
ers are being excluded. Under the described conditions, it is atypi-
cal to observe either pre-dauer arrest or partial dauers in wild-type
worms but the prevalence of such animals may be increased in
some mutant backgrounds.
1. Remove the plates from the incubator and inspect them using
a dissecting microscope. Provided with sufficient heat-killed
food, wild-type worms will develop from embryo to late L4
stage within 72 h at the assay temperature of 25 C. In a prop-
erly executed assay, 010 % of wild-type worms should arrest as
dauers under negative control or no-pheromone conditions.
When characterizing dauer formation in non-wild-type worms,
it is important to allow sufficient time for their development,
which might proceed at a slower rate than in wild-type worms.
Therefore, the assay is routinely scored 84 h after egg laying.
This additional time has no effect on the results for wild-type
worms since dauers remain arrested and animals that have
bypassed the dauer pathway remain easily distinguishable. If
scoring at 84 h, ignore any next-generation L1 larvae that may
have hatched.
2. Count the number of dauer larvae and the number of young
adult worms on the plate. The counting process is simplified
by drawing a grid on the base of the plate. Note that some
worms may be burrowed into the agar and that dauers may be
present on the sidewalls of the plate.
3. Compare the number of worms counted to the number of
eggs recorded on day 5. In wild-type worms, at least 95 % of
eggs should hatch and develop normally. Discard data from
280 Scott J. Neal et al.

plates with significant contamination and from those with too

many (>85) or too few (<65) worms.
4. The number of dauers is presented as a proportion of the total
number of worms (dauers and adults) on the plate, and is cor-
related with the log of the concentration of pheromone in the
assay plate. A complete dataset should consist of replicate data
from at least three independent experiments performed on dif-
ferent days.

4 Notes

1. Noble agar increases the rate of dauer formation by about 10 %

and gives more consistent results. The concentration of agar
may be increased up to 2 % to suppress burrowing behavior (see
Note 7), although this may affect the baseline rate of dauer
formation.
2. Culturing animals at 20 C is optimal for growth and fecun-
dity, especially for mutant strains with compromised health.
We have not observed any quantitative differences in wild-type
dauer formation rates when the parental worms are cultured at
20 C versus 25 C.
3. Depending on the actual dimensions of the Petri plates used, it
might be necessary to increase the volume of assay agar in
order to fully cover the bottom surface of the plate. To achieve
consistent results the agar must evenly cover the bottom sur-
face of the plate. However, if the agar is too thin it may dry
excessively or crack over the course of the assay. Remember to
adjust the pheromone volume to maintain the desired
concentration.
4. It is useful to array the assay plates on a small flat tray to facili-
tate their movement from the bench to the fume hood when it
is time to dry the food spots.
5. When dispensing the molten agar on top of the 100 L of
pheromone in the assay plate it is not necessary to stir or swirl
the media. The mixing that occurs during dispensation is suf-
ficient to distribute the pheromone throughout the medium.
Swirling the cooling medium tends to leave a meniscus of dried
agar around the plate, allowing worms to crawl up the side-
walls and eventually desiccate.
6. Plates should be prepared by mid-afternoon and be given
1618 h to dry completely on the bench. There should be no
condensation on the lid when the assay is started. If the assay
plates are too humid the egg-laying worms tend to move off
the food and crawl up the wall of the plate and/or burrow (see
Note 7).
 C. elegans Dauer Formation Assay 281

7. If an egg-laying worm burrows into the agar it does not have

access to the food on the surface and will typically lay fewer
eggs. Eggs laid in the agar are also difficult to see and count
and will develop abnormally with respect to this assay. Extensive
burrowing on a plate also typically coincides with reduced rates
of dauer formation. Occasionally an egg-laying worm will bur-
row into the agar and must be removed prior to counting the
eggs. Briefly vortexing or tapping the plate will sometimes
draw the worm to the surface. The worm should only be
extracted/dug from the plate as a last resort since the damage
to the agar will encourage the larvae to burrow as well.
8. It is not advisable to offset the lids of the assay plates to expe-
dite drying. This may lead to dust particles landing on the
plate and increases in fungal contamination. Overdrying of the
plates may lead to cracking of the agar and will cause the worms
to burrow (see Note 7).
9. The shape of the tip of the platinum wire used to transfer
worms is important in this assayit must not scratch or gouge
the assay agar surface during transfer and it must also carry
over minimal live bacteria (see Note 11). A smoothed spatula-
shaped wire tip is useful since this allows worms to be scooped
from underneath without the use of bacterial glue. This is
also useful when the egg-laying worms must be removed from
the assay plate without removing significant amounts of the
heat-killed bacteria.
10. The use of five gravid hermaphrodites in this assay is intended
to allow the correct number of embryos to be deposited on a
plate in a reasonable time frame in order to have a growth-
synchronized brood. Using varying number of egg-layers will
affect the assay in two ways: it will affect the amount of food
that is consumed prior to the hatch of the brood and it will
alter the concentration of endogenous pheromone. Also note
that extended egg-laying sessions may result in the earliest-laid
embryos hatching before the eggs are counted.
11. Live OP50 is more attractive to egg-laying worms than the
heat-killed OP50. Eggs will tend to be tightly clustered in
patches of live food whereas they are normally more randomly
distributed on plates featuring only heat-killed food. Live food
is a potent suppressor of dauer formation in this assay and will
negatively impact results.
12. Care should be taken to only count fertilized eggs when assess-
ing the number of embryos on each assay plate. One-day-old
adult wild-type worms will lay few if any unfertilized oocytes
whereas some mutant strains will lay many. Since population
density is an important factor in this assay as is the concentra-
tion of available food, it is important that the correct number
282 Scott J. Neal et al.

of L1 larvae hatch on each assay plate. Inspect the full surface

of the plate, including around the edges of the agar, when
counting fertilized eggs.
13. Temperature significantly modulates dauer formation and
must be carefully controlled. A sensitive thermometer should
be used to calibrate the 25 C incubator and in addition, the
placement of assay plates within the incubator should be
consistent. Since humidity affects dauer formation to some
extent, care should also be taken to house the plates in similar
containers and to not overly stack the assay plates within the
container. Natural convection incubators are used as opposed
to forced-air incubators.

Acknowledgments

Funding for this work was provided by the National Science

Foundation (IOS 0842452 and IOS 1256488 to P.S.), the Human
Frontiers Science Program (RGY0042/2010 to P.S.), and the
DGIST Convergence Science Center Program of the Ministry of
Education, Science and Technology (11-BD-04 to K.K.). S.J.N.
was supported by a postgraduate scholarship (PGS-D3) from the
Natural Sciences and Engineering Research Council of Canada and
by the Brandeis National Committee.

References

1. Cassada RC, Russell RL (1975) The dauer 7. Riddle DL, Swanson MM, Albert PS (1981)
larva, a post-embryonic developmental variant Interacting genes in nematode dauer larva for-
of the nematode Caenorhabditis elegans. Dev mation. Nature 290:668671
Biol 46:326342 8. Thomas JH, Birnby DA, Vowels JJ (1993)
2. Golden JW, Riddle DL (1982) A pheromone Evidence for parallel processing of sensory
influences larval development in the nematode information controlling dauer formation in
Caenorhabditis elegans. Science 218:578580 Caenorhabditis elegans. Genetics 134:
3. Golden JW, Riddle DL (1984) The 11051117
Caenorhabditis elegans dauer larva: develop- 9. Vowels JJ, Thomas JH (1992) Genetic analysis
mental effects of pheromone, food, and tem- of chemosensory control of dauer formation in
perature. Dev Biol 102:368378 Caenorhabditis elegans. Genetics 130:
4. Golden JW, Riddle DL (1984) A 105123
Caenorhabditis elegans dauer-inducing phero- 10. Riddle DL, Albert PS (1997) Genetic and
mone and an antagonistic component of the environmental regulation of dauer larva devel-
food supply. J Chem Ecol 10:12651280 opment. In: Riddle DS, Blumenthal T, Meyer
5. Schaedel ON, Gerisch B, Antebi A, Sternberg BJ, Priess JR (eds) C. elegans II. Cold Spring
PW (2012) Hormonal signal amplification Harbor Press, Cold Spring Harbor, Plainview,
mediates environmental conditions during NY, pp 739768
development and controls an irreversible com- 11. Hu PJ (2007) Dauer. WormBook:119.
mitment to adulthood. PLoS Biol http://www.wormbook.org/chapters/www_
10(4):e1001306 dauer/dauer.html
6. Klass M, Hirsh D (1976) Non-ageing devel- 12. Albert PS, Riddle DL (1988) Mutants of
opmental variant of Caenorhabditis elegans. Caenorhabditis elegans that form dauer-like
Nature 260:523525 larvae. Dev Biol 126:270293
 C. elegans Dauer Formation Assay 283

13. Fielenbach N, Antebi A (2008) C. elegans 21. Ailion M, Thomas JH (2000) Dauer forma-
dauer formation and the molecular basis of tion induced by high temperatures in
plasticity. Genes Dev 22:21492165 Caenorhabditis elegans. Genetics 156:
14. Butcher RA, Fujita M, Schroeder FC, Clardy J 10471067
(2007) Small molecule signaling of dauer for- 22. Lee J, Kim KY, Joo HJ, Kim H, Jeong PY, Paik
mation in C. elegans. Nat Chem Biol YK (2011) Methods for evaluating the
3:420422 Caenorhabditis elegans dauer state: standard
15. Butcher RA, Ragains JR, Kim E, Clardy J dauer-formation assay using synthetic daumo-
(2008) A potent dauer pheromone compo- nes and proteomic analysis of O-GlcNAc mod-
nent in Caenorhabditis elegans that acts syner- ifications. Methods Cell Biol 106:445460
gistically with other components. Proc Natl 23. Gu SG, Pak J, Guang S, Maniar JM, Kennedy S,
Acad Sci U S A 105:1428814292 Fire A (2012) Amplification of siRNA in
16. Butcher RA, Ragains JR, Clardy J (2009) An Caenorhabditis elegans generates a transgenera-
indole-containing dauer pheromone compo- tional sequence-targeted histone H3 lysine 9
nent with unusual dauer inhibitory activity at methylation footprint. Nat Genet 44:157164
higher concentrations. Org Lett 11:31003103 24. Rechavi O, Minevich G, Hobert O (2011)
17. Srinivasan J, von Reuss SH, Bose N, Zaslaver A, Transgenerational inheritance of an acquired
Mahanti P, Ho MC, ODoherty OG, Edison small RNA-based antiviral response in C. ele-
AS, Sternberg PW, Schroeder FC (2012) A gans. Cell 147:12481256
modular library of small molecule signals regu- 25. Greer EL, Maures TJ, Ucar D, Hauswirth AG,
lates social behaviors in Caenorhabditis elegans. Mancini E, Lim JP, Benayoun BA, Shi Y,
PLoS Biol 10:e1001237 Brunet A (2011) Transgenerational epigenetic
18. Pungaliya C, Srinivasan J, Fox BW, Malik RU, inheritance of longevity in Caenorhabditis ele-
Ludewig AH, Sternberg PW, Schroeder FC gans. Nature 479:365371
(2009) A shortcut to identifying small mole- 26. Katz DJ, Edwards TM, Reinke V, Kelly WG
cule signals that regulate behavior and devel- (2009) A C. elegans LSD1 demethylase contrib-
opment in Caenorhabditis elegans. Proc Natl utes to germline immortality by reprogram-
Acad Sci U S A 106:77087713 ming epigenetic memory. Cell 137:308320
19. Jeong PY, Jung M, Yim YH, Kim H, Park M, 27. Johnson TE, Mitchell DH, Kline S, Kemal R,
Hong E, Lee W, Kim YH, Kim K, Paik YK Foy J (1984) Arresting development arrests
(2005) Chemical structure and biological aging in the nematode Caenorhabditis elegans.
activity of the Caenorhabditis elegans dauer- Mech Ageing Dev 28:2340
inducing pheromone. Nature 433:541545 28. Swanson MM, Riddle DL (1981) Critical peri-
20. von Reuss SH, Bose N, Srinivasan J, Yim JJ, ods in the development of the Caenorhabditis
Judkins JC, Sternberg PW, Schroeder FC elegans dauer larva. Dev Biol 84:2740
(2012) Comparative metabolomics reveals 29. Popham JD, Webster JM (1979) Aspects of
biogenesis of ascarosides, a modular library of the fine structure of the dauer larva of the
small-molecule signals in C. elegans. J Am nematode Caenorhabditis elegans. Can J Zool
Chem Soc 134:18171824 57:794800
 Chapter 21

Acute Behavioral Responses to Pheromones in C. elegans

(Adult Behaviors: Attraction, Repulsion)
Heeun Jang and Cornelia I. Bargmann

Abstract
The pheromone drop test is a simple and robust behavioral assay to quantify acute avoidance of pheromones
in C. elegans, and the suppression of avoidance by attractive pheromones. In the pheromone drop test,
water-soluble C. elegans pheromones are individually applied to animals that are freely moving on a large
plate. Upon encountering a repellent, each C. elegans animal may or may not try to escape by making a
long reversal. The fraction of animals that make a long reversal response indicates the repulsiveness of a
given pheromone to a specific genotype/strain of C. elegans. Performing the drop test in the presence of
bacterial food enhances the avoidance response to pheromones. Attraction to pheromones can be assayed
by the suppression of reversals to repulsive pheromones or by the suppression of the basal reversal rate to
buffer.

Key words C. elegans, Pheromone, Ascaroside, Long reversal, Repulsion, Suppression of reversal,
Attraction

1 Introduction

The pheromone drop test is adapted from a more general drop test
for acute avoidance of water-soluble repellents [1, 2]. C. elegans
moves forward through sinusoidal locomotion and occasionally
changes direction by making a transient reversal or by making a
sharper acute turn. A long reversal, often followed by a sharp turn
of the entire body into an omega shape, is a common escape strat-
egy upon encountering repellents. In the pheromone drop test, a
population of animals is transferred to a test plate on which they
crawl freely. A volume of a few nanoliters of pheromone dissolved
in buffer is delivered to the side of each animal while it moves for-
ward. As the drop touches the animal, it surrounds the entire ani-
mal by capillary action and is sensed by the amphids, the C. elegans
sensory organs in the anterior. If the worm senses a repulsive stim-
ulus, it makes a long reversal. The presence or the absence of the
long reversal response in each animal is recorded, and the fraction

Kazushige Touhara (ed.), Pheromone Signaling: Methods and Protocols, Methods in Molecular Biology, vol. 1068,
DOI 10.1007/978-1-62703-619-1_21, Springer Science+Business Media, LLC 2013

285
286 Heeun Jang and Cornelia I. Bargmann

of animals with a long reversal response among the total number of

animals tested is calculated.
The known C. elegans pheromones belong to a family of small
molecules with an ascarylose sugar backbone, and are called asca-
rosides. Some of the identified ascarosides including ascr #3 (C9)
are repulsive to wild-type hermaphrodites in the drop test [3]. For
a negative control and basal level correction, buffer alone is also
tested. For positive controls, a high-osmolarity glycerol solution
can be used as a repellent, and isoamyl alcohol, an attractive odor,
can be used as an attractant.

2 Materials

Prepare all solutions using ultrapure water that is distilled, filter

purified with 0.2 m membrane, and autoclaved.
1. M13 buffer: 30 mM TrisHCl, pH 7.0, 100 mM NaCl,
10 mM KCl. To make a 10 M13 stock, weigh 3.634 g Tris
HCl, 5.844 g NaCl, and 745.5 mg KCl, and dissolve in water.
Adjust pH to 7.0, and make up to 100 mL with water. Filter
through a 0.2 m membrane. Store at room temperature.
Dilute 1:10 in water with pheromone immediately before use.
2. Pheromone ascaroside stock: C. elegans pheromones are not
commercially available as of this writing, but methods for puri-
fication and synthesis of individual ascarosides have been devel-
oped (see chapter by Rebecca Butcher). Dissolve each
synthesized ascaroside in ethanol to the final concentration of
10 mM. Tightly seal with Parafilm in small aliquots in glass
vials and store at 4 C.
3. Glycerol solution: Prepare 1 M glycerol. Weigh 9.209 g glyc-
erol. Dissolve in water, and make up to 100 mL with water.
Store at room temperature.
4. Nematode growth media (NGM) plates: Prepare agar, pep-
tone, 5 mg/mL cholesterol in ethanol, 1 M CaCl2, 1 M
MgSO4, 1 M KH2PO4, and 1 M K2HPO4. To make 5 mg/mL
cholesterol, add 2.5 g of cholesterol to 500 mL of ethanol, fol-
lowed by overnight incubation at 37 C. Prepare 1 M KPO4
buffer (pH 6.0) by mixing 13.2 mL of 1 M KH2PO4 with
86.8 mL of 1 M K2HPO4 in a separate sterile beaker, and use
fresh within 12 days. To make 1 L of NGM, add 1 mL of
5 mg/mL cholesterol in ethanol, 3 g NaCl, 22 g agar, and
2.5 g peptone to a 2 L Erlenmeyer flask. Add 975 mL water
and stir well. Cover flask with aluminum foil, and autoclave for
50 min. Cool flask to 52 C with stirring. Add 1 mL of 1 M
CaCl2 and 1 mL of 1 M MgSO4 and mix well. Add 25 mL of
1 M KPO4 (pH 6.0) to the flask. Using sterile procedures and
a peristaltic pump, dispense 24 mL of the NGM solution into
 Acute Behavioral Responses to Pheromones in C. elegans 287

a 90 mm diameter Petri plate, and 10 mL into a 55 mm diameter

Petri plate [4]. Make NGM plates in large quantities and store
in the cold room at 4 C.
5. OP50 culture: From a single colony, grow a culture of the
E. coli strain OP50 in 100 mL LB media at room temperature
for 34 days. Measure OD 600 and use at a density range of
OD 600 = 0.40.7. Store the bottle at 4 C for up to 2 weeks.
6. Platinum wire pick: Prepare platinum wire 0.255 mm in diam-
eter (Tritech Research) and Borosilicate Glass Pasteur Pipette
5.75 in. in length (Fisherbrand) or equivalent. Cut 23 cm of
platinum wire and insert one end into the narrow end of a glass
Pasteur pipette, and flame the glass to melt it around the plati-
num wire. Flatten a few mm at the free end by pressing the
wire between pliers, leaving a free flat wire tip.
7. Glass capillary: 10 L Glass microcapillary pipettes (Kimble) or
equivalent.
8. Filtered plastic tips: 1200 L Graduated filter tips (USA
Scientific, INC) or equivalent.
9. Unfiltered plastic tips: 1200 L Natural beveled tips (USA
Scientific, INC) or equivalent.
10. Tubing: Rubber tubing about 4060 cm in length and 1.5 mm
in inner diameter.

3 Methods

3.1 Pheromone Drop 1. Four days prior to the assay, pick three animals in the L4 larval
Test (Off Food) stage onto a new NGM plate (40 mm diameter) that was previ-
ously seeded with 100200 L E. coli OP50 culture. Seed
plates 13 days before adding worms. The progeny of these
animals will be used for the drop test.
2. Grow the animals for 4 days at 20 C (see Note 1).
3. On the day of the assay, take out unseeded NGM plates
(90 mm diameter) from 4 C. Dry at room temperature for
12 h with the lid open to remove excessive moisture. Assays
will be performed on these plates (see Note 2).
4. Remove the animal growth plates from the 20 C incubator.
The majority of the animals should have reached the young
adult stage.
5. Transfer 30 young adult animals to a new unseeded NGM
plate (55 mm diameter) at room temperature with the plati-
num wire, transferring as little food as possible (see Note 3).
6. Let the animals crawl away from any food that was transferred
with them.
288 Heeun Jang and Cornelia I. Bargmann

Fig. 1 The pheromone delivery system. The filtered tip (f) connects the upper end
of the tubing to the mouth and prevents saliva from going into the tubing. The
unfiltered tip (u) connects the lower end of the tubing to the drawn-out glass
microcapillary that contains pheromones diluted in M13 buffer (c)

7. Transfer animals to a second unseeded NGM assay plate with

the platinum wire, using no food. Animals should be food-free
(see Note 3).
8. Wait for 3060 min to precondition the worms.
9. During preconditioning, make a serial dilution of ascaroside in
M13 buffer in Eppendorf tubes (see Note 4).
10. Pull the glass capillary by hand on a flame and break it to
reduce the diameter of the tip.
11. Set up the pheromone delivery system using a pulled glass cap-
illary, a rubber tube, an unfiltered tip, and a filtered tip (see
Fig. 1). Push the broad end of the capillary into the unfiltered
tip firmly to seal the joint.
12. Place the narrow end of the capillary into the ascaroside solu-
tion. The ascaroside solution will enter the glass capillary by
capillary action.
13. Apply a droplet of ascaroside by mouth pressure through the
filtered tip to a forward moving worm. Score the positive and
negative responses. A long reversal is defined as a backward
movement that equals or is larger than the half-length of the
worm, initiated within 4 s after the stimulus touches the worm.
Most long reversals begin within a second (see Fig. 2).
14. Repeat step 13 for additional worms. Moving from one side of
the plate to the other side and testing only 20 worms per plate
help prevent testing the same animal twice.
15. The fraction reversing is (number of animals that make a long
reversal)/(number of total animals tested). A fraction revers-
ing of 1 represents complete avoidance and 0 represents no
response (see Note 5).
16. The same assay is performed with a drop of M13 buffer alone
on a similar number of animals. The fraction reversing is
obtained as above. The fraction reversing of wild-type N2 her-
maphrodite animals to buffer is typically about 0.10.2.
 Acute Behavioral Responses to Pheromones in C. elegans 289

Fig. 2 The long reversal response of C. elegans in the pheromone drop test. Time runs from left to right in the
top row and then the bottom row; total time elapsed is about 5 s. Top left panel: 10 nM of ascr #3 (C9) is
applied to the side of a wild-type N2 hermaphrodite worm moving forward in the direction of the arrow (toward
lower left). As the stimulus touches the worm (second panel), the animal initiates a long reversal toward the
upper right (arrows, third to fifth panels). Bottom right panel: Animal ends the long reversal with a sharp turn,
and changes direction of movement toward the lower right (arrow)

17. The avoidance index is the effect size of (fraction reversing to

pheromone) (fraction reversing to buffer alone). An avoid-
ance index of 1 represents complete avoidance, but is rarely
observed because of background response to buffer. Typical
repulsion value ranges from 0.3 to 0.7. A neutral response is
zero. An attraction response is represented as a negative num-
ber, but cannot easily be tested in the absence of food due to
the low basal reversal rate to buffer.
18. As a positive control, you may perform the same assay with
high-osmolarity glycerol solution in the capillary (1 M). The
fraction reversing of wild-type N2 hermaphrodite animals is
about 0.9. Therefore, the avoidance index of glycerol (with
subtraction of the buffer-alone response) in wild-type N2 her-
maphrodites is about 0.70.8.
290 Heeun Jang and Cornelia I. Bargmann

3.2 Pheromone Drop Acute pheromone avoidance is enhanced 10100-fold by the

Test (on Food) presence of bacterial food. For example, wild-type N2 hermaphro-
dite worms avoid ascr #3 (C9) at 100 nM or higher concentrations
of food, and 1 nM or higher concentration on food. The basal
avoidance response to buffer alone is also generally increased.
1. Four days prior to the assay, pick three animals in the L4 larval
stage onto a new NGM plate (40 mm diameter) that was previ-
ously seeded with 100200 L E. coli OP50 culture. Seed
plates 13 days before adding worms. The progeny of these
animals will be used for the drop test.
2. Grow the animals for 4 days at 20 C (see Note 1).
3. The day before the assay, take out unseeded NGM plates from
4 C. Dry at room temperature for 12 h with the lid open to
remove excessive moisture. Assays will be performed on these
plates.
4. Pour about 6 mL OP50 culture in LB media onto each dried
NGM plate, swirl, and remove excessive media. This will allow
you to obtain a homogenous bacterial lawn.
5. Incubate the seeded assay plates with the lid on at 37 C for
1418 h.
6. On the day of the assay, remove the assay plates from the 37 C
incubator and incubate at room temperature for 12 h to
equilibrate. Remove moisture from the lid of the plate.
7. Remove the animal growth plates from the 20 C incubator.
The majority of the animals should have reached the young
adult stage.
8. Transfer 30 young adult animals to the assay plate with the
platinum wire, using minimal food.
9. Wait for 3060 min to precondition the worms.
10. During preconditioning, make a serial dilution of ascaroside in
M13 buffer in Eppendorf tubes (see Note 4).
11. Pull the glass capillary by hand on a flame and break it to
reduce the diameter of the tip.
12. Set up the pheromone delivery system using a pulled glass
capillary, a rubber tube, an unfiltered tip, and a filtered tip (see
Fig. 1). Push the broad end of the capillary into the unfiltered
tip firmly to seal the joint.
13. Place the narrow end of the capillary into the ascaroside
solution. The ascaroside solution will enter the glass capillary
by capillary action.
14. Apply a droplet of ascaroside by mouth pressure through the
filtered tip to a forward moving worm. Score the positive and
negative responses. A long reversal is defined as a backward
movement that equals or is larger than the half-length of the
 Acute Behavioral Responses to Pheromones in C. elegans 291

worm, initiated within 4 s after the stimulus touches the worm.

Most long reversals begin within a second (see Fig. 2).
15. Repeat step 14 for additional worms. Moving from one side of
the plate to the other side and testing only 20 worms help pre-
vent testing the same animal twice.
16. The fraction reversing is (number of animals that make a long
reversal)/(number of total animals tested). A fraction revers-
ing of 1 represents complete avoidance and 0 represents no
response (see Note 5).
17. The same assay is performed with a drop of M13 buffer alone
on a similar number of animals. The fraction reversing is
obtained. The fraction reversing to buffer of wild-type N2 her-
maphrodite animals on food is typically about 0.3.
18. The avoidance index is the effect size of (fraction reversing to
pheromone) (fraction reversing to buffer only). An avoidance
index of 1 represents complete avoidance, but is rarely observed
because of background response to buffer. Typical repulsion value
ranges from 0.2 to 0.6. A neutral response is zero. An attraction
response is represented as a negative number (see Subheading 3.3).
19. As a positive control, you may perform the same assay with
high-osmolarity glycerol solution (1 M). The fraction revers-
ing of wild-type N2 hermaphrodite animals is about 0.7.
Therefore, the avoidance index of glycerol (with subtraction of
buffer-alone response) in wild-type N2 hermaphrodite animals
is about 0.4.

3.3 Pheromone Drop According to the biased random walk model for C. elegans chemo-
Test on Food for taxis, the suppression of reversals may indicate attraction [5, 6].
Measuring Attraction Attraction to pheromones can be measured as a reduction in spon-
taneous reversal responses in the drop test on food, taking advan-
tage of the increased basal reversal rate to buffer in the presence of
food. When given attractive ascaroside(s) in a drop test, the worm
will make fewer reversals compared to the buffer alone. This will
result in a negative value of the avoidance index. For example, for
wild-type male animals in the presence of food, the fraction revers-
ing to 100 nM ascr #3 (C9) in buffer is 0.3, and the fraction revers-
ing to buffer alone is 0.5. The resulting avoidance index is about
0.2, indicating attraction of males to 100 nM ascr #3 (C9) [3].
The maximal attraction response is defined by the response to
buffer alone or another control; typical attraction values on food
range from 0.1 to 0.4.
As a positive control for reversal suppression, the attractant
isoamyl alcohol can be used. A 1:10,000 dilution of isoamyl alco-
hol suppresses the basal reversal frequency to M13 buffer on food.
Alternative assays for pheromone attraction or repulsion have
been developed based on the accumulation of animals on agar
impregnated with pheromones, compared to accumulation on
agar alone [7, 8].
292 Heeun Jang and Cornelia I. Bargmann

4 Notes

1. Pheromone drop test results can be sensitive to the growth

conditions and the density of the worms. When strains with
smaller brood size or slower growth rate are tested along with
other strains, try to achieve similar population density on the
day of the assay by picking more parental animals or picking a
few days earlier.
2. Dryness of the plate is best if the moving worms can leave a
visible track on the agar that will disappear in few minutes.
3. Using a platinum wire pick with an extensively flattened, wide
end helps to scoop up worms without food. Briefly flame-
sterilize the platinum wire between animals.
4. Prepare a fresh pheromone solution each day of the assay.
5. Animals can also exhibit other behavioral responses in the drop
test, including short reversals, acceleration, or dwelling (short
back-and-forth movements). These responses can be scored
additionally and analyzed as needed.

References
1. Hilliard MA, Bargmann CI, Bazzicalupo P 5. Pierce-Shimomura JT, Morse TM, Lockery SR
(2002) C. elegans responds to chemical repel- (1999) The fundamental role of pirouettes in
lents by integrating sensory inputs from the Caenorhabditis elegans chemotaxis. J Neurosci
head and the tail. Curr Biol 12:730734 19:95579569
2. Hart AC (July 3, 2006) Behavior. In: 6. Luo L et al (2008) Olfactory behavior of swim-
WormBook (ed) The C. elegans research com- ming C. elegans analyzed by measuring motile
munity, WormBook. doi: 10.1895/worm- responses to temporal variations of odorants.
book.1.87.1, http://www.wormbook.org J Neurophysiol 99:26172625
3. Jang H et al (2012) Neuromodulatory state 7. Srinivasan J et al (2008) A blend of small mol-
and sex specify alternative behaviors through ecules regulates both mating and development
antagonistic synaptic pathways in C. elegans. in Caenorhabditis elegans. Nature 454:
Neuron 75(4):585592 11151118
4. Stiernagle T (February 11, 2006) Maintenance of 8. Macosko EZ et al (2009) A hub-and-spoke cir-
C. elegans. In: WormBook (ed) The C. elegans cuit drives pheromone attraction and social
research community, WormBook. doi/10.1895/ behaviour in C. elegans. Nature 458:
wormbook.1.101.1, http://www.wormbook.org 11711175
 Chapter 22

Behavioral Analysis of Pheromones in Fish

Peter W. Sorensen

Abstract
Pheromones are chemicals that pass between members of the same species which have inherent meaning.
Because most fish pheromones are mixtures, and their actions can be complex, behavioral assays are
required to identify them. This chapter describes a few strategies and two specific methods (one for mea-
suring attraction and another for sexual arousal) that can serve this purpose in fishes that live in nonflowing
water such as the carps.

Key words Fish pheromone, Bioassay, Carp, Attraction, Arousal, Priming, Releasing

1 Introduction

Pheromones can be defined as chemicals that serve as evolved cues

which pass between members of the same species and elicit adaptive
behavioral and/or endocrinological responses [1, 2]. Pheromones
are thus defined and identified by their biological activity and bioas-
says are required to first prove their existence and their quantify them
[3]. This is especially true for pheromones in fishes as they commonly
use mixtures of relatively unspecialized bodily metabolites as phero-
mones, making simple biochemical approaches ineffective on their
own. Pheromones have traditionally been placed into two categories
based on their activity: primers that evoke critical developmental
and/or endocrinological responses, and releasers that evoke dra-
matic and rapid and adaptive behavioral responses [1, 2]. Recent
studies suggest that some pheromones may have both types of actions
although generally just one of these actions may dominate [4].
Although priming pheromones which modulate growth rate, body
shape, and rate of sexual maturation have been described in a wide
variety of fishes [4], only one priming pheromone, 17,20-
dihydroxy-4-pregnen-3-one (17,20P), has been unequivocally
identified [3]. Accordingly, this chapter focuses on behaviorally active
releaser pheromones for which about half a dozen have been identi-
fied and which are also generally of greater interest [4].

Kazushige Touhara (ed.), Pheromone Signaling: Methods and Protocols, Methods in Molecular Biology, vol. 1068,
DOI 10.1007/978-1-62703-619-1_22, Springer Science+Business Media, LLC 2013

293
294 Peter W. Sorensen

Behavioral actions of fish pheromones are generally both com-

plex as well as species and life stage specific [2, 3]. Releasing effects
of pheromones in fishes can include (1) species recognition and
aggregation [6]; (2) alarm [7]; (3) sexual attraction [24]; and (4)
sexual arousal [24]. Depending on the species and situation, these
activities may not be mutually exclusive (i.e., a sexual signal could
also be species specific). Pheromones may also complement the
activities of other types of waterborne chemicals that fish detect
and learn to recognize. To effectively measure a pheromone, a bio-
assay must identify and focus on the unique (distinctive) activities
it stimulates in conspecifics; assays are thus species and situation
specific. This chapter briefly reviews variables associated with strat-
egies for bioassay design and then provides details on sex phero-
mones (and assays) used for carps, large riverine fishes in the
minnow family (Cyprinidae) which include the goldfish (Carassius
auratus) and common carp (Cyprinus carpio). These species are
useful models both because their behaviors and physiologies typify
many fishes and their pheromones are well known [24].

1.1 Fish Behavior An assay must test relevant aspects of a species behavior in realistic
and Ecological but simple (i.e., quantifiable) manners. Developing a new assay can
Context, and Their be challenging because there are close to 30,000 species of fish, and
Roles in Pheromone all have their own distinctive ecologies, and physiologies. Because a
Function as Well as good assay should only be measuring one behavioral cue (i.e., the
Assay Design pheromone), all other variables (e.g., ecological requirements of
the test species) must be understood and controlled. Depending on
the species, fish pheromone function may be influenced by several
ecological factors including (1) photoperiod/season (reproduction
may be highly seasonal); (2) time of day (fish often reproduce only
at certain times of day); (3) light and sound levels and spectra; (4)
water chemistry and salinity; (5) water temperature; (6) depth; (7)
water flow; and (8) substrate (for hiding or constructing nests). Not
all variables are important to all species and it can be insightful to
study wild fishes where they evolved to before examining and assay-
ing pheromones in the laboratory. Much work has been performed
in the carps which need 18 C water temperature to be reproduc-
tively active (more detail below) and still-to slowly flowing water
with spawning substrate. Another example is the sea lamprey,
Petromyzon marinus, which only responds to its migratory phero-
mone in natural river waters with appropriate chemistries and tem-
peratures, and then at night when still maturing [8]. In contrast,
sexually mature, sea lamprey respond only to a gill-derived sex
pheromone but at any time of day [9].
Not only is it important to understand the ecological require-
ments for fish when testing their behavior, but it is also essential to
understand physiological underpinnings of sexual drive: fish do not
respond to sexual cues unless sexually receptive. In some species
sexual receptivity is short-lived, and it nearly always is species and
 Behavioral Analysis of Pheromones in Fish 295

situation specific [4]. Thus, in most male fishes, sexual receptivity

correlates with spermiation and surges in hormones; for example,
sexual activity of male common carp is greatly enhanced by expo-
sure to 17,20P [4]. In most externally fertilizing female fishes
including the carps, female sexual receptivity generally coincides
with ovulation and elevated levels of circulating prostaglandin F2
(PGF2).
A third set of factors that determines both pheromonal func-
tion and assay design is the suite of social behaviors exhibited by a
species. For example, whether a fish spawns in large groups, or in
pairs, or builds and defends its own nests, correlates with whether
its pheromone might be expected to stimulate aggregation; phero-
mone assays should examine relevant attributes. In the case of the
carps, small groups of males aggressively compete for access to
receptive females while releasing pheromones in their urine; bioas-
says for this species thus test carps in groups (see below). Context,
or the spectrum of other sensory cues presented to animals, is also
highly relevant. For example, male common carp only respond
strongly to sex pheromones when presented in the context of body
odor (in which it is normally found) and it must be presented in
sex pheromone assay [10]. Finally, exposure regime and how pher-
omones are released are also very important. For example, most
carp pheromones are released in urine (vs. gills or feces) whose
release is pulsed and are thus encountered a discrete pulse with
high-concentration gradients that contain temporal information.
Bioassays for carp sex pheromones add pheromone as a concen-
trated stream (see below).

1.2 A Note on the Like many fishes, both the goldfish and common carp (herein con-
Reproductive Biology sidered carps) are seasonal spawners which typically spawn once
and Chemical Ecology a year [4, 11]. Final maturation in female carps is triggered by ris-
of Goldfish and ing water temperatures, spawning substrate (floating vegetation),
Common Carp and hormonally derived preovulatory priming pheromones (e.g.,
17,20P), so females tend to ovulate in a reasonably synchronized
fashion at daybreak (very low light). Upon ovulation, females
experience a surge in PGF2 which stimulates their sexual receptiv-
ity and is then released in pulses as a urinary sex pheromone (assays
for which we describe herein). Males appear to have evolved to find
(be attracted to) this pheromone and then respond strongly to it
when very close (in the presence of strong-concentration gradi-
ents) by exhibiting strong courtship and competitive behaviors.
Interestingly, responses to this cue are also short-lived, perhaps
because males that respond for an overly long length of time attract
the attention of predators. The hormones and pheromones used
by the carps seem to typify many fishes, including all members of
the family Cyprinidiae, and their reproductive behaviors are rela-
tively typical of this group; variants of these assays should work well
for many other fishes including the zebrafish, Danio rerio.
296 Peter W. Sorensen

1.3 Introduction Not surprisingly, many species of fish including carps are attracted
to the Nature, and to conspecifics in particular stages of reproductive maturity, most
Measurement of notably ovulation. Aggregation and sex pheromones often (but
Sexual Attraction not always) play a key role in attraction which is commonly mea-
sured. Attraction is a consequence of changes in movement pat-
terns which result in a change in a fishs distribution. These changes
are generally species specific and may also include increased
activity, or arousal (discussed below) as well as movement direction.
A key factor to consider in attraction assays is water flow; the carp
have evolved in relatively still waters and evolved the ability to
locate the diffuse pheromone plumes found within them. In con-
trast, other species have evolved in flowing waters (e.g., sea lam-
prey, trout) and to move upstream (i.e., exhibit positive rheotaxis)
in response to the relatively well-defined filamentous pheromone
plumes found within them. Herein, we describe a still-water assay.

