Biofouling

ISSN: 0892-7014 (Print) 1029-2454 (Online) Journal homepage: www.tandfonline.com/journals/gbif20

The Biofilm Matrix

David G Allison

To cite this article: David G Allison (2003) The Biofilm Matrix, Biofouling, 19:2, 139-150, DOI:
10.1080/0892701031000072190
To link to this article: https://doi.org/10.1080/0892701031000072190

Published online: 09 Sep 2010.

Submit your article to this journal

Article views: 1349

View related articles

Citing articles: 29 View citing articles

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=gbif20
Biofouling, 2003 Vol 19 (2), pp 139–150

The Biofilm Matrix

DAVID G ALLISON*

School of Pharmacy and Pharmaceutical Sciences, University of Manchester, Manchester M13 9PL, UK

(Received 21 May 2002; in final form 1 October 2002)

The extracellular matrix is a complex and extremely arrangement of different species within a biofilm
important component of all biofilms, providing architec- (Costerton et al., 1994). Such interactions give the
tural structure and mechanical stability to the attached biofilm community metabolic and physiological
population. The matrix is composed of cells, water and
secreted/released extracellular macromolecules. In capabilities which are not possible for the individual,
addition, a range of enzymic and regulatory activities can unattached cells (Gilbert et al., 1997). Biofilms are
be found within the matrix. Together, these different generally considered to be problematic and can have
components and activities are likely to interact and in so consequences that directly affect society, the scale of
doing create a series of local environments within the which is often overlooked. Indeed, in many instances
matrix which co-exist as a functional consortium. The
matrix architecture is also subject to a number of extrinsic the persistent and problematic nature of biofilms is
factors, including fluctuations in nutrient and gaseous attributed to the surrounding matrix (Allison et al.,
levels and fluid shear. Together, these intrinsic and 2000). It is surprising, therefore, that for so
extrinsic factors combine to produce a dynamic, hetero- ubiquitous and important an occurrence, under-
geneous microenvironment for the attached and enveloped standing about the properties, formation and
cells.
structure of bacterial biofilms and their components
is far from complete. More so, biofilms have been
Keywords: biofilm; matrix; exopolysaccharide
recognized for over 60 y (Henrici, 1932; Zobell &
Allen, 1935).
At this juncture, it is worth clarifying matrix
terminology. A significant feature of the matrix is the
INTRODUCTION presence of biosynthetic microbial polymers lying
outside the integral cell surface components of the
Characteristic to many biofilms is the production of resident bacteria. These extracellular biopolymers
an extracellular matrix that envelops the attached include exopolysaccharides, nucleic acids, proteins,
cells. This is generally composed of water and glycoproteins and phospholipids. Since polysaccha-
microbial macromolecules and provides a complex rides were identified as being a common component,
array of microenvironments surrounding the the term ‘glycocalyx’ was introduced to describe
attached cells. Moreover, since the matrix structure the gelatinous mass surrounding attached cells
and integrity is heavily influenced by changes in the (Costerton et al., 1981). When applied to eukaryotic
surrounding macro-environment and is constantly cells, the glycocalyx suggests a defined structure.
changing the biofilm matrix, it may be considered as This is not appropriate for microbial biofilms, where
a dynamic environment (Sutherland, 2001). Attach- not only the extracellular biopolymers but the
ment to a surface is thought to initiate a cascade of resident cells are constantly changing. Hence,
physiological changes in the cells which leads, in biofilm matrix is a better descriptor, implying a
part, to the overproduction of exopolysaccharides multi-component, dynamic heterogeneous system.
(Allison & Sutherland, 1987; Davies & Geesey, 1995). In a similar vein, the abbreviation EPS has been
These exopolysaccharides not only help attach cells used interchangeably for exopolysaccharides, exo-
to the colonized surface, but also facilitate the spatial polymers and extracellular polymeric substances

*Fax: 0161-275-2396; e-mail: david.allison@man.ac.uk

ISSN 0892-7014 print/ISSN 1029-2454 online q 2003 Taylor & Francis Ltd
DOI: 10.1080/0892701031000072190
140 D G ALLISON

(Wingender et al., 1999a). To use EPS as an acronym upon a combination of intrinsic factors such as the
for the all-embracing term extracellular polymeric genotype of the attached cells and extrinsic factors
substances gives the impression of uncertainty that include the surrounding physico-chemical
regarding matrix composition. At a gross level, this environment (Wimpenny, 2000). Thus, it is probable
is not the case; the composition of many different that biofilm matrices, even those produced by
biofilm matrices has been studied (see below). In this identical organisms, will vary greatly in their
respect, the individual polymeric components could composition and in their physical properties.
be prefixed with the term “matrix” (e.g. matrix The composition of the matrix may be influenced
proteins) to differentiate them from intracellular or by different processes, namely active secretion of
cell-bound materials. Thus, although exopolysac- biopolymers and enzymes, shedding of cell surface
charides may not be the most abundant component material such as lipopolysaccharide, cell lysis and
of the matrix, a traditional standpoint will be adsorption of macromolecules from the surrounding
adopted in this article whereby EPS will specifically environment. It is worth noting that matrix material,
represent exopolysaccharides. particularly EPS, shed from biofilms can be adsorbed
Very often, biofilms are composed of mixed at places distinct from their source. Enzymes altering
communities of microorganisms and the products the macromolecules within the matrix will also have
of their metabolism. Because of this inherent a very marked influence on its physical properties
complexity, this can make isolation and characteri- and composition. Different enzymes involved in EPS
zation of matrix components extremely difficult. degradation (hydrolases, lyases, glycosidases,
Detailed analyses of the biofilm matrix has been esterases and other enzymes) are abundant in
hampered by a number of factors, not least the biofilms as extracellular proteins (Sutherland,
availability of sufficiently sensitive and specific tools 1999b; Wingender et al., 1999b). Most of these
for probing the matrix structure. In addition, it is enzymes form an integrated part of the biofilm
difficult to assess accurately composition if some matrix and serve to release cells from the attached
components are only present in low quantities and community and may provide low-molecular weight
are otherwise masked by copious quantities of more breakdown products available as carbon and energy
dominant macromolecules. That is not to say, sources for metabolism by the immobilized bacteria
however, that minority components are not import- (Frølund et al., 1995). The presence of glycosyltrans-
ant and do not play a significant role in developing ferases in oral biofilms will result in levans and
and maintaining matrix architecture. dextrans being synthesized in the presence of
Whether matrix polymers differ from those sucrose. Hence, when sucrose is present the EPS
associated with planktonically grown cells, and element of matrix composition will be constantly
also whether the matrix polymers, which bind cells changing. Moreover, enzymes degrading both these
to other cells, differ from “foot-print polymers” polymers may also be present. Thus, new surfaces
which cement the primary colonizers to the are created and others masked, which in turn will
substratum is uncertain at present (Sutherland, have a significant influence on successional cell
1995). From the limited number of studies that attachment (Sutherland, 2001).
have attempted to characterize the matrix associated Depolymerization of structural polymers of the
with bacterial biofilms, it is clear that they vary biofilm matrix with subsequent release of cells
greatly in their composition and physical properties. allows for the dispersal of biofilm bacteria and
However, they may be no different from the colonization of new sites. In this manner, enhanced
extensive range of polymers derived from planktonic expression of alginate lyase in Pseudomonas aerugi-
cultures which have now been characterized and nosa led to an increased rate of detachment of cells
probably few, if any, are biofilm-specific (Sutherland, from agar-grown biofilms, suggesting a role for this
1997). As yet, the cryptic capability of biofilm enzyme in biofilm sloughing (Boyd & Chakrabarty,
organisms to synthesize novel polymers only in the 1994). Similarly, the presence of EPS degrading
biofilm mode has not been successfully demon- enzymes present in dense biofilms of Pseudomonas
strated (Sutherland, 1999a). fluorescens was proposed as a mechanism for
detachment of biofilm cells under starvation con-
ditions (Allison et al., 1998). It should also be
MATRIX COMPOSITION remembered that, in multi-species biofilms, the
collective action of several different enzymes may
Since biofilms are found in almost every environ- result in the degradation or alteration of EPS which
ment, both natural and artificial, in which moisture are resistant to discrete enzymes.
and microorganisms are present, it becomes extre- Water is by far the major component of the biofilm
mely difficult to generalize about their structure matrix, accounting for up to 97% of the mass. The
and physiological activities. As summarized by water may be bound within the capsules of the
Sutherland (2001), matrix composition is based bacterial cells or can exist as a solvent whose physical
 THE BIOFILM MATRIX 141