1.4 Introduction Upon encountering pheromones released by sexually mature and

to the Nature and active female conspecifics, many mature fish become active (aroused),
Measurement of and start courting females (exhibit species-specific and stereotypical
Sexual Arousal sets of behaviors designed to stimulate them). Although species spe-
cific, these behaviors typically include increases in nudging (physical
contact) females vents (or sometimes gills) where urine laden pher-
omones are generally released, and chasing and pushing females into
areas where they might spawn (for carp this is floating vegetation).
Time of day, water temperature, light, and the presence of spawning
substrate are important contextual cues. We have developed an assay
for arousal that works well for still-water fishes that spawn in open
water. It models the sudden appearance of a pheromone-laden
plume in regions of the tanks with floating vegetation and is rela-
tively facile and quick to implement Variations of this assay should be
relevant for many related species that spawn and use pheromones in
similar manners (see below).

2 Materials

2.1 An Assay 1. Quiet, dark area with a controlled photoperiod, dimmable

to Test for Sexual overhead lights (see Note 1).
Attraction in Carps 2. 12 m circular tank (test arena) with a gravel-covered bottom,
central drain, predetermined test areas, and water inputs at
opposite ends (see Notes 2 and 3).
3. Overhead infrared lights and an overhead low-light camera (see
Note 4).
4. Video monitor, DVD recorder, and optional data acquisition
system (see Note 5).
5. At least 40 sexually mature, receptive male fish (see Note 6).
 Behavioral Analysis of Pheromones in Fish 297

6. Two peristaltic pumps, odorless Tygon tubing, beakers to hold

odors, and airstones (see Note 7).
7. Control, pheromone, and food odors (see Note 8).
8. A supply of dechlorinated or natural well water at 1820 C.

2.2 An Assay to Test 1. Quiet, dark area with a controlled photoperiod and overhead
for Sexual Arousal in lights (see Note 9).
Carps and Other 2. Glass aquaria (typically 70100 L) with gravel on bottoms (see
Still-Water Fishes Notes 10 and 11).
3. At least 50 sexually mature, receptive male fish (see Note 12).
4. Peristaltic pumps, odorless Tygon tubing, beakers, and air-
stones or filters (see Note 13).
5. Control, pheromone, and food odors (see Note 14).
6. A supply of dechlorinated or well water at 1820 C.

3 Methods

3.1 An Assay to Test Carps and their relatives are extremely sensitive to low-frequency
for Sexual Attraction sounds and lights, so a relatively isolated area must be identified for
this assay. It is desirable that behavioral scoring be conducted
blind (i.e., by individuals unaware of treatment identity, to pre-
vent bias). This can be accomplished by having someone other
than the observer make up test odors, or by recording all trials.
Four to six trials can often be run a day in the following manner
(11 trials is usually a good sample size for an experiment):
1. Drain test assay arena and refill with high-quality water at the
correct temperature.
2. Gently move and add test subjected to the assay tank, allow
them to acclimate for a predetermined time, and turn off water
(see Note 15).
3. Turn peristaltic pumps on and start observing fish distribution
across test areas in the arena(s) for the Pretest period. (see
Note 16).
4. After the pretest period, determine which test area the fish had
spent most of their time in and then change odor inputs so
that the test odor (pheromone) is added to the area where
they had spent the least amount of time. An appropriate con-
trol (or another odor for head-to-head tests to measure relative
strength) should simultaneously be added to the other area
(see Note 17).
5. Observe and record the distribution of all fish for the Test
Period following the same procedures used for the pretest
period.
298 Peter W. Sorensen

6. Add food odor to the least used side and watch for a final
positive control period (see Note 18).
7. Repeat at least ten more times using different groups of fish.
8. Place fish in a recovery tank for future experiments.
9. Analyze data (see Note 19).

3.2 An Assay to Test Carps and their relatives are scramble-spawners, so when they
for Sexual Arousal become aroused, inspect and chase each other actively. Specific
behavioral attributes that you may want to score should be deter-
mined in advance by watching naturally spawning fishes. It is desir-
able that behavioral scoring be conducted blind (i.e., by
individuals unaware of treatment identity, to prevent possible bias).
Trials can be recorded but it is usually much easier to watch. Up to
11 trials (a convenient number which also has good statistical
power) can be run a day following these protocols:
1. Drain test tanks and refill with high-quality chlorine-free water
at the correct temperature; water inflow may be allowed to
continue throughout the experiment.
2. Gently move and add groups of test subjects to tanks and allow
them to acclimate for a predetermined time (see Note 20).
3. Briefly monitor behavioral activity of the test fish (see Notes 21
and 22).
4. If appropriate, start the Pretest: turn peristaltic pump on to
add control blank odor and then start observing fish behav-
ior and activity. Watch fish for a predetermined test period
while noting and recording behavioral attributes of interest
(see Notes 23 and 24).
5. At the conclusion of pretest period, change odor input so that
pheromone is added instead of the control (see Note 25).
6. Start the Test Period and watch the distribution of all fol-
lowing the same procedures used for the pretest period.
7. Add food odor to the least used side and watch for a final
period (see Note 26).
8. Repeat at least ten more times using different groups of fish.
9. Analyze data (see Note 27).

4 Notes

1. It is important to test fish in areas that are quiet and vibration

free as many fishes are very sensitive to these. Light levels
should be less than 1 lx and follow natural photoperiods dur-
ing the spawning season (16 L:8D for carps).
 Behavioral Analysis of Pheromones in Fish 299

Opaque Plexiglas

Test Area Test Area

Neutral Area

Drain
Odor inlet Odor inlet
and airstone and airstone

Perforated panel

150 cm

Fig. 1 Schematic diagram of the pheromone attraction assay for carp viewed
from above

2. An assay tank (arena) design for common carp is shown in

Fig. 1. It can be modified to simultaneously optimize both
odor dispersal (this can be studied by injecting fluorescent
dyes) and the swimming patterns of fish (it is desirable that
test fish swim throughout the entire maze during the pretest
[this might be evaluated in pilot studies, see below]). Thus,
while we sometimes use partitions in test arenas to reduce
odor dispersal from one side to other, for fishes that do not
swim spontaneously (Asian carps [Hypophthalmicthys sp.]), we
remove the partitions and shorten test periods. The test period
also need not be longer than the period of time that fish might
be expected to respond to the presence of a sex pheromone in
the absence of a pheromone donor (in goldfish this is 15 min
[this may be determined empirically]). Gravel can also be
added to the bottom of assay tanks used for bottom-feeding
fish as it makes them more comfortable and prone to explore
all regions of the tank. Generally we set water depth at 10 cm
because fish are often comfortable with this depth and the
cameras can see into it. Dark cloth curtains are placed around
arenas. Water inlets should be designed so that they can easily
and quietly be turned off; otherwise supply incoming dechlo-
rinated or tap water at rates between 0.1 and 1.0 L/min. More
details of the assay tank design are found in refs. 10, 12.
300 Peter W. Sorensen

3. We recommend setting up at least four mazes so that experiments

can proceed quickly and multiple trials can be run in a day.
4. We use pairs of high intensity deep-infrared 12VDC LED lights
(most fish see minimally into the infrared) as light sources. These
are placed on an overhead frame and an oblique crossing pattern
to minimize glare. Matching low-light security cameras with
good-quality polarized optics with high resolution are used to
watch fish. A fish eye lens may be useful. If you are interested in
fish that do not require dim lighting, full-spectrum and brighter
lights might also allow you to track individual fishes using their
pigmentation or even colored tags become possible.
5. A simple black-and-white monitor is generally adequate for
low-light infrared observation. It should be placed at least a
meter away from the assay tanks. We often mark regions of
tank on the screen so that it can be scored live but always
record to a DVD to allow for reanalysis. Automated data acqui-
sition systems may be helpful, especially if multiple arenas and
color are used because these systems can track individuals.
However, we have had some difficulties with these (they may
have trouble tracking individual fish under dim light when they
cross), so it is good to test such systems.
6. It is essential to use healthy, sexually mature, and receptive fish
as subjects. Ideally fish will be obtained directly from the wild
or suppliers as pet store fish are often treated with antifungal
products that destroy their olfactory systems (electro-
olfactogram [EOG] recording can confirm this [13]). We use
fully spermiated males that we know will spawn with females if
offered the opportunity. All-male groups are used as it reduces
variance (sometimes a single female is distracting). In some
cases, we prime males by pre-exposing the priming phero-
mones (0.1 nM 17,20P) the night before because this greatly
enhances their tendency to respond to sex pheromones and
females [10]. If necessary, fish may be reused but we wait at
least 3 weeks before doing so [10] and we do while tracking
them and alternating treatments, so experience is unlikely to
be a factor. Typically subjects are kept in conditions identical to
those of the assay arena. All experiments are run at standard
and relevant times of day and spawning temperatures (e.g.,
1820 C for goldfish and common carp). We starve fish for
24 h to assure a constant level of motivation.
7. Odors are pumped into each end of the assay tanks using high-
quality peristaltic pumps (typically 10 ml/min). Test and con-
trol odors are typically drawn from pre-cleaned (with clean tank
water) 4-L beakers via plastic tubing and then injected into the
tanks via PE tubing associated with airstones. Airflow through
the latter is adjusted to provide a minimal but constant level of
controlled mixing. Generally pumps run throughout the entire
 Behavioral Analysis of Pheromones in Fish 301

experiment; we simply alternate between blanks, controls, and

test odor inputs.
8. Food odor (paper-filtered odor of fish food) is often added at
the experiments conclusion to confirm the health and respon-
siveness of the fish [10, 12]. Odors are always freshly made up
(that day) in the same water that is used to supply the test
assays. If solvent carriers (e.g., ethanol) are needed to dissolve
pheromones, the carrier is kept to a concentration of less than
0.1 ml/L to reduce the chance of it being detected and then it
is as a control treatment. Beakers that house the odors must be
cleaned and then rinsed extensively with water use with the
assay to remove residual soap and amino acids (EOG recording
can confirm cleanliness).
9. Conditions are similar to those required for attraction (see
Note 1) except that lighting levels are usually greater to allow
fish to be observed from the side and the holding facility may
be more modest.
10. Standard 20 gal (80 L) glass aquaria work well for fishes less
than 10 cm in length. We cover all sides except the front
which we mark with lines using electric tape (to note movement).
A drain is often drilled in the top edge of one side and water
flow comes through the top (which may be screened; egg
crate from lighting fixtures works well) (Fig. 2). We typically
set tanks at a height of 1.5 m so that they can be easily viewed
by observers on stools and install black fabric viewing cur-
tains in front of them with holes for viewing. More detail is
found in refs. 5, 13.
11. We recommend setting up 610 aquaria as testing is relatively
rapid and multiple tests can be performed in a day.
12. We preselect males for these relatively simple experiments by
determining that they will spawn with PGF2-injected female
first [13]. They could also be pretreated with priming
pheromone (see Note 6). Generally we starve fish for 24 h
before testing to ensure constant levels of motivation. By
restricting tests to male-only groups, variance in responsive-
ness is reduced.
13. Odor addition is similar to the attraction assay (see Note 7)
except that there is only one odor injection location per tank
and we generally place it in floating artificial vegetation (green
yarn tied to floats for carps).
14. Odors are made up exactly as for the attraction assay (see Note 8).
15. Pilot tests that do not add any odors should be conducted to
determine acclimation time and group size: the objective is to
find a combination of both so that all fish are exploring all areas
of the test arena at least once every 5 min before trials begin.
Usually movement patterns follow a bell curve with acclimation
302 Peter W. Sorensen

Fig. 2 Schematic diagram of tank assay to measure sexual arousal. It is viewed

from the side. For fish that routinely move the length of the aquarium (carp), we
draw lines on the long side and watch from that side

timewe find that 1 h works well for groups of five common

carp and 20 min for groups of three silver carp. Pilot tests
should also be conducted with dyes to confirm that odors stay
in one-half of the arena(s) for at least 10 min (pump rates and
aeration can be adjusted) and that turning peristaltic pumps
on/off does not startle fish.
16. Pretest and test period duration (which should be the same)
and group size are determined by pilot tests as the period of
time needed to ensure that most fish in most trails actively visit
all areas of the arena and spend at least a third of their time in a
region where the odor could be added (we typically use five fish
and 1015 min for carp). During the pretest the location of all
fish throughout the test arena is noted at preset intervals that
allow one to gauge overall distribution (we use 15-s intervals).
If desired, information on swimming activity (and perhaps some
readily discernible behavioral attributes) can also be garnered
 Behavioral Analysis of Pheromones in Fish 303

by replaying DVDs and noting fish crossing into specific

regions as designated by lines drawn in the assay arena or on
monitor [12]. It is desirable to record data in short intervals
(5 min) so that data can be reanalyzed in different ways if
responses prove to be short-lived.
17. Odor is added to the least used side to create a consistent con-
servative bias. Odors are changed by changing inlet tubes in
beakers used to feed experiment. A control is always tested as a
treatment group. If fish show a strong bias in the maze (75 %
time during control) or do not move, the trial is terminate and
not scored.
18. This step is an optional positive control to confirm fish respon-
siveness (they should always feed).
19. Each trial is independently analyzed live or ideally by reviewing
DVDs [6, 10]. To accomplish this, the number of carp observed
in each test area of each maze at predetermined intervals dur-
ing both the pretest period and then the test period is noted.
Fish position is based on the location of its eye. The number of
fish is then summed for each period, and divided by the total
possible score to yield percent time that all fish spent in the
experimental area (the area or the side to which the test odor
was added) during each period. Pretest values are then sub-
tracted from test values for each trial to yield relative attrac-
tion, a measure of change in time spent in the test area during
odor addition. The use of relative attraction normalizes the
data and removes any possible side bias. (Note: If the assay
design includes a neural area, then expected time in each area
is not 50 %.) An average value for relative attraction is next
calculated for all trials of each experiment and differences
between treatments compared using one-way ANOVAs. If sig-
nificance is indicated (P < 0.05), individual values are compared
to each other and the blank control using a HolmSidak
adjustment. In addition, values for blank control are also tested
against zero using t-tests to confirm a lack of maze bias. Head-
to-head experiments (i.e., one odor versus another) are evalu-
ated by comparing relative attraction values with t-tests after
confirming normality with a KolmogorovSmirnov test.
20. Pilot tests that do not add any odors should be conducted to
determine acclimation time and group size: the objective is to
find a combination of both so that all fish are exploring all
areas of the tanks and only interacting weekly when trials begin.
Usually movement patterns follow a bell curve with acclima-
tion timewe find that adding groups of five male fish the
night before experiments works well. Pilot tests should also be
conducted with dyes to confirm that odors stay in a well-
defined area of the tank. Modifying aeration rate and method
can achieve this purpose. Adding artificial vegetation can also
304 Peter W. Sorensen

help as does adding more fish (fish seem to learn individual

identities and their reproductive condition if small groups are
used, and then may cease responding to new cues).
21. Experiments should use a set temperature and be conducted at
a set time of day to mimic natural spawning conditions (an
hour after lights on for carps and 1820 C).
22. If test fish are extremely active at the time the experiment is
due to start (i.e., fish are chasing each other more than 25 % of
the time), the trial is aborted. In very mature goldfish this is
relatively common and longer acclimation times may reduce
the frequency with which it occurs.
23. Test period duration is determined by pilot tests as the period
of time needed to ensure that most fish in most trails actively
visit all areas of the tank and inspect each other a few times (we
typically use 10 min for carp). The duration of the test period
should match that of the pretest but might be the duration of
the response they exhibit to pheromones in the absence of a
female in pilot experiments.
24. Pilot tests should determine behavioral attributes of interest.
These ideally will be determined by watching naturally ovu-
lated and spermiated fishes spawn in advance, ideally in the
wild. For carps, it is relevant to record locomotor activity
(the number of times fish cross lines drawn across face of test
aquaria using their pupils as reference point), nudges
(instances when fish come into physical contact with each other
[inspecting their vents or gills]), pushes (when fish displace
each other), chasing (when fish swim quickly after another
[either chasing incidents or time spent chasing can be moni-
tored]), and feeding activity (when carp pick up stones and
other bottom; a positive control as aroused fish do not feed as
much). More detail on these attributes is found in refs. 1214.
Activity may be recorded using a manual recorder or a com-
puter data recorder. It is desirable to record in short segments
(5 min) so that data can be reanalyzed in different ways if
responses are short-lived.
25. Odors are changed by changing inlet tubes in the supply
beakers.
26. This step is an optional positive control which is not analyzed
and may be omitted if one is sure that the fish are healthy and
responsive.
27. Each trial and set of behaviors is independently scored and then
analyzed. One way is to use nonparametric procedures [13] if
experimental variance is very high. Another way is to normalize
the data and examine changes in behaviors. Here, one measures
the number of times each behavioral action is performed by all
fish in each group (i.e., fish identity is not noted) for each trail,
 Behavioral Analysis of Pheromones in Fish 305

first during the pretest period and then the test period. Percent
change is then calculated for each behavior and treatment, and
then comparing all of these values (including a blank control)
by one-way ANOVA if data are first found to be normally dis-
tributed and to meet assumptions of homogeneity of variance.
Blank controls must be included as a treatment group and post
hoc follow-up tests performed as for the attraction tests (see
Note 19). If desired, possible pump effects and maze bias effects
can be evaluated using raw data in Wilcoxon rank sum tests.

References
1. Wyatt T (2003) Pheromones and animal steroids functions as a migratory pheromone in
behaviour. Cambridge University Press, the sea lamprey. Nat Chem Biol 1:324328
Cambridge 9. Li W, Scott AP, Siefkas MJ, Yan H, Liu Q, Yun
2. Sorensen PW (in press) Introduction to phero- S-S, Gage DA (2002) Bile acid secreted by
mones and related cues in fish In: Sorensen male sea lamprey that acts as sex pheromone.
PW, Wisenden, BW (eds) Fish pheromones and Science 296:138141
related cues. Wiley Blackwell, Iowa 10. Lim H, Sorensen P (2011) Polar metabolites
3. Sorensen PW, Hoye TR (2010) Pheromones in synergize the activity of prostaglandin F2 in a
vertebrates. In: Lew M, Hung-Wen L (eds) species-specific hormonal sex pheromone
Comprehensive natural products II. Elsevier, released by ovulated common carp. J Chem
Oxford, pp 225262 Ecol 37:695704
4. Stacey NE, Sorensen PW (2011) Hormonal 11. Kobayashi M, Sorensen PW, Stacey NE (2002)
pheromones. In: Encyclopedia of fish physiol- Hormonal and pheromonal control of spawn-
ogy: from genome to environment, vol 2. ing in goldfish. Fish Physiol Biochem 26:71
Academic, San Diego, pp 15131562 84, (22):17
5. Sorensen PW, Stacey NE, Chamberlain KJ 12. Maniak PJ, Lossing R, Sorensen PW (2000)
(1989) Differing behavioral and endocrinolog- Injured Eurasian ruffe, Gymnocephalus cernuss,
ical effects of two female sex pheromones on release an alarm pheromone which may prove
male goldfish. Horm Behav 23:317332 useful in their control. J Great Lakes Res 26:
6. Levesque H, Scaffidi D, Polkinghorne CA, 183195
Sorensen PW (2011) A multi-component spe- 13. Sorensen PW, Hara TJ, Stacey NE, Goetz FW
cies identifying pheromone in the goldfish. J (1988) F prostaglandins function as potent
Chem Ecol 37:219227 olfactory stimulants that comprise the postovu-
7. Mathuru AS, Kibat C, Cheong WF, Shui S, latory female sex pheromone in goldfish. Biol
Wenk MR, Friedrich RW, Jusuthasan S (2012) Reprod 39:10391050
Chondroitin fragments are odorants that trig- 14. DeFraipont M, Sorensen PW (1993) Exposure
ger fear behavior in fish. Curr Biol 22:17 to the pheromone 17,20-dihydroxy-4-
8. Sorensen PW, Fine JM, Dvornikovs V, Jeffrey pregnen-3-one enhances the behavioural spawn-
CS, Shao F, Wang JZ, Vrieze LA, Anderson ing success, sperm production and sperm motility
KR, Hoye TR (2005) Mixture of new sulfated of male goldfish. Anim Behav 46:245256
 Chapter 23

Analysis of Male Aggressive and Sexual Behavior in Mice

Takefumi Kikusui

Abstract
Pheromone and odor signals play a pivotal role in male mouse reproductive behaviors, such as sexual and
aggressive behavior. There are several methods used to assess male behaviors, each of which examines a
unique aspect of the biological function of mice. There are two major ways of assessing male aggressive
behavior in mice, one is using isolation-induced aggression, and the other is territorial aggression in pair-
housed males. To analyze male sexual behavior, a female mouse that is hormone-primed with estradiol and
progesterone is usually introduced into a male home range, and mounting, intromission, and ejaculation
behaviors are observed for 1 h. Here, we summarize the detailed protocols for assessing male behaviors.

Key words Male aggression, Sexual behavior, Mounting, Attacking, Ejaculation, Territorial behavior

1 Introduction

Male and female reproductive behavior is mainly regulated by olfac-

tory information. Chemical signals, including pheromones, are the
most important factors that stimulate aggression and sexual behavior
in male mice [1]. Aggressive behavior in mice is defined as physical
attacking, and including the potential to bite other animals.
Predation, infanticide, defense, and offense are types of aggressive
behaviors in mice [2]. However, this section is concerned with offen-
sive attacking and territorial marking as forms of territorial defense.
Whereas male chemosignals induce aggressive behavior in male
mice, female chemosignals enhance male sexual behavior. Male
sexual behavior includes precopulatory behavior (sniffing the body
and anogenital area and ultrasonic vocalization) and copulatory
behavior (mounting, intromission, and ejaculation), both of which
are highly dependent on female chemosignals.
Male mice emit ultrasonic vocalizations in the presence of
females or when they are stimulated by a females urinary phero-
mones [3]. Recent studies have demonstrated that ultrasonic song
vocalizations of male mice have behavioral features similar to the
songs of songbird, including discrete syllables with temporal

Kazushige Touhara (ed.), Pheromone Signaling: Methods and Protocols, Methods in Molecular Biology, vol. 1068,
DOI 10.1007/978-1-62703-619-1_23, Springer Science+Business Media, LLC 2013

307
308 Takefumi Kikusui

sequencing, repeated phrases, and variability among individuals [4].

Sexual activity and motivation are related to ultrasonic vocalization
calls, and these calls can stimulate female sexual behavior. Therefore,
we also describe methods of analysis of ultrasonic song
vocalizations.

2 Materials

2.1 Male Aggressive Strain differences that affect the level of offensive attack in mice
Behavior were initially reported 50 years ago [5]. Extensive information on
strain distributions and male offense are available [68]. For exam-
2.1.1 Subject Males
ple, it has been reported that DBA/1 and DBA/2 males are more
in Aggression Analysis
aggressive than C57BL/6 or C57BL/10 males [9]. It is notable
that these strain differences in aggression depend on life history,
test situation, and opponent type [10] (see Note 1). Molecular
biologists and behavioral neuroscientists are encouraged to consult
the referenced literature, at the beginning stages of their collabora-
tion, to choose the optimal strain, with background genes that are
most appropriate for their specific research goals. This is particu-
larly important in genetically engineered mouse studies, for which
there is no ideal strain recommended across all behavioral para-
digms for all null mutations. Normally, C57BL/6 or 129SV strains
are used to generate transgenic lines, but 129SV males show higher
aggression than C57BL/6 males [11]; additionally, there have
been reports indicating that gene(s) on the Y-chromosome can
contribute to strain differences in inter-male aggression [12].
Thus, it is important to consider purifying the genetic background
of transgenic mice in order to assess the aggression levels in trans-
genic mice (see Note 2).

2.1.2 Opponent Mice One of the most important issues in aggression analysis is selecting
in Aggression Analysis the appropriate opponent mouse. The size, age, strain, and behav-
ior of an opponent mouse can lead to a completely different type
of aggression behavior in the subject mouse [13]. If the opponent
mouse attacks back, the subject resident mouse thereafter tends to
show reduced aggression. The opponent should have the follow-
ing characteristics: (1) very low aggressiveness, (2) younger or
lighter than the subject mouse, (3) sexually mature intact male
(older than 67 weeks), (4) group housed with male siblings, and
(5) not sexually and aggressively experienced.
Male chemosignals are the most important factors for inducing
aggression, and these chemicals are dependent on gonadal
testosterone secretion [14]. Therefore, intact male mice are prefer-
able as the opponents. However, testosterone can enhance the
opponents aggressiveness. To solve this issue, the most commonly
used opponents are anosmic mice, juvenile and subordinate mice,
and A/J strain males (see Note 3). Anosmic mice have undergone
 Analysis of Male Aggressive and Sexual Behavior in Mice 309

surgical removal of their olfactory bulbs, whereas juvenile and

submissive intruders are those that have experienced being bitten;
neither of these groups attack the residents and are used as oppo-
nents. Olfactory information is necessary for the induction of
aggression in mice, as supported by the fact that olfactory balbec-
tomy completely abolishes aggression [15, 16]. Therefore, anos-
mic mice are not aggressive, even though they are gonadally intact
and can emit male chemosignals. A/J mice are extremely unag-
gressive [17]; therefore, using an A/J mouse as the opponent is
also preferable. However, even though they are not aggressive,
each type of opponent shows different attack responses and bite
avoidance behavior. These differences in opponent behavior lead
to differences in the aggression levels of the resident mice [13].
The housing conditions of an opponent also influence its
aggression. Isolation for more than 24 weeks causes an increase in
male aggressiveness; therefore, the opponent should be housed
with siblings.

2.1.3 Housing and There are many factors that influence male aggression, including
Hormonal Conditions hormone levels, prior experience of aggression, isolation and
in Aggression Analysis grouping, dominance and subordination, a high or a low incidence
of the stress experience, and food deprivation. There are two major
sets of conditions for housing the subject mice used for aggression
analysis, namely, social isolation-induced aggression and pair-
housed territorial aggression. It has been found that fighting and
threatening behavior increases with progressive isolation, up to an
asymptote at 5658 days. In pair-housed territorial aggression, the
residents usually show stable aggression 48 weeks after pairing
[18]. The earlier the age at which differential housing is imposed,
the greater are the behavioral differences between animals under
each housing condition [19].
Isolation confers certain behavioral characteristics, similar to
those of pair-housed resident mice, such as aggressiveness and the
tendency to attack other males. However, isolated mice are rela-
tively less aggressive than pair-housed resident mice. There is a
hypothesis that aggressive behavior induced in isolation is related
to pathological changes [20]. Therefore, one should be cautious in
using mice with isolation-induced aggression because this type of
aggression may be dependent on isolation-related stress responses,
as well as on chemosignal information (see Note 4).
Either isolation-induced aggression or pair-housed territorial
aggression in laboratory mice can be dramatically reduced by cas-
tration. Decline by castration may, however, be restored or even
accentuated by androgen replacement [9]. A recent transgenic mouse
study demonstrated that both estrogen receptors and androgen
receptors are involved in the maintenance of aggression [21]. Both
adult androgen levels and neonatal testosterone levels influence
aggression, and are referred to as organizational effects. Other
310 Takefumi Kikusui

hormones, such as glucocorticoid, can also modulate the aggres-

sion levels. Mice that have been isolated for longer periods, and
have fought in aggression tests, have higher corticosterone titers
than comparable group-housed animals. Ideally, these hormone
levels should be equal in each mouse to assess the effects of chemo-
signals on aggression. At the very least, researchers need to con-
sider the individual differences in aggressive behavior in their
experiments.

2.2 Male Sexual When assessing male sexual behavior, sexually nave males are nor-
Behavior mally used as the subject because the experience of mating changes
male sexual motivation and chemosensory sensitivity. For example,
2.2.1 The Subject Male
if almond odor is paired with the reward state induced by ejacula-
in Sexual Behavior Analysis
tion in males, they will ejaculate more frequently with almond-
scented females when given the choice of copulating with scented
and unscented females [22].
There are large strain differences in sexual behavior in mice
[23]. These differences in inbred mouse strains represent a won-
derful opportunity for genetic analyses. However, not all strain dif-
ferences are solely genetic in origin, and the early developmental
environment can shape male sexual behavior. Thomas E. McGill
initially compared 16 characteristics of male sexual behaviors
among 3 inbred strains (C57BL/6, BALB/c, and DBA/2J).
C57BL/6J males had shorter latencies between mounting and
mounting with thrusting and shorter intervals between intromis-
sions. Furthermore, these males displayed more intromissions and
thrusts than DBA/2J males. Interestingly, one existing hypothesis
suggests that genes on the Y chromosome regulate sexual behavior
independently of gonadal hormones [24]. One particular F1
hybrid, C57BL/6J dams by DBA/2J sires, resulted in a male
hybrid (B6D2F1) capable of sustaining copulatory behavior for up
to a year after castration. Males produced by the reciprocal cross
(DBA/2J dams and C57BL/6J sires) did not have this character-
istic, nor did F1 males from BALB/c by DBA/2J matings [25],
suggesting that the interaction of genes on the Y chromosome and
other genes on autosomes modulate male sexual behavior.
As described above in the aggression section, C57BL/6 or
129SV strains are usually used to generate transgenic lines, and
thus it is appropriate to consider that the degree of genetic back-
ground of the transgenic mice is critically related to male sexual
behavior. For example, estrogen receptor-alpha knockout mice, in
a mixed background of 129SV and C57BL/6, showed copulation
with females and deposited sperm plugs; however, backcrossing
these to a C57BL6 line diminished this behavior [23]. Thus, it is
important to consider purifying the genetic background of trans-
genic mice when assessing aggression levels in mice (see Note 2).
 Analysis of Male Aggressive and Sexual Behavior in Mice 311

2.2.2 The Opponent The most common method of sexual behavior analysis is to use a
Female in Sexual Behavior hormone-primed female as an opponent. Subject males are initially
Analysis tested with ovariectomized females, which are treated with estra-
diol and progesterone beforehand, of the same strain, or with
BALB/c female mice, which tend to show greater sexual receptiv-
ity. The dosage and timing of steroid treatment is as follows: estra-
diol benzoate (EB) dissolved in corn oil, ranging from 0.5 to
10 g, injected twice at 48 and 24 h before testing, followed by
progesterone 200500 g dissolved in corn oil, injected once
between 12 and 4 h before testing. These treatments fully recover
female receptivity for male copulatory behavior. Using sexually
experienced females is preferable, because sexually nave females
show relatively lower receptivity and higher avoidance behavior
with the male mice (see Note 5). Estrogen can be administered to
ovariectomized females by subcutaneous implantation of an
estrogen-containing tube, which is usually a silicon tube, 1 cm in
length and 1 mm in inner diameter, filled with 17-estradiol of
200500 g/mL mixed in silicon adhesive or oil. However, pro-
gesterone should be administered by injection (see Note 6).
Alternatively, an intact female can be an opponent of male mice.
In this case, researchers need to monitor vaginal smears to confirm
the estrous cycle. Female mice normally tend to display signs of
estrus, including mating behavior, every 4 or 5 days, and the estrous
cycle has been divided into as few as four phases: diestrus, proestrus,
estrus, and metestrus. Cellular characteristics of vaginal smears
reflect changes in the structure of the vaginal epithelium during the
cycle. In the stage of metestrus, a large number of cornified cells
and leukocytes are observed. In the diestrus stage, the vaginal con-
tents consistently lack cornified cells, whereas leukocytes are very
plentiful. In proestrus, the vaginal smear is devoid of leukocytes and
characterized by nucleated epithelial cells. The stage of estrus is
characterized by marked cornification of the cells and the disappear-
ance of leukocytes. At the end of estrus, the cornified layer is
sloughed off, and invasion by leukocytes occurs [26]. The advan-
tage of using an intact female is that the displayed behavior reflects
a natural behavior. Conversely, there is a disadvantage that not all
females show receptive behaviors towards male mice, and this prob-
ably causes variation in male sexual behavior (see Note 5).

2.2.3 Equipment for Ultrasound recording: All experiments are performed in a sound-
Analyzing Male Ultrasonic proof chamber under a dim red light. Ultrasonic sounds are detected
Song Vocalization using a condenser microphone, such as UltraSoundGate CM16/
CMPA (Avisoft Bioacoustics, Berlin, Germany), designed for record-
ings between 10 and 200 kHz. The microphone was connected to
an A/D converter (for example, UltraSoundGate 116; Avisoft
Bioacoustics, Berlin, Germany) with a sampling rate of 300 kHz,
and acoustic signals were transmitted to a sound analysis system
(for example, SASLab Pro; Avisoft Bioacoustics, Berlin, Germany).
312 Takefumi Kikusui

During the recording, a subject male mouse was housed individually

in a testing cage for 0.52 h. The test cage was placed in the sound-
proof chamber, and a female mouse, devocalized by unilateral sec-
tioning of the inferior laryngeal nerve [27], was introduced into the
test cage. The ultrasound was recorded and the data were later ana-
lyzed using audio software (for example, SASLab Pro; Avisoft
Bioacoustics, Berlin, Germany and Adobe Audition) (see Note 7).

2.2.4 Housing and The subject mice are singly or group housed before the test. As
Hormonal Conditions mentioned above, a long period of isolation, such as more than
in Male Sexual Behavior 58 weeks, is related to pathological changes in behavior [20].
Analysis Therefore, individual housing for less than 5 weeks is preferable.
Group housing of male siblings cannot result in certain problems
associated with social dominance and subordination within the
group; however, unfamiliar males housed in groups are not suit-
able for behavioral analysis because the social hierarchy changes
their sexual behavior considerably, such that it is higher in domi-
nants and lower in subordinates [28].
The degree of sexual behavior is considerably dependent on
androgen levels, both in neonates and adults. Male precopulatory
and copulatory behavior is critically reduced by castration, and
almost fully recovered by testosterone treatment [23]. Therefore,
comparison of male sexual behavior needs an equal level of circu-
lating testosterone both in neonates and adults, particularly in
transgenic mice. Ideally, these hormone levels should be equal in
each mouse to assess the effects of chemosignals on sexual behav-
ior. Researchers need to consider the individual differences in
sexual behavior in their experiments.

3 Methods

3.1 Standard Generally, one resident male is assigned as a subject, and its behav-
Procedures ior toward the opponent is analyzed. Socially grouped male aggres-
in Male Aggression siveness is not so common in studies on the neurobiology of
aggression, but are sometimes conducted from an ethological
point of view [29]. One study found that attack by dominant male
colony mice on intruders included chasing and lateral attack behav-
iors, whereas the corresponding intruder behaviors were flight,
boxing, and checking [30]. Mice did not show a significant con-
straint on bites to ventral areas, and the rat defensive behavior of
lying on the back was rare; the corresponding on-top behavior
of attackers was almost absent in mice.
There are two major procedures for assessing aggression. One is
evaluating the development of aggression in nave mice, and the
other is analyzing aggressive behavior in trained resident mice. In
both procedures, the opponent of an intruder is not used repeatedly,
and the same intruder is not used twice in the same day because of
 Analysis of Male Aggressive and Sexual Behavior in Mice 313

ethical issues. It is also preferable to change the intruder for every

testing session in subject mice because familiar intruders are attacked
less vigorously than unfamiliar mice. It seems likely that the defeated
mouse becomes less potent as a stimulus eliciting attack as the resi-
dent becomes habituated to it.

3.1.1 Behavioral The behavioral inventory included agonistic and nonagonistic ele-
Parameters in Male ments [31]: (1) attack bites and leaps, sideways movements and tail
Aggression rattles, anogenital contacts, nosing, and pursuit recorded for the
resident; (2) defensive upright posture, escape leaps, and supine
posture recorded for the intruder; (3) locomotion, rearing, self-
grooming, and mutual upright postures were recorded for both
animals. Tail rattling, which is fast lateral rattling of the tail, and
sideways movements, which is a lateral rotation of the body, accom-
panied by piloerection and short steps, are a prominent feature of
aggressive encounters between mice and is thought to be a threat-
ening behavior. Measuring locomotion, rearing and self-grooming
can exclude other physiological or neurobiological factors that
modulate aggression; for example, muscle relaxant effects can
decrease aggressive behavior, but these effects also can be detect-
able in locomotor activity during the behavioral test.

3.1.2 Aggression The subject male mouse is placed in his home cage because anxiety
Development would inhibit aggression (see Note 1). An intruder mouse is intro-
duced in the cage and the behavior of the resident is analyzed.
Usually, a test session lasts 510 min if the resident does not show
attack biting toward the intruder. Once the resident mouse shows
attacking behaviors, the test is ended 35 min after the first attack or
until 1020 attacking bites, due to ethical issues. This procedure is
repeated 2 or 3 days apart and 25 times. The repetition of the test
session is necessary because sometimes the initial aggressive behavior
would also be related to anxiety or adapting to the new environ-
ment. The comparable parameters are dependent on the procedures.
If the researchers end the test sessions by the duration time, the
number of attacking bites is the most important parameter, which
usually increases by repetition of the session. When the bite numbers
ends the test session, the time to the end of testing is the index to
compare, and it decreases by the repetition of the sessions.