properties are determined by the solutes dissolved in heteropolysaccharides are almost all composed of
it (Sutherland, 2001). Water binding and mobility repeating backbone units varying in size from
within the biofilm matrix are integral to the diffusion disaccharide to octasaccharide. Structural diversity
processes that occur within the biofilm and, there- and, hence, rheological properties is increased by
fore, form part of the fine structure of the biofilm non-carbohydrate substituents (e.g. acetyl, pyruvate,
(Schmitt & Flemming, 1999). Resident cells, which sulphate groups) and linkage types. It is also worth
may include many different species, only account for noting that several species of bacteria are able to
, 5% of the matrix. Water and cells apart, the other synthesize more than one chemically distinct EPS,
components of the matrix include, in varying although normally only one type is expressed under
amounts, EPS (1 – 2%), globular glycoproteins and any set of specific growth conditions. Moreover,
proteins, which include lytic products and secreted research in many laboratories has focused on
enzymes (1 – 2%), nucleic acids from lysed cells characterizing and understanding alginate pro-
(1 – 2%), lipids, phospholipids and sequestered ions duction in P. aeruginosa biofilms. Whilst of interest
from the surrounding environment (Goodwin & to some situations, particularly infections of the
Foster, 1989; Sand & Gehrke, 1999; Flemming et al., cystic fibrotic lung, it should be noted that bacterial
2000; Flemming & Wingender, 2001). In addition, alginate is a very distinct EPS and is not representa-
trace amounts of humic acids have been detected in tive of EPS found in the diverse and complex range
the matrices of some soil and water biofilms (Burns, of environments where biofilms are found. This
1989; Jahn & Nielsen, 1998). It is important to note oversight has perhaps led to the misconception that
that these estimates are really only snapshots of all biofilm EPS are similar.
different biofilms at different points in time. Specific Polysaccharide chains vary in size from 103 –
composition for any biofilm will vary depending 108 kDa and contain sub-unit configurations which
upon the organism(s) present, their physiological may also be both functionally- and species-specific
status, the nature of the growth environment, bulk (Sutherland, 1985). Depending upon the components
fluid-flow dynamics, the substratum and the of the repeat units polysaccharides are usually
prevailing physical conditions. negatively charged, sometimes neutral or rarely
positively charged. Furthermore, polysaccharides
may be hydrophilic but can also have hydrophobic
EPS Composition
properties (Neu & Poralla, 1990). Indeed, many
Common to virtually all biofilm matrices are EPS. polymers are heterogeneous with respect to lipophi-
Although not a major component in terms of amount licity and hydrophobicity, let alone a fully-formed,
present, the EPS is, however, regarded as the major multi-component biofilm matrix (Sutherland, 1997).
structural component of the matrix, providing a Polymer hydrophobicity can, therefore, play an
framework for the biofilm complex. In essence, the important part in determining the behaviour of the
EPS provides the skeleton into which microbial cells polysaccharide at the cell surface or at an interface.
and their bioactive products are inserted. However, biochemical information of this nature has
At an individual cell level, EPS occur in two basic almost exclusively been provided from studies on
forms, viz. capsular, whereby the EPS is intimately EPS isolated from culture medium. At a cellular
associated with the cell surface, and as slime which is level, structure– function relationships for most EPS,
only loosely associated with the cell. Differentiation as reviewed by Sutherland (1997), have received little
between the two forms can often be difficult, since systematic investigation. Few EPS known to be
cells producing large quantities of capsule may implicated in the adhesion process have been
‘release’ some material at the periphery, giving the comprehensively analysed. Moreover, EPS-produ-
appearance of slime production. cing adherent microorganisms are often cultivated in
Chemically, bacterial EPS are highly hetero- nutrient-rich growth media, irrespective of their
geneous polymers containing a number of distinct natural habitat. This creates a nutritional environ-
monosaccharides and non-carbohydrate substitu- ment, which, for the majority of attached popu-
ents, many of which are strain specific (Sutherland, lations, is significantly different from that of a biofilm
1985; Whitfield, 1988). As with all polysaccharides, in situ. It is well documented that bacterial cell
those produced by microorganisms can be divided surface phenotype can change markedly in response
into homopolysaccharides and heteropolysaccha- to changes in the surrounding growth environment,
rides. Most homopolysaccharides are neutral glu- particularly those brought about by growth rate and
cans, whilst the majority of heteropolysaccharides nutrient limitation (Ellwood & Tempest, 1972; Brown
appear to be polyanionic (Sutherland, 1990). Homo- & Williams, 1985; Brown & Gilbert, 1993). Exopoly-
polysaccharides can possess three different struc- saccharides are also subject to environmental
tures, namely linear molecules comprised of a single modulation with respect to composition and mol-
linkage type, linear repeat units possessing a one- ecular mass (Tait et al., 1986) which in turn can affect
sugar side chain, and branched structures. Microbial their capacity to interact with other polymers and
142 D G ALLISON