3.1.3 Aggression As a result of repletion of aggression experience, aggression in the

Analysis in Trained resident mice becomes stable after approximately 45 repeated tests
Residents (see Note 1). Under conditions of pair housing and experiencing
aggression, the resident male can discriminate their paired female
from an unfamiliar one, and they are more sensitive to individual
differences, such as showing vigorous aggression toward an unfa-
miliar castrated intruder [32] (see Note 8). This discriminative abil-
ity is based on the individual signature in urine, and this signature is
acquired during the developmental period, although the responsible
molecules have yet to be determined [33].
314 Takefumi Kikusui

Using the trained resident, the durations of aggressive (sideways

threats, attack bite, and tail rattles) and nonaggressive (anogenital
contact, locomotion, rearing, digging, self grooming, and sniffing)
behaviors are measured. To measure the residents aggressiveness,
the number of attack bites and the latency to the first attack bites
are the most important parameters [33]. In this test procedure, all
intruders avoided bloodshed, indicating that the intruders were
not fiercely attacked.

3.2 Standard A test chamber is usually the home cage of the subject male mice and
Procedures in Male is placed on a shelf to allow for ventral viewing. Each of the males
Sexual Behavior was individually housed in the test cage 124 h prior to the court-
ship testing. After 24 h, they were paired with estrogen- and proges-
terone-treated ovariectomized females, and their behavior was
video-recorded for 1 h during the dark phase under a dim red light.

3.2.1 Behavioral The total numbers of male mounts, female lordosis behavior
Parameters in Male towards the mounts, intromission, and ejaculation behavior of
Sexual Behavior males towards the females are indispensable parameters to score.
Latency, frequency, duration, and other measures have been used
as indices for these behaviors. Mount was defined as a male using
both forepaws to climb onto a female from behind for copulation.
Lordosis response, which is a great indicator of female receptivity,
is defined as a female with all four paws grounded, the hind region
elevated from the floor of the test chamber, no evidence of attempt
to escape or exhibition of a defensive upright posture, and the back
slightly arched (see Note 5). Intromission was defined as a male
pelvic thrust with a stable frequency continuously for more than
6 s and demonstration of the females anogenital area elevated over
the ground when finished. The total duration of anogenital sniff-
ing of the subject males toward females, the approach and sniffing
behavior of the females toward males, the grooming behavior of
the subject males, the grooming behavior of the females, and the
rearing behavior of the females are also scored according to the
following definitions, and these parameters reflect the ability of
locomotion, exploring, recognition, and motivational state of the
subject mice. Anogenital sniffing is defined as a male mouse actively
approaching, touching, and sniffing the females anogenital area.
Approach and sniffing is defined as a female stretching out,
approaching and sniffing a stud males body, including the head
and anogenital area. Grooming is defined as a male self-grooming
its body, including the face, anogenital area, and body trunk.

3.3 Ultrasound The following is an example of the sound analysis methods, and
Analysis of Male the details of the settings can be modified. Spectrograms were gen-
Ultrasonic Song erated with an FFT-length of 1,024 points and a time-window
Vocalization overlap of 75 % (100 % frame, Hamming window). The spectro-
gram was produced at a frequency resolution of 488 Hz and a time
 Analysis of Male Aggressive and Sexual Behavior in Mice 315

resolution of 1 ms. A lower cut-off frequency of 2030 kHz was

used to reduce background noise outside the relevant frequency
band. Parameters analyzed for each subject included the number of
syllables, duration of syllables, and qualitative and quantitative
analyses of sound frequencies, measured in terms of frequency at
the maximum of the spectrum (see Note 9).
Waveform patterns of calls collected from the subject mice can
be analyzed in detail. Each syllable is identified as belonging to 1 of
10 distinct categories, based on internal pitch change, length, and
shape. The classification of the ten categories of ultrasonic vocaliza-
tion syllables is described elsewhere [34]. To date, there is no auto-
matic categorization program available; therefore, researchers blind
to the experimental treatment need to categorize the syllables using
their own discretion. In order to confirm the categorization, a like-
lihood ratio test, which examines whether there is a systematic dif-
ference between the blinded experimenters, is performed by a
generalized linear model that consists of response variables.

3.4 Methods for As described above, chemosignals of male mice play a critical role
Chemosignal Inputs in the induction of male aggressive behavior, whereas female che-
and Male Aggressive mosignals induce male sexual behavior. In particular, urine con-
Behavior and Sexual tains maleness and femaleness signals and stimulates male aggression
Behavior and sexual behavior, respectively (see Note 10). To assess the abil-
ity of urine to induce male behavior, the following behavioral assay
is indispensable. Urine, collected by massaging the bladder region
to induce urination of gently scruffed mice, is combined from mul-
tiple animals, in order to minimize the individual signatures and
social ranks that are found in urine. If the researchers test female
urine, they need to collect urine from more than ten females, in
order to standardize the estrous cycle. Alternatively, they can mon-
itor the estrous cycle by vaginal smears and collect the urine from
a specific estrous stage of females. To swab urine or other materials
onto the opponent, a cotton fluff (3 3 mm) is soaked in 60 L of
aqueous sample and applied to the back and anogenital area of the
opponent before the behavioral testing. As an alternative, samples
are dropped onto a cotton swab and presented to the subject mice
before the behavioral testing [35].

4 Notes

1. Generally, not all male mice are aggressive, with 7080 % of

males displaying typical aggression. The reason is unknown, but
because younger mice show more problems, adults of more than
10-weeks-old are preferable. Social contact experience, consisting
of 5- to 10-min encounters with an unfamiliar male and female
once a day for 35 days also provides stability to the results.
2. The optimal testing procedure of the transgenic mice study is
to breed heterogeneous males and females and to assign all the
316 Takefumi Kikusui

male siblings as the subjects. Therefore, the ratio of the testing

mice becomes WT 1:Hetero 2:Knockout 1. This procedure
can minimize so-called litter effects [36].
3. In the event that the opponent intruder attacks the subject resi-
dent, the data of this pair can be excluded because the aggres-
siveness of the subject mouse is clearly inhibited by the fight.
4. From our experience, some socially isolated males show very
quick and reflex-like aggression towards the intruder. The
latency of the first attack is less than a few seconds. In this case,
the aggressor would not discriminate the intruders sex and
individuality but simply attack what they encounter. It is pref-
erable to exclude these aggressors because their behavior is
mostly one of being conditioned to attack the intruder and is
not dependent on chemical discrimination.
5. If the female opponent shows escaping and upright posturing
towards the subject male, the data of this pair can be excluded
because the sexual behavior of the subject mice is clearly inhib-
ited by rejection from the female.
6. A hormone-primed female vaginal smear is a good indicator of
the success of the treatment. The smear reflects the stage of
estrus, which is characterized by marked cornification of the
cells and the disappearance of leukocytes.
7. Minimization of the noise level is important. The bedding
should be filter paper-type, which does not make excessive
noise when the mice are walking or digging. Further, filtering
the sound frequency, cut-off at 20 kHz, is a useful method for
increasing the sound/noise ratio.
8. The average of the number of trained aggressors is approxi-
mately 7080 %; that is, 2030 % of mice do not develop aggres-
sion. These mice, of course, can be excluded from the study.
9. Audition software, such as Avisoft Pro (Avisoft Vioacoustics)
and Adobe Audition (Adobe), have an automatic measure-
ment of these parameters. See the manufacturers guidelines.
10. Fresh urine is the most effective, but urine that has been fro-
zen at 20 C is still bioactive. Do not store the urine at higher
temperatures because volatile components will disappear.

References
1. Mugford RA, Nowell NW (1970) Pheromones ultrasonic courtship (Mus musculus). Anim
and their effect on aggression in mice. Nature Behav 2:333341
524:967968 4. Holy TE, Guo Z (2005) Ultrasonic songs of
2. Scott JP (1966) Agonistic behavior of mice and male mice. PLoS Biol 1:e386
rats: a review. Am Zool 6:683701 5. Scott J (1942) Genetic differences in the social
3. Nyby J, Wysocki CJ, Whitney G, Dizinno G behavior of inbred strains of mice. J Hered
(1977) Pheromonal regulation of male mouse 33:1115
 Analysis of Male Aggressive and Sexual Behavior in Mice 317

6. Crawley JN, Belknap JK, Collins A, Crabbe JC, duration of individual housing/grouping on
Frankel W, Henderson N, Hitzemann RJ, lntermale fighting behavior and adrenocortical
Maxson SC, Miner LL, Silva AJ et al (1997) activity in TO strain mice. Aggressive Behav
Behavioral phenotypes of inbred mouse strains: 4:307323
implications and recommendations for molecu- 20. Valzelli L (1973) The isolation syndrome in
lar studies. Psychopharmacology (Berl) mice. Psychopharmacologia 31:305320
132:107124 21. Sato T, Matsumoto T, Kawano H, Watanabe T,
7. Simon NG (1979) The genetics of intermale Uematsu Y, Sekine K, Fukuda T, Aihara K,
aggressive behavior in mice: recent research Krust A, Yamada T (2004) Brain masculiniza-
and alternative strategies. Neurosci Biobehav tion requires androgen receptor function. Proc
Rev 2:97106 Natl Acad Sci U S A 6:1673
8. Jones SE, Brain PF (1987) Performances of 22. Kippin TE, Pfaus JG (2001) The nature of the
inbred and outbred laboratory mice in putative conditioned response mediating olfactory con-
tests of aggression. Behav Genet 1:8796 ditioned ejaculatory preference in the male rat.
9. Selmanoff MK, Abreu E, Goldman BD, Behav Brain Res 1:1124
Ginsburg BE (1977) Manipulation of aggres- 23. Burns-Cusato M, Scordalakes EM, Rissman EF
sive behavior in adult DBA/2/Bg and (2004) Of mice and missing data: what we
C57BL/10/Bg male mice implanted with tes- know (and need to learn) about male sexual
tosterone in Silastic tubing. Horm Behav 3: behavior. Physiol Behav 2:217232
377390 24. De Vries GJ, Rissman EF, Simerly RB, Yang LY,
10. Maxson SC, Canastar A (2003) Conceptual Scordalakes EM, Auger CJ, Swain A, Lovell-
and methodological issues in the genetics of Badge R, Burgoyne PS, Arnold AP (2002) A
mouse agonistic behavior. Horm Behav 3: model system for study of sex chromosome
258262 effects on sexually dimorphic neural and behav-
11. Abramov U, Puussaar T, Raud S, Kurrikoff K, ioral traits. J Neurosci 20:90059014
Vasar E (2008) Behavioural differences 25. McGill TE, Manning A (1976) Genotype and
between C57BL/6 and 129S6/SvEv strains retention of the ejaculatory reflex in castrated
are reinforced by environmental enrichment. male mice. Anim Behav 3:507518
Neurosci Lett 3:223227 26. Green EL (1975) Biology of the laboratory
12. Shrenker P, Maxson SC (1983) The genetics of mouse. Dover Publications, New York, NY
hormonal influences on male sexual behavior of 27. Pomerantz SM, Nunez AA, Bean NJ (1983)
mice and rats. Neurosci Biobehav Rev 3: Female behavior is affected by male ultrasonic
349359 vocalizations in house mice. Physiol Behav
13. Brain PF, Benton D, Childs G, Parmigiani S 1:9196
(1981) The effect of the type of opponent in 28. Wolff RJ (1985) Mating behaviour and female
tests of murine aggression. Behav Processes choice: their relation to social structure in wild
4:319327 caught House mice (Mus musculus) housed in
14. Novotny M, Harvey S, Jemiolo B, Alberts J a seminatural environment. J Zool 1:4351
(1985) Synthetic pheromones that promote 29. Ginsburg B, Allee W (1942) Some effects of
inter-male aggression in mice. Proc Natl Acad conditioning on social dominance and subordi-
Sci U S A 82:20592061 nation in inbred strains of mice. Physiol Zool
15. Ropartz P (1968) The relation between olfac- 15:484506
tory stimulation and aggressive behaviour in 30. Brain P, McAllister KH, Walmsley S (1989)
mice. Anim Behav 1:97100 Drug effects on social behavior. Neuromethods
16. Rowe FA, Edwards DA (1971) Olfactory bulb 13:687739
removal: influences on the aggressive behaviors 31. Kletzkin M (1969) An experimental analysis of
of male mice. Physiol Behav 6:889892 aggressive-defensive behaviour in mice. In:
17. Brodkin ES, Goforth SA, Keene AH, Fossella Garattini S, Sigg EB (eds) Aggressive behavior.
JA, Silver LM (2002) Identification of quanti- Wiley, New York, NY, pp 253262
tative trait Loci that affect aggressive behavior 32. Nakamura K, Kikusui T, Takeuchi Y, Mori Y
in mice. J Neurosci 3:11651170 (2007) The critical role of familiar urine odor
18. Miczek KA, ODonnell JM (1978) Intruder- in diminishing territorial aggression toward a
evoked aggression in isolated and nonisolated castrated intruder in mice. Physiol Behav 23:
mice: effects of psychomotor stimulants and 512517
L-dopa. Psychopharmacology (Berl) 57:4755 33. Nakamura K, Kikusui T, Takeuchi Y, Mori Y
19. Goldsmith JF, Brain PF, Benton D (1976) (2008) Influences of pre- and postnatal early
Effects of age at differential housing and the life environments on the inhibitory properties
318 Takefumi Kikusui

of familiar urine odors in male mouse aggression. Touhara K (2010) The male mouse pheromone
Chem Senses 6:541551 ESP1 enhances female sexual receptive behav-
34. Kikusui T, Nakanishi K, Nakagawa R, Nagasawa iour through a specific vomeronasal receptor.
M, Mogi K, Okanoya K (2011) Cross fostering Nature 7302:118
experiments suggest that mice songs are innate. 36. Mendl M, Paul E (1991) Litter composition
PLoS One 3:e17721 affects parental care, offspring growth and the
35. Haga S, Hattori T, Sato T, Sato K, Matsuda S, development of aggressive behaviour in wild
Kobayakawa R, Sakano H, Yoshihara Y, Kikusui T, house mice. Behaviour 116:90108
 Chapter 24

Assessment of Urinary Pheromone Discrimination, Partner

Preference, and Mating Behaviors in Female Mice
Olivier Brock, Julie Bakker, and Michael J. Baum

Abstract
Behavioral testing methods are described for determining whether female mice can discriminate between
volatile urinary pheromones of conspecifics of the same vs. opposite sex and/or in different endocrine
conditions, for determining sexual partner preference, for quantifying receptive (lordosis) behavior, and
for monitoring the expression of male-typical mounting behavior in female mice.

Key words Estradiol, Progesterone, Main olfactory system, Vomeronasal organ, Lordosis, Mounting
behavior

1 Introduction

Modern studies of the roles of genes, hormones, and pheromones

in both the development and the adult expression of courtship
behaviors in female mice rely on behavioral testing methods that
produce results which are replicable among laboratories around the
world. Studies conducted in both the laboratories of co-authors,
Baum and Bakker [1, 2], over the past decade have assessed the
capacity of female mice to discriminate urinary odors from male vs.
female (or from testes-intact males vs. castrated males) conspecifics.
The results obtained have been used to assess sex differences and/
or gonadal hormone effects on pheromone detection and to assess
the effects of lesioning either the main olfactory epithelium and/or
the vomeronasal organ on sex discrimination capacity. We have also
assessed the preference of female mice to approach and investigate
volatile body (or urinary) odors and/or the sound and sight of male
vs. female (or testes intact vs. castrated male) [3, 4]. The results
obtained have been used to assess aspects of brain sexual differentia-
tion in female mice, including the development of circuits that
underlie females normal preference to seek out a male reproductive
partner. After a female approaches a male, she must be receptive to
the males mounts and intromissions in order both to receive

Kazushige Touhara (ed.), Pheromone Signaling: Methods and Protocols, Methods in Molecular Biology, vol. 1068,
DOI 10.1007/978-1-62703-619-1_24, Springer Science+Business Media, LLC 2013

319
320 Olivier Brock et al.

sperm from the male and the vaginal-cervical stimulation needed

to establish pregnancy. We [4, 5], and others [69], have developed
a sequence of tests for lordosis behavior to assess females receptiv-
ity. Finally, a full assessment of courtship capacity in female mice
includes their display of male-like mounting and pelvic thrusting
behaviors directed towards female conspecifics. Proestrous female
mice show considerable mounting behavior directed towards other
females; the functional significance of this mounting behavior is not
known. We have studied female mounting behavior in studies
designed to assess the role of perinatal estradiol signaling in male-
typical sexual differentiation [4] and the possible role of VNO/
pheromone vs. sex hormone signaling in the expression of this
male-typical behavior in females [10].

2 Materials

2.1 Urinary Odor 1. Silastic capsules for chronic administration of estradiol to ovari-
Discrimination Tests ectomized females are prepared using SILASTIC tubing
(Silclear Tubing, Medical Grade Silicone Tubing, length
15 m, Ref 2110150949, Degania Silicone, Israel) by cutting
1-cm-long capsules (inner diameter, 1.57 mm; outer diameter,
2.41 mm). Close one end of the capsule with SILASTIC medi-
cal adhesive (Type A, Dow Corning Corporation, USA) for no
more than 2 mm. Let the adhesive dry for 24 h. Fill the capsule
with a mixture (1:1) of crystalline 17-estradiol (cat. 101656,
MP Biomedicals) and cholesterol (C8667, Sigma) for 5 mm.
Close the remaining open end of the capsule with SILASTIC
medical adhesive. Let the capsule dry for 24 h, and pre-incubate
the capsule in saline at 37 C for 24 h before s.c. implantation
into subjects. The amount of steroid released from a SILASTIC
implant is directly proportional to the surface of the capsule and
inversely related to its thickness [11]. Our capsules produce
circulating levels of estradiol around 150 pg/ml in ovariecto-
mized female mice. This concentration is within the range of
values observed during proestrus in ovary-intact, cycling female
mice [12]. The capsules can be stored for several months in the
dark. They have to be protected from the light to prevent
degradation. Once implanted, these Silastic implants release a
constant amount of estradiol for up to 2 months. It is preferable
to remove the Silastic implant after this period to avoid side
effects like vaginal inflammation.
2. In order to induce full behavioral estrus, progesterone must be
administered s.c. to ovariectomized females previously given
estradiol. Prepare a solution of progesterone by mixing pro-
gesterone (P0130, Sigma) in sesame oil to get a final concen-
tration of 500 g/100 l. Let the solution stir overnight. The
progesterone solution can be kept several months in the dark.
 Assessment of Urinary Pheromone Discrimination, Partner Preference 321

3. These tests are carried out in standard clear plastic colony cages
(29 18 13 cm) in which the subject has lived alone for 48 h
prior to the test without a change in bedding. The food is
removed from the food hopper and the water bottle is removed
a few minutes prior to the behavioral test.
4. Urine is collected from testes-intact male mice of the same
strain as the female subjects and from castrated males as well as
cycling females when they are in vaginal proestrus or estrus.
Estrous female urine can also be collected from ovariectomized
females that are primed with estradiol followed by progester-
one to induce estrus (see details below). Urine can be collected
by holding the mice by the scruff of the neck over a funnel and
by pushing gently on their belly. Alternatively, a metabolic cage
can be used to collect urine. Urine from four mice of the same
sex/endocrine condition is pooled and mixed by vortexing.
Aliquots of the pools of urine are then frozen at 80 C in
separate vials for later use.
5. Square pieces of filter paper (5 5 mm) are taped to plastic
weigh boats (4.5 4.5 cm); a sufficient number of these uri-
nary odor presentation weigh boats are prepared in advance to
accommodate the total number of presentations of water or
urine in the actual tests.

2.2 Partner To assess partner preference using either visual, auditory, and/or
Preference Tests olfactory stimuli, we use a Plexiglas box (60 cm long 30 cm
high 13 cm wide) that is divided into three compartments by
placing either opaque (black) or transparent partitions and has a
perforated top. Each compartment is thus 20 cm in length. The
partitions contain perforated holes (0.5 cm diameter) at a height of
8 cm to facilitate the diffusion of odors from the two side compart-
ments to the middle compartment. Backside, floor, left and right
sides are made of black Plexiglas; front side and top are made of
transparent Plexiglas (see Note 1).

2.3 Lordosis Tests Lordosis tests are usually conducted in a Plexiglas aquarium (35 cm
long 25 cm high 19 cm wide) with a perforated top and whose
floor is covered with fresh sawdust. We also have used the home
cage (with food and water removed) of individual stimulus males in
which to assess females lordosis capacity after they are introduced.

2.4 Female Mounting 1. These tests are carried out in standard colony cages in which
Behavior Tests the subject has lived alone for 48 h prior to testing.
2. Female stimulus mice of the same strain as the female subjects
are ovariectomized at least 2 weeks prior to use. These stimu-
lus females are given a s.c. silastic implant of diluted estradiol
which is left in place for the duration of the study. Stimulus
females are given a s.c. injection of progesterone 34 h prior to
the onset of behavioral testing.
322 Olivier Brock et al.

3 Methods

All behavioral testing is carried out under either dim yellow lighting
or under red light in a separate room from the mouse colony during
the dark portion of the 12 light/12 dark day/night cycle. Mice are
singly housed beginning at least 1 week prior to any of the behav-
ioral tests. A video camera with infrared night vision detection is
used to record all behavioral tests. We either score behavior on
line as it occurs or by analyzing videos of the behavior after the
fact. We use Pocket PC and Noldus Observer Software to record
the frequencies and durations of the particular behaviors under
study. We use Sigma Stat software to carry out statistical analyses of
the results. Whenever possible, the treatment condition of differ-
ent subjects is concealed from the investigator who scores the
behavior under study.

3.1 Urinary Odor 1. Female subjects are typically ovariectomized several weeks
Discrimination Tests prior to assessing their capacity to detect/discriminate differ-
ent volatile urinary odors. Ovariectomy is performed under
general anesthesia using either a mixture of ketamine (80 mg/
kg per mouse) and medetomidine (Domitor, Pfizer, 1 mg/kg)
or continuous inhalation of 1 % isoflurane. Place the female
gently on her side and make a small incision, first in the skin,
then through the muscle layer. Lift the ovary and make a small
suture to clamp of the uterus from the ovary and to avoid any
bleeding. Then remove the ovary and suture the muscle layer,
followed by the skin. Then place the female on her other side
to repeat the same procedure. If females are going to be treated
with estradiol by means of a SILASTIC capsule, make a small
incision in the skin right below the neck, insert the capsule s.c.
and then suture the skin (see Note 2). Mice receive atipamezole
(Antisedan, Pfizer, 4 mg/kg s.c.) at the end of surgery in order
to antagonize medetomidine-induced effects, thereby acceler-
ating their recovery, an analgesic (Temgesic, Schering-Plough,
0.05 mg/kg s.c.) and are placed on a heating pad until they are
mobile. Females should be allowed to recover from the surgery
for at least 2 weeks before starting behavioral testing.
2. Beginning 2 weeks after ovariectomy, subjects can be tested in
the absence of hormone replacement if the investigator is not
interested in the activational effects of sex hormones on
females urinary odor discrimination ability. Otherwise, the
female subjects should be brought into behavioral estrus at the
time of odor discrimination assessment. Subjects that received
a s.c. Silastic implant of estradiol at the time of ovariectomy can
be given a single s.c. injection of progesterone (500 g/
mouse) 3 h prior to the onset of testing to induce full behav-
ioral estrus.
 Assessment of Urinary Pheromone Discrimination, Partner Preference 323

3. Aliquots of urine are thawed whereupon 10 L of a particular

kind of urine is pipetted onto the filter paper attached to a
weigh boat just prior to presentation. The weigh boat is placed
behind the wire mesh of the cages food chamber so that the
subject is able to approach the stimulus with its nose coming
within 5 mm without direct contact. No direct contact with
the urinary stimulus with either snout or paws is possible. As a
result, the subject is only exposed to volatiles emitted from the
respective urinary stimuli; presumably only the main olfactory
epithelium (and not the VNO) is involved in the detection of
these odorants.
4. Tests of urinary odor discrimination begin with three consecu-
tive 2-min presentations (1-min intervals between each stimu-
lus) of deionized water. Beginning 1-min after presentation of
the third water stimulus the subject is presented with three
consecutive 2-min presentations of testes-intact male urine (at
1-min intervals; urine #1) followed by three consecutive 2-min
presentations of either castrated male urine or estrous female
urine (depending on the study; urine #2). This presentation
sequence of type of urine can be varied either across subjects in
a particular group during one test day. Alternatively, subjects
can be retested with the reverse sequence of urine types on a
separate day.
5. An observer uses a computer and Noldus software to record
the number of seconds that each subject spends with its nose
against the wire mesh grid abutting the odor stimulus during
each 2-min stimulus presentation. Nonparametric one-tailed
within-groups Wilcoxon tests are used to compare the time
spent investigating the third presentation of water vs. the first
presentation of urine #1 and to compare time investigating the
third presentation of urine #1 vs. the first presentation of urine
#2. Significant differences in each of these comparisons indi-
cate whether or not subjects could perceive the newly intro-
duced odorant from the previous one (see Note 3).
6. Subjects performance in habituation/discrimination tests
relies on the intrinsic motivation that mice have to explore uri-
nary odors from conspecifics. As such, observed group
differences in detection performance in this type of test may
reflect a complex interaction between social motivation and
odor perception (see Note 4 for further discussion).

3.2 Partner 1. Allow each subject to adapt to the three compartment box
Preference Tests once on the day before the onset of preference testing by plac-
ing them in the middle compartment for 10 min (with no
stimulus animals placed in the two side compartments).
2. On the day of testing, place stimulus animals (e.g., an estrous
female vs. a testes-intact male or, alternatively, a testes-intact
324 Olivier Brock et al.

male vs. a castrated male) along with their own bedding to

make the stimulus as odorous as possible into the two respec-
tive side compartments of the apparatus (see Note 5).
3. Then introduce the experimental subject into the middle com-
partment containing no sawdust, and observe it for 9 min.
Record the time the subject spends poking its nose through
the holes of each side partition or actively sniffing the bottom
of the partition. Each test session is divided into three 3-min
intervals to determine whether investigation times decrease
during the test (see Note 6).
4. Prior to testing each new subject, clean the middle compart-
ment with Norvanol while leaving the soiled bedding in the
two side compartments. Place a new experimental subject in
the middle compartment, and repeat the procedure until all
experimental animals are tested (see Note 7).
5. For each female subject, calculate a preference score (PS) for
the male by dividing the time spent investigating the testes-
intact male compartment minus the time spent investigating
the compartment housing either the estrous female or the cas-
trated male (depending on which of these alternative choices
to the testes-intact male is given to the female subject) divided
by the total time spent investigating both compartments.

3.3 Lordosis Tests 1. At least 1 week before testing experimental female mice, make
sure that a sufficient number of high quality stimulus males are
available (see Note 8). Stimulus males are given mating experi-
ence, as follows: Place a stimulus male in the Plexiglas testing
aquarium for 30 min. Then introduce a stimulus female, which
has been implanted with an estradiol capsule for at least 1 week
and primed with progesterone (500 g/mouse; injected i.p.)
3 h before testing, into the aquarium. In order to optimize its
utility as a stimulus male, he should mount the stimulus female
within 5 min and also display intromissions with the female.
After 10 min, or a maximum of 10 intromissions, remove the
stimulus male and place him back into his homecage (see Note
9). Repeat this procedure on 2 or 3 consecutive days; males
that mate each day will later be useful for lordosis testing.
2. At the beginning of each lordosis test, place a sexually experi-
enced male mouse alone in the test aquarium for 15 min of
habituation. No habituation is needed if tests are to be con-
ducted in the males home cage.
3. Beginning 3 h after receiving a s.c. progesterone injection
(500 g ip), place a female subject in the test aquarium and
record her lordosis responses displayed in response to the
mounts (with pelvic thrusting) received from the stimulus
male. The test lasts until the female receives 10 mounts or
 Assessment of Urinary Pheromone Discrimination, Partner Preference 325

10 min have elapsed (see Note 10) after which the subject is
replaced in her homecage (see Note 11). Repeat this process,
using fresh stimulus males, as needed in order to sustain the
mounting attempts needed to assess females lordosis respon-
siveness (see Note 12).
4. After each test, calculate a subjects lordosis quotient (LQ) by
dividing the number of lordosis responses displayed by the
female subject by the number of mounts received (100). A
minimum of four tests is normally required to observe signifi-
cant levels of lordosis behavior and to reach plateau values
(LQ = 6070 %; [5]). Unlike rats, female mice rarely achieve
LQs of 100 %. Lordosis testing should be performed at 34 day
intervals (see Note 13).

3.4 Female Mounting 1. Ovariectomized female subjects are given testosterone either
Behavior Tests via daily s.c. injections (testosterone propionate dissolved in
sesame oil; TP, 3 mg/kg) or via s.c. implantation of a Silastic
capsule (inner D 1.57 mm; outer D 2.41 mm) containing a
5 mm length of undiluted powdered testosterone (see Note 14)
for at least 7 days prior to the onset of tests for mounting
behavior.
2. An ovariectomized stimulus female that is brought into behav-
ioral estrus via estradiol and progesterone treatment is placed
into the home cage of the testosterone-primed female subject
for 30 min. An observer scores the occurrence of ano-genital
investigations as well as mounts accompanied by pelvic thrust-
ing movements directed towards the receptive stimulus female
(both the frequency and duration in sec. of each of these behav-
iors is recorded). The number of pelvic thrusting movements
associated with each mount is also recorded (see Note 15).
3. The following variables are summarized and subjected to sta-
tistical analyses: the percentage of female subjects in each treat-
ment group that show mounts directed towards a stimulus
mouse during any test, the total frequency and total duration
(in sec.) of mounts directed towards a stimulus mouse during
a 30-min test (individual means are computed, followed by
grand group means if more than one behavioral test is given),
and the total number of pelvic thrusts displayed by the subject
over a 30 min test.

4 Notes

1. Another device commonly used in the literature to assess odor

preferences is an enclosed, Plexiglas Y-Maze [13, 14]. It con-
sists of the stem of the Y (55 cm long) and two arms (65 cm
long) that diverge at 60 angle from the stem. All parts of the
326 Olivier Brock et al.

maze are 9 cm high and 9 cm wide. A removable perforated

Plexiglas door at the distal end of the each arm separates the
goal box from the rest of the maze. We use either an opaque
(to prevent visual cues) or a clear (to allow visual cues) Plexiglas
door. A perforated Plexiglas door is placed at the other side of
the goal box. The start box with a removable perforated
Plexiglas door is located at the base of the stem. An electric fan
is placed behind the start box, from which it is separated by a
wire screen. The perforated doors and the fan make it possible
to pull air over odor stimuli placed in the distal goal boxes
through the entire maze into the start box. We decided not to
use this device anymore for one major reason: the behavior
observed in the Y-maze can be confounded by differences in
exploratory activity. For example, after gonadectomy, C57Bl6
male and female mice become anxious in the Y-Maze and show
a strong decrease in exploratory activity (most of the time,
mice even do not approach the goal boxes). By contrast,
gonadectomized CD1 mice are hyperactive in the Y-Maze and
show very little interest in remaining in proximity to a specific
stimulus.
2. After surgery, house female mice individually for a few days in
order to avoid that they will nibble each others sutures which
will increase the risk of prematurely losing their capsule.
3. Subjects detection thresholds for different urinary odors can
be established using a variation on this method. On successive
days subjects ability to detect volatile odorants emitted from
progressively lower dilutions of one type of urine (e.g., testes-
intact male urine) following three presentations of deionized
water can be assessed. In two studies [1, 2] long-term gonad-
ectomized female subjects (no hormone replacement) success-
fully detected the urinary volatiles emitted from a 1:80 dilution
of male urine (but not from a 1:160 dilution of male urine)
whereas long-term castrated males (no hormone replacement)
successfully detected volatiles emitted from a 1:20 (but not
1:40) dilution of male urine.
4. If the investigator is concerned that the outcome of a habitua-
tion/dishabituation study reflects group differences in sexual
motivation as much as differences in odor perception capacity,
studies may conducted in which hunger serves as the motiva-
tion for subjects performance in an olfactory discrimination
task. In one such two-choice odor detection task [15] the abil-
ity of food-deprived mice to use decreasing concentrations of
male urine as a discriminative stimulus to find a food pellet
buried in sand was assessed. In another instance [16], a go/
no-go task (urinary volatiles were presented consecutively
using an 8-channel liquid dilution olfactometer) was used to
compare the ability of mice to discriminate volatiles emitted
Assessment of Urinary Pheromone Discrimination, Partner Preference 327

from the urine of testes-intact males, castrated males, and

ovariectomized females given either estradiol alone or
estradiol + progesterone.
5. To assess partner preference based on subjects visual, audi-
tory, and olfactory senses, use transparent partitions. To assess
preference based on auditory and/or olfactory stimuli, use
opaque partitions. In the majority of published experiments
female subjects are given a choice between a testes-intact
male and an estrous female. Depending of the experimental
question, a choice between a testes-intact male and a cas-
trated male or an estrous female and an ovariectomized
female can be offered. Anesthetized stimulus animals can also
be used to present the subject with only olfactory cues that
signal a partner preference.
6. Experimental subjects of some strains tend to investigate more
during the first few minutes of the test. However, if no decrease
in investigation time is observed across the entire test, the total
test time spent investigating the two stimuli is calculated.
7. After being used in several test sessions, some stimulus animals
will sleep in the corner of their respective side compartments.
These stimuli should be replaced with new, active (awake)
stimulus animals.
8. Note that not all male mice are good breeders. Some make
no effort to mount females or are very aggressive (attack) the
female subjects. Thus before starting lordosis tests, make sure
that you have a large batch of sexually active male mice which
are not aggressive towards females.
9. Once male mice have experienced sexual behavior with a
female, house them individually, otherwise they will fight to
the death.
10. The receptivity of a female mouse can be scored as follows: (a)
the female runs away and/or responds aggressively in response
to a males mounting attempt; (b) the female does not run
away but rejects (kicks at the male) the mount; (c) the female
accepts the mount but does not respond by showing lordosis
(concave curvature of the back); (d) female responds by show-
ing full lordosis without any resistance directed towards the
stimulus male. In some strains (or transgenic models), female
mice can require several mounting attempts in order to reach
stage d. In these particular cases, you can extend the test to
20 mounts or 15 min. The LQ should be based on the number
of times the female subject displays full lordosis (d).
11. During the course of lordosis testing, gently separate the male
from the female after the first intromission occurs in order to
retard the occurrence of ejaculation. Once a male has ejacu-
lated, he will not show any interest towards the subsequent
328 Olivier Brock et al.

female subjects. If a male ejaculates with a female, replace him

immediately with a new stimulus male in order to obtain the
requisite number of mounts needed to compute the LQ for
that test.
12. Do not change the bedding on the floor of the test aquarium
between female subjects. Change it only before each daily test
session.
13. You can easily score two females at the same time. The optimal
procedure is to use three aquariums, each containing one male.
Introduce one female into each of the first two aquariums.
If one of these females is not mounted by the stimulus male,
switch her into the third aquarium.
14. We have used both daily injections of TP [10] and a s.c. Silastic
capsule [4] of testosterone to elevate circulating plasma levels
of testosterone up to that characteristic of testes-intact males.
Depending on whether subjects are to be tested under other
endocrine conditions, investigators may choose either of these
methods to administer testosterone.
15. Depending on the aim of our studies, we have administered
16 tests of mounting behavior in female subjects that were
ovary intact or ovariectomized and treated with estradiol alone,
estradiol + progesterone, or testosterone. In some instances we
have also recorded mounting behavior that female subjects
direct towards stimulus males that had been castrated several
weeks previously. These males had urine from testes-intact
males smeared on their backsides. These castrated males rarely
attempted to mount our female subjects.

Acknowledgment

Preparation of this review was supported, in part, by FRSM grant

3.4571.10 to J.B.

References
1. Baum MJ, Keverne EB (2002) Sex difference 4. Brock O, Baum MJ, Bakker J (2011) The
in attraction thresholds for volatile odors from development of female sexual behavior requires
male and estrous female mouse urine. Horm prepubertal estradiol. J Neurosci 31:
Behav 41:213219 55745578
2. Pierman S, Douhard Q, Balthazart J, Baum 5. Bakker J, Honda S, Harada N, Balthazart J
MJ, Bakker J (2006) Attraction thresholds and (2002) The aromatase knock-out mouse pro-
sex discrimination of urinary odorants in male vides new evidence that estradiol is required
and female aromatase knockout (ArKO) mice. during development in the female for the
Horm Behav 49:96104 expression of sociosexual behaviors in adult-
3. Brock O, Bakker J (2011) Potential contribu- hood. J Neurosci 22:91049112
tion of prenatal estrogens to the sexual differ- 6. Mani SK, Reyna AM, Chen JZ, Mulac-
entiation of mate preferences in mice. Horm Jericevic B, Conneely OM (2006) Differential
Behav 59:8389 response of progesterone receptor isoforms in
 Assessment of Urinary Pheromone Discrimination, Partner Preference 329

hormone-dependent and -independent facili- one propionate in the castrated male rat.
tation of female sexual receptivity. Mol Horm Behav 8:7787
Endocrinol 20:13221332 12. Wersinger SR, Haisenleder DJ, Lubahn DB,
7. Ogawa S, Chan J, Chester AE, Gustafsson JA, Rissman EF (1999) Steroid feedback on
Korach KS, Pfaff DW (1999) Survival of repro- gonadotropin release and pituitary gonadotro-
ductive behaviors in estrogen receptor beta pin subunit mRNA in mice lacking a functional
gene- deficient (betaERKO) male and female estrogen receptor alpha. Endocrine 11:
mice. Proc Natl Acad Sci U S A 96: 137143
1288712892 13. Keller M, Douhard Q, Baum MJ, Bakker J
8. Ogawa S, Eng V, Taylor J, Lubahn DB, Korach (2006) Destruction of the main olfactory epi-
KS, Pfaff DW (1998) Roles of estrogen thelium reduces female sexual behavior and
receptor-alpha gene expression in reproduc- olfactory investigation in female mice. Chem
tion- related behaviors in female mice. Senses 31:315323
Endocrinology 139:50705081 14. Pankevich DE, Cherry JA, Baum MJ (2006)
9. Rissman EF, Early AH, Taylor JA, Korach KS, Effect of vomeronasal organ removal from
Lubahn DB (1997) Estrogen receptors are male mice on their preference for and neural
essential for female sexual receptivity. Fos responses to female urinary odors. Behav
Endocrinology 138:507510 Neurosci 120:925936
10. Martel KL, Baum MJ (2009) Adult testoster- 15. Sorwell KG, Wesson DW, Baum MJ (2008)
one treatment but not surgical disruption of Sexually dimorphic enhancement by estradiol
vomeronasal function augments male-typical of male urinary odor detection thresholds in
sexual behavior in female mice. J Neurosci mice. Behav Neurosci 122:788793
29:76587666 16. Wesson DW, Keller M, Douhard Q, Baum MJ,
11. Smith ER, Damassa DA, Davidson JM (1977) Bakker J (2006) Enhanced urinary odor dis-
Plasma testosterone and sexual behavior fol- crimination in female aromatase knockout
lowing intracerebral implantation of testoster- (ArKO) mice. Horm Behav 49:580586
 Chapter 25

Assessing Postpartum Maternal Care, Alloparental

Behavior, and Infanticide in Mice: With Notes
on Chemosensory Influences
Kumi O. Kuroda and Yousuke Tsuneoka

Abstract
Chemosensory signaling influences maternal care and other innate behaviors toward conspecific young
animals in rodents. In this chapter, we describe basic protocols for assessment of postpartum maternal
behavior and other pup-directed behaviors in laboratory mice. The specific aim of this protocol is to screen
out the abnormal phenotypes in parenting of genetic mutant mice under the standard housing condition.
The possible underlying mechanisms for a given abnormality in the mother-young interaction are briefly
suggested as well.

Key words Parental care, Paternal behavior, Infanticide, Mus musculus, Olfaction, Pup retrieval assay,
Nest building

1 Introduction

1.1 Parental Maternal behavior is defined as the collection of behaviors by the

Behavior and Related mother that can increase offspring survival [1, 2]. Similar nurtur-
Pup-Directed ing behaviors as maternal behaviors, called as paternal behavior by
Behaviors in Mice fathers and alloparental behavior by older conspecifics, are widely
seen in mammals. In this review, we will collectively refer to these
maternal, paternal and alloparental behaviors as parental behav-
ior. In addition to these nurturing behaviors, mice also perform
negative pup-directed behaviors, such as nonresponding (the sub-
ject mouse sniffs the pup at first, but stay away for most of the
time), avoidance (moving away from the pup or burying the pup in
the bedding) and infanticide (pup biting/killing, often but not
necessarily combined with cannibalism). Infanticide is most fre-
quent in virgin (sexually nave) male mice. In C57BL/6J, a stan-
dard inbred strain with high sociability, the 7080 % of virgin males
commit infanticide even after repeated pup exposure (see Note 1).
Once the same male has mated with a female and cohabitates with
the pregnant mate, however, he eventually stops infanticide by the

Kazushige Touhara (ed.), Pheromone Signaling: Methods and Protocols, Methods in Molecular Biology, vol. 1068,
DOI 10.1007/978-1-62703-619-1_25, Springer Science+Business Media, LLC 2013

331
332 Kumi O. Kuroda and Yousuke Tsuneoka

time of delivery of his biological offspring. And at this time, the

father mice provide paternal care even toward nonbiological off-
spring [3, 4] (see Note 2).

1.2 Components Among the components of rodent parental behavior listed in

of Mouse Parental Table 1, pup-retrieval behavior is widely used as an index of paren-
Behaviors tal responsiveness in both rats and mice (e.g., [57]). The latency
to retrieve each pup is easily and unambiguously measurable, and
can be assessed not only in postpartum mothers but also in non-
lactating females and males, as discussed in Subheading 3.2.
Assessment of the nest quality is also a preferred measure in mice.
For postpartum mothers, their ability of placentophagia and provi-
sion of maternal milk should be assessed as well. Therefore in this
chapter, the basic protocols for evaluation of postpartum maternal
behaviors and pup-retrieval behavior of mon-mothers are intro-
duced. The presented protocol is compatible with the common
practices of mutant mouse husbandry in the SPF condition, so that
it may be useful for the initial screening of parental responsiveness
in mutant mouse strains. For more detailed information on paren-
tal behaviors in rats and other mammalian species, please refer the
previous literature [812], and for further discussions and for listing
gene mutant strains implicated in this topic, refer [13].

1.3 Effects of Pup-directed behaviors described above are largely dependent on

Chemosensory olfaction, in many mammalian species including rodents [14]. In
Signaling on the particular, the mouse parental behavior is totally dependent on the
Pup-Directed main olfactory function (see Notes 3 and 4). Anosmia caused either
Behaviors by postnatal surgery or by genetic mutation strongly inhibits most
of parental behaviors in both virgin and postpartum mother mice.
For the best studied example, the Adcy3 gene encodes the type III
adenylyl cyclase, which coupled with Golf, both of which are
required for sensory transduction of the main olfactory epithelium.
Adcy3 homozygous mutant (/) mice were anosmic, although
some odorants could be detected through the VNO [15]. Adcy3
(/) mutants are initially smaller than the wild-type littermates
but catch up after 3 months of age. Adcy3 (/) females are fertile
and show normal placentophagia after parturition, but also exhibit
severe deficits in pup retrieval and nest building, causing the majority
of their pups death within 2 days [16]. Maternal aggression of
Adcy3 (/) mothers is severely disturbed too. Furthermore,
Adcy3 (/) virgin females as well are almost devoid of a pup
retrieval response. Consistently, the anosmic Golf homozygous
mutant mothers fail to retrieve pups and to crouch over the pups,
so that all pups of four litters died without milk in their stomachs
by postnatal day 2 [17]. The authors did not report the infanticide
in these studies.
The accessory olfactory function is not required for pup retrieval
behavior or nest building, but is required for maternal aggression
 Assessing Postpartum Maternal Care, Alloparental Behavior, and Infanticide in Mice 333

Table 1
Components of maternal behaviors in laboratory mice and rats

General References for

Description references non-mothers
Pup-directed behaviors
Nursing Provision of the opportunity to suckle by [60] Rat female [61].
exposing the nipples and being immobile in
various nursing postures, such as high
crouch, low crouch or lying positions.
Retrieval Picking up the pup gently by a part of the body [35, 62] Rat [5, 63].
(most commonly, the dorsal skin) with the Mouse female
incisors and carries it to the nest site. [38]; male [3].
Grouping Gathering the pups together into one quadrant [64, 65]
so that they touched one another.
Anogenital Licking the anogenital region of the pup, [66] Rat female [67].
licking which is stimulatory for pups urination and
defecation.
Body licking Licking the pups body generically except for
the anogenital region.
Tactile Any contact with pups, such as stepping on [28, 68] Mouse pups with
stimulation pups or resting in contact with pups. rat aunt [69].
Non-pup directed behaviors
Placentophagia Elimination and ingestion of placenta, [70, 71] Rat female [72].
umbilical cords, amniotic membrane and Mouse female
fluids attached to pups body, by vigorous [73].
licking after delivery. This makes the pups
body dry and clean.
Nest building Transporting the nesting materials toward the [74, 75] Mouse female
nest or manipulating the material to shape [51].
the enclosed nest edge.a
Defense of the Protection of the pups from intruders, [76, 77] Mouse female
young predators and environmental hazards. It is [78].
called as maternal aggression, if the target
of maternal defense is unfamiliar conspecifics.
a
Nonpregnant mice produces a small sleeping nest, while females at late pregnancy make a bigger and more
complex nest, termed a brood nest. Brood nest building starts from one to a few days before parturition,
continues for the first 2 weeks of lactation and declines

against intruders as shown by surgical VNO removal in mice [18].

This finding is supported by the reports of two genetic mutant
mouse strains, Trpc2 (transient receptor potential cation channel,
subfamily C, member 2) mutant [19] and Del(6)1Mom mutant
(lacked a cluster of vomeronasal pheromone V1 receptor family)
[20]; these mutants were fertile and showed normal pup retrieval
behavior, but exhibited less maternal aggression. In addition, the
334 Kumi O. Kuroda and Yousuke Tsuneoka

VNO dysfunction abolished the infanticide in virgin male mice

(Tachikawa, manuscript in preparation), as well as the aversive
responses toward pups in rats [21, 22]. Surgical ablation of VNO
and the Trpc2 mutant mice also caused the reduction of inter-male
aggression as well, although Del(6)1Mom mutants display normal
inter-male aggression, suggesting the possible correlation of mater-
nal aggression and infanticide (pup-directed aggression) with the
inter-male aggression. Clearly, all of these aggressive behaviors
toward conspecific animals are largely dependent on the phero-
mone signaling.

2 Materials

2.1 Cage System, The standard shoebox breeding cages (approx. size of
Bedding and Nest 265 mm 205 mm, 140 mm high) with automated ventilation and
Materials water-supply in the SPF condition can be used for the behavioral
assay. For breeding the subject mice, normal cage bedding materi-
als can be used. For the assay of the pup-directed behaviors, how-
ever, we use paper chips made from purified pulp paper-pulp (e.g.,
alpha-dri, Shepherd) as cage beddings. Wood-chip bedding may
affect the outcome of parental behavior, as reported for the mice
lacking Fyn tyrosine kinase [23, 24] and FosB transcription factor
[25]. Hexanal, a volatile substance contained in plants and causing
a grassy odor, was the responsible chemical component for this
effect at least in fyn (/) [24]. The woods used to make the same
wood-chip bedding product often vary by season. Contents of
chemicals such as hexanal significantly vary between the types of
wood used, and between the treatment of chips (autoclaving, addi-
tion of pesticides and so on). As mouse pup-directed behaviors are
very sensitive to chemosensory signals, the wood chips are less suit-
able than quality-controlled, purified paper chips.
The addition of nest material makes it easy to identify the loca-
tion and quality of the nest [26]. Normal cotton pads, balls, or thin
paper strips can be used, but the compressed cotton piece (e.g.,
Nestlet, Ancare, Bellmore, NY) is ideal. Adult mice normally bite
and tear this densely packed cotton sheet extensively into fluffier
pieces to make their nest within a couple of days. If this square
piece of Nestlet has not been torn and remains in its original form,
it is highly suspected that the subject mouse should have some sort
of health problems, or has serious defects with nest building
behavior.

2.2 Genetic The confounding effects of genetic background on behavioral

Background phenotypes have been acknowledged [27], complicating the inter-
pretation of findings. Also for maternal behavior, strain difference
among different substrains has been reported [28, 29]. If the original
 Assessing Postpartum Maternal Care, Alloparental Behavior, and Infanticide in Mice 335

mutant strain was constructed using 129Sv-derived embryonic

stem cells, it is often preferable to backcross the mutant strain into
C57BL/6, a standard congenic strain for neuroscience research
(see Note 5). Backcrossing is usually performed by mating a het-
erozygous female with a C57BL/6 male, and the heterozygous
female offspring is selected to mate with a C57BL/6 male again.
After five generations of backcross, the 96.875 % of the whole
genome is C57BL/6 background. However, genetic complica-
tions due the flanking-gene effect; that is, the closely linked genes
surrounding the targeted locus tend to remain even after 20 gen-
erations of backcrossing, and affects the apparent phenotypic dif-
ferences between the mutant (the flanking genes are of the original
background) and the wild-type (the flanking genes are of C57BL/6
derived). This problem can be properly addressed by the specific
breeding strategies proposed previously [30].

2.3 Production Here we describe the procedure for detecting the autosomal reces-
of Subject Animals sive phenotypes in adult parental behaviors of conventional tar-
geted disruption (knockout, KO) of single genes as an example.
The protocol should be appropriately modified for mutations that
follow the other patterns of inheritance, such as X-linked or domi-
nant mutations, conditional KOs, or parent-of-origin specific gene
expression (genomic imprinting).
1. Cohabitate one to three heterozygous (+/) female mice of
1012 weeks old with one (+/) male of 1224 weeks old in
one breeding cage (see Note 6).
2. According to the reproductive success, make a few to several
breeding cages at a time, aiming at production of 6 ~ 8 homo-
zygous (/) and the wild-type (+/) subject females of simi-
lar ages (see Note 7), to be tested at a time (see Note 8).
3. Remove visibly pregnant females into a separate cage and check
for delivery every morning.
4. The pups should be weaned at a determined time window after
the birth (see Note 9) and group-housed until the behavioral
experiments.
5. (a) To test adult virgin mice for their pup-directed behaviors,
single-house subject females and males at 10 weeks and
12 weeks of age, respectively, in a new cage containing paper
bedding with a cotton square as nest material described in
Subheading 2.1. Measure the body weight of each animal at
this point would be informative.
(b) To test postpartum maternal behaviors, follow the next
Subheading 2.4 to make the females pregnant, and single-
house the pregnant females as above.
336 Kumi O. Kuroda and Yousuke Tsuneoka

2.4 Breeding 1. To compare the effects of maternal genotype on maternal

Strategy for behaviors, crosses between (+/+) males with (/) subject
Postpartum Maternal females and between (/) males with (+/+) subject females
Behavior Tests should be conducted, resulting in all the pups having the same
genotype (+/). If this is not possible because of the infertility
of (/) males, then crosses between (+/+) females with (+/)
males can be conducted. The principle is to avoid the use of
(/) pups as stimuli for testing maternal behavior, and also
trying to equalize the pups genotype for (/) and the con-
trol (+/+) subject mothers. Mating should preferably start
with the 1012 weeks of age for primiparous maternal behav-
ior testing. Checking the females for a vaginal plug early in the
morning following mating provides information about the
possible delivery date and also the sexual behavior of this
female, although this is not essential for maternal behavior
assessment itself.
2. Once the female gets visibly pregnant, it is isolated in a new
cage containing paper bedding with a cotton square as
described above.
3. After the female has been moved into this cage, the bedding
should not be changed during the first week of lactation, so as
not to disturb the mother and pups. In case the bedding gets
too dirty during this first week, one can remove the dirty part
of the bedding manually, and add the same amount of new
bedding, but one should avoid doing this during the peripar-
tum period.

3 Methods

3.1 Assessing The postpartum examinations should be done in this presented

Postpartum Maternal order, to minimize the stress of the mother. Throughout these
Behavior on the First observations, we try not to take mouse mothers out of the cage
Morning After Delivery (see also [11]). To collect pups for retrieval test or for pup examina-
tion, if she is in the nest over pups, we wait for her to move away
or gently push her away. Nor do we suspend her by the tail com-
pletely in the air, but let her forelimbs attached on the ground or
the cage cover. Suspending a peripartum female by the tail may
induce a stress reaction, which sometimes result in the mother
destroying her nest or attacking her pups.
1. Parturition check: Around the estimated day of delivery, the
cage should be checked for parturition every morning (see Note
10). With our animal husbandry of a 12:12-h light/dark cycle
with lights on from 08:00 to 20:00, we check for delivery
between 9 and 10 am every morning. This is because, after this
time, the pups born in the previous dark phase will get weaker if
the milk intake is not sufficient by any reason. If the pups get too
Assessing Postpartum Maternal Care, Alloparental Behavior, and Infanticide in Mice 337

weak and hypothermic, the maternal responses will decrease

and sometimes the mother may cannibalize the dying pups,
obscuring the cause of pups death. For the same reason, if one
wants to cross foster the pups to rescue them, the success rate
decreases after this time. In the afternoon it would be difficult to
make a foster mother accept these weakened and chilled pups.
On the other hand, if the mother is still in delivery, one
should not disturb the cage. It is normal for the nest to get
destroyed during delivery, as the mother circulates and moves
restlessly in the small cage during delivery of each pup. It is bet-
ter to wait for about an hour after the delivery of the last pup,
until the mother settles down and finishes placentophagia,
rebuilding the nest, pup retrieval, and stays with the pups in the
nest to nurse them, before starting the ratings described below.
2. Nest grading:(see Note 11) If the parturition has been ended,
first evaluate the nest quality from outside of the cage, so as
not to disturb the mother. The nest is rated as 0 when there is
no nest, or the nest location is unclear because the nest material
is distributed randomly in the cage. The nest is rated as 1 when
the nest is flat and not well focused. Still the grade-1 nest
should be able to be identified in the cage. The nest is 2 when
the nest is similar to a shallow soup bowl [31]. The nest is graded
as 3 when the nest is shaped into a hollow surrounded by a
continuous bank (designated as incomplete dome or half of a
sphere in [31]). With the optimal nest material such as fluffy
paper strips, a completely enclosed nest, rated as 4 can be
achieved (full dome [31]). In such nest, the pups are scarcely
visible from any directions. The bedding is gathered toward
the corner of the nest site, so that the nest floor is higher than
the floor of the dirty corner of the cage, which the mother uses
as an area for defecation and urination. With about 250 mL
(one cup) of the paper-chip bedding and one piece of
5 cm 5 cm Nestlet per a shoebox mouse cage as described
here, however, this grade-4 nest is rather rare. Increasing the
amount of nest material may cause mechanical troubles of
automated water-supply and individual cage ventilation sys-
tems. If there are two clear nests in one cage, the grading can
be made separately for each nest.
3. Pup grouping evaluation: Pups location in relation to the nest
and other pups (=if they are in body contact with other pups)
should be briefly recorded, by examining them from outside of
the cage (e.g., three alive pups in the nest and grouped, two
dead? pups buried under the bank of the nest, one alive pup
outside of the nest) (status of each pup will be examined later
in more detail, so that a brief note is sufficient at this moment).
Ideally all the pups should be in the nest and tightly grouped
(huddled) with each other. When any the pups are out of nest or
338 Kumi O. Kuroda and Yousuke Tsuneoka

buried in the bottom or the bank of the nest (better observable

from the bottom of the clear cage), the pups condition (e.g.,
healthy and pinky, pale, or dead) should be examined thor-
oughly later as in the pup examination paragraph. Again, it is
not abnormal maternal behavior to leave dead or dying pups out
of the nest, and/or cannibalize them.
4. Pup retrieval observation: To quantify pup retrieval latency,
remove the cage top, and gently take three healthy pups from
the nest, and place one pup in each corner of the cage outside
the nest (Fig. 1a). Then return the cage cover. The female and
pups are observed continuously for 10 min and the following
measures were recorded: latency to sniff a pup for the first time
(see Note 12), to retrieve each pup into the nest, group all
pups, and crouch over the pups continuously for >1 min. Pup
carrying to the other place than the nest should be recorded
separately. When the mother has finished retrieval and group-
ing, and has crouched over all the pups in the nest for more
than 1 min, it is called as full maternal behavior (note that
the criteria are slightly different in each literature [3234]).
The latency to show nest building or pup licking can also be
recorded, although because of the small size, these behaviors
in mice are more difficult to be distinguished from pup
regrouping, pup sniffing or self-grooming compared with
those in rats.
If most of the pups are found in the nest at the initial check
from outside, one can assume that the mother has already
exhibited pup retrieval, because newborn pups are not able to
group themselves together. When the pup are found grouped
but the mother does not retrieve any pup during the pup
retrieval observation, it is suspected that the mother is stressed
by the current experimental maneuver. In that case, gently

Fig. 1 Pup retrieval assay (a) and the milk band (b)
Assessing Postpartum Maternal Care, Alloparental Behavior, and Infanticide in Mice 339

placing the cage back to the husbandry shelf, where it tends to

be darker, and observe another 15 min for retrieval and full
maternal behavior. If the mother shows retrieval, the delay of
retrieval may have been caused by stress hypersensitivity rather
than by lack of maternal responsivity itself.
In case where pups are visibly unhealthy because of inap-
propriate placentophagia, lack of milk intake, low body tem-
perature or bleeding, these pups cannot be used for the stimulus
pups in the pup retrieval test. In this case, the test can be made
using the healthy pups (properly cleaned, milk in stomach and
warm, preferably of the same genetic background) of the similar
age from another mother (see Note 13). It is often seen that the
mutant mother with insufficient milk secretion appears to aban-
don her weakened pups, but when fresh healthy donor pups are
introduced, the same mother shows a quick retrieval response.
Formally in this case, the control mothers should also be
exposed with donor pups for retrieval scoring, to equalize the
experimental conditions between mutant and control.
5. Pup examination: Finally, all live or dead pups are taken out of the
cage, and are investigated and classified either as alive with milk
in the stomach (Fig. 1b), alive without milk, or dead. In
addition, it is recorded whether there are any remaining amniotic
membranes, umbilical cord, or placenta attached to the body.
When such sticky fetal tissue is remaining, the pups skin may be
covered by bedding materials (see [13]). Also the body of the pups
should be briefly observed for possible bite marks or injuries. If
cannibalization has occurred, any remaining pup bodies are care-
fully sought for in the bedding and are recorded.
6. Maternal body examination: If the pups are without milk, the
maternal nipple should be examined briefly by holding its tail
and lifting its hindlimbs slightly in the air, with forelimbs still on
the cage top. At the same time the vaginal opening is checked
for bleeding and any obstruction that could be caused by a dead
fetus or other remaining tissue, which are the signs of parturi-
tion problems. If the mother has nursed the young and the
pups can properly suckle, the nipples should be somewhat pro-
truded from the ventral skin surface covered by short hair (eas-
ily identified by comparing the ventral surface of a lactating
female with another female that is definitely not nursing). If the
pups have been suckling vigorously because milk letdown is
limited, the tip of the nipple is overly elongated and may show
bleeding. Together with observations of maternal crouching
over the pups in the nest, these nipple observations help to con-
firm that maternal nursing behavior has been performed.
7. Interpretation of the results. Interpret the above observatory
data using the flowchart presented in Fig. 2. Analyze the laten-
cies of each behavior during pup retrieval observations, and
results from pup and maternal body examinations.
340 Kumi O. Kuroda and Yousuke Tsuneoka

*Do NOT follow the sequential order of this flow chart! First finish collecting all the information in the temporal order described
in the section 3.1 and interpret these information following this flow chart.

<START>
YES
Some pups are KO and/or Change breeding If the mother looks healthy and finished the labor, one
with phenotypic expression strategy to exclude can perform "maternal retrieval assay using donor pups"
pup factors
NO

Mother is visibly unhealthy,

has excessive bleeding Labor complication Perform "non-maternal retrieval assay" (subheading 3.2) instead.
or remaining fetuses at
Assess prostaglandin signaling
vaginal opening

Wait until the labor

Mother is still under labor If she is still in labor after a couple of hours, labor
ends and assess again complication is suspected.

Most pups are Severe placento- Assess norepinephrine signaling

with the intact amnion phagia defects susp. Perform "maternal retrieval assay using donor pups"

There are bite marks or Within two hours of first light Infanticide susp.
injuries on the body of pups phase after the delivery?
Perform "maternal retrieval assay using donor pups"
to confirm the behavior

Within two hours of first light

Most of pups are dead phase after the delivery? Too late to determine the
cause of pups' death

Start observation earlier next time

Perform "maternal retrieval assay using donor pups"

Maternal nipples Milk production/ Histological analysis of the mammary gland

Pups show milk bands show signs Assess the prolactin or oxytocin signaling function
ejection defects susp.
of suckling Perform "maternal retrieval assay using donor pups"

Number of live pups are

only one or two Litter size effect susp.

Perform "maternal retrieval assay using donor pups"

At initial observation there

Maternal retrieval assay was a good nest where Mother is anosmic Pup retieval defect
(Own or donor pups) Otherwise
most pups were retrieved caused by anosmia

Full maternal
Primary pup retieval
behavior defect susp.
Maternal stress-
vulnerability susp.
Normal maternal Perform "non-maternal retrieval assay"
behavior (subheading 3.2) as well
Try maternal retrieval test in less stressful condition

Fig. 2 Flowchart of maternal behavior assessment after the delivery

3.2 Parental The pup retrieval in virgin female mice can be tested much more
Retrieval Behavior easily and quickly than that of postpartum mothers in mice or that
and Infanticide Toward of virgin female rats, since the virgin female mice are parental
Donor Pups by within half an hour at the first pup exposure [35, 36]. It is mini-
Non-maternal Mice mally 2-day procedure from single housing to testing for 30 min
next day, if the donor pups can be provided from breeding colony.
In addition the pup retrieval in virgin females is free of any physical
or endocrine complications by parturition. It should also be noted
that the virgin females oparental behaviors are independent of
hypophyseal hormones both in mice [37, 38] and in rats [5].
 Assessing Postpartum Maternal Care, Alloparental Behavior, and Infanticide in Mice 341

1. To quantify parental responsiveness of non-parturient animals,

male or female subject mice are individually housed for 2 days
(at least 24 h) prior of an experiment, in a new cage containing
paper bedding with a cotton square as nest material as described
above (see Note 14).
2. On the test day, the nest site and quality should be recorded as
described above.
3. Then each animal is exposed to three 1- to 6-day-old donor
pups (see Note 15). One pup is placed in each corner of the
cage distant from the nest.
4. The cages are continually observed for the next 30 min and the
following measures are recorded: latency to sniff a pup for the
first time, to retrieve each pup into the nest, and to crouch over
all the pups continuously for more than 1 min (see Note 16).
If the subject mouse has performed all of these behaviors
within 30 min, it is labeled as fully parental. If only part of
these behaviors are seen by the end of the 30-min session, for
example only one of the three pups is retrieved to the nest, or
the subject does not retrieve at all but instead collects nest
material to one of the pups and crouches over it, the subject is
designated as partially parental. If the subject only sniffs the
pups but does not show any retrieving or crouching responses
throughout, it is labeled as nonresponding. If the subject does
not approach and sniff any of the pups, it should be examined
whether some kind of health problem or sensory dysfunction
exists. If any of the pups is attacked during the test, which is
normally observed during the first few minutes after the pup
exposure, stop the observation and rescue all the pups imme-
diately. The severely wounded pup is euthanized as described
previously [39]. This subject is deemed as infanticidal (see
Note 17). Otherwise, pups are left in the cage for 30 min.
5. After 30 min of observation, the pups are removed and, if sus-
pected, examined for any bite marks.
6. If many of the subjects are nonresponding or partially parental,
the same 30-min pup exposure session can be repeated for sev-
eral days to examine whether changes in response subsequently
develop by pup sensitization [7, 16]. On the other hand, once
an animal becomes fully parental for 2 successive days, it sel-
dom goes back to partially parental or infanticidal.

4 Notes

1. We found that the proportion of infanticidal virgin males is

highly dependent on the rearing environment of the males as
well as the testing conditions [25]. For example, the C57BL/6
342 Kumi O. Kuroda and Yousuke Tsuneoka

males weaned at 21 days rather than 28 days are less infanticidal

in our experimental conditions.
2. Evidence suggests that such infanticide seen in many wild
mammalian species is adaptive in terms of inclusive fitness; that
is, it is beneficial for the survival of their own biological off-
spring at the expense of nonbiological offspring that are poten-
tial competitors for environmental resources [40]. Especially
for species forming a harem (polygyny) of a small number of
males with multiple females such as langurs, lions, and mice, a
new male tends to kill all the existing young upon take over a
harem; such infanticide brings the females lactation to an end,
which hastens the occurrence of ovulation [41]. The new alpha
male stops infanticide by the time that its biological offspring
are born. This timing is clearly correlated with the gestation
period of the particular species [42]. These findings support
the idea that infanticide of non-offspring young is a valid and
adaptive reproductive strategy selected through evolution.
3. Care should be taken for the species differences of the olfac-
tory function in parental behaviors. Unlike in mice, maternal
behavior in rats is under multisensory control, in terms that the
maternal behavior is not dependent on any single sensory
modalities (olfactory, visual, auditory, and somatosensory)
[43, 44]. Elimination of any two, or all three of these sensory
systems did not completely abolish the behavior, although
some deficits in maternal retrieving were observed. More
importantly, as in other mammalian species such as hamsters,
rabbits and sheep, olfactory information from pups is necessary
for the typical pup avoidant reaction that occurs in virgin
female rats (see [14]). Olfactory bulbectomy, complete vom-
eronasal nerve cuts, pharmacological blockade of transmission
in the accessory olfactory bulb, or destruction of the olfactory
epithelium by intranasal application of zinc sulfate, all reduce
the latency to the onset of parental responsiveness in virgin
female rats. Therefore the high dependence of the mouse
parental behaviors on olfaction may be rather atypical in mam-
malian species.
4. It should be noted that anosmic neonatal pups show defects
in nipple finding and subsequent suckling, resulting in neo-
natal lethality [15, 45]. This means that the mother-pup
interactions can be disturbed bidirectionally by anosmia.
Therefore the proper breeding strategies are needed to seg-
regate pup factors from maternal factors: that is, the observed
mother-pup dyad should contain the homozygous mutant
only in one side, either in the mother or in pups. Of addi-
tional note, the vomeronasal organ does not seem required
for suckling in pups [46].
Assessing Postpartum Maternal Care, Alloparental Behavior, and Infanticide in Mice 343

5. Milk-production deficit can be exaggerated by backcrossing

the mutant lines from 129Sv into C57BL/6 as discussed [13].
6. Whenever appropriate, the subject experimental mice and the
control mice should be the littermates (reared by the same
mother) in general. That is, for production of experimental
homozygous (/), heterozygous (+/) and control (+/+)
mice should be produced by crossing (+/) females and (+/)
males, in order to minimize the potential confounding influ-
ences of background genes from breeder parents and rearing
environmental factors [47]. Moreover, maternal behavior is
known to transmit across generations nongenomically (future
maternal behavior is affected by the maternal behavior received
as neonates) [48].
7. If the preliminary analyses confirms that the given genetic
mutation is complete recessive (i.e., (+/) mothers and pups
are essentially not different from (+/+)), use of the control
females of (+/) versus (/) subject females can be con-
ducted for time/cost savings. In this case, produce (/) and
(+/) littermate females by crossing a (/) male with (+/)
females.
8. N = 1020 mice per genotype are commonly needed to detect
moderate behavioral differences using standard multivariate
statistical analyses such as multiple and repeated-measures
analysis of variance, followed by appropriate post hoc tests
[49]. It is acceptable to start with about N = 46 animals/
group for the initial screening, and then repeat at the same size
or increased size for replications. These data can be combined,
as long as the data are not significantly different across cohorts.
As in any other research area, confirmation of the initially
found behavioral phenotype across independent cohorts of
mice provides compelling evidence for the functional outcome
of the mutation [49].
9. The wild type and mutants on C57BL/6J background are
weaned at 4 weeks of age in our laboratory, as well as in the
Jackson Laboratory (see Breeding strategies for maintaining
colonies of laboratory mice, The Jackson Laboratory Resource
Manual, 2009). This is because at 3 weeks of age, C57BL/6J
pups appears still small.
10. The pregnant mouse mothers deliver pups most typically dur-
ing the dark phase of 20 day after copulation, some variations
(typically a half or 1 day delay) may occur [50] (see also
Biology of the laboratory mouse, The Jackson Laboratory,
Adapted for the Web at http://www.informatics.jax.org/
greenbook/index.shtml).
344 Kumi O. Kuroda and Yousuke Tsuneoka

11. The method described here is according to the previous

publications [31], with some simplifications to be compatible
to other measures of maternal behavior assessments in mice
[13]. See also [26, 5154].
12. This measure of sniffing latency is useful to show the possible
insensitivity or retardation of reaction to the pup introduction
in the experimental animals.
13. It should be noted that laboratory mice do not selectively care
their own young, but also retrieve alien young, provided their
age is comparable. This nonselective caring may be related to
the trait of communal nesting and nursing in feral Mus muscu-
lus species [55].
14. Because 1-month of social isolation has been shown to be
stressful and may increase aggressive behavior in male mice
[47, 56], they are group housed prior to the experiment, and
they are not isolated for more than 2 weeks. The cage bedding
should not be changed before the testing.
15. The pups with fur are not optimal stimuli for elicitation of
retrieving response.
16. If one wishes to gain more information of other components
of parental behavior than retrieval, the observer can simultane-
ously score the behavior of each mouse once every 15 s during
the 30 min test for a total 120 observations, as either crouch-
ing, sniffing and/or licking, nest building, or other behaviors.
Similar methods were described in [57].
17. In case that infanticide is expected as dominant response, there
is a technique to minimize harming the stimulus pups [58,
59]; in essence, instead of placing pups into the subjects home
cage directly, first place a pup contained in a wire-mesh tube or
a small metal tea strainer. If the subject mouse is observed to
start biting the container aggressively with eyes squinted and
tail rattled, it is highly probable to be infanticidal. If the subject
mouse does not show any signs of attack, the wire-covered pup
is taken away and there naked pups are introduced in the cage
as described above to further test pup retrieval behavior.