cations (Sutherland, 1990). Thus, in order to fully Suitable physical methods include attenuated total
characterize any biofilm polymer, it is essential to be reflection/Fourier transform-infrared spectroscopy
able to mimic in situ growth conditions that lead to (ATR-FT-IR), which records the appearance of
EPS production. In this respect, previous studies on specific chemical groups on the surface of an internal
EPS synthesis and composition that have not taken reflective element. Whilst useful for studying cell
these factors into consideration should be reviewed adhesion, biofilm development and adsorption of
with caution. isolated polysaccharides (Nivens et al., 1995; Schmitt
et al., 1995), studies are limited to certain attachment
surfaces. Similarly, NMR spectroscopy can also be
MATRIX ARCHITECTURE used as a non-invasive technique to study intact
biofilms, 1H NMR having been used to demonstrate
The biofilm matrix provides the possibility that the mass distribution and gather flow velocity infor-
resident microorganisms can form stable aggregates mation (Lewandowski et al., 1993). The potential of
of different cell types, leading to the development of this technique, however, to measure EPS com-
a functional, synergistic microconsortium. The ponents in situ in complex biofilm communities has
spatial arrangement of microorganisms gives rise to not been assessed.
nutrient and gaseous gradients as well as those of In recent years, confocal laser scanning
electron acceptors, products and pH. Thus, aerobic microscopy (CLSM) has developed as a non-
and anaerobic (anoxic) habitats can arise in close destructive, three-dimensional optical sectioning
proximity, and, as a consequence, the development technology for the visualization of fully hydrated,
of large variability of species takes place. Moreover, viable biofilm systems (Neu & Lawrence, 1999). Use
since the structure is not rigid, the potential is there of such technology has provided new information on
for organisms to move in it, thereby facilitating biofilm ultrastructure and confirmed their hetero-
genetic exchange. A few studies have demonstrated geneity (Stoodley et al., 1999a). Increased application
improved gene exchange in pure culture biofilms of this tool has occurred in conjunction with a wide
(Angles et al., 1993; Lisle & Rose, 1995). In general, range of fluorescent probes, providing an ideal
however, the effects on plasmid transfer of differ- method by which to study the spatial distribution of
ences in localization of donor and recipient cells biofilm properties.
within mixed-species biofilms is still poorly under- The in situ analysis of the biofilm matrix relies on
stood (Angles & Goodman, 2000). Further, it has probes that possess varying degrees of specificity for
been shown that the nature of the surface may be representative molecules or characteristic chemical
important in determining the structures of biofilms, groups. The surface chemistry of bacterial cells and
which may in turn affect plasmid transfer (Dalton the surrounding matrix can be analysed using a
et al., 1994). Lorenz et al. (1988) demonstrated variety of probes including those for EPS, proteins
transformation efficiency in Bacillus subtilus to be and nucleic acids (Neu & Lawrence, 1999). Some of
50-fold greater at solid/liquid interfaces compared these are highly specific for EPS, such as antibodies
to liquid and 3200 times higher when only cells and lectins. Studies by Costerton et al. (1981) used
attached to mineral surfaces were considered. antibodies prepared against a planktonically pro-
However, their studies on two Gram-negative duced EPS to reveal the presence of similar material
bacteria did not show these increased transformation in the biofilm matrix. Although this study was a
rates when grown as biofilms. Research on gene defining demonstration of matrix composition, the
transfer within biofilms is still in its infancy. Most of use of antibodies as specific probes has its
the studies of plasmid transfer in biofilms involve limitations. First, they have to be produced in animal
the addition of donor and recipient cells simul- or cell cultures against an isolated polysaccharide.
taneously. This situation may not be typical of This EPS should be in a very pure form to ensure
natural environments where either donor or recipi- high specificity and little cross-reactivity. This raises
ents may already be established in a biofilm. Very questions about their applicability to complex EPS
little has been done to elucidate the role of EPS in this containing biofilm matrices, particularly in multi-
process. The evidence to date would suggest more species populations.
of an indirect role through, for example, alteration of Fluor-conjugated lectins can be used to map the
the hydrodynamic conditions in a biofilm or chemical composition of the EPS material associated
maintenance of the plasmid within the attached with bacterial cells. Lectins are a group of diverse
population. proteins which bind to specific configurations of
Understanding the mechanical and architectural sugar residues. The majority of the lectins described
properties of the matrix is very much dependent in the literature are isolated from plants and other
upon having suitable methods to permit accurate eukaryotic organisms. These may not be the ideal
analysis of matrix structure and composition. Ideally, source for probing the EPS component of the biofilm
an in situ, non-destructive approach should be used. matrix, particularly within bacterial biofilm systems.
 THE BIOFILM MATRIX 143