Acknowledgments

We thank Kashiko Tachikawa and Yoshihiro Yoshihara for helpful

discussion and comments on this topic. This work was supported
by RIKEN, the Uehara Memorial Foundation, and the JSPS
Grant-in-Aid programs. We would appreciate if the readers kindly
let us know if any corrections or comments on this article at:
oyako@brain.riken.jp.
 Assessing Postpartum Maternal Care, Alloparental Behavior, and Infanticide in Mice 345

References

1. Krasnegor NA, Bridges RS (1990) Mammalian anosmia in transgenic mice. Neuron 27(3):487
parenting: biochemical, neurobiological, and 497, S0896-6273(00)00060-X [pii]
behavioral determinants. Oxford University 16. Wang Z, Storm DR (2011) Maternal behavior
Press, New York, NY is impaired in female mice lacking type 3
2. Numan M, Insel TR (2003) The neurobiology adenylyl cyclase. Neuropsychopharmacology
of parental behavior. Springer, New York, NY 36(4):772781
3. vom Saal FS, Howard LS (1982) The regula- 17. Belluscio L, Gold GH, Nemes A, Axel R
tion of infanticide and parental behavior: impli- (1998) Mice deficient in G(olf) are anosmic.
cations for reproductive success in male mice. Neuron 20(1):6981
Science 215(4537):12701272 18. Bean NJ, Wysocki CJ (1989) Vomeronasal
4. Priestnall R, Young S (1978) An observational organ removal and female mouse aggression:
study of caretaking behavior of male and female the role of experience. Physiol Behav 45(5):
mice housed together. Dev Psychobiol 11(1): 875882
2330 19. Leypold BG, Yu CR, Leinders-Zufall T, Kim
5. Rosenblatt JS (1967) Nonhormonal basis of MM, Zufall F, Axel R (2002) Altered sexual and
maternal behavior in the rat. Science social behaviors in trp2 mutant mice. Proc Natl
156(781):15121514 Acad Sci U S A 99(9):63766381. doi:10.1073/
6. Numan M, McSparren J, Numan MJ (1990) pnas.082127599, 082127599 [pii]
Dorsolateral connections of the medial preop- 20. Del Punta K, Leinders-Zufall T, Rodriguez I,
tic area and maternal behavior in rats. Behav Jukam D, Wysocki CJ, Ogawa S, Zufall F,
Neurosci 104(6):964979 Mombaerts P (2002) Deficient pheromone
7. Brown JR, Ye H, Bronson RT, Dikkes P, responses in mice lacking a cluster of vomerona-
Greenberg ME (1996) A defect in nurturing in sal receptor genes. Nature 419(6902):7074
mice lacking the immediate early gene fosB. 21. Mennella JA, Moltz H (1988) Infanticide in
Cell 86(2):297309 the male rat: the role of the vomeronasal organ.
8. Elwood RW (1983) Parental behaviour of Physiol Behav 42(3):303306
rodents. Wiley, New York, NY 22. Fleming AS, Gavarth K, Sarker J (1992) Effects
9. Sluckin W, Herbert M (1986) Parental of transections to the vomeronasal nerves or to
Behaviour. Basil Blackwell, Oxford the main olfactory bulbs on the initiation and
10. Rosenblatt JS, Snowdon CT (eds) (1996) long-term retention of maternal behavior in
Parental care: evolution, mechanism, and adap- primiparous rats. Behav Neural Biol 57(3):
tive significance, vol 25, Advances in the study 177188
of behavior. Academic, San Diego, CA 23. Yagi T, Aizawa S, Tokunaga T, Shigetani Y,
11. Lonstein JS, Fleming AS (2002) Parental Takeda N, Ikawa Y (1993) A role for Fyn tyro-
behaviors in rats and mice. Curr Protoc sine kinase in the suckling behaviour of neona-
Neurosci Chapter 8:8.15.1118.15.26 tal mice. Nature 366(6457):742745.
doi:10.1038/366742a0
12. Capone F, Bonsignore L, Cirulli F (2005)
UNIT 13.9 methods in the analysis of maternal 24. Hamaguchi-Hamada K, Sanbo C, Hamada S,
behavior in the rodent. Curr Protoc Toxicol Yagi T (2004) Exposure to hexanal odor influ-
13(9):116 ences maternal behavior and induces neonatal
death in Fyn tyrosine kinase-deficient mice.
13. Kuroda KO, Tachikawa K, Yoshida S, Tsuneoka Neurosci Res 48(3):259267
Y, Numan M (2011) Neuromolecular basis of
parental behavior in laboratory mice and rats: 25. Kuroda KO, Meaney MJ, Uetani N, Kato T
with special emphasis on technical issues of using (2008) Neurobehavioral basis of the impaired
mouse genetics. Prog Neuropsychopharmacol nurturing in mice lacking the immediate early
Biol Psychiatry 35(5):12051231. doi: gene FosB. Brain Res 1211:5771
10.1016/j.pnpbp.2011.02.008, 26. Deacon RM (2006) Assessing nest building in
S0278-5846(11)00062-5 [pii] mice. Nat Protoc 1(3):11171119
14. Levy F, Keller M (2009) Olfactory mediation 27. Doetschman T (2009) Influence of genetic
of maternal behavior in selected mammalian background on genetically engineered mouse
species. Behav Brain Res 200(2):336345 phenotypes. Methods Mol Biol 530:423433
15. Wong ST, Trinh K, Hacker B, Chan GC, Lowe 28. Brown RE, Mathieson WB, Stapleton J,
G, Gaggar A, Xia Z, Gold GH, Storm DR Neumann PE (1999) Maternal behavior in
(2000) Disruption of the type III adenylyl female C57BL/6J and DBA/2J inbred mice.
cyclase gene leads to peripheral and behavioral Physiol Behav 67(4):599605
346 Kumi O. Kuroda and Yousuke Tsuneoka

29. Shoji H, Kato K (2006) Maternal behavior of 43. Beach FA, Jaynes J (1956) Studies of Maternal
primiparous females in inbred strains of mice: a Retrieving in Rats: III. Sensory cues involved
detailed descriptive analysis. Physiol Behav in the lactating females response to her young.
89(3):320328 Behavior 10:104125
30. Wolfer DP, Crusio WE, Lipp HP (2002) 44. Herrenkohl LR, Rosenberg PA (1972)
Knockout mice: simple solutions to the prob- Exteroceptive stimulation of maternal behavior
lems of genetic background and flanking genes. in the naive rat. Physiol Behav 8(4):595598
Trends Neurosci 25(7):336340 45. Tasaki T, Sohr R, Xia Z, Hellweg R, Hortnagl
31. Hess SE, Rohr S, Dufour BD, Gaskill BN, H, Varshavsky A, Kwon YT (2007) Biochemical
Pajor EA, Garner JP (2008) Home and genetic studies of UBR3, a ubiquitin ligase
improvement: C57BL/6J mice given more with a function in olfactory and other sensory
naturalistic nesting materials build better nests. systems. J Biol Chem 282(25):1851018520
J Am Assoc Lab Anim Sci 47(6):2531 46. Kimchi T, Xu J, Dulac C (2007) A functional
32. Moltz H, Lubin M, Leon M, Numan M (1970) circuit underlying male sexual behaviour in the
Hormonal induction of maternal behavior in female mouse brain. Nature 448(7157):
the ovariectomized nulliparous rat. Physiol 10091014. doi:10.1038/nature06089,
Behav 5(12):13731377 nature06089 [pii]
33. Pedersen CA, Prange AJ Jr (1979) Induction 47. Crawley JN (2007) Whats wrong with my
of maternal behavior in virgin rats after intra- mouse? Behavioral phenotyping of trangenic
cerebroventricular administration of oxytocin. and knockout mice, 2nd edn. Wiley-Liss, New
Proc Natl Acad Sci U S A 76(12):66616665 York, NY
34. Bridges RS, DiBiase R, Loundes DD, Doherty 48. Francis D, Diorio J, Liu D, Meaney MJ (1999)
PC (1985) Prolactin stimulation of maternal Nongenomic transmission across generations
behavior in female rats. Science 227(4688): of maternal behavior and stress responses in the
782784 rat. Science 286(5442):11551158
35. Noirot E (1972) The onset of maternal behav- 49. Crawley JN (2008) Behavioral phenotyping
ior in rats, hamsters, and mice a selective review. strategies for mutant mice. Neuron 57(6):
In: Lehrman DS, Hinde RA, Shaw E (eds) 809818
Advances in the study of behavior, vol 4. 50. Hedrich H (ed) (2004) The laboratory mouse,
Elsevier, New York, NY, pp 107145 The hand book of experimental animals.
36. Lonstein JS, De Vries GJ (2000) Sex differ- Elsevier, Amsterdam
ences in the parental behavior of rodents. 51. Gandelman R (1973) Induction of maternal
Neurosci Biobehav Rev 24(6):669686 nest building in virgin female mice by the pre-
37. Leblond CP (1940) Extra-hormonal factors in sentation of young. Horm Behav 4(3):
maternal behavior. Proc Soc Exp Biol Med 191197
38:6670 52. Slotnick BM, Nigrosh BJ (1975) Maternal
38. Leblond CP, Nelson WO (1937) Maternal behavior of mice with cingulate cortical, amyg-
behavior in hypophysectomized male and dala, or septal lesions. J Comp Physiol Psychol
female mice. Am J Physiol 120:167172 88(1):118127
39. Perrigo G, Belvin L, Quindry P, Kadir T, 53. Numan M, Callahan EC (1980) The connec-
Becker J, van Look C, Niewoehner J, vom Saal tions of the medial preoptic region and mater-
FS (1993) Genetic mediation of infanticide and nal behavior in the rat. Physiol Behav 25(5):
parental behavior in male and female domestic 653665
and wild stock house mice. Behav Genet 54. Mann PE (1993) Measurement of maternal
23(6):525531 behavior. Methods Neurosci 14:343358
40. Trivers RL (1972) Parental investment and 55. Manning CJ, Wakeland EK, Potts WK (1992)
sexual selection. In: Campbell B (ed) Sexual Communal nesting patterns in mice implicate
selection and the descent of man 1871-1971. MHC genes in kin recognition. Nature
Aldine-Atherton, Chicago, IL, pp 136172 360(6404):581583
41. Hrdy SB (1974) Male-male competition and 56. Valzelli L, Bernasconi S, Gomba P (1974)
infanticide among the langurs (Presbytis entel- Effect of isolation on some behavioral charac-
lus) of Abu, Rajasthan. Folia Primatol (Basel) teristics in three strains of mice. Biol Psychiatry
22(1):1958 9(3):329334
42. Parmigiani S, Vom Saal FS (eds) (1990) 57. Liu D, Diorio J, Tannenbaum B, Caldji C,
Infanticide and parental care. Harwood aca- Francis D, Freedman A, Sharma S, Pearson D,
demic publishers, London Plotsky PM, Meaney MJ (1997) Maternal care,
 Assessing Postpartum Maternal Care, Alloparental Behavior, and Infanticide in Mice 347

hippocampal glucocorticoid receptors, and 68. Ambrose A (ed) (1969) Stimulation in early
hypothalamic-pituitary-adrenal responses to infancy. Academic, New York, NY
stress. Science 277(5332):16591662 69. Rosenberg KM, Denenberg VH, Zarrow MX
58. Perrigo G, Belvin L, vom Saal F (1990) (1970) Mice (Mus musculus) reared with rat
Humane behavioral assays in the laboratory: an aunts: the role of rat-mouse contact in mediat-
alternative test for assessing how rodents ing behavioural and physiological changes in
behave toward young. Humane Innov Alternat the mouse. Anim Behav 18(1):138143
Anim Experiment 4:208209 70. Kristal MB (1980) Placentophagia: a biobehav-
59. Perrigo G, Bryant W, vom Saal F (1989) The ioral enigma (or De gustibus non disputandum
use of live pups in a humane, injury-free test for est). Neurosci Biobehav Rev 4(2):141150
infanticidal behaviour in male mice. Anim 71. Kristal MB (1991) Enhancement of opioid-
Behav 38:897904 mediated analgesia: a solution to the enigma of
60. Stern JM, Lonstein JS (2001) Neural media- placentophagia. Neurosci Biobehav Rev
tion of nursing and related maternal behaviors. 15(3):425435, S0149-7634(05)80035-1 [pii]
Prog Brain Res 133(64):263278 72. Kristal MB, Graber GC (1976) Placentophagia
61. Lonstein JS, Wagner CK, De Vries GJ (1999) in nonpregnant rats: influence of estrous cycle
Comparison of the nursing and other paren- stage and birthplace. Physiol Behav
tal behaviors of nulliparous and lactating female 17(4):599605
rats. Horm Behav 36(3):242251 73. Kristal MB, Eleftheriou BE (1975)
62. Wiesner BP, Sheard NM (1933) Maternal Placentophagia in nonpregnant nulliparous
behaviour in the Rat. Oliver and Boyd, London mice: a genetic investigation. Behav Biol 13(1):
63. Bridges RS, Zarrow MX, Goldman BD, 113119
Denenberg VH (1974) A developmental study 74. Koller G (1952) Der Nestbau der Weissen
of maternal responsiveness in the rat. Physiol Maus und seine hormonale Auslosung. Zool
Behav 12(1):149151 Anz Suppl 17:160168
64. Fahrbach SE, Morrell JI, Pfaff DW (1984) 75. Denenberg VH, Taylor RE, Zarrow MX
Oxytocin induction of short-latency maternal (1969) Maternal behavior in the rat: an investi-
behavior in nulliparous, estrogen-primed gation and quantification of nest building.
female rats. Horm Behav 18(3):267286, Behaviour 34(1):116
0018-506X(84)90016-3 [pii] 76. Lonstein JS, Gammie SC (2002) Sensory, hor-
65. Kinsley CH, Bridges RS (1988) Parity- monal, and neural control of maternal aggres-
associated reductions in behavioral sensitivity sion in laboratory rodents. Neurosci Biobehav
to opiates. Biol Reprod 39(2):270278 Rev 26(8):869888
66. Moore CL (1992) The role of maternal stimu- 77. Gammie SC (2005) Current models and future
lation in the development of sexual behavior directions for understanding the neural circuit-
and its neural basis. Ann N Y Acad Sci 662: ries of maternal behaviors in rodents. Behav
160177 Cogn Neurosci Rev 4(2):119135
67. Gubernick DJ, Alberts JR (1985) Maternal 78. Svare B, Gandelman R (1976) Suckling stimu-
licking by virgin and lactating rats: water trans- lation induces aggression in virgin female mice.
fer from pups. Physiol Behav 34(4):501506 Nature 260(5552):606608
 Chapter 26

Testing Smell When It Is Really Vital: Behavioral

Assays of Social Odors in the Neonatal Mouse
Benoist Schaal, Syrina Al An, and Bruno Patris

Abstract
The initial interactions of mouse newborns with their mother are crucial for their survival. These interactions
rapidly end in the pups reaching nipples and getting milk. While we realize that olfaction is clearly pre-
vailing in the success of these first suckling episodes, we still understand little about the nature and range
of the natural odorants involved. Here we non-exhaustively describe some experimental principles and
methods to assay the behavior of newly born and infant mice exposed to different odor stimuli from con-
specifics. Testing neonatal and young mice with chemostimuli which they are evolutionarily or develop-
mentally canalized to detect may be a productive way to trace unanticipated odor signals. Moreover,
testing neonates also may also lead to characterize unsuspected strategies of murine females to produce and
release odor messages.

Key words Mouse (Mus musculus), Newborn, Infant, Olfaction, Social odor, Pheromones, Behavior

1 Introduction

The survival of newborns mammals depends on the success of their

first suckling of colostrum and milk, ensuring continued supply in
hydration, energy, nutrients, passive immunization, as well as the
transfer of growth factors, hormones and regulatory peptides, and
nonpathogenic bacteria. Through milk, mothers also convey sen-
sory information that impinges on the current and future adaptive
competence of their offspring. Thus, after successful birth, it is
imperative for neonates to perceive their mother and locate her
mammae. The corollary of this inescapable bottleneck of the mam-
malian condition is that neonates must have been evolutionarily
tailored to use any sensory cue susceptible to speedily bring them
to those goals. Thus, testing neonatal and infant animals with stim-
uli which they are canalized to detect may be a productive way to
trace unanticipated odor signals. Testing neonates also leads to
characterize strategies of females to produce and release odor mes-
sages that evolved to match perceptual biases in their newborns.

Kazushige Touhara (ed.), Pheromone Signaling: Methods and Protocols, Methods in Molecular Biology, vol. 1068,
DOI 10.1007/978-1-62703-619-1_26, Springer Science+Business Media, LLC 2013

349
350 Benoist Schaal et al.

Despite we know that these interactions with the mother are

crucial in the viability of newborns [13], we have still much to
understand about the nature and range of the stimuli that direct
their initial behavior. Here, we will address first general advantages
and difficulties in investigating early chemosensory development in
mammals, emphasizing a species that has become a privileged
mammalian model in biology, the mouse. Then, we will provide
selected methods and protocols that have been devised to quantify
chemosensory abilities of mouse newborns and infants toward nat-
ural odors conveyed by biological secretions of conspecifics.

1.1 Studying A range of psychobiological features make mammalian neonates

Chemoreception in and infants particularly profitable to investigate in the context of
Newborn and Young olfactory exchanges in general, and of pheromonal communication
Animals: Advantages in particular:
and Problems 1. Pronounced taxonomic differences in neonatal development
are associated with qualitative and quantitative differences in
initial capacities. While some species give birth to newborns of
the altricial type, others deliver precocial newborns, and still
others engender newborn of an intermediate type. The most
obvious criteria underlying this altricialityprecociality scale
relate to neonatal thermoregulatory, locomotor, and sensory
abilities. Regarding sensory abilities, precocial newborns have
a complete (although still maturing) set of active sensory sys-
tems, while those of altricial newborns are only partially func-
tional. In mammals, altriciality means that somesthesis and
chemoreception are the only sensory means to control neona-
tal behavior. In contrast, precocial newborns additionally dis-
pose of a good level of visual and auditory awareness. Thus,
relying predominantly on chemosensory processes to survive,
altricial newborns are outstanding models to assess the earliest
olfaction-guided responses.
2. Mammalian newborns, especially those of the altricial type, are
obviously limited in their behavior, but they are far from being
incomplete, incompetent or hindered adults. In much con-
trast, they are endowed with specialized responses that are best
expressed in their natural/usual ecologies. Neonatal abilities to
move and orient, explore and choose, grasp and suck are
indeed fully adjusted to the sequence of operations these inex-
perienced organisms have to achieve in their species-typical
environment. Each developmental phase has its typical ways to
generate behavior adaptively, and the specialized responses of
neonates have once been considered as automatic and reflex-
like, conceptualized as fixed action patterns prompted by
specific releasing stimuli. Although these conceptions have
become more flexible with advancing psychobiological research
(e.g., [4, 5]), neonatal sensorimotor abilities are first organized
 Testing Smell When It Is Really Vital 351

around stimulus-response loops that are ready to operate right at

birth, and much remains to be understood about the sensory trig-
gers of these initial functional loops.
3. Neonatal mammals rapidly learn and escape sensorimotor
automatisms in the regulation of their behavior. Newborn
rodents are especially skilled to learn about the odor environ-
ment [68]: (1) they acquire any odorant which they are
merely exposed to or which is associated with the mother, and
such acquisitions can take only minutes to establish; (2) such
learning is functional from birth onwards, although its efficacy
increases with maturation and experience; (3) neonatal olfac-
tory sensitivity and preferences are partially induced prenatally
[911], and thus, at first presentation of an odorant, neonatal
reactions may trace back into fetal acquisition [10, 12, 13]; (4)
multiple, redundant reinforcing processes and agents afforded
by the mothers body or her milk promote rapid odor learning
in neonates (e.g., [7]). In sum, these neonatal learning abilities
are of great empirical interest to investigate early plasticity of
the senses, brain and behavior, but on the other side the fast
odor learning of neonates may obscure the salience of odor signals
that do not depend on exposure effects. This is especially inconve-
nient to track behaviorally active odorants such as pheromones
which, by definition, imply no or minimal induction by expo-
sure effects or learning (cf. below).
4. It ensues from the previous point that newborn rodents are
practical to investigate because their olfactory experience can be
somehow titrated. Their odor memory stores can be pre-shaped
by controlled odor exposure during prenatal development.
They can be gestated by a female transferring them experimen-
tal odorants through the placenta and then cross-fostered to a
lactating dam transferring the same or a different odorants into
her milk. They can be prevented from any postnatal contact
with their mother, and tested for odorant stimuli before they
had any direct exposure to them postnatally. Neonatal rodents
are thus suitable to appreciate whether a natural odorant can be
behaviorally active in absence of previous exposure.
5. In many altricial species, newborns and juvenile individuals are
easily tractable and amenable to approach an attachment
figure (e.g., the mother or sibling) or salient sensory traits
disembodied from her (odorants carried in biological
secretions). However, optimal species-specific responsiveness of
neonate animals will be obtained only if a range of basic ethological
principles are applied (e.g., [14]) that take into account: (1)
age-dependency of test conditions (in terms of perceptual,
motor and cognitive abilities of young organisms; existence of
sensitive periods), (2) state-dependency of responses (separation
stress; thermoregulation; food deprivation; circadian rhythms),
352 Benoist Schaal et al.

and (3) the social-ecological validity of the stimuli being

assayed, of the testing context and of target responses. Although
such principles may be general in the investigation of any
animal, they appear especially sensitive with more physiologically
unstable and psychologically susceptible newborns and infants.

1.2 Mammalian Newborn mammals are canalized into adaptive responses by two
Newborns and the types of nonexclusive mechanisms, predisposed processes which do
Pheromone Concept: (apparently) not depend on previous exposure effects, and highly
Methodological efficient learning processes which fine-tune them to the current or
Consequences prospective environment. Associative or non-associative learning
mechanisms work in extracting arbitrary and incidental informa-
tion from the current odor environment to convert them into
meaningful odor cues. By contrast, predisposed mechanisms rely on
hardwired stimulus-response loops that can operate at the very first
stimulus presentation and do apparently not need induction by
previous contact with it. Such predisposed stimuli have been selected
because they bear an inherent signal value to the neonatal receiver
[15]. While cues tend to be lastingly on, signals can be switched
on-off according to the emitters behavior or condition [16].
Any biological secretion may convey numerous cues, including
informative olfactants derived from the environment or current
physiological or psychological states of a conspecific. In contrast,
signals are less frequent. Therefore, in the complex mixtures given
off in body secretions or excretions, rare odor signals will unavoid-
ably be embedded in an abundance of odor cues. Thus, it is always a
challenging endeavor to separate rare or unique odor signals from
affluent odor cues in a chemically complex body secretion.
Before sorting odor signals from odor cues, a number of steps
have to be fulfilled. First, an age- and state-adapted behavioral assay
has to be validated. Second, behaviorally active biological substrates
must be identified, and the emitters state in which this substrate
bears maximal activity characterized. Then, the active substrate has
to be reduced in complexity to target fractions that remain active on
the target response, e.g., extricating composite substrates; separat-
ing volatile from nonvolatile compounds, or hydrosoluble from
liposoluble compounds. Then, comes the difficult task of designing
techniques that combine chemical separation and functional
responses to pinpoint (the) active compound(s) in these fractions.
Once chemically identified, candidate compounds have then to be
related in activity to the original odor substrate, and then screened
through a set of criteria to assess whether they are mere cues or if
they qualify as signal(s), i.e., pheromone(s) (see below). Finally,
once a pheromone is identified, its communicative function and
adaptive meaning between emitting and receiving organisms need
to be ascertained. The successful attainment of each of these steps
depends on the elaboration of reliable behavioral assays.
 Testing Smell When It Is Really Vital 353

Regarding the pheromone concept, its initial definition [17]

now appears minimalist, encompassing almost any olfactant
exchanged between conspecifics, be it a cue or a predisposed sig-
nal. To avoid this latent confusion and in an attempt to critically
resize the concept in relation with the cognitive complexity of
mammalian behavior, Beauchamp, Doty, Moulton and Mugford
[18] (see also [1922]) proposed criteria implying that any candi-
date pheromone should: (1) be chemically simple, in the sense
that the active fraction is composed of one chemical or a small set
of chemicals in a given ratio; (2) release in a conspecific receiver
unambiguous behavioral/physiological responses which are mor-
phologically invariant in a same context, and (3) obvious in terms
of adaptive function; (4) elicit these responses in a selective way
(thus tested against a set of reference compounds), and (5) be
tested in various species to establish specificity at taxonomic level
(genus, species, subspecies); finally, (6) the candidate stimulus-
response link should minimally or not at all depend on previous
exposure effects and learning. Thus, in principle, to qualify a can-
didate compound as a pheromone one should make sure that the
responsiveness to it cannot be ascribed to prenatal induction by
direct exposure, to facilitated learning during the natal process, or
to rapid learning immediately after birth. This latter criterion is
generally overlooked in most cases of mammalian pheromones
described in adult animals [1820, 23]. However, investigating
young organisms should make its assessment feasible.
The above methodological steps and concepts have been suc-
cessfully applied to mammalian neonates. In fact, the most com-
plete (although yet incomplete!) demonstrations to date of
mammalian pheromones according to Beauchamp et al.s opera-
tional criteria [18] have made use of newborns. Let us develop two
examples which are borrowed beyond the mouse literature. First,
in the newborn rat, pup saliva and pup salivary gland extract were
shown to be reliable elicitors of nipple grasping [24]. Subjected to
gas chromatography-mass spectrometry (GC-MS), these substrates
were shown to carry dimethyl disulfide (DMDS). Synthetic DMDS
was indeed effective in eliciting the grasping of nipples that were
washed before the test, although its releasing potency was about
50 % relative to that of intact nipples [24]. Thus, DMDS explained
only in part the elicitation of nipple grasping in rat pups, unknown
compounds from the original mixture coating nipples being prob-
ably involved. DMDS is special, however, as it also reduces aversive
responses to perioral stimulation in the fetus. This effect, mediated
by opioidergic processes, suggests unconditional reward properties
of DMDS already established in late gestation [25]. As DMDS was
reported to be undetectable by GC in amniotic fluid [26], its
behavioral activity may not depend on prenatal exposure. In these
studies, however, pups were aged postnatal days (PND) 35,
implying that the activity of DMDS could derive from exposure to
354 Benoist Schaal et al.

saliva prior to testing. Thus, it is difficult to decide whether DMDS

is an odor cue acquired in the first suckling episodes or if it is an
evolved signal to promote nipple attachment. Further, to be cate-
gorized as a pheromone, its species-specificity remains to be estab-
lished and it needs to be assayed against reference compounds to
eliminate the possibility of unspecific arousal effects.
The second example also goes beyond the murine focus of the
present chapter, but remains in scope as it involves newborns. The
extensive corpus of behavioral investigation by P Schley and R
Hudson on odor cues that direct rabbit pups to the nipple and
milk [2729] paved the way for the chemical characterization of
the behaviorally active volatiles involved in milk. A GC-MS-
olfactometry analysis of rabbit milk effluvium resulted in identify-
ing 2-methyl-but-2-enal (2MB2) as accounting for the bioactivity
of milk, as it released a similar response at the highest possible rate
in very different contexts of stimulation [30]. 2MB2 was then
assayed to verify whether it qualified as a pheromone. In chemical
terms, it is the simplest (viz., monomolecular) possible stimulus
bearing a strong potency to elicit nipple grasping actions in pups,
although other, unidentified compounds may act in the same direc-
tion. Further, 2MB2 triggers the criterion response regardless of
mode of presentation (at least during the first day 10 days post-
birth) in a highly selective manner (tested against 40 reference
odorants, [31]). Its activity is optimal within a concentration range
(109 to 105 g/mL, [32]), giving flexibility for intensity fluctua-
tions of the signal. The species-level bioactivity of 2MB2 was estab-
lished through its independence from maternal diet and genetic
background, and its inertness for neonatal rats, mice, cats, and even
brown hares [30]. Finally, to be behaviorally active, 2MB2 does not
appear to require contingent labor-related arousal, suckling, or con-
tact with the mother, and it is resistant to exposure deprivation
during the first 6 days of life. Further, it is active in pups delivered
12 days before term, and GC-MS failed detecting it in amniotic
fluid and blood plasma of lactating does [30]. These results allowed
qualifying 2MB2 from rabbit milk as a pheromone. As it was
detected in ejected milk it was called mammary pheromone.
Regarding its functional involvement, 2MB2 was shown to
control mutually beneficial functions as it contributes to the rabbit
doe-offspring reciprocity in eliciting pups contagious directional
actions toward the female, and guidance to nipples; on the does
side, it boosts tactile stimulations from pups that trigger, synchro-
nize and sustain lactational physiology. Those pups that do not
react to 2MB2 on PND 1 have low survival chance at weaning
[34]. Otherwise, maximal activity of 2MB2 closely matches the
period of strict milk-dependence of pups (up to PND 10) and
declines when they begin eating solid food [33]. 2MB2 elicits then
an automatic drive to grasp a nipple at any time during the first
critical postnatal days, and it is only after PND 10 that its activity
 Testing Smell When It Is Really Vital 355

relaxes and comes under prandial and circadian control [35].

Finally, 2MB2 also works as a primary reinforcing stimulus trans-
ferring its behavioral activity to any (initially neutral) odorant after
a single, brief association [36].
The above examples designate mammalian female-neonate
units as promising model systems to uncover odor cues and signals
that regulate their vital exchanges, and then to characterize molec-
ular and neurobehavioral processes underlying these ways of com-
munication. Similar advances are needed to understand the system
of cues and signals that rule early relationships between murine
mothers and their newborns.

2 The Mouse Pup as a Model to Track Behaviorally Significant Odorants

2.1 Nasal Mouse pups are born altricial. Accordingly, they are raised in a
Chemoreception nest in quasi-permanent contact with the mother (and, some-
in the Context of times, father) and, in the case of wild mice, with other related
Murine Developmental females, as communal nursing is frequent in this species [37].
Milestones Newborn mice are essentially guided by somesthesis and chemo-
reception in their early orientation responses. Audition comes into
play by PND 9, and vision by PND 1415 [38, 39]. Before PND
9, mouse pups are thus only governed by tactile, thermal and
chemical cues (nasal and oral chemoreception). Murine nasal che-
moreception is based on multiple receptive systems. These include
the main olfactory epithelium, the vomeronasal organ, which are
part of the main and accessory systems, respectively, the trigemi-
nal system, the Grueneberg ganglion, and the septal organ of
Masera [40, 41]. Much remains to be learned about the func-
tional onset of these distinct nasal systems, especially the last three
of them. The main olfactory system sets on prenatally [42], thus
contributing very early to the experiential induction of behavioral
phenotypes [43]. But the vomeronasal system does not appear to
be functional before birth and seems to remain immature until
puberty ([44, 45]; but see [46]). Regarding motor development,
mouse pups are crawlers during the first week of like, and begin to
walk by PND 7 and to run and jump by PND 15 [47, 48].
Inventing behavioral assays will obviously depend on these states
of motor development.
The general role of olfaction in neonatal rodents (i.e., rat and
mouse) was ascertained through its abolishment by bulbar lesion
[4951], by ZnSO4 peripheral denervation [52], or by genetic
disruption inducing early dysfunction in olfactory transduction
(e.g., [53]). Neonatal rodents are also disturbed when the normal
odor cuing system of the dam is disrupted. For example, washing
the abdomen of a lactating dam results in a drastic reduction of
nipple seizing performance in mouse pups [55, 56] as well as in
rat pups [54].
356 Benoist Schaal et al.

2.2 Assaying Social The mouse being a privileged model species in various subdisciplines
Odor-Based Behavior of biology it has received extensive coverage regarding methods
in Infant Mice and protocols to assay behavior in adult (e.g., [5763]) and in
young individuals (e.g., [47, 64, 65]). Here, we will focus on
behavioral assays that have been devised to measure responsiveness
of newborn and infant mice when exposed to biological substrates
collected from conspecifics. After presenting some general condi-
tions for testing young mice, we provide a brief survey of bioassays
adapted to three age-periods, newborn mice (PND 02), previsual
infant mice (from PND 3 to 14), and visual preweaning infant
mice (PND 1421).

2.2.1 General Conditions 1. Temperature: Mouse pups have poor thermoregulatory abilities
up to PND 1228 [66, 67], and they then mainly rely on
thermotaxis and huddling to maintain body temperature [68].
In general, breeding and experimental rooms are kept between
20 and 24 C. When pups have to be separated from the dam
before testing, hypothermia can be prevented by keeping the
male with the litter and/or putting the home cage on a heating
plate (Gestigkeit, Germany; set at the temperature that ensures
that the nest is stabilized at 3032 C). All tests should be real-
ized in conditions that reproduce those prevailing in the nest
(3032 C), which can be obtained by placing the test device
on a heating plate.
2. Lighting: Experimental rooms are generally lit up following a
given reversed lightdark cycle (e.g., 12:12 h). Before pups
have opened eyes, they can be assayed under artificial light.
But, when vision gets functional it is best to test them under
red light during the dark phase.
3. Noise: As the auditory system is functional on PND 9 and as
pups emit and likely detect ultrasonic vocalizations during the
first two weeks [6972], experiments should be conducted in
a dedicated, quiet room.
4. Parasite odors: Odor cues given off by murine conspecifics, the
experimental material or experimenters themselves may inter-
fere with target odor stimuli and thereby negatively influence
the responses of neonates. Thus, to avoid interference with
such unwanted odors, any behavioral testing should be con-
ducted in a dedicated room. Experimenters are recommended
to limit their body odors in wearing a scentless clean blouse
that remains in the testing room, avoiding perfume, coffee or
tobacco, and handling animals with gloved hands (do not use
latex gloves). Finally, the experimental material should be thor-
oughly washed twice, with (70 %) ethanol and then with water,
and finally dried at each testing session. It is often necessary to
visually mark animals after a treatment, a test or whatever: in all
cases, take care to select a scentless marker.
 Testing Smell When It Is Really Vital 357

5. Stress associated with separation, cold, novelty and handling: All

sorts of stressors related to social isolation can adversely affect
newborn and infant mice. When the protocol necessitates sep-
aration from the dam, keeping the male with the litter in their
home cage can limit social as well as cold stress. Cold stress is
controlled in maintaining the home cage or the experimental
device on a heating plate [73, 74, 76, 77] or under 60-W red
bulbs suspended 20 cm above the device [78, 79]. The han-
dling method is essential to preserve mice from fear [80], and
it is thus recommended to handle pups with extreme delicacy
and consistency across experimenters to limit artifactual
response variability. A protocol to familiarize pups with the
experimental device can have positive effects on stress responses.
For example, pups can be allowed to explore the device for
13 min to reduce neophobia before to the actual test.
6. Motivation of animals: To optimize arousal and motivation,
pups are often deprived of contact with the mother and, hence,
of suckling, prior behavioral testing. Several separation sched-
ules were used without apparent immediate harm to pups or
dams (i.e., 1 h [81]; 2 h [76]; 4 h [73, 74]). During the first 2
postnatal weeks, mouse pups stay in the nest and huddle and
exhibit low levels of arousal and activity. In an attempt to stan-
dardize the pups arousal level right before the tests, some
experimenters apply a stimulation protocol, for example by
stroking them along the back (e.g., ten strokes) with a soft
paintbrush [73].
7. Litter effects. Neonatal mice come in litters, and littermates are
more similar to one another than to members of other litters
because of differences in genetic and environmental back-
ground [82]. The first way to reduce litter effects is thus
through rigorous husbandry methods, such as using inbred
strains, diminishing the number of breeding males, and stan-
dardizing housing conditions and pup development in equal-
izing litter size (to control for competition between pups).
Litters effects are further limited when testing. Not only
should every pup be assayed only once, but the number of lit-
termates per test condition should be reduced to preserve
independence of data. The optimum is one littermate per con-
dition, or one male and one female (e.g., [83]), but most
studies are constrained to be less conservative because of lim-
ited number of litters in their breeding colony (e.g., 4-litter-
mates/condition [73, 74]; pups from three litters distributed
in two or more test conditions [56]). Finally, data should be
assessed statistically for potential litter effects in treating each
litter as the statistical unit [77, 83] or in using the litter as a
covariable [43, 73, 74].
358 Benoist Schaal et al.

2.2.2 Sampling Odorant Many different types of stimuli have been used to assess behavioral
Stimuli from Conspecifics: responses in newborn and infant mice. Before entering into more
Collection and detailed analyses, experimenters often use first entire animals as
Conservation stimuli to verify whether these are olfactorily differentiable to pups
along a selected dimension. Then, more analytic experiments eval-
uate which biological substrates from distinct body regions of con-
specifics are efficient in reproducing the responses elicited by
olfactorily intact entire animals. Some of these substrates are exem-
plified below. In certain cases, artificial arbitrary odorants are used
as controls.
1. Live animals as stimuli. Entire animals can be used to assess
whether neonates detect olfactory correlates of age, sex, physi-
ological states, genetic relatedness or merely familiarity. Pups
are then exposed to two animals simultaneously (a female vs. a
male, a lactating vs. a non-lactating female, the mother vs. the
father, etc.). For example, Breen and Leshner [84] presented
mouse pups with live lactating vs. non-lactating dams in a two-
choice device which allowed to hide them while only blowing
their body odor toward the tested pup. In similar tests in recent
studies, two stimulus-females were sedated (intraperitoneal
injection of 45 mg/kg body weight of ketamine and 9 mg/kg
Xylazine; [75]) to be simultaneously exposed to pups in a
choice-arena. Entire animals can also be used to analyze intra-
individual discrimination by pups, scrutinizing whether limited
body regions emit volatiles. For example, nipples were espe-
cially reactogenic for newly born pups and were the target of
numerous studies in the mouse [55, 56, 75]. In this case also,
the stimulus females were anesthetized (e.g., 100 mg/kg ket-
amine, 10 mg/kg xylazine, and 0.2 mg/kg acepromazine
[56]; 50 mg/kg of pentobarbital sodium [55]). Other sources
of salient odorous substrates are the face, mouth, pedal, or
ano-genital regions. If some regions are not of direct interest
for the bioassay, care should be taken to avoid their inclusion
in the presentation device.
2. Odor-active biological substrates. To achieve normal life-
sustenance activities, any living organism produces innumera-
ble chemosensorily active compounds conveyed in excretions,
secretions and wastes. Mus musculus carries an extremely
diverse network of scent producing sources that varies along all
dimensions of biological functioning (e.g., individual, sex, age,
kin, social status, physiological state, stress-anxiety, pathology
and parasite load) (e.g., [8587, 104]). Here, we highlight
some of the biological substrates that are of interest in the con-
text of infants nosing to their dams. But many other biological
substrates from the dam, but also from the father or litter-
mates, will be interesting to screen for behavioral activity of
newborns.
 Testing Smell When It Is Really Vital 359

(a) Skin secretions. The skin surface of adult mice emits a range
of unspecified volatile profiles that pups can regionally dis-
criminate. For example, they olfactorily differentiate back
from ventrum. Smears of dorsal and abdominal skin secre-
tions can be collected by rubbing one edge of a small petri
dish on the mouses abdomen [77], or by rubbing a
humidified cotton-tip over dorsal fur or nipples [73].
(b) Amniotic fluid. At parturition, dams self-lick and thereby
spread traces of amniotic fluid, blood, saliva on their
abdominal fur and around nipples. This fluid is then avidly
attended by newly born pups [56, 75, 88]. Its collection
necessitates anesthesia and rapid euthanization of the
female [75] before externalizing uterine horns and amni-
otic sacs using established surgical techniques. Amniotic
fluid can then be aspired through the amnion (untainted
with blood) with a 2-mL glass Pasteur pipette.
(c) Saliva. For the same reasons as amniotic fluid, saliva is
omnipresent over the nursing dams fur, and in addition is
of both maternal and infant origin. Separate sampling of
maternal and pup saliva consists in delicately introducing
the tip of a glass stick or a humidified small cotton-tip into
the animals mouth (adult or pup; glass stick diameter:
0.4 cm and 0.1 cm, respectively) and let it there for
1015 s [73, 74]. Saliva can also be collected from anes-
thetized dams injected with 0.1 mg pilocarpine to stimu-
late salivation [56].
(d) Milk is an obvious odorivector substrate in lactating
females. What is less obvious is that milk is in constant
change with advancing lactation, and with genotypes,
parity, sex of offspring, mothers diet, etc. Thus, its col-
lection should be done as a function of this compositional
variability. It may be noted that pups respond preferen-
tially to the odor of a milk collected in dams which lacta-
tion age matches their own age (i.e., PND-2 pups prefer
milk of lactation day 2 rather milk from lactation day 15
[75]). To optimize the milking procedure (see also [90]),
it is better to separate dams from their litter for 2 h.
Donor dams are then anesthetized and injected intraperi-
toneally with 0.15 mL oxytocin, to stimulate milk ejec-
tion in association with gentle massages on the mammary
areas (anesthesia does not appear to notably affect the
odor of milk [74]). Milk is aspired from each nipple with
a glass Pasteur pipette (2 mL) to be aliquoted into glass
vessels kept on ice. Milk is then either immediately deep-
frozen (80 C) or (much better) used fresh [7375].
Milking should not be repeated on a same donor female
during a same lactation cycle.
360 Benoist Schaal et al.

(e) Urine, feces. Females can be put on a clean glass surface

(at ambient temperature) to stimulate spontaneous urina-
tion. Abdominal massage can ease micturition [56]. Fresh
urine is then pipetted into glass vessels. Urine and feces are
also often collected in a less controlled fashion, in using
the soiled bedding material from the nest or after letting
the animals defecate/urinate in a test box [84]. Soiled
bedding is then often used to test olfactory discrimination,
in contrasting own litters vs. alien litters beddings (e.g.,
[72, 83, 91]).
(f) Sample conservation issues. To limit alteration (by oxidi-
zation, evaporation) of the odor of the above substrates,
collection operations should be as swift as possible and
recipient vessels kept on ice during collection. Ideally,
the substrates should be tested immediately after sam-
pling. If not, immediate storing at 20 C, or better at
80 C, is mandatory (but even at this temperature sam-
ples remain instable and they should be used as soon as
possible). It is advisable to aliquot the samples in smallest
possible glass vessels in volumes necessary for a single
test. To reduce individual variability of samples, and if
individual variation is not an issue under test, pools of
them should be constituted.
3. Artificial odorants. These odorants are of interest to under-
stand olfaction in neonates for multiple reasons. First, artificial
odorants are often used in social contexts to assess the effects
of experience on emerging preferences, with the assumption
that the learning processes involved with artificial odorants
simulate those that operate in the natural situation. Such odor
probes contributed, for example, to highlight that fetuses
learn odors in utero [42, 92], that neonatal behavior depends
on fetal learning [3, 12], and that neonates learn any odor
associated with the mother as subsequently positive (e.g., [7,
93]). Numerous pure odorant compounds were used with
mouse pups in such learning experiments associating them
with nursing or other aspects of maternal care (ano-genital
licking or its experimental simulation with soft paintbrush) (for
example, n-hexanoic acid and citronellal [76], peppermint and
lemon [79], +carvone and citral [72], benzaldehyde and limo-
nene [94], or acetophenone and isopropyl tiglate [43]). Care
should be taken when selecting such odorants that they bear
little trigeminal properties and that they are as much as possi-
ble matched in intensity (at least for a panel of human noses).
Second, artificial odorants constitute useful controls to test
whether a response is specifically caused by a candidate chemo-
signal and does not result from nonspecific arousal or novelty
effects that would be elicited by any odorant.
 Testing Smell When It Is Really Vital 361

2.2.3 Perinatal and 1. Test paradigms

Neonatal Mice (PND 02) Behavioral assays generally aim to examine an organisms dis-
criminative responsiveness to stimuli that differ along qualita-
tive or quantitative dimensions controlled by the experimenter.
Optimally, they intend to elicit specific responses that the ani-
mals display spontaneously in a given natural situation, rather
than only a nonspecific response (such as nasal flaring or sniff-
ing). A diversity of methods has been achieved to investigate
olfaction-related abilities in neonatal and infant mice. The
most used paradigms rely on differential orientation responses
of the body or head, or on responses that are more specific to
neonates, involving oral reactivity (oral seizing of a nipple or a
nipple-shaped object, sucking).
(a) Differential attraction. In their initial stage of limited
motility, newly born pups can be put in assays requiring
head orientation without additional motor demand. Such
assays, first used in neonatal rats [95] were applied to
mouse pups on PND 0, several minutes or hours after
birth, or on PND 12 [74, 77]. Their principle is that the
pup is set between two odorant stimuli and ensuing head-
turning responses are timed in latency and duration. These
paired-choice tests can be run in two modes, either in con-
trasting two odorant stimuli (O1 vs. O2) to assess relative
orientation, or in contrasting an odorant and a control
stimulus (O1 vs. Control) to assess absolute orientation.
To make sure that the outcome of a relative orientation
assay results from attraction to O1 and not from avoidance
of O2, it should always be arranged with two absolute ori-
entation assays testing either odorants against a scentless
control. Thus, absolute attraction assays evaluate in the
same time detection and intrinsic attraction/avoidance of
an odorant against the control, and help interpreting the
results of relative attraction assays.
(b) Oral responses. Beyond head orientation, newborns display
oral behavior toward certain odor cues, interpreted as reflect-
ing attraction and appetence. Following a short sequence of
bilateral rooting motions, these oral responses involve jaw
extension, grasping and initiating rhythmic sucking actions.
An oral response assay initially developed in the rat newborn
consisted in the presentation of a natural nipple to elicit oral
grasping and sucking [54, 89]. This paradigm was also
applied to the mouse newborn [55, 56, 75].
2. Devices and procedures
(a) Head-orientation assays. Several variants of head-orientation
assays were developed. One variant consisted in a heating
plate (Gestigkeit, Germany) covered with a blotting paper
on which two scentless auto-adhesive labels were stuck
362 Benoist Schaal et al.

symmetrically, at a distance of 2 mm (for PND-0 pups) or

3 mm (for PND-2 pups) on each side of a midline. A cer-
tain volume (510 L) of each stimulus was pipetted on
either label (Fig. 1) [74]. Then, the assay began when the
body and head of the pup were aligned along the midline
of the device. The assays lasted 3 min for PND-0 pups and
or 2 min for PND-2 pups, during which they were video-
taped. In subsequent analyses of these videotapes, the pup
orientation toward each odor stimulus was timed using a
PC-based recording software (e.g., The Observer,
Wageningen, The Netherlands). The tested pup was con-
sidered to be oriented toward a stimulus when both of its
nares crossed the midline of the device in direction of that
stimulus [74]. In the other variant, two scented dishes
were placed 2.5 cm apart at equidistance from a line drawn
on clean tissue paper [77]. The pup was immobilized
between the tips of the experimenters thumb and index
finger and aligned on the line at a distance of 4 cm from
the odor sources. It was held in this way until its head and
shoulders moved enough that their nose touched one of
the dishes, signifying the choice [77]. These assays came
out with latency to orient, total duration spent oriented to
each odor stimulus, or the proportion of litters in which
pups chose one odor stimulus.
(b) Oral response assays. Nipple attachment and suckling assays
were adapted from tests on rat pups by Teicher and Blass
[54, 89] to assess neonatal mice on PND 12 and earlier
when exposed to a nipple in the context of a females
abdomen. Prior to the assay, nipples of stimulus-females
were treated to suppress (washing, masking) their natural
odor, as indicated by the abolishment of the typical oral

Fig. 1 Head-orientation assay adapted for mouse pups aged postnatal day 02.
For the testing, the device is set on a heating plate to keep the pup warm. From
[74]; reproduced by authorization from Public Library of Science
 Testing Smell When It Is Really Vital 363

grasping and suckling responses of pups. In studies using

lactating dams, the washing procedure consisted in rub-
bing nipples for 30 s with 50 L distilled water and drying
them with a cotton swab [56]. Other studies avoided the
nipple washing step in using non-lactating females whose
nipples were shown to be naturally behaviorally inert to
newborn pups [75]. When neutral nipples are achieved by
either method, different target substrates can be painted
onto them by rotating a glass stick soaked with the sub-
strate for 10 s [75] or by painting them with a soft brush
[56]. Whether the substrates of interest reinstate the typi-
cal form and level of response is then directly tested in
presenting the treated nipples to the pups.
Two options were used to present pups to nipples. First,
stimulus-females were anesthetized and laid supine, and the pups
were vertically held by an experimenter at 1 cm from the females
nipple for 2 min to let them approach and orally seize the nipple
[56]. In the second option, the stimulus-females were sedated and
hand-held by an experimenter A by pinching the inter-scapular
skin between thumb and index to vertically expose their abdomen
(Fig. 2) [75]. The pups to be tested were held in the same way by
an experimenter B. The 2-min test began when a pups muzzle was
set in slight contact with the nipple of the stimulus-female. Both
assays were videotaped for subsequent blind coding of latency to

Fig. 2 Nipple seizing assay for newly born mouse pups. A 1-day-old pup is presented
to a nipple in the context of an unfamiliar, behaviorally neutral non-lactating
females abdomen. From [75]; reproduced by authorization from Wiley
364 Benoist Schaal et al.

the first nipple grasping [56], total duration of nipple grasping

and frequency of mouth actions (sucking) during grasping [75].
A device eliciting oral grasping of an artificial (plastic) nipple was
developed to investigate fetal and neonatal rats [9698], and
should be feasible in mouse newborns as well.

2.2.4 Previsual Mouse 1. Test paradigms

Pups (PND 314) During the first week, mouse pups become rapidly mobile. By
PND 3 they are sufficiently active to be introduced into test
arenas for short assays measuring differential attraction towards
stimuli. By that age, they also emit increasingly frequent ultra-
sonic vocalizations in response to isolation from the dam/lit-
ter/nest. These vocalizations are attenuated or stopped when
returned to the normal situation or when familiar odor stimuli
are restituted.
(a) Differential attraction. Several studies have set up double-
choice tests to evaluate differential exploration of social
odorants or of artificial odorants with which pups had dis-
tinctive experience (e.g., on PND 6 [73, 74]; on PND
2P12 [91]).
(b) Detectiondifferential stress buffering effect. The relief
effect familiar odors have on ultrasonic vocalizations
expressed in stressed pups was used as an index of detection
and differential stress buffering potency of odors in several
recent studies (e.g., on PND 8 [72], PND 2P12 [91]).
2. Devices and procedure
(a) Differential investigation: paired-odor choice test. The most
usual test device used to investigate attraction/preference
behavior in rodents is the double-choice arena. Several
such arenas have been successfully applied in infant mice.
The principle is based on a rectangular polypropylene box
with a mesh floor above the bottom, under which both
stimuli are presented (Fig. 3). Several dimensions of the
box were used, some designed for younger pups (e.g.,
Length width height: 7.1 5 7 cm [73, 74]) and some
for older pups (e.g., 20 36 10 cm [78]; 19 32 13 cm
[76]). Under the mesh floor, two equal compartments are
separated by a barrier to avoid mixing of odors. It may be
practical to delimit a neutral zone below the floor to
increase the separation of both odor stimuli (as, e.g., in
[78, 79]). Also, to limit the mixing of odors that vertically
convect above the mesh floor, it is better not to cover the
test arena. In certain cases (when only minute quantities of
the stimulus are available), both stimuli can be presented
in the corners of the arena above the mesh floor to take
advantage of pups tendency to explore corners [73]. The
stimuli can be composed of live animals (Fig. 3) to assess
 Testing Smell When It Is Really Vital 365

Fig. 3 Two-odor choice-arenas used to assess differential investigation of paired odor stimuli in mouse pups
aged 6 days old. During testing, the device is set on a heating plate to keep the pup warm. Left : On the bottom
of the device, below the grid, two live (anesthetized) mice are presented simultaneously to assess pups dif-
ferential investigation of adult conspecifics in different physiological stages. From Al An et al. [73]; reproduced
by authorization from Wiley. Right : Mouse pup roaming in the device on the bottom of which two stimuli are
presented below the grid, placed symmetrically from the midline. From [74]; reproduced by authorization from
Public Library of Science

their relative attraction as a function of physiological state,

or of biological substrates (disposed in one or several petri
dish(es) set symmetrically relative to the midline).
Once stimuli are placed beneath the mesh, the pup is
placed on the midline of the arena, and its roaming in the
two halves of the arena can be videotaped for 2 min [73],
3 min [76, 78], 6 min [91], or 5 1 min [79]. Subsequently
the duration of investigative roaming in either scented
compartment of the arena is blindly analyzed. A clear,
reproducible criterion of orientation should be decided: an
animal was considered oriented over one half of the arena
when its entire muzzle including both eyes crossed the
midline [73, 74] or when the anterior snout of the animal
was in that half of the apparatus [91]. When the stimuli
were smeared on corners of the test arena, the orienting
criterion toward a stimulus corner was when the tip of the
pups muzzle was within the 1-cm radius to it (a line being
drawn on the TV screen for analysis) [73].
(b) Ultrasonic vocalization test. Each tested pup is placed in a
chamber where odorants are delivered in a time-controlled
fashion. The stimuli can be conveyed by an olfactometer
or, more simply, by petri dishes placed under the mesh set
above the bottom of the chamber [72, 91]. The test con-
sists of an initial isolation period of 90 s for baseline
recording of ultrasonic vocalization (USV) followed by
60-s odor exposure [72, 91]. The baseline level of USV is
recorded some time after introducing the pup in the test
366 Benoist Schaal et al.

chamber to avoid the enhanced emission related to handling.

USV are detected with an ultrasonic microphone (SM2
Microphone, Ultra Sound Advice) located at 10 cm above
of the pup. This microphone is coupled to a bat detector
(S25 Bat Detector, Ultrasound advice; frequency range
10180 kHz) that converts ultrasounds into the audible
frequency range. Afterwards, USV are extracted and auto-
matically counted using an in-house software (cf. [85]).
The rate of USV emitted during the last 30 s of isolation is
compared with that emitted during the first 30 s of odor
exposure [72, 91]. Another procedure [71] consists in a
2-min odor-free familiarization period, followed by a 1-min
odor exposure in placing a petri dish containing an odor
stimulus under the mesh, and next followed by replacing
this first dish by a dish containing the other odor for 1 min.
The total duration of USVs during each 1-min odor expo-
sure was measured.

2.2.5 Visual Preweaning After PND 14, mouse pups have mobility and reactiveness levels that
Mouse Pups (PND 1421) compare to those of adults. They begin to gnaw at nest materials and
solid foods, to be fully weaned by PND 21. As early as PND14, and
increasingly so around weaning, they can thus be submitted to a
variety of bioassays devised for adult mice. Pups can then investigate
their environment in moving around, in approaching a target area or
object, in digging, in feeding or drinking. Different kinds of situa-
tions include attraction-preferences assays evaluating differential
investigation of social odors sources presented at opposed sides of
two-choice arenas or Y-mazes, or in water or feeding trays. In such
assays, when one odor stimulus is not differentiated from the other,
explanations may pertain to non-preference because the stimuli were
not detected, because they were not discriminated, or finally because
they had equivalent reinforcing value despite they were discrimi-
nated. Thus, in addition to differential attraction or preference
assays, discrimination assays are required to reach a conclusion on
whether two odorants are differentiated although they bear similar
reinforcing value (see, e.g., [99, 102]). At this age (and perhaps even
earlier), mouse pups can also be subjected to assays evaluating con-
ditioned preferences (see, e.g., [99, 102]).
Here, we will only summarize assays that evaluate differential
attraction. Multiple versions of two-odor choice assays have been
designed to test differential investigative responses in preweanling
mice. One case is an arena made of a rectangular polycarbonate/
Plexiglas box (16 11 13.5 cm [74]; 29 22 10 cm [83]) with a
stainless steel mesh placed 0.7 cm above the bottom. Under the
mesh, two equal compartments (or three compartments of which
one is a neutral area), are delimited by a polypropylene barrier
(Fig. 4) [74, 83]. Both stimuli are either directly inserted into each
compartment [83], or carried by rectangular polypropylene plates
(7.8 11 cm, 0.2 cm thickness [74]). At the beginning of each test,
 Testing Smell When It Is Really Vital 367

Fig. 4 The test box used to assess differential attraction to paired odor stimuli for
mouse pups aged 15 days. From [75]; reproduced by authorization from Wiley

the pup is placed along the midline of the arena between the two
test odors. Each mouse is allowed to move freely in the arena,
while it is videotaped for 2 min [74] or 3 min [83] to code the
duration of investigation in each compartment. A mouse pup is
defined as investigating one of the odors when its entire muzzle
has crossed the midline [74] or its snout and forepaws have entered
the area [83]. A pup is excluded from further analyses when it per-
formed less than three midline crossings or when it urinated [83].
In other studies, the choice device takes the form of a Y-maze
[94]. Petri dishes containing odorized filter paper are placed at the
end of each short arms of the Y-maze (height: 13 cm, width:
6.5 cm, long arm length: 21 cm, short arms length: 15 cm). Both
dishes have a porous cap in order to prevent any contact with the
odorant. The tested pup is placed in the neutral zone of the Y-maze
and left to roam for 3 min while videotaped. The amount of time
the pup spent investigating each short arm is recorded. Another
interesting device has been designed to automatically record the
choice behavior of adult mice [99]. It is constituted of a hole-
board (40 40 cm) in which five holes are opened to receive glass
vessels in which odorized swabs are placed and covered with bed-
ding material. Depending on the test 25 holes can be made acces-
sible to the tested animal, so that different paradigms can be
applied, including detection tasks (in varying the concentrations of
the odorant), discrimination tasks (using a habituationdishabitu-
ation procedure), memory tasks, and preference tasks. When the
mouse introduces its muzzle into a hole, sensors connected to a
computer are triggered to time duration of sniffing and digging
activity [99]. So far, this apparatus has not been used with young
mice, but it holds promise to investigate the development of dif-
ferential responsiveness to social odors.
368 Benoist Schaal et al.

3 Conclusion

How limited our knowledge is on odor-regulated behaviors in

mammalian infants is astounding in regard of their critical impor-
tance for survival and typical development. After the promising
wave of data gathered by developmental psychobiologists in the
1970s and 1980s (e.g., [1, 6, 8, 26, 100]), the area is worthy of
rekindled interest. Such effort promises indeed to be scientifically
rewarding to understand proximate mechanisms of chemo-emission
and chemoreception in murine rodents, especially in a time of
explosive genetic and molecular neurogenetic research in relation
with chemosensory processes in mice. The present methodological
survey focusing on the infant mouse will undoubtedly be enriched
by borrowing additional methods and ideas from similar research
on other rodents (for reviews, e.g., [100102]) and other animals
at large (e.g., [103, 104]).

Acknowledgments

This work was supported by the Centre National de la Recherche

Scientifique (INSB, INEE) and the Regional Council of Burgundy.

References

1. Blass EM, Teicher MH (1980) Suckling. Handbook of behavioral neurobiology, vol 9,

Science 210:1522 Developmental psychobiology and behavioral
2. Schaal B (2010) Mammary odor cues and ecology. Plenum, New York, pp 347388
pheromones: mammalian infant-directed 8. Alberts JR (1987) Early learning and ontoge-
communication about maternal state, mam- netic adaptation. In: Krasnegor NA, Blass EM,
mae, and milk. Vitam Horm 83:81134 Hofer MA, Smotherman WP (eds) Perinatal
3. Alberts JR, Ronca AE (2012) The experi- development: a psychobiological perspective.
ence of being born: a natural context for Academic, Orlando, FL, pp 1137
learning to suckle. Int J Pediatr. doi: 9. Smotherman WP, Robinson SR (1987)
10.1155/2012/129328 Psychobiology of fetal experience in the rat. In:
4. Blass EM (ed) (1986) Handbook of behav- Krasnegor NA, Blass EM, Hofer MA,
ioral neurobiology. Developmental psychobi- Smotherman WP (eds) Perinatal development:
ology and behavioral ecology, vol 9. Plenum, a psychobiological perspective. Academic,
New York Orlando, FL, pp 3960
5. Blumberg MS, Freeman JH, Robinson SR 10. Schaal B, Orgeur P (1992) Olfaction in utero:
(eds) (2010) Oxford handbook of develop- can the rodent model be generalized? Quart J
mental behavioral neuroscience. Oxford Exp Psychol 44B:245278
University Press, New York 11. Youngentob SL, Kent PF, Sheehe PR et al
6. Rosenblatt JS (1983) Olfaction mediates (2007) Experience-induced fetal plasticity:
developmental transitions in the altricial new- the effect of gestational ethanol exposure on
born of selected species of mammals. Dev the behavioral and neurophysiologic olfactory
Psychobiol 16:347375 response to ethanol odor in early postnatal
7. Brake SC, Shair H, Hofer MA (1986) and adult rays. Behav Neurosci 121:
Exploiting the nursing niche: the infants 12931305
sucking and feeding in the context of the 12. Pedersen PA, Blass EM (1982) Prenatal and
mother-infant interaction. In: Blass EM (ed) postnatal determinants of the 1st suckling
 Testing Smell When It Is Really Vital 369

episode in albino rats. Dev Psychobiol 15: world lagomorph conference. University of
349355 Guelph, Guelph, Canada, pp 291294
13. Schaal B (2005) From amnion to colostrum 28. Hudson R, Distel H (1983) Nipple location
to milk: odor bridging in early developmental by newborn rabbits: Evidence for pheromonal
transitions. In: Hopkins B, Johnson SP (eds) guidance. Behaviour 82:260275
Prenatal development of postnatal functions. 29. Keil W, von Stralendorff F, Hudson R (1990)
Praeger, London, pp 51102 A behavioral bioassay for analysis of rabbit
14. Calamandrei G (2004) Ethological and meth- nipple-search pheromone. Physiol Behav 47:
odological considerations in the use of new- 525529
born rodents in biomedical research. Ann Ist 30. Schaal B, Coureaud G, Langlois D et al
Super Sanita 40:195200 (2003) The mammary pheromone of the rab-
15. Maynard-Smith J, Harper D (2003) Animal bit: chemical and behavioural characterisa-
signals. Oxford University Press, Oxford, UK tion. Nature 424:6872
16. Hauser MD (ed) (1996) The evolution of 31. Coureaud G, Langlois D, Perrier G et al
communication. MIT Press, Cambridge, MA (2003) A single key-odorant accounts for the
17. Karlson P, Lscher M (1959) Pheromones: pheromonal effect of rabbit milk: Further test
a new term for a class of biologically active of the mammary pheromones activity against
substances. Nature 183:5556 a wide sample of volatiles from milk.
18. Beauchamp GK, Doty RL, Moulton DG et al ChemoEcol 13:187192
(1976) The pheromone concept in mammals: 32. Coureaud G, Langlois D, Sicard G et al
a critique. In: Doty RL (ed) Mammalian (2004) Newborn rabbit reactivity to the mam-
olfaction, reproductive processes, and behav- mary pheromone: concentration-response
ior. Academic, New York, pp 143160 relationship. Chem Senses 29:341350
19. Johnston RE (2000) Chemical communica- 33. Coureaud G, Rdel H, Kurz CA et al (2008)
tion and pheromones: The types of chemical Age dependent responsiveness to the mam-
signals and the role of the vomeronasal sys- mary pheromone in domestic and wild rab-
tem. In: Finger TE, Silver WL, Restrepo D bits. ChemoEcol 18:5259
(eds) The neurobiology of taste and smell. 34. Coureaud G, Fortun-Lamothe L, Langlois D
Wiley, New York, pp 101127 et al (2007) The reactivity of neonatal rabbits
20. Doty RL (2003) Mammalian pheromones: to the mammary pheromone as a probe for
fact or fantasy? In: Doty RL (ed) Handbook viability. Animal 1:10261032
of olfaction and gustation, 2nd edn. Marcel 35. Montigny D, Coureaud G, Schaal B (2006)
Dekker, New York, pp 345383 Shift from automatism to prandial control in
21. Schaal B (2008) Social odors and phero- the response of newborn rabbits to the mam-
mones in mammals. Biofutur 27:4145 mary pheromone. Physiol Behav 89:742749
22. Wyatt TD (2010) Pheromones and signature 36. Coureaud G, Moncomble AS, Montigny D
mixtures: defining species-wide signals and et al (2006) A pheromone that rapidly pro-
variable cues for identity in both invertebrates motes learning in the newborn. Curr Biol
and vertebrates. J Comp Physiol A. 16:19561961
doi:10.1007/s00359-010-0564-y 37. Knig B, Markl H (1987) Maternal care in
23. Doty RL (2010) The great pheromone myth. house mice. I. The weaning strategy as a
John Hopkins University, Baltimore, MD means for parental manipulation of offspring
24. Pedersen PA, Blass EM (1981) Olfactory con- quality. Behav Ecol Sociobiol 20:19
trol over suckling in albino rats. In: Aslin RN, 38. Ehret G (1976) Development of absolute
Alberts JR, Petersen MR (eds) Development of auditory thresholds in the house mouse (Mus
perception, psychobiological perspectives, vol musculus). J Am Audiol Soc 1:179184
1, Audition somatic perception and the chemi- 39. Weber EM, Olsson IA (2008) Maternal
cal senses. Academic, New York, pp 359381 behaviour in Mus musculus sp. An ethological
25. Smotherman WP, Robinson SR (1992) review. Appl Anim Behav Sci 114:122
Dimethyl disulfide mimics the effect of milk 40. Breer H, Fleischer J, Strotmann J (2006) The
on fetal behavior and responsiveness to cuta- sense of smell: multiple olfactory subsystems.
neous stimuli. Physiol Behav 52:761765 Cell Mol Life Sci 63:14651475
26. Blass EM (1990) Suckling: determinants, 41. Rodriguez I, Boehm U (2008) Pheromone
changes, mechanisms, and lasting impres- sensing in mice. Results Probl Cell Differ.
sions. Dev Psychol 26:520533 doi:10.1007/400_2008_8
27. Schley P (1979) Olfaction and suckling 42. Coppola DM, Millar LC (1997) Olfaction in
behavior in young rabbits. In: Myers K, utero: behavioral studies of the mouse fetus.
MacInnes CD (eds) Proceedings of the 1st Behav Proc 39:5368
370 Benoist Schaal et al.

43. Todrank J, Heth G, Restrepo D (2010) procedures. Neurobehav Toxicol Teratol 7:

Effects of in utero odorant exposure on neu- 579586
roanatomical development of the olfactory 59. Crawley JN, Paylor R (1997) A proposed test
bulb and odor preferences. Proc Roy Soc B battery and constellations of specific behav-
278:19491955 ioral paradigms to investigate the behavioral
44. Coppola DM, OConnell RJ (1989) Stimulus phenotypes of transgenic and knockout mice.
access to olfactory vomeronasal receptors in Horm Behav 31:197211
utero. Neurosci Lett 106:241248 60. Brown RE, Stanford L, Schellinck HM (2000)
45. Coppola DM, Budde J, Millar L (1993) The Developing standardized tests for knockout
Vomeronasal duct has a protracted postnatal and mutant mice. ILAR J 41:163174
development in the mouse. J Morphol 218: 61. Schellinck HM, Forrestel CA, LoLordo VM
5964 (2001) A simple and reliable test of olfactory
46. Tarozzo G, Cappello P, De Andre M et al learning and memory in mice. Chem Senses
(1998) Prenatal differentiation of mouse 26:663672
vomeronasal neurones. Eur J Neurosci 10: 62. van Meer P, Raber J (2005) Mouse behav-
392396 ioural analysis in systems biology. Biochem J
47. Fox WM (1965) Reflex-ontogeny and behav- 389:593610
ioural development of the mouse. Anim 63. Youngentob SL (2005) A method for the
Behav 13:234241 rapid automated assessment of olfactory func-
48. Crawley JN (2000) Whats wrong with my tion. Chem Senses 30:219229
mouse? Behavioral phenotyping of transgenic 64. Cuomo V, De Salvia MA, Petruzzi S et al
and knockout mice. Wiley-Liss, New York (1996) Appropriate end points for the charac-
49. Kovach JK, Kling A (1967) Mechanisms of terization of behavioral changes in develop-
neonate sucking behaviour in the kitten. mental toxicology. Environ Health Perspect
Anim Behav 15:91101 104:307315
50. Singh PJ, Tobach E (1975) Olfactory bulbec- 65. Branchi I, Ricceri L (2002) Transgenic and
tomy and nursing behavior in rat pups. Dev knock-out mouse pups: the growing need
Psychobiol 8:151164 for behavioral analysis. Genes Brain Behav 1:
51. Teicher MH, Flaum LE, Williams M et al 135141
(1978) Survival, growth and suckling behav- 66. Clough G (1982) Environmental effects on
ior of neonatally bulbectomized rats. Physiol animals used in biomedical research. Biol Rev
Behav 21:553561 57:487523
52. Hofer MA, Shair H, Singh P (1976) Evidence 67. Berry R, Bronson FH (1992) Life history and
that maternal ventral skin substances promote bioeconomy of the house mouse. Biol Rev
suckling in infant rats. Physiol Behav 17: 67:519550
131136 68. Smith JS (1981) Mouse model for abortive
53. Wong ST, Trinh K, Hacker B et al (2000) rabies infection in the central nervous system.
Disruption of the type III adenylyl cyclase Infect Immun 16:297308
gene leads to peripheral and behavioral anos- 69. Elwood RW, Keeling F (1982) Temporal
mia in transgenic mice. Neuron 27:487497 organization of ultrasonic vocalizations in
54. Teicher MH, Blass EM (1976) Suckling in infant mice. Dev Psychobiol 15:221227
newborn rats: eliminated by nipple lavage, 70. Branchi I, Santucci D, Vitale A et al (1998)
reinstated by pup saliva. Science 193:422425 Ultrasonic vocalizations by infant laboratory
55. Hongo T, Hakuba A, Shiota K et al (2000) mice: a preliminary spectrographic character-
Suckling dysfunction caused by defects in the ization under different conditions. Dev
olfactory system in genetic arhinencephaly Psychobiol 33:249256
mice. Biol Neonate 78:293299 71. Branchi I, Santucci D, Alleva E (2001) Ultrasonic
56. Logan DW, Brunet LJ, Webb WR et al vocalisation emitted by infant rodents: a tool for
(2012) Learned recognition of maternal sig- assessment of neurobehavioural development.
nature odors mediates the first suckling epi- Behav Brain Res 125:4956
sode in mice. Curr Biol. doi:10.1016/j.cub. 72. Lemasson M, Delb C, Gheusi G et al (2005)
2012.08.041 Use of ultrasonic vocalizations to assess olfactory
57. Butcher RE, Vorhees CV (1979) A prelimi- detection in mouse pups treated with
nary test battery for the investigation of the 3-methylindole. Behav Proc 68:1323
behavioral teratology of selected psychotropic 73. Al An S, Chrati A, Schaal B et al (2011)
drugs. Neurobehav Toxicol 1:207212 Orientation of newborn mice to lactating females:
58. Adams J (1985) Collaborative behavioral ter- biological substrates of semiochemical interest.
atology study: protocol design and testing Dev Psychobiol. doi:10.1002/dev.21003
 Testing Smell When It Is Really Vital 371

74. Al An S, Belin L, Patris B et al (2012) An 90. DePeters E, Hovey R (2009) Methods for
odor timer in milk? Synchrony in the odor of collecting milk from mice. J Mammary Gland
milk effluvium and neonatal chemosensation Biol 14:397400
in the mouse. PLoS One 7:e47228 91. de Chevigny A, Lemasson M, Saghatelyan A
75. Al An S, Belin L, Schaal B et al (2012) How et al (2006) Delayed onset of odor detection
does a newly born mouse get to the nipple? in neonatal mice lacking tenascin-C. Mol Cell
Odor substrates eliciting first nipple grasping Neurosci 32:174186
and sucking responses. Dev Psychobiol. 92. SmothermanW P (1982) Odor aversion learn-
doi:10.1002/dev.21082 ing by the rat fetus. Physiol Behav 29:769771
76. Armstrong CM, DeVito LM, Cleland TA 93. Fillion TJ, Blass EM (1986) Infantile experience
(2006) One-trial associative odor learning in with suckling odors determines adult sexual
neonatal mice. Chem Senses 31:343349 behavior in male rats. Science 231:729731
77. Todrank J, Busquet N, Baudoin C et al 94. Hellier JL, Arevalo NL, Smith L et al (2012)
(2005) Preferences of newborn mice for 7-Nicotinic acetylcholine receptor: role in
odours indicating closer genetic relatedness: early odor learning preference in mice. PLoS
is experience necessary? Proc Roy Soc B 272: ONE 7(4):e35251
20832088 95. Hepper PG (1988) Adaptive fetal learning:
78. Alleva E, Calamandrei G (1986) Odor-aversion prenatal exposure to garlic affects postnatal
learning and retention span in neonatal mouse preferences. Anim Behav 36:935936
pups. Behav Neural Biol 46:348357 96. Petrov ES, Varlinskaya EI, Smotherman WP
79. Bouslama M, Durand E, Chauviere L et al (1997) The newborn rat ingests fluids
(2005) Olfactory classical conditioning in through a surrogate nipple: a new technique
newborn mice. Behav Brain Res 161: for the study of early suckling behavior.
102106 Physiol Behav 62:11551158
80. Hurst JL, West RS (2010) Taming anxiety in 97. Cheslock S, Varlinskaya E, Petrov E et al
laboratory mice. Nat Methods 7:825826 (2000) Rapid and robust olfactory condition-
81. Durand E, Dauger S, Vardon G et al (2003) ing with milk before suckling experience: pro-
Classical conditioning of breathing pattern motion of nipple attachment in the newborn
after two acquisition trials in 2-day-old mice. rat. Behav Neurosci 114:484495
J Appl Physiol 94:812818 98. Smotherman WP, Goffman D, Petrov E et al
82. Zorrilla EP (1997) Multiparous species pres- (1997) Oral grasping of a surrogate nipple by
ent problems (and possibilities) to develop- the newborn rat. Dev Psychobiol 31:317
mentalists. Dev Psychobiol 30:141150 99. Mandairon N, Sultan S, Rey N et al (2009) A
83. Thomas NR, Fonken LK, LeBlanc M et al computer-assisted odorized hole-board for
(2010) Maternal separation alters social odor testing olfactory perception in mice.
preference development in infant mice (Mus J Neurosci Methods 180:296303
musculus). J Comp Psychol 124:295301 100. Shair HN, Barr GA, Hofer MA (eds) (1991)
84. Breen MF, Leshner AI (1977) Maternal pher- Developmental psychobiology. New methods
omone: a demonstration of its existence in the and changing concepts. Oxford University
mouse (Mus musculus). Physiol Behav 18: Press, New York
527529 101. Slotnick BM, Schellinck HM (2002) Methods
85. Hurst JL, Beynon RJ (2004) Scent wars: the in olfactory research with rodents. In: Simon
chemobiology of competitive signalling in SA, Nicolelis M (eds) Frontiers and methods
mice. Bioessays 26:12881298 in chemosenses. CRC, New York, pp 2161
86. Brennan PA, Zufall F (2006) Pheromonal 102. Schelling HM, Price SR, Wong JM (2008)
communication in vertebrates. Nature Using ethologically relevant tasks to study
444:308315 olfactory communication in rodents. In:
87. Ferrero D, Liberles SD (2010) The secret Hurst JL, Beynon RJ, Roberts SC, Wyatt TD
codes of mammalian scents. WIREs Syst Biol (eds) Chemical signals in vertebrates, vol 11.
Med 2:2333 Springer, New York, pp 7180
88. Kodama N (1990) Preference for amniotic 103. Haynes KF, Millar JG (eds) (1998) Methods
fluid in newborn mice. Annual Meeting of the in chemical ecology, vol 2, Bioassay methods.
Int Soc Dev Psychobiol, Cambridge, UK Klwer Academic, Norwell, MA
89. Teicher MH, Blass EM (1977) First suckling 104. Brown RE (1985) The rodents II: suborder
response of the newborn albino rat: the roles Myomorpha. In: Brown RE, Macdonald DW
of olfaction and amniotic fluid. Science 198: (eds) Social odours in mammals, vol 1.
635636 Clarendon, Oxford, pp 345457
 Chapter 27

An Assay for Human Chemosignals

Idan Frumin and Noam Sobel

Abstract
Like all mammals, humans use chemosignals. Nevertheless, only few such chemosignals have been identified.
Here we describe an experimental arrangement that casts a wide net for the possible chemosignaling
functions of target molecules. This experimental arrangement can be used in concert with various methods
for measuring human behavioral and brain responses, including psychophysiology and brain imaging.
Moreover, many of the methodological issues we describe are relevant to any study with human
chemosignals.

Key words Human, Chemosignals, Pheromones, fMRI, Psychophysiology, Psychophysics

1 Introduction

All mammals communicate using chemosignals, and humans are

no different. Mammalian chemosignaling is especially prominent
in reproduction-related behaviors, and this too is true for humans.
For example, the clearest case of chemical communication in
humans is the phenomenon of menstrual synchrony, whereby
women who live in close proximity, such as roommates in dorms,
synchronize their menstrual cycle over time [1]. This effect is
mediated by an odor in sweat. This was verified in a series of studies
where experimenters obtained underarm sweat extracts from donor
women during either the ovulatory or follicular menstrual phase.
These extracts were then deposited on the upper lips of recipient
women, where follicular sweat accelerated ovulation, and ovula-
tory sweat delayed it [2, 3]. Moreover, variation in menstrual tim-
ing can be increased by the odor of other lactating women [4], or
regulated by the odor of male hormones [5, 6].
A second human reproduction-related chemosignaling behav-
ior relates to mate selection. The human genome includes a region
called Human Leukocyte Antigen (HLA), or more broadly termed
Major Histocompatibility Complex (MHC), which consists of
many genes related to the immune system, in addition to olfactory

Kazushige Touhara (ed.), Pheromone Signaling: Methods and Protocols, Methods in Molecular Biology, vol. 1068,
DOI 10.1007/978-1-62703-619-1_27, Springer Science+Business Media, LLC 2013

373
374 Idan Frumin and Noam Sobel

receptor genes and pseudogenes. Several studies have found that

women can use smell to discriminate between men as a function of
similarity between their own, and the mens HLA alleles [711].
The ideal smell of genetic makeup remains controversial, yet
most evidence suggests that women prefer an odor of a man with
HLA alleles not identical to their own, but at the same time not
too different [11, 12]. In turn, this preference may be for MHC
heterozygosity rather than dissimilarity [13]. This chemosignaling
dependent mate preference is plastic. For example, single women
preferred odors of MHC-similar men, while women in relation-
ships preferred odors of MHC-dissimilar men [14]. Moreover,
olfactory mate preferences are influenced by the menstrual cycle
[1518] and by hormone-based contraceptives [7, 8, 19]. Finally,
olfactory influences on mate preferences are not restricted to
women. Men can detect an HLA odor different from their own
when taken from either men or women odor donors, and rate the
similar odor as more pleasant for both of the sexes [8, 13]. In addi-
tion, men preferred the scent of common over rare MHC alleles
[13]. Moreover, unrelated to HLA similarity, male raters can detect
the menstrual phase of female body odor donors. The follicular
phase is rated as more pleasant and sexy than the luteal phase [18],
an effect that is diminished when the women use hormonal contra-
ceptives [13, 20].
These behavioral results are echoed in hormone expression.
Men exposed to the scent of an ovulating woman subsequently
displayed higher levels of testosterone than did men exposed to the
scent of a non-ovulating woman or a control scent [21]. Moreover,
a recent study on chemosignals in human tears revealed a host of
influences on sexual arousal [22]. Sniffing negative-emotion-
related odorless tears obtained from women donors induced reduc-
tions in sexual appeal attributed by men to pictures of womens
faces. Sniffing tears also reduced self-rated sexual arousal, reduced
physiological measures of arousal, and reduced levels of testoster-
one (recently also seen by Oh and colleagues [23]). Finally, func-
tional magnetic resonance imaging revealed that sniffing womens
tears selectively reduced activity in brain substrates of sexual arousal
in men [22].
Human chemosignaling is not restricted to reproduction-
related behavior. Although many types of social chemosignaling
have been examined [24], here we will detail one particular case,
and that is the ability of humans to smell fear. Fear or distress che-
mosignals are prevalent throughout animal species [25, 26]. In an
initial study in humans, Chen and Haviland-Jones [27] collected
underarm odors on gauze pads from young women and men after
they watched funny or frightening movies. They later asked other
women and men to determine by smell, which was the odor of
people when they were happy or afraid. Women correctly
identified happiness in men and women, and fear in men. Men
 An Assay for Human Chemosignals 375

correctly identified happiness in women and fear in men. A similar

result was later obtained in a study that examined women only [28].
Moreover, women had improved performance in a cognitive verbal
task after smelling fear sweat versus neutral sweat [29], and the
smell of fearful sweat biased women toward interpreting ambigu-
ous expressions as more fearful, but had no effect when the facial
emotion was more discernible [30]. Also, subjects had an increased
startle reflex when exposed to anxiety-related sweat versus sports-
related sweat [31]. Finally, imaging studies have revealed disso-
ciable brain representations after smelling anxiety sweat versus
sports-related sweat [32]. These differences are particularly pro-
nounced in the amygdala, a brain substrate common to olfaction,
fear responses, and emotional regulation of behavior [33]. Taken
together, this body of research strongly suggests that humans can
discriminate the scent of fear from other body odors, and it is not
unlikely that this influences behavior.
How can we assay whether a given substance is, or contains, a
human chemosignal? The rational for how to do this is simple:
Once we have identified a chemosignal and a behavior we think it
relates to, we can measure that behavior, or its neural substrates,
with and without exposure to the chemosignal. Thus, if we have
identified a fear-related chemosignal, we can measure fear in the
presence of the chemosignal versus the presence of an unrelated
control substance, or measure brain activity in the amygdala for
example, again in the presence of the chemosignal versus the pres-
ence of an unrelated control substance. However, in many cases we
may have a potential chemosignal in hand without a clear notion
regarding its expected influence. With this in mind, we have devel-
oped a behavioral assay that provides a rather widely cast net. This
is an experiment aimed at probing for a host of potential psycho-
logical, physiological, and brain responses. In this chapter we will
describe this assay. The human responses within this assay can be
measured with several standard methods, for example psychophysi-
ology and brain imaging. The general application and analysis of
psychophysiology and brain imaging has been detailed in various
chapters of this series [3437], and is indeed beyond the scope of
one text. Therefore, here we will concentrate on the unique aspects
of assaying human chemosignals with these methods. To reiterate,
this chapter is not intended to teach psychophysiology and brain
imaging, but rather how to bring human chemosignals into this
environment. We also detail various alternatives regarding major
design aspects of such experiments. For example, for stimuli one
can use the full human-derived media that presumably contains the
chemosignals (sweat, tears, etc.), or individual synthetic molecules
that have been dubbed putative human chemosignals. Also, one can
measure the direct impact of the stimuli, or the influence of the
stimuli on some task, such as emotional appraisal, or startle response.
An additional major design aspect with extreme alternatives has to
376 Idan Frumin and Noam Sobel

do with stimulus delivery. If one wants to deliver the stimuli with

millisecond temporal resolution, one needs an olfactometer. These
devices, which can be self-built [3840] or bought [41], are com-
plicated and expensive, and are therefore in the hands of relatively
few labs. With this in mind, in this chapter we will restrict our
description to methods that do not call for an olfactometer. Finally,
Subheading 4 contains various insights from our experience with
apparently small decisions that sometimes make all the difference
between a successful and unsuccessful experiment. We wish you
good luck with yours.

2 Materials

2.1 General 1. Ethical approval for the procedures from appropriate authori-
ties (Helsinki or IRB committee).
2. Human volunteers (~30 per study).
3. General questionnaires. These should include a comprehensive
demographics questionnaire, and the Ekman mood question-
naire [42] (see Note 1). Questionnaires should be made exe-
cutable on-screen using presentation software (Fig. 1).
4. Well-ventilated room, subserved by Carbon and HEPA filtra-
tion, ideally coated with odorant non-adherent material such
as stainless steel (see Note 2). The room should be observable
from a neighboring control room through one-way mirror
and/or video monitors such that subjects can be left alone in
the room during the experiments. An intercom between exper-
imental and neighboring control room is helpful.
5. A subject-chair that is both comfortable and adjustable, ideally
a dentist-type patient chair (Fig. 2). The chair should be
equipped with a wide armrest that can be refitted for either the
left or right arm. This armrest is for the non-dominant hand

Fig. 1 Schematic of on-screen Visual Analog Scale (VAS). This graphic is pre-
sented on the screen in front of the subject. The subject uses the mouse to drag
the marker horizontally to a position that reflects their self-assessment of the
current mood in question (in this case happy). Once the marker is in the appro-
priate place, the subject clicks the mouse to enter their judgment, and the next
mood question appears (e.g., happy is replaced with sad). This continues for
the 17 mood questions (see Note 1). Although this may seem crude, it is in fact
informative and reliable
 An Assay for Human Chemosignals 377

Fig. 2 Subject set-up in chair. Subject comfortably seated in stainless steel room.
Visible transducers include body temperature (temp.), ear pulse (EP), nasal res-
piration (Cannula), thoracic respiration belt (TR), abdominal respiration belt (AR),
and inset highlights skin conductance sensors (SCR), finger pulse (FP), and blood
pressure (BP). Note monitor in easy viewing angle, and one-way mirror behind
monitor, which allows viewing from neighboring experimenter control room

later fitted with physiological transducers. The opposing armrest

should have a keyboard and mouse holder, to allow subject
responses. Display hardware, e.g., computer monitor, should
ideally be situated in comfortable viewing angle from the chair.

2.2 Collecting 1. Scentless soap.

Body-Odor or Sweat 2. Cotton pads or cotton shirts.
3. Medical adhesive tape.
4. Sealable aluminum-lined plastic bags (sized to contain afore-
mentioned pads/shirts) (see Note 3).
5. Refrigeration for the samples at 4 C or below (see Note 4).
6. For sweat: An emotional setting for the active condition (see
Subheading 3.1.2), and a treadmill/exercise bicycle for the
control.

2.3 Collecting Tears In all cases:

1. Vials/Tubes (preferably glass, wide opening (12 cm
diameter)).
2. Saline/physiological solution, or ringer solution (see Note 5).
378 Idan Frumin and Noam Sobel

2.3.1 Emotional Tears 1. A computer or TV/DVD set for screening movies (head-
phones optional).
2. A sad movie (ideally chosen by the subject) (see Note 6).
3. A small make-up mirror, or similar.

2.3.2 Trigeminal/ 1. Nasal endoscope (ideally <2 mm OD flexible) (see Note 7).
Reflexive Tears 2. Glass Capillaries (50200 l) (e.g., Hirschmann Micropipettes
or Drummond Microcaps).
3. A rubber bulb or micropipette pump to fit the aforementioned
capillaries.

2.4 Preparing 1. Synthetic chemosignal, currently commonly used include:

Synthetic Putative (a) Androstadienone (androsta-4,16,-dien-3-one; CAS#
Chemosignals 4075-07-4) (see Note 8).
(b) Estratetraenol (1,3,5(10),16-estratetraen-3-ol; CAS#
1150-90-9).
(c) Androstenone (5-androst-16-en-3-one ; CAS# 18339-
16-7) (see Note 9).
2. DiluentMost commonly propylene glycol (1,2 propanediol;
CAS# 57-55-6) (see Note 10).
3. Clove oil/Eugenol (4-Allyl-2-methoxyphenol; CAS# 97-53-0)
(see Note 11).
4. Analytical scale.
5. Chemical hood.
6. Glass vials and working tools such as spatula (see Notes 2 and 17).
7. Vortex and/or magnetic stirrer (see Note 12).

2.5 Delivering 1. Band-aids.

Chemosignals 2. Pipette.
3. Opaque wide-mouthed jars (ideally glass) with both sealed and
mesh covers.
4. Absorbent material, either cotton pads or PTFE beads (see
Note 13).

2.6 Measuring 1. A computer with stimulus-presentation software, either desig-

Psychophysical and nated (e.g., Psyscope [43], E-Prime [44]) or programmable
Psychophysiological (e.g., Matlab [45], LabVIEW [46]). The computer should
Responses output to two separate monitors, one in front of the subject
chair, and one in the experimenter control room.
2. Pre-rated sets of short emotional movie clips. We use 10-min
segments of the nature film Deep Blue as neutral, 5-min
segments of the film Nine and a half weeks as erotic, a 5-min
segment from the movie The Champ as sad, and 5-min seg-
ments from Monty Python as funny (see Notes 6 and 14).
 An Assay for Human Chemosignals 379

3. A human approved psychophysiology rig (e.g., AD Instruments

[47]) consisting of:
4. 16-channel (or more) biosignal amplifier.
5. Skin conductance (SCR) transducer with two bipolar finger
Ag/AgCl Electrodes (surface: 1 cm2).
6. Electrocardiogram transducer (ECG) with three circular Ag/
AgCl conductive adhesive electrodes (0.9 cm diameter).
7. Finger and/or ear pulse transducer (FP/EP) using an IR ple-
thysmograph (size: 15 mm 15 mm 6.3 mm).
8. Skin surface temperature transducer (ST) made of a small
ceramic-encapsulated metal oxide semiconductor (9.5 mm in
length, 2 mm in diameter), designed to operate from 0 to
50 C.
9. Abdominal and thoracic respiration belt transducers (AR and
TR) (30 cm rest length, 10 cm maximum elongation, 4.5 cm
in width) mounted on Velcro belts. The transducers contain a
piezoelectric device that responds linearly to changes in length
(sensitivity, 4.5 1 mV/mm).
10. Motion transducer attached to the subject chair (e.g., high-
sensitivity (2,500 mV/g) (EGCS, Entran Devices, Fairfield,
NJ [48])) (see Note 15).
11. Continuous blood-pressure (BP) finger-cuff monitor (e.g.,
Finapres Ohmeda 2300 [49]) (see Note 16).
12. Nasal airflow transducer/spirometer and nasal cannula to fit.
13. Biosignal presentation and analysis software (e.g., AD
Instruments Chart [47]).

2.7 Measuring 1. Bioassy kits for measuring hormones in saliva (e.g., Salimetrics
Endocrine Responses [50] Testosterone and Cortisol kits [51, 52]).
2. Collection tubes (e.g., 15 ml conical PP tubes). Optional
short straws.
3. Refrigerator + Freezer.
4. Polypropylene (PP) Microtubes (e.g., Eppendorf conical
1.5 ml microtubes).
5. Centrifuge (suitable for microtubes (1.5 ml)).
6. Precision pipettes for 20 l range and for 200 l range.
7. Multichannel pipette for 200 l range.
8. Plate agitator/rotator.
9. Spectrophotometric plate reader, to match kit instructions
(e.g., for Salimetrics kits450 nm filter).
10. Computer software capable of 4-parameter sigmoid minus
curve fit (4PL logistics non-linear regression) (e.g., SigmaPlot
[53], Microsoft Excel Solver tool [54] or MyAssays add-in [55],
Matlab [45]).
380 Idan Frumin and Noam Sobel

2.8 Imaging Neural 1. An MRI machine (either 1.5 or 3 T) with gradient capability
Responses for functional MRI (fMRI) and appropriate head-coil for func-
tional brain imaging.
2. MRI compatible projection, and MRI compatible earphones.
3. High-powered personal computer with extensive memory and
imaging analysis software (e.g., SPM [56], BrainVoyager [57],
MR Vista [58]).

3 Methods

First, one must select which human chemosignal one intends to

study, and collect/prepare it as follows (see Note 17):

3.1 Collecting 1. Supply the subject with scentless soap and either a shirt or pads
and Preparing in a sealed plastic bag (see Note 3).
the Chemosignal 2. Instruct the subject to avoid eating extremely odorous foods
3.1.1 Collecting that may influence body odor, such as fenugreek, asparagus
Body-Odor and garlic.
3. Instruct the subject to wash with the scentless soap before bed-
time, not to wear any perfume or deodorant, and wear the
supplied shirt over night.
4. In the morning, the subject is asked to put the shirt in the bag
and reseal it, and put it in refrigeration for later collection by
the experimenter (see Note 4).

3.1.2 Collecting 1. Fit the subject with cotton pads placed under the armpits and
Axillary Sweat secured in place using medical adhesive tape. Alternatively, the
subject wears a cotton shirt whose armpits are later cut out.
2. Place the subject in the sweat generating condition.
Control sweat: Subject is placed on either a treadmill or exercise
bicycle for ~30 min.
Chemosignal sweat: For the condition of interest, the subject is
placed in an appropriate setting. For example, for fear
sweat we recommend either tandem skydiving or bungee
jumping. Stress sweat can be typically obtained from col-
lege students during a statistics exam.
3. Immediately after the sweat is obtained (either control or che-
mosignal), the relevant emotion is assessed. For example, for
fear, after the activity, subjects are requested to rate their level
of fear on a 10-point scale ranging from not afraid at all to
the most afraid I have ever been.
4. Immediately after the activity, the pads or the cut armpit areas
of the shirt are stored in a sealed plastic bag and kept under
refrigeration (see Note 4).
 An Assay for Human Chemosignals 381

Fig. 3 Collecting tears. To collect tears, the subject watches a sad movie of their
own choice in isolation, and once tears start to trickle, they collect them directly
into a vial, assisted by a small mirror. Observe the process from neighboring
room via one-way mirror or video monitor to assure reliability.

3.1.3 Collecting Tears Tear collection technique may influence tear fluid content [59,
60]. Ask donors to avoid use of creams, lotions or makeup. Instruct
subjects to wash their face without using soap before the collection
begins.

Emotional Tears 1. Seat the subject in a comfortable room, equipped with a display
and Control device such as a computer, DVD or projector.
2. Instruct the subject on how to collect tearsby using a mirror
and a collection vial, capturing the teardrops directly into the
vial as they visibly trickle down the cheek (Fig. 3).
3. To demonstrate this to the subject, use a pipette to apply drops
of saline or ringer solution under the subjects eye, and let the
subject capture these drops into the collection vial. This trick-
led saline will later serve as the control compound.
4. Leave the subject alone in the room, and project the pre-
selected sad movie (see Note 6). Continue to monitor the sub-
ject through one-way mirror or video.
5. Ask the subject to call for you once they have obtained tears in
the vial, ideally up to the 1 ml mark.
6. Use the tears as soon as possible, but always before 3 h.

Trigeminal/Reflexive Tears 1. Seat the subject in a comfortable chair that allows tilting of the
head backwards and sideways.
2. An experienced ENT physician should insert a sterilized flexi-
ble endoscope into the naris, and make contact with the sep-
tum near the inferior turbinate (see Note 7).
3. The contact of the endoscope tip with the trigeminal nerve end-
ings-rich tissue should elicit a tearing reflex in the ipsilateral eye.
382 Idan Frumin and Noam Sobel

4. As soon as tears begin to collect in the eye, tilt the head of the
subject gently sideways, while keeping the endoscope in place
and avoiding harsh movements. The tilt should be in the direc-
tion of the narisi.e., for a right naris and right eye tilt the
head to the right (see Note 18).
5. Use a glass capillary tube to capture the tears as they drop out
of the eye (see Note 19).
6. Immediately empty the capillary into a vial.

3.1.4 Synthetic Putative The chosen chemosignal can be presented in either solid form or
Chemosignals solution (see Note 20). Conduct all the following procedures
inside a chemical hood to avoid dust and smell contamination of
the surrounding area (see Notes 2 and 17).

Preparation in Solution 1. Weigh the desired amount of solid chemosignal (e.g., for a
2 mM concentration of 10 ml androstadienone solution take
~5 mg solid androstadienone crystals (m.w. 270.4 g/mol) and
dissolve in 10 ml of propylene glycol).
2. To generate masked-odor solution (as in [61, 62]), use 1 %
w/w eugenol in propylene glycol as the diluent (see Note 11).
3. Rigorously vortex or magnetically stir the solution until no
crystals are visible. This might require a couple of hours.
Mild heat (<35 C) can be applied to facilitate dissolution
(see Note 12).
4. The above stock solution can be further diluted to any desired
concentration by using propylene glycol or propylene gly-
col + 1 % eugenol (e.g., for 10 ml 250 M final concentration
take 1.25 ml of the above stock solution and mix it with
9.75 ml diluent). For threshold concentrations see ref. [63].
5. Keep the stock and diluted solutions in an airtight glass vial,
preferably under refrigeration and/or well-ventilated storage.

3.2 Narrative of The most powerful designs are within-subjects, that is, the same
Psychophysiology subjects are tested in two conditions (chemosignal and control),
Experiment counter-balanced for order, and double-blind as to compound
identity. In other words, neither the participant, nor the experi-
menter interacting with them, should know which experiment is
with the chemosignal or with the control. We recommend that for
each subject, the two experiments be conducted day after day and
at the same time (see Note 21), so as to minimize external sources
of variance. In overview, for each experiment you will obtain a
baseline for all measures, then you will conduct a stimulus expo-
sure, and then continue to monitor the response over about an
hour. Together, this makes for nearly a 2-h experiment, which you
will repeat day after day.
 An Assay for Human Chemosignals 383

3.3 Psycho- 1. A same-sex experimenter should greet the subject (i.e., women
physiology Set-Up subject by women experimenter).
2. Obtain informed consent.
3. Escort the subject to the experimental room, and leave them
there alone to complete an initial on-screen baseline mood
questionnaire.
4. Return to the room, and provide the subject with a collection
tube and collect a baseline saliva sample (see Notes 22 and 26).
5. Set the remaining collection tubes, all clearly consecutively
numbered, within a holder in easy reach of the subjects domi-
nant hand.
6. The same-sex experimenter now applies the various psycho-
physiology transducers to the subject:
Skin Conductance Response (SCR): Place the SCR electrodes
on the second phalanx of the index and the third digit of
the non-dominant hand and attach with Velcro strap.
Electrocardiogram (ECG): Paste the two signal electrodes on
the left and right lower rib, above the abdomen. Paste the
ground electrode on the left ankle.
Finger pulse/ear pulse (FP/EP): Using a Velcro strap, place
one plethysmograph on the pinky finger of the non-domi-
nant hand. Using an ear-clip, place another plethysmo-
graph on the ear lobe on the side of the non-dominant
hand.
Skin temperature (ST): Using medical adhesive tape, paste the
thermistor directly below the non-dominant axilla.
Abdominal and thoracic respiration (AR and TR): Place AR
belt around the belt line, and the TR belt around the chest
of the subject, just below the axilla.
Continuous blood pressure (BP): Place the BP finger-cuff on the
third finger of the non-dominant hand, i.e., the finger
between the two fingers with SCR electrodes.
Spirometer: Fit a nasal cannula to the subject.
Motion: This is the only transducer not fitted to the subject,
but rather to the armrest of the subject chair.
7. Leave the subject alone in the experimental room, instructing
them that if an experimenter later enters, they are not to engage
in conversation with them.
8. From the control room, observe that all physiological variables
are reading properly at 1 kHz, and if any are particularly noisy,
return to the subject room to reconnect sensors as necessary
(see Note 23). Start the recording at 1 kHz (make sure you
have sufficient memory for a 2-h recording).
384 Idan Frumin and Noam Sobel

3.4 Psycho- 1. Commence with 10-min baseline recording. During these

physiology Experiment 10 min project a baseline nature film.
2. Project the mood questionnaire.
3. Project instruction to spit into the second collection tube.
4. An opposite sex experimenter now enters the room with a jar
containing the stimulus. The experimenter stands next to the
subject, opens the jar such that it remains with mesh cover
only. At this point, the presentation software should sound an
auditory instruction as follows: At the tone, sniff for the dura-
tion of the tone, three, two, one, TONE. Tone duration
should be 1.6 s. At the two count, the experimenter should
bring the open jar to the nose of the subject, allowing them to
take a sniff. The experimenter should pull the jar away at the
end of the tone. At this point, consecutive on-screen VAS
scales should appear, requesting the subject to rate intensity
and pleasantness of the stimulus. The above constitutes one
trial. The entire exposure should consist of ten trials, with 30-s
inter-trial intervals.
5. After the ten sniffs, the experimenter removes the nasal can-
nula from the subject, and pastes the pre-prepared stimulus-
containing (~100 l) band-aid to the upper lip of the subject,
to allow continued exposure throughout the experiment
(Fig. 4). The experimenter then exits the room.
6. The subject receives an on-screen mood questionnaire.
7. The subject receives on-screen instructions to spit into the
third collection tube.

Fig. 4 Continued exposure to chemosignals. (a) Use a simple band-aid. (b, c) Fold
the band-aid backwards to form a pad with adhesive on the back. (d) Use a pipette
to deposit chemosignal onto the pad (~100 l). (e) Paste the pad on the upper lip
 An Assay for Human Chemosignals 385

8. Project one of three mood-induction movies (erotic/sad/

funny, counterbalanced for order across subjects).
9. The subject receives an on-screen mood questionnaire.
10. The subject receives on-screen instructions to spit into the
fourth collection tube.
11. Project a baseline nature film.
12. Project the second of three movies (erotic/sad/funny).
13. The subject receives an on-screen mood questionnaire.
14. The subject receives on-screen instructions to spit into the fifth
collection tube.
15. Project a baseline nature film.
16. Project the third of three movies (erotic/sad/funny).
17. The subject receives an on-screen mood questionnaire.
18. The subject receives on-screen instructions to spit into the
sixth collection tube.
19. A same-sex experimenter enters the room, disconnects the sub-
ject from the various transducers, and invites them to return at
the same time tomorrow. The full timeline for this experiment
is in Fig. 5.
20. Freeze saliva samples at or below 20 C. Samples can be
transferred to 1.5 ml microtubes at this stage to save space
(see Note 24).
21. On the next day, the procedures are repeated exactly, yet using
the second compound (chemosignal/control). The order of
the three mood-induction film clips should be counter-
balanced across subjects.

3.5 Narrative of Those who have an fMRI-compatible olfactometer (likely less than
Imaging Experiment ten labs in the world) can investigate the event-related response to
the chemosignal alone. Most labs, however, lacking such an olfac-
tometer, are primarily restricted to examining the brain response to
a task or set of stimuli from either the auditory or visual domains,
yet under two separate conditions: after exposure to chemosignals

Fig. 5 Schematic timeline of psychophysiology experiment. M mood questionnaire, S saliva collection, Set-up
hook up all physiological transducers, VN emotionally neutral video clip taken from a nature film, Exp exposure
to chemosignal or control, starting with ten sniffs, and then placing of band-aid on upper lip, VE1VE3 the three
mood-induction video clips; sad, happy, and sexually arousing counter balanced for order across subjects
386 Idan Frumin and Noam Sobel

versus after exposure to control. We further recommend indepen-

dent localizer tasks before exposure to either compound. The
imaging task can include complex event-related designs, yet here
we detail the simplest block-design approach.

3.6 Imaging 1. Screen subject for MRI compatibility.

Experiment 2. Place subject in scanner.
3. Obtain full-head T1 weighted high-resolution (1 mm3) 3D
acquisition.
4. Obtain a T1 weighted reduced resolution full-brain acquisition
at identical parameters (resolution and orientation) to those
later used for functional scans. This is for later drawing of
regions of interest (ROIs), and realignment.
5. Set parameters for full-brain functional scans (typically T2*
weighted). Do not use TRs greater than 2 s, and use low TE to
reduce loss in ventral temporal susceptibility regions.
6. Commence block-design localizer: project alternating 30-s
movie segments containing either neutral or emotion-of-
interest content. For example, to study sexual behavior related
chemosignals, alternate neutral and erotic clips. To study
fear responses, alternate neutral and frightening clips.
Provide 6 alternations across conditions.
7. Extract scanner bed, but instruct subject to remain motionless.
8. With the subject supine in the scanner bed, conduct stimulus
exposure as in the psychophysiology experiment. Specifically:
9. An opposite sex experimenter stands next to the subject, opens
the jar such that it remains with mesh cover only. At this point,
the presentation software should sound an auditory instruc-
tion as follows: At the tone, sniff for the duration of the tone,
three, two, one, TONE. Tone duration should be 1.6 s. At
the two count, the experimenter should bring the open jar
to the nose of the subject, allowing them to take a sniff. The
experimenter should pull the jar away at the end of the tone.
At this point, consecutive on-screen VAS scales should appear,
requesting the subject to rate intensity and pleasantness of the
stimulus. The above constitutes one trial. The entire exposure
should consist of ten trials, with 30-s inter-trial intervals.
10. After the ten sniffs, the experimenter pastes the pre-prepared
stimulus-containing (~100 l) band-aid to the upper lip of the
subject, to allow continued exposure throughout the experi-
ment (Fig. 4). The subject is then reinserted into the scanner.
11. Again obtain a T1 weighted reduced resolution full-brain
acquisition exactly as in step #4.
12. Again set parameters for full-brain functional scans exactly as in
step #5.
 An Assay for Human Chemosignals 387

Fig. 6 Schematic timeline of imaging experiment. T1HR high-resolution T1 full

head acquisition, T1LR low resolution T1 acquisition at same orientation as later
functional scans, T2* functional scans with alternating video clips, neutral and
emotion of interest, Exp extraction from scanner, and exposure to chemosignal
or control, starting with ten sniffs, and then placing of band-aid on upper lip

13. Commence block-design experiment exactly as in step #6.

14. Extract subject, and invite them to return at the exact same
time on the next day. The full time line for this experiment is
in Fig. 6.
15. Repeat all of the above on the next day, yet using the other
(chemosignal/control) compound.

3.7 Psycho- For each subject, for each day, and for each physiological measure,
physiology Analysis mood question, and hormone in saliva:
Pointers
1. Extract the baseline values obtained during the 10-min
baseline.
2. Extract the values obtained during the ten-sniff stimulus
exposure.
3. Extract the values obtained during each type of emotional
setting.
4. For each subject, compute a percent change score from base-
line for each measure on each day.
5. Compare the change from baseline in the chemosignal condi-
tion to the change from baseline in the control condition (see
Note 25).

3.8 Endocrine 1. Thaw saliva, vortex for a few seconds, and centrifuge at
Analysis Pointers 1,500 g for 15 min to precipitate particulate material and
proteins. Use only the clear supernatant, carefully avoiding dis-
ruption of the pellet.
2. Use your specific chosen kit instructions for reagent prepara-
tion and needed volumes for analysis.
3. Analysis is best performed in triplicates for each time point of
each subject.
4. Arrange the layout of the plate in advance. Leave space as
instructed by the kit protocol for standard curve measures and
388 Idan Frumin and Noam Sobel

controls, and design the rest of the space so that each triplicate
spans two rows and columns, in an L-shaped pattern. This is
done in order to avoid intra-column differences across the plate.
5. Upon completion of the kit procedure, you should have a
read-out of the plate in relative optical density (OD) values.
Use the readouts of the standard curve to fit a 4PL logistics
line, using one of the suggested computer programs (see
Subheading 2.7, item 10).
6. Average the triplicates and notice any aberrant results. Omit
outliers as necessary.

3.9 Imaging 1. Conduct the standard pre-processing steps of your analysis

Analysis Pointers scheme (i.e., motion correction, realignment, etc.).
2. Combine the localizer scans (step #6) from Day 1 and 2, and
conduct a standard linear contrast to generate functional
regions of interest (fROIs). For example, extract areas signifi-
cantly more or less active in the contrast of neutral versus
erotic.
3. Extract the time course within these fROIs from the later con-
ducted task (step #13).
4. Compute the difference in percent signal change across the
two conditions (e.g., neutral and erotic) within the fROI.
5. Repeat the above for the second day experiment.
6. Compare the change from baseline across the 2 days, chemo-
signal versus control.

4 Notes

1. The Ekman mood scale originally contains the following 16

variables: Amused; Content; Happy; Calm; Confident;
Interested; Angry; Anxious; Annoyed; Bored; Contemptuous;
Embarrassed; Stressed; Afraid; Disgusted and Sad. To these we
have added a 17th variable: Sexually aroused. To obtain mood
state, the 17 variables are displayed in succession with a visual-
analogue scale (VAS) ranging between very and not at all,
as in Fig. 1 [22, 64, 65].
2. If you intend to conduct a single study with human chemosig-
nals, then you can get passed without stainless steel-coated
rooms. In contrast, if you are building a lab that will conduct
many such studies over time (years), then experimental rooms
coated with nonporous material (e.g., Stainless steel, PTFE)
are important, as without them you will eventually have con-
tamination across studies.
3. Plastic bags used to store shirts should be odorless, non-
absorbing, non-diffusing and properly sealable. We have found
 An Assay for Human Chemosignals 389

that aluminum-lined anti-static plastic bags (such as the ones

used to store electronic components) work well for this pur-
pose. These are usually sealable using a cheap heat-sealer.
Alternatively, zip-lock style sealing is also optional, provided
that the seal is hermetic. Amber/Opaque glass jars are also a
viable option.
4. Methods for storing chemosignals are critical. In the case of
tears, we used only fresh non-refrigerated tears obtained within
3 h or less of their use. Moreover, in a small pilot study we found
that storing them for longer periods altered the effects. As to
sweat and body-odor, the highest recommended temperature
for keeping biological samples is generally 4 C (regular home
fridge temperature). This is a fairly bacteriostatic temperature
(i.e., prevents bacterial growth), it slows enzymatic reactions
and also reduces evaporation (and hence also loss of volatiles)
but does not allow long-term storage without degradation due
to oxidation or eventual bacterial/fungal spoilage. Freezing
samples at the usual 20 C (regular home freezer, but avoid
no-frost devices) or 80 C (laboratory deep freezer) would
result in better sample preservation, but care should be given to
the rate of freezing: Slow freezing, by simply putting the sam-
ples into the freezer chamber, may result in the formation of
large crystals of ice. The growing ice crystals might in turn dis-
rupt the membranes of cells found in the sample, resulting in
altered composition of the sample due to lost cellular compart-
mentalization and subsequent sample degradation. Therefore,
unless otherwise specified, it is advised to flash-freeze samples
prior to long term storage, by transferring the samples into cryo-
genic tubes and dropping them into liquid nitrogen, followed
by storage in freezers. Having said all this, be advised that some
plastic materials tend to adsorb lipid components in a way which
is aggravated by flash-freezing. Therefore the composition of
lipids in the sample might be altered.
5. We use medical grade sterile saline solution (0.9 % w/v Sodium
Chloride for IV injection). Alternatives can be Ringer solution
(different compositions are available) or a more specialized
artificial tears (see [66, 67]).
6. To obtain tear donors, we first post an add asking for individuals
who cry easily. When they call in, we ask them which film made/
makes them cry, and we obtain this film from the video library.
Our experience is that only about 10 % of those who think they
cry easily, in fact do so. Such donors, however, can return to lab
on a regular basis, and will typically generate 1 ml of tears in
response to the same film segment again and again.
7. Flexible thin endoscopes allow the most comfortable means of
eliciting reflexive tears. However, other more low-tech
alternatives, such as a thin long shaft medical transfer swab
(e.g., Copan Italia 160C) have also proven useful.
390 Idan Frumin and Noam Sobel

8. We have found that for obtaining good quality steroid chemo-

signal molecules, Steraloids (Newport, RI) is a good source.
9. Androstenone is a swine pheromone. Pheromones are presum-
ably species-specific. Nevertheless, androstenone has been
studied in various human chemosignaling contexts. Like with
all steroids, be sure to check purity.
10. While Propylene Glycol (PG; 1,2 propanediol) is a widely used
diluting solvent in the odor and fragrance world, Dipropylene
Glycol (DPG; 4-Oxa-2,6-heptandiol and 4-Oxa-1,7-
heptandiol mixture) is also suitable as an alternative. DPG is
very similar to PG, although it has lower freezing and higher
boiling points, and a lower vapor pressure. Both are almost
odorless, and effectively dissolve most moderate to highly
hydrophilic compounds and essential oils. Mineral oil can also
be considered, as the solubility of steroidal compounds may be
higher in hydrophobic solvents, but it is less commonly used.
11. McClintock and colleagues used clove oil to mask the odor of
androstadienone [61, 62]. Clove oil is comprised mainly of
eugenol, but with varying concentrations (6095 %, depend-
ing on the natural source). To ensure consistent outcome,
pure solid eugenol dissolved in the diluent is preferable. Use
1 % w/v eugenol dissolved in propylene glycol. The resulting
solution has a recognizable yellow tint, so be sure to use
opaque jars for delivery.
12. Dissolution of solid steroid chemosignal compounds might
prove hard at higher concentrations (i.e., stock solutions).
Possible aids are the use of a sonicator (either ultrasonic bath
or probe) and the application of gentle heat (~35 C) com-
bined with a lengthy stir or vortex.
13. PTFE beads (we usually use 23 mm in diameter) are a cheap,
inert and simple means of enlarging the effective surface area
of solutions. Wet a bed of beads inside a jar to avoid using the
liquid solution directly.
14. In different cultures, it may be worth validating these sources
in a separate short validation study. For example, American
subjects rated clips from Mr. Bean as funny, but Israeli sub-
jects did not. In both cultures, however, The Champ induced
sadness, and Nine and a half weeks induced sexual arousal in
both sexes. Note that more explicit film clips often induce dis-
gust rather than arousal.
15. The motion sensor is best fitted to the non-dominant hand
armrest. This way, the motion trace can later be used to iden-
tify motion-artifacts in the physiological data.
16. Finapres Ohmeda 2300 are the best continuous finger-cuff PB
monitors, yet they are no longer produced. Thus, each time one
shows up on eBay (typically from hospital surplus equipment
 An Assay for Human Chemosignals 391

dealers, as they were used in child anesthesiology), all the

psychophysiologists fight over it. Keep your eyes out for one.
17. We strongly recommend that when working with chemosig-
nals, one applies contamination prevention standards typical of
work with radioactive material. Under our working hypothesis
of extreme sensitivity to chemosignals, with functional thresh-
old far below conscious detection threshold, the possibility of
contaminating a study is ever present. With this notion in
mind, it is also advised to work with single-use tools whenever
possible (i.e., disposable plastic spatulas, plastic weighing
boats, etc.).
18. When collecting reflex tears, make sure to tilt the subjects
head in the correct direction, so that tears wont flow toward
the medial canthus and thus drain into the nasolacrimal duct.
19. Hold the glass capillary tube in parallel to the ground or in a
small tilt so that the fluid flows freely into the tube aided by
gravity and capillary force. Do not over-fill the capillary tube
and use the marking to assess the amount of fluid withdrawn.
20. Although many studies have used solid androstadienone crys-
tals as the stimulus, this results in a headspace concentration
that is most likely non-biological.
21. Various hormones peak at various times of day. Thus, if you
select particular hormones to follow, you should first investi-
gate this phenomenon for the selected hormone, as it may
imply particular times of day to avoid, e.g., early morning for
testosterone.
22. You should leave a glass of water at reach of the dominant
hand of the subject, and suggest that they take a sip after each
saliva donation (i.e., ~15 min before their next donation).
Otherwise, they might run out of saliva across the 2-h study.
23. A trick to know whether your SCR recording is working is to
abruptly bang (once) on the door of the subject room. This
should startle the subject, and you should see a clear SCR
deviation.
24. Saliva samples should be frozen at 20 C or below, and not
only for long term storage. The freezing itself precipitates pro-
teins (e.g., mucins) out of the saliva liquid, allowing for the
separation by centrifugation of the potentially interfering fac-
tors for the analysis. However, multiple freezethaw cycles
should be avoided [51, 52].
25. Note that this analysis scheme is susceptible to the risks associ-
ated with multiple comparisons. This is inherent to casting a
wide net. A solution is to first study a small group of subject in
a separate pilot study, in order to identify the impact of a given
chemosignal. For example, a chemosignal may have profound
392 Idan Frumin and Noam Sobel

effects on SCR but not on BP. You should then continue the
main study collecting only the measures of interest identified
in the pilot. This way you will avoid multiple comparisons.
A second alternative is to combine several of the measures into
composite measures, as in references [22, 64, 65, 68].
26. Steroid hormones secretion is episodic in nature. Thus, a mini-
mum of three samples per experiment is advised [51, 52]. We
found that taking five samples is better, dispersed throughout
the length of the experiment [22]. To achieve an average base-
line, equal volumes of samples can be physically pooled, but to
detect fluctuations stemming from experimental intervention
take individual samples across different time points.

Acknowledgment

This work was supported by the James S. McDonnell Foundation.

References

1. McClintock MK (1971) Menstrual synchorony 9. Ober C, Weitkamp LR, Cox N, Dytch H,

and suppression. Nature 229(5282):244245 Kostyu D, Elias S (1997) HLA and mate
2. Stern K, McClintock MK (1998) Regulation of choice in humans. Am J Hum Genet 61(3):
ovulation by human pheromones. Nature 497504, doi:S0002-9297(07)64312-2 [pii]
392(6672):177179. doi:10.1038/32408 10.1086/515511
3. Russell MJ, Switz GM, Thompson K (1980) 10. Eggert F, Muller-Ruchholtz W, Ferstl R (1998)
Olfactory influences on the human menstrual Olfactory cues associated with the major histo-
cycle. Pharmacol Biochem Behav 13(5): compatibility complex. Genetica 104(3):191197
737738 11. Jacob S, McClintock MK, Zelano B, Ober C
4. Jacob S, Spencer NA, Bullivant SB, Sellergren (2002) Paternally inherited HLA alleles are
SA, Mennella JA, McClintock MK (2004) associated with womens choice of male odor.
Effects of breastfeeding chemosignals on the Nat Genet 30(2):175179, doi:10.1038/ng830
human menstrual cycle. Hum Reprod 19(2): ng830 [pii]
422429 12. Roberts T, Roiser JP (2010) In the nose of the
5. Cutler WB, Preti G, Krieger A, Huggins GR, beholder: are olfactory influences on human mate
Garcia CR, Lawley HJ (1986) Human axillary choice driven by variation in immune system genes
secretions influence women's menstrual cycles: or sex hormone levels? Exp Biol Med (Maywood)
the role of donor extract from men. Horm 235(11):12771281, doi:ebm.2010.010112 [pii]
Behav 20(4):463473 10.1258/ebm.2010.010112
6. Wysocki CJ, Preti G (2004) Facts, fallacies, 13. Thornhill R, Gangestad SW, Miller R, Scheyd
fears, and frustrations with human phero- G, McCollough JK, Franklin M (2003) Major
mones. Anat Rec A Discov Mol Cell Evol Biol histocompatibility complex genes, symmetry,
281(1):12011211. doi:10.1002/ar.a.20125 and body scent attractiveness in men and
7. Wedekind C, Seebeck T, Bettens F, Paepke AJ women. Behav Ecol 14(5):668678
(1995) MHC-dependent mate preferences in 14. Roberts SC, Little AC (2008) Good genes,
humans. Proc Biol Sci 260(1359):245249. complementary genes and human mate prefer-
doi:10.1098/rspb.1995.0087 ences. Genetica 132(3):309321.
8. Wedekind C, Furi S (1997) Body odour prefer- doi:10.1007/s10709-007-9174-1
ences in men and women: do they aim for spe- 15. Gangestad SW, Cousins AJ (2001) Adaptive
cific MHC combinations or simply design, female mate preferences, and shifts
heterozygosity? Proc Biol Sci 264(1387):1471 across the menstrual cycle. Annu Rev Sex Res
1479. doi:10.1098/rspb.1997.0204 12:145185
 An Assay for Human Chemosignals 393

16. Havlicek J, Roberts SC, Flegr J (2005) 28. Ackerl K, Atzmueller M, Grammer K (2002)
Womens preference for dominant male The scent of fear. Neuroendocrinol Lett 23(2):
odour: effects of menstrual cycle and rela- 7984
tionship status. Biol Lett 1(3):256259, 29. Chen D, Katdare A, Lucas N (2006)
doi:A79R9E6B7CCHUEN7 [pii]10.1098/ Chemosignals of fear enhance cognitive perfor-
rsbl.2005.0332 mance in humans. Chem Senses 31(5):415
17. Little AC, Jones BC, Burriss RP (2007) 30. Zhou W, Chen D (2009) Fear-related chemo-
Preferences for masculinity in male bodies signals modulate recognition of fear in
change across the menstrual cycle. Horm ambiguous facial expressions. Psychol Sci
Behav 51(5):633639 20(2):177
18. Singh D, Bronstad PM (2001) Female body 31. Prehn A, Ohrt A, Sojka B, Ferstl R, Pause BM
odour is a potential cue to ovulation. Proc Biol (2006) Chemosensory anxiety signals augment
Sci 268(1469):797801. doi:10.1098/rspb. the startle reflex in humans. Neurosci Lett
2001.1589 394(2):127130
19. Roberts SC, Gosling LM, Carter V, Petrie M 32. Prehn-Kristensen A, Wiesner C, Bergmann
(2008) MHC-correlated odour preferences in TO, Wolff S, Jansen O, Mehdorn HM, Ferstl
humans and the use of oral contraceptives. R, Pause BM (2009) Induction of empathy by
Proc Biol Sci 275(1652):27152722, the smell of anxiety. PLoS One 4(6):e5987
doi:F542428772R96X64 [pii]10.1098/rspb.
2008.0825 33. Mujica-Parodi LR, Strey HH, Frederick B,
Savoy R, Cox D, Botanov Y, Tolkunov D,
20. Kuukasjarvi S, Eriksson CJP, Koskela E, Rubin D, Weber J (2009) Chemosensory cues
Mappes T, Nissinen K, Rantala MJ (2004) to conspecific emotional stress activate amyg-
Attractiveness of womens body odors over the dala in humans. PLoS One 4(7):113123
menstrual cycle: the role of oral contraceptives
and receiver sex. Behav Ecol 15(4):579584 34. Magnetic Resonance Neuroimaging: Methods
and Protocols (2011). Magn Reson
21. Miller SL, Maner JK (2010) Scent of a woman:
Neuroimaging Methods Protoc 711:1598.
mens testosterone responses to olfactory ovu-
doi:10.1007/978-1-61737-992-5
lation cues. Psychol Sci 21(2):276283, doi:09
56797609357733[pii]10.1177/0956797609 35. Analgesia: Methods and Protocols (2010).
357733 Analgesia Methods Protoc 617:1560.
doi:10.1007/978-1-60327-323-7
22. Gelstein S, Yeshurun Y, Rozenkrantz L,
Shushan S, Frumin I, Roth Y, Sobel N (2011) 36. fMRI Techniques and Protocols (2009). Fmri
Human tears contain a chemosignal. Science Tech Protoc 41:1843. doi:10.1007/978-1-
331(6014):226230, doi:science.1198331[pii] 60327-919-2
10.1126/science.1198331 37. Boulton Aa, Baker Gb, Hiscock M (1991).
23. Oh TJ, Kim MY, Park KS, Cho YM (2012) Neuromethods, vol 17 Neuropsychology.
Effects of chemosignals from sad tears and Perceptual and motor skills 72(2):703
postprandial plasma on appetite and food 38. Johnson BN, Sobel N (2007) Methods for
intake in humans. PLoS One 7(8):e42352, building an olfactometer with known concen-
doi:10.1371/journal.pone.0042352PONE- tration outcomes. J Neurosci Methods
D-11-08927 [pii] 160(2):231245, doi:S0165-0270(06)00461-4
24. Meredith M (2001) Human vomeronasal [pii]10.1016/j.jneumeth.2006.09.008
organ function: a critical review of best and 39. Sobel N, Prabhakaran V, Desmond JE, Glover
worst cases. Chem Senses 26(4):433 GH, Sullivan EV, Gabrieli JD (1997) A method
25. Hauser R, Marczak M, Karaszewski B, for functional magnetic resonance imaging of
Wiergowski M, Kaliszan M, Penkowski M, olfaction. J Neurosci Methods 78(12):
Kernbach-Wighton G, Jankowski Z, Namiesnik 115123
J (2008) A preliminary study for identifying 40. Lundstrom JN, Gordon AR, Alden EC,
olfactory markers of fear in the rat. Lab Anim Boesveldt S, Albrecht J (2010) Methods for
(N Y) 37(2):7680, doi:laban0208-76 building an inexpensive computer-controlled
[pii]10.1038/laban0208-76 olfactometer for temporally-precise experi-
26. Pageat P, Gaultier E (2003) Current research ments. Int J Psychophysiol 78(2):179189,
in canine and feline pheromones. Vet Clin doi:S0167-8760(10)00666-5[pii]10.1016/j.
North Am Small Anim Pract 33(2):187211 ijpsycho.2010.07.007
27. Chen D, Haviland-Jones J (2000) Human 41. Yantis S, Johnson DN (1990) Mechanisms of
olfactory communication of emotion. Percept attentional priority. J Exp Psychol Hum Percept
Mot Skills 91(3 Pt 1):771 Perform 16(4):812825
394 Idan Frumin and Noam Sobel

42. Levenson RW, Ekman P, Friesen WV (1990) ganglion cells in the spotted hyena (Crocuta
Voluntary facial action generates emotion- crocuta). Brain Behav Evol 62(4):182192
specific autonomic nervous system activity. 58. Drea CM, Vignieri SN, Kim HS, Weldele ML,
Psychophysiology 27(4):363384 Glickman SE (2002) Responses to olfactory
43. Baylis GC, Driver J (1993) Visual attention stimuli in spotted hyenas (Crocuta crocuta): II.
and objects: evidence for hierarchical coding of Discrimination of conspecific scent. J Comp
location. J Exp Psychol Hum Percept Perform Psychol 116(4):342349
19(3):451470 59. Jones DT, Monroy D, Pflugfelder SC (1997) A
44. Willis CM, Church SM, Guest CM, Cook WA, novel method of tear collection: comparison of
McCarthy N, Bransbury AJ, Church MRT, glass capillary micropipettes with porous poly-
Church JCT (2004) Olfactory detection of ester rods. Cornea 16(4):450458
human bladder cancer by dogs: proof of prin- 60. Stuchell RN, Feldman JJ, Farris RL, Mandel
ciple study. Br Med J 329(7468):712 ID (1984) The effect of collection technique
45. Mathwroks Matlab. http://www.mathworks. on tear composition. Invest Ophthalmol Vis
com/products/matlab/. Accessed 16 Sep 2012 Sci 25(3):374377
46. National Instruments LabVIEW. http://www. 61. Hummer TA, McClintock MK (2009) Putative
ni.com/labview/. Accessed 16 Sep 2012 human pheromone androstadienone attunes
47. AD Instruments. http://www.adinstruments. the mind specifically to emotional information.
com/. Accessed 16 Sep 2012 Horm Behav 55(4):548559, doi:S0018-
48. EntranMeasurment specialists. http://www. 506X(09)00005-1[pii] 10.1016/j.yhbeh.
meas-spec.com/entran.aspx. Accessed 16 Sep 2009.01.002
2012 62. Jacob S, McClintock MK (2000) Psychological
49. Hauser MD, Kralik J (1997) Life beyond the state and mood effects of steroidal chemosignals
mirror: a reply to Anderson & Gallup. Anim in women and men. Horm Behav 37(1):5778,
Behav 54(6):15681571 doi:10.1006/hbeh.1999.1559S0018-
50. Swartz KB (1997) What is mirror self- 506X(99)91559-3 [pii]
recognition in nonhuman primates, and what is 63. Lundstrom JN, Hummel T, Olsson MJ (2003)
it not? Ann N Y Acad Sci 818:6471 Individual differences in sensitivity to the odor
51. Salimetrics (2011) High sensitivity salivary cor- of 4,16-androstadien-3-one. Chem Senses
tisol enzyme immunassay kit protocol. http:// 28(7):643650
www.salimetrics.com/documents/Cortisol_ 64. Bensafi M, Brown WM, Khan R, Levenson B,
Kit_Insert.pdf. Accessed 16 Sep 2012 Sobel N (2004) Sniffing human sex-steroid
52. Salimetrics (2011) Salivary testosterone derived compounds modulates mood, memory
enzyme immunassay kit protocol. http:// and autonomic nervous system function in spe-
www.salimetrics.com/documents/Testo_Kit_ cific behavioral contexts. Behav Brain Res
Insert.pdf. Accessed 16 Sep 2012 152(1):1122, doi:10.1016/j.bbr.2003.09.
53. Anderson JR, Gallup GG (1997) Self- 009 S0166432803003309[pii]
recognition in Saguinus? A critical essay. Anim 65. Bensafi M, Tsutsui T, Khan R, Levenson RW,
Behav 54:15631567 Sobel N (2004) Sniffing a human sex-steroid
54. Delfour F, Marten K (2001) Mirror image pro- derived compound affects mood and auto-
cessing in three marine mammal species: killer nomic arousal in a dose-dependent manner.
whales (Orcinus orca), false killer whales Psychoneuroendocrinology 29(10):1290
(Pseudorca crassidens) and California sea lions 1299, doi:10.1016/j.psyneuen.2004.03.007
(Zalophus californianus). Behav Processes S0306453004000460[pii]
53(3):181190 66. Frey WH II, DeSota-Johnson D, Hoffman C,
55. Drea CM, Vignieri SN, Cunningham SB, McCall JT (1981) Effect of stimulus on the
Glickman SE (2002) Responses to olfactory chemical composition of human tears. Am J
stimuli in spotted hyenas (Crocuta crocuta): I. Ophthalmol 92(4):559567
Investigation of environmental odors and the 67. Van Haeringen NJ (1981) Clinical biochemis-
function of rolling. J Comp Psychol try of tears. Surv Ophthalmol 26(2):8496
116(4):331341 68. Wyart C, Webster WW, Chen JH, Wilson SR,
56. Bekoff M, Sherman PW (2004) Reflections on McClary A, Khan RM, Sobel N (2007)
animal selves. Trends Ecol Evol 19(4):176180 Smelling a single component of male sweat
57. Calderone JB, Reese BE, Jacobs GH (2003) alters levels of cortisol in women. J Neurosci
Topography of photoreceptors and retinal 27(6):12611265
 INDEX

A C
Accessory olfactory bulb (AOB) .............................. 237249, Caenorhabditis elegans .................... 7192, 273282, 285292
251252, 254, 342 Calcium dye indicators ............................................. 123, 191
Acetonitrile................................ 49, 50, 58, 60, 78, 81, 8385 Calcium imaging ..................................... 122124, 126127,
Adaptation.........................................222, 228, 231, 234, 235 179187, 189199, 201, 211218, 222, 223
Adenylyl cyclase ....................................................... 108, 332 Calreticulin ............................................................... 122, 123
Absorption spectra ..................................................... 52, 191 Cannula ..................... 238240, 243244, 377, 379, 383, 384
Agar solution ............................................ 224226, 234, 235 Capillary columns ............................................5, 6, 35, 36, 38
Aggregation ..............................................................294296 Capillary tube ............................................... 24, 27, 382, 391
Aggression .........................121, 307310, 312316, 332334 Carp...................................................56, 62, 65, 68, 294304
Aggressive behavior ............................33, 307310, 312315, Castrated ..................... 32, 313, 319, 321, 323, 324, 326328
334, 344 Central nervous system (CNS) ................................. 247, 248
Amniotic fluid .......................................................... 353, 354 Centrifuge ..................................................49, 76, 77, 80, 81,
Amphibian ................................................................. 96, 103 83, 85, 88, 113, 114, 124126, 193, 379, 387
Amygdala ................................................................. 247, 375 c-Fos .................................. 48, 50, 51, 53, 248, 249, 254, 256
Androgen.................................................................. 309, 312 Channel .......................87, 108, 115, 116, 137, 144, 145, 152,
Androstenone ........................................................... 378, 390 183, 190, 203, 222, 262, 268, 269, 326, 333, 379
Anion-exchange ...........................................................4850 Channelrhodopsin ............................................................262
Antennae ...............................................46, 8, 157175, 267 Chaperone ................................................................ 122, 123
Antennal lobe ................................................... 107, 179187 Chasing ............................................................ 296, 304, 312
Antibody.............................................57, 58, 6365, 68, 249, Chemical communication....................4, 2931, 33, 179, 373
252254, 256, 262, 265, 266 Chemical signals ......................................33, 55, 71, 157, 307
Anxiety ............................................................. 313, 358, 375 Chemoreception ....................................... 350352, 355, 368
AOB. See Accessory olfactory bulb (AOB) Chemosensory ..........................................211, 222, 235, 310,
Ascarosides ...........................................7192, 274, 275, 286, 331334, 350, 358, 368
288, 290, 291 Chemosensory receptor .............................. 95104, 134, 232
Attractants .................................................... 4, 157, 286, 291 Chemosignals ...................................................... 30, 34, 121,
Attraction .................................................. 55, 285292, 294, 222, 307310, 312, 315, 373392
296299, 301, 303, 305, 361, 364367 Chemotaxis.......................................................................291
Autosampler ......................................... 26, 36, 38, 58, 61, 67, Chiral center ...................................................................5, 11
78, 81, 8385 Chiral columns .................................................................5, 7
Avoidance .......................... 285, 288291, 309, 311, 331, 361 Chromatography ....................................................... 5, 8, 11,
Avoidance index ....................................................... 289, 291 1518, 2021, 2943, 48, 56, 57, 60, 71, 7476,
81, 84, 85, 8991, 159, 165, 353
B Chromosome engineering ....................................... 133, 134,
Bacterial artificial chromosomes (BACs) ......................... 134, 137138, 147150, 152
135, 138, 140, 151 Cis-vaccenyl acetatte (cVA) ...................15, 17, 180, 182185
Bedding ............................................248, 316, 321, 324, 328, Collagenase .............................................. 110, 111, 114, 117
331, 334337, 339, 341, 344, 360, 367 Complementary RNA (cRNA) ........ 108111, 113116, 118
BLAST......................................................... 97, 98, 101103 Confocal microscope ........................................ 129, 265, 266
Bleeding ................................................................... 322, 339 Congenic ..........................................................................335
Blood Pressure (BP) ................................. 377, 379, 383, 392 Conspecific .......................................179, 294, 296, 319, 320,
Brain imaging ........................................................... 375, 380 323, 331, 333, 334, 350, 352, 353, 356, 358, 365

Kazushige Touhara (ed.), Pheromone Signaling: Methods and Protocols, Methods in Molecular Biology, vol. 1068,
DOI 10.1007/978-1-62703-619-1, Springer Science+Business Media, LLC 2013

395
 PHEROMONE SIGNALING: METHODS AND PROTOCOLS
396 Index

Correlation spectroscopy (COSY ) ......................... 75, 8588 Estrus ..........................................29, 311, 316, 320322, 325
COSY. See Correlation spectroscopy (COSY ) Evaporator ...................................................17, 18, 25, 58, 77
Courtship ............................ 16, 261263, 266, 295, 314, 320 Exocrine gland-secreting peptide 1 (ESP1)....................... 48,
Cre-loxP ............................................133, 147, 148, 150, 153 5053, 128, 248, 250, 254, 256
cRNA. See Complementary RNA (cRNA) Exocrine gland-secreting peptide (ESP) family................129
Cuticle ..............................................1517, 25, 27, 169, 173, Expression vector...............................109, 116, 124, 137, 148
175, 183, 184, 186, 267 Extracellular recording .....................................................238
cVA. See Cis-vaccenyl acetatte (cVA) Extra-orbital lacrimal gland (ELG).............................. 4852

D F
Dauer.................................................................. 71, 273282 Feeding activity ................................................................304
DEAE column ....................................................... 49, 50, 53 Field potential ..........................................................221235
Deorphanization............................................... 122, 123, 189 Fish pheromone.............................................. 5557, 61, 294
Detabases.................................................... 21, 101, 102, 134 Flame ionization detector (FID) ...................................... 5, 6
Diimide reduction ..............................................................10 Fluo-4, 123125, 127, 128, 130
Dimethyl disulfide (DMDS) .......................7, 9, 10, 353, 354 Fluorescence .................................................... 123, 127, 128,
Discrimination................... 316, 319328, 358, 360, 366, 367 130, 185, 190, 191, 196, 197, 202, 206, 209, 217,
Dissection ...........................................48, 140, 181, 183184, 230, 262, 264, 267, 269
186, 192194, 197, 202204, 208, 215, 216, 225, Fluorescence resonance energy
230233, 237, 238, 240245, 249, 265 transfer (FRET).............................................. 262
DMDS. See Dimethyl disulfide (DMDS) Fluorescent microscopy ............................................ 125, 264
DMSO ...................................... 118, 124, 137, 193, 202, 206 Fly ............................................................... 1527, 165, 170,
Drosophila melanogaster...................................... 15, 16, 22, 25, 179, 180, 183, 185, 186, 261271
170, 179, 186, 261 Formyl peptide receptors (FPRs).............................. 121, 211
Fragment ions ............................................... 810, 12, 13, 67
E Frontal cortex ................................................................... 242
EAG. See Electroantennography (EAG) Fume hood ................................................. 20, 116, 278, 280
Ejaculation................................................ 307, 310, 314, 327 Fura-2, 191193, 195, 198
Ekman mood scale ...........................................................388
G
Electrical recording................................... 201, 230, 237246
Electroantennography (EAG) ............. 5, 8, 18, 107, 157175 G418, 135, 136, 140, 143153
Electrodes ...............................................6, 24, 116, 158160, GAL4-UAS ............................................................. 185, 261
162165, 167, 169175, 225227, 230, 231, Gas chromatography (GC)........................................ 57, 11,
233235, 237239, 244246 16, 18, 2943, 56, 159, 165, 353
Electron ionization (EI) ......................................... 22, 41, 42 Gas chromatography mass
Electron ionization-mass spectrometry (EI-MS) .................8 spectrometry (GC-MS) ....................... 5, 6, 811,
Electro-olfactogram (EOG) ................ 62, 221, 222, 300, 301 16, 18, 20, 2223, 25, 26, 2943, 353, 354
Electrophysiology ..................................... 157160, 169173 GC. See Gas chromatography (GC)
Electroporation..................................133, 137, 143144, 148 GCaMP ............................................179, 185, 186, 212, 217
Electrovomeronasogram (EVG) ............................... 221235 GC-MS. See Gas chromatography mass spectrometry
ELG. See Extra-orbital lacrimal gland (ELG) (GC-MS)
ELISA. See Enzyme-linked immunosorbent assay (ELISA) Gene targeting .................................. 133141, 143148, 152
Embryonic fibroblasts (EFs)............................. 133, 140141 GFP....................134, 139, 140, 191, 213, 262, 263, 265, 266
Embryonic stem (ES) cells ................133, 134, 136153, 335 Glass bottomed dish ................................................. 126, 129
Emotion .......................................................... 374, 375, 377, Glass capillaries ..........................................11, 113, 160, 162,
378, 380, 381, 386, 387 225228, 230, 234, 235, 264, 267, 287, 288, 290,
Enzyme-linked immunosorbent 378, 382, 391
assay (ELISA) ................................. 5659, 6265 Glass pipette ...................... 180, 182186, 204, 226, 228230
EOG. See Electro-olfactogram (EOG) Goldfish........................... 56, 62, 65, 294, 295, 299, 300, 304
ES cells. See Embryonic stem (ES) cells Gonadal hormone..................................................... 310, 319
ESP1. See Exocrine gland-secreting peptide 1 (ESP1) G protein .................................................................. 122, 222
Estradiol ............................ 311, 320322, 324, 325, 327, 328 G protein-coupled receptor (GPCR) .................. 95, 99, 103,
Estrogen ........................................................... 309311, 314 108, 121, 122, 211
 PHEROMONE SIGNALING: METHODS AND PROTOCOLS
Index
397

H Liquid nitrogen ..........................................35, 36, 4042, 49,

50, 135, 142, 143, 389
Habituation .............................................. 323, 324, 326, 367 Locomotor activity ................................................... 304, 313
Headspace ............................................3335, 38, 39, 42, 43, Lordosis .....................................314, 320, 321, 324325, 327
159, 168, 171, 185, 391 Lyophilizer ....................................................... 77, 78, 85, 89
HEK293T ........................................................ 122, 123, 129
Hermaphrodites ................................278, 281, 286, 288291 M
Heterologous expression ................................... 108, 121130
Major histocompatiibility
Hexane ....................................................611, 16, 20, 22, 25
complex (MHC) ....................... 48, 231, 373, 374
High-performance liquid
Major urinary proteins (MUPs) ................................... 31, 47
chromatography (HPLC) .................. 5, 7, 11, 34,
Male aggression .........................308, 309, 312313, 315, 334
35, 37, 4853, 57, 66, 67, 78, 214, 216
Mammals........................................................29, 57, 96, 103,
HLA. See Human Leukocyte Antigen (HLA)
331, 349352, 372
Homologous recombination .................................... 133, 134,
Mammary pheromone ......................................................354
138, 139, 146, 147, 152
MARCM. See Mosaic Analysis with a Repressible Cell
HPLC. See High-performance liquid chromatography
Marker (MARCM)
(HPLC)
Mass-to-charge (m/z)................................... 8, 13, 16, 60, 61,
Human Leukocyte Antigen (HLA) ......................... 373, 374
67, 74, 75, 84
Hydrocarbons ........................ 3, 912, 1517, 22, 25, 26, 270
Mate recognition ..............................................................121
Hypothalamus ..................................................................247
Maternal behavior .................................... 331340, 342344
I Mating behavior ....................................... 7, 8, 311, 319328
Menstrual synchrony ........................................................373
Immediate early genes ..............................................247257 Methanol .......................................... 7, 1621, 25, 34, 35, 37,
Immunofluorescence ........................................................190 3942, 50, 58, 59, 77, 78, 81, 8385,
Immunohistochemistry ............................................ 265, 266 87, 88, 90, 91, 202, 206
Immunostaining ............................................... 248256, 265 4-Methyl-1,2,4-triazoline-3,
Infanticide ........................................................ 307, 331344 5-dione (MTAD).......................................... 7, 10
Insect ........................................ 3, 4, 6, 12, 15, 16, 1819, 23, MHC. See Major histocompatiibility
24, 29, 107118, 157175, 179187, 224, 225, complex (MHC)
231, 232, 261270 Microelectrodes ................................................ 167, 238, 244
In situ hybridizations ........................................................190 Microinjection ..........................................108, 111, 114115,
Intact genes ........................................................................97 117, 118, 159
Internal ribosome entry site (IRES) ......... 134, 139, 140, 213 Micromanipulator ........................................... 111, 115, 117,
Interspecies ............................................................... 189, 248 158, 160, 161, 164, 172, 214, 229,
Intraspecies .......................................................................189 233, 264, 265, 268
Intromission .............................. 307, 310, 314, 319, 324, 327 Micropipettte................................................... 111, 113, 160,
Intruders ............................................309, 312314, 316, 333 162164, 167, 175, 227, 228, 267, 378
IRES. See Internal ribosome entry site (IRES) Microscope ...................................................76, 79, 125, 126,
Isoproterenol ............................................................129130 129, 142, 144, 145, 158, 160164, 167, 169, 173,
184, 193, 195, 196, 203, 204, 208, 213, 214, 226,
L
229231, 233, 239, 244, 250, 253, 256, 265267,
Lacrimal glands ............................................................30, 48 274, 279
Lactating ..................................................339, 351, 354, 355, Microvilli .......................................................... 201, 211, 233
358, 359, 363, 373 Mosaic Analysis with a Repressible Cell
Larval stages ............................................. 273, 277, 287, 290 Marker (MARCM) ........................ 261, 263, 266
LC-ESI-MS.......................................................................74 Moth .......................................................313, 157, 165, 172
LC-MS/MS. See Liquid chromatography-mass Mounting .................................................160, 161, 168, 173,
spectrometry (LC-MS/MS) 201, 203, 204, 207, 224225,
Lipid pheromones ..............................................................15 231, 233, 250, 253, 256, 265, 267, 268,
Lipocalin ............................................................................30 307, 310, 320, 321, 325, 327, 328
Liquid chromatography .................................... 30, 55, 57, 74 Multi-electrode array recording ........................................238
Liquid chromatography-mass Multigene family ................................................................95
spectrometry (LC-MS/MS) ...................... 30, 74, MUPs. See Major urinary ptoteins (MUPs)
75, 7778, 8185, 92 m/z. See Mass-to-charge (m/z)
 PHEROMONE SIGNALING: METHODS AND PROTOCOLS
398 Index

N Predator ...............................................55, 157, 248, 295, 333

Preference .................................................244, 319328, 351,
Needle ................................................25, 111, 113115, 117, 360, 364, 366, 367, 374
118, 202, 214, 215, 224229, 233, Preference score ................................................................324
248250, 263, 265, 267 Pregnancy block................................................................121
Nematode ................................................................. 206, 275 Preputial gland.......................................30, 33, 3637, 39, 40
Nematode growth media (NGM) ............................... 76, 79, Priming pheromones ........................................ 293, 295, 300
275, 277, 286288, 290 Proestrus ........................................................... 311, 320, 321
Neomycin .........................................................................149 Progesterone .....................................311, 314, 320322, 324,
Neonate ....................................................312, 343, 349351, 325, 327, 328
353, 355, 356, 358, 360, 361 Prostaglandin ............................... 5557, 62, 65, 68, 295, 340
Neural circuits ..................................................................179 Pseudo-molecular ions..................................................61, 67
Neuroepthelium........................................ 201, 211, 216, 218 Pseudogenes ..........................................97, 98, 100102, 374
Newborn....................................338, 349358, 361, 363, 364 Puberty .............................................................................335
NGM. See Nematode growth media (NGM) Puller ......................... 111, 113115, 160, 163, 225228, 244
Nipple .............................................................. 183, 333, 339, Pup ................................................... 331344, 353, 355367
342, 353, 354, 358, 359, 361364 Pup-directed behaviors .............................................331335
NMR. See Nuclear magnetic resonance (NMR) Puromycin ......................... 123126, 129, 137, 147150, 153
Nuclear magnetic resonance (NMR) ................ 15, 31, 7191
R
O
Rabbit ................................................249, 265, 266, 342, 354
Odor cues ................................................. 352, 354356, 361 Recording electrode ......................................... 158, 159, 164,
Odor delivery............................................................180183 165, 167170, 172, 174, 175, 225230, 233235
Olfaction ........................................................... 30, 122, 332, Releasing ................................. 43, 63, 68, 294, 295, 350, 353
342, 350, 355, 360, 361, 375 Reproduction .............................................. 79, 294, 373, 374
Olfactometer .....................................325, 326, 365, 376, 385 Repulsion..................................................................285292
Olfactory ..........................................29, 30, 56, 95, 107118, Residents ...........................................308, 309, 312314, 316
134, 158, 159, 212, 217, 222, 230, 237247, 250, Retention times ........................... 6, 22, 38, 57, 61, 67, 74, 85
251, 307, 309, 319, 321, 323, 326, 332, 342, 350, Retrieval assay ..................................................................338
351, 355, 358, 360, 373, 374 Reverse-phase ...............................................................4851
Olfactory bulbectomy .......................................................342 Ringer's solution ............................................... 337, 381, 389
Olfactory bulbs ............................50, 237246, 251, 309, 342 Rodent ................................... 2931, 332, 351, 355, 364, 368
Olfactory epithelium ......................................30, 95, 96, 221,
222, 319, 323, 332, 342, 355 S
Olfactory receptors ............................................................. 95
Saline solution ...................................................... 6, 163, 389
Olfactory sensory neurons ........................................ 158, 212
Saliva .............................................. 30, 33, 34, 37, 3941, 43,
Orco ..................................................108, 112, 115, 118, 134
288, 353, 359, 379, 383, 385, 387, 391
Ovariectomized ................. 311, 314, 320322, 325, 327, 328
SCR. See Skin Conductance Response (SCR)
Ovulation ................................................. 295, 296, 342, 373
Sensory epithelium ............ 201, 221, 222, 226229, 232235
P Serum ...................................................33, 34, 37, 3941, 64,
123, 124, 135, 192, 249, 265, 266
Parturition ......................... 332, 333, 336, 337, 339, 340, 359 Seven TM. See Seven transmembrane (TM)
Patch-clamping ........................................................ 223, 238 Seven transmembrane (TM) .................................. 22, 95, 98
Paternal behavior ..............................................................331 Sex pheromone ...........................................3, 4, 7, 11, 12, 15,
Peristaltic pump ........................................112, 113, 117, 129, 17, 48, 62, 294296, 299, 300
242, 286, 297, 298, 300, 302 Sexual attraction ............................................... 294, 296297
Pheromone gland ....................................................... 5, 7, 12 Sexual behavior ..................................307316, 327, 336, 386
Pheromone receptors .................107, 108, 121130, 222, 248 Silica gel ........................................................... 15, 17, 21, 22
Photobleaching................................................. 191, 208, 209 SIM analyses ................................................................62, 67
Phychophysiology..................................... 375, 379, 382387 SIM windows ......................................................... 60, 61, 67
Pipettes ...............................................68, 111, 161164, 166, Single sensillum recordings .............................. 107, 157175
173, 225229, 235, 245, 287, 379 Site-derected mutagenesis ................................................108
Pipette tip .....................................................59, 68, 145, 149, Skin Conductance Response (SCR) ........................ 377, 379,
161, 162, 173, 174, 195, 224, 225, 227, 228, 230 383, 391, 393
 PHEROMONE SIGNALING: METHODS AND PROTOCOLS
Index
399
Slice preparation ............................................... 190, 211219 Trypsin .............................................123, 124, 126, 127, 129,
Sniffing .............................. 307, 314, 324, 338, 361, 367, 374 135, 136, 138, 140145, 147, 149, 151
Social behavior .................................................................295 Two-electrode voltage-clamp
Social odor ................................................................349368 recording (TEVC) .......................... 108, 115, 117
Solid phase extraction (SPE) ........................................ 5658 Two-photon microscopy ...................................................202
Solid phase micro-extraction (SPME) ............6, 7, 16, 31, 33
Solvent extraction ...........................................................7, 31 U
Sorptive stir bar extraction (SBSE) ............ 31, 33, 36, 3941 Ultrasonic vocalization .............. 307, 308, 315, 356, 364, 365
Southern blotting ..............................138, 139, 146148, 152 Untranslated region (UTR) ...................................... 134, 140
SPE. See Solid phase extraction (SPE) Urinary .................................................31, 32, 3436, 38, 47,
Species recognition ...........................................................294 198, 199, 295, 307, 319328
Speedvac .....................................................25, 49, 51, 53, 77, Urine ..............................................3036, 3840, 47, 57, 65,
78, 81, 83, 85, 88, 89 198, 199, 213, 218, 231, 295, 296, 313, 315, 316,
SPME. See Solid phase micro-extraction (SPME) 321, 323, 326, 327, 354356, 360, 368
Steroid hormones ..................................................... 273, 392 UTR. See Untranslated region (UTR)
Submaxillary glands ............................................................30 UV light ......................................................... 18, 21, 26, 199
Suckling............................. 339, 342, 348, 354, 357, 362, 363
Sulfated steroids ...............................................................207 V
Supernatant ..........................................49, 7981, 83, 85, 88,
Vertebrate ........................................................... 95104, 121
113, 114, 125, 141, 144, 146, 148, 151, 195, 277,
Vibratome................................................................. 191, 213
278, 387
VNO. See Vomeronasal organ (VNO)
Sweat .................................................373, 375, 377, 380, 389
Vomeronasal ...............................................30, 47, 48, 51, 95,
SynaptopHluorin ..............................................................180
96, 121123, 189199, 201208, 211218,
Syringe .................................................16, 1820, 22, 25, 26,
221235, 237, 238, 240244, 247257, 319, 333,
58, 66, 113, 115, 173, 203, 214, 215, 218,
342, 355
224229, 239, 244, 248250, 263, 267
Vomeronasal epithelium ...................................................201
Vomeronasal organ (VNO) ......................................... 30, 47,
T
96, 121, 189, 192, 194, 197, 201209, 211218,
TBLASTN searches ........................................... 97, 100, 101 221235, 237249, 251254, 319, 320, 323,
Tears ...................................................48, 110, 114, 183, 186, 332334, 342, 355
194, 232, 334, 374, 375, 377, 378, Vomeronasal receptors ................................................ 95, 121
381382, 389, 391 Vomeronasal sensory neurons ...................................... 47, 48,
Tenax .......................................................6, 33, 35, 39, 41, 42 51, 122, 123, 190, 195
Territorial marking ...........................................................307
Testosterone .............................. 312, 325, 328, 374, 379, 391 W
TEVC. See Two-electrode Wind tunnel .....................................................................6, 8
voltage-clamp recording (TEVC)
Thin layer chromatography (TLC) ........................ 1526, 60 X
TLC plate............................................................... 16, 2025
Xenopus laevis oocytes .....................................................108
Transgenic mice .........................211219, 308, 310, 312, 315
X-ray crystallography..........................................................31
Trigeminal .........................................355, 360, 378, 381382
Triple quadrupole mass Y
spectrometer ........................61, 65, 74, 75, 78, 92
Truncated gene ...................................................................97 Y-Maze..................................................... 325, 326, 366, 367