Rather, it may be more appropriate to apply Despite the great potential of using lectins as
bacterial-specific lectins that recognize unique specific probes to estimate glycoconjugate distri-
monosaccharides, atypical oligosaccharide bution in biofilm systems, questions still remain
sequences or even a certain type of polysaccharide. regarding the identity of binding sites, the influence
Most lectins are available commercially as purified of the fluorescent conjugates on the behaviour of the
proteins or even fluorescently labelled. In general, lectins and the interactions between lectins in
the specificity of lectins is ascribed to the manufac- multiple-lectin staining. Progress with this method
turer and is based on the influence of lectins on blood of EPS analysis will depend on a more detailed
cell agglutination. characterization of various lectins, a greater under-
In conjunction with CLSM, lectins have proved standing of the nature of the lectin-binding sites for
valuable tools in the study of the 3-dimensional proteins and carbohydrates and defining the natural
structure of biofilms (Kolari et al., 1998; Lawrence binding specificity and properties of multiple lectin-
et al., 1998). An example of the use of lectin-CLSM binding sites relevant to microbial and biofilm
analysis is shown in Figure 1. In this study, applications.
P. aeruginosa biofilms grown in a glass flow-cell Non-destructive studies of mixed-species biofilms
using an artificial groundwater medium were ideally require the reliable localization of members of
stained with fluorescent probes specific for nucleic specific bacterial populations in relation to others, as
acids (SYTO 59, red) and b1 – 3 and 1 –4 EPS linkages well as information on the activity of individual cells.
(FITCA, gold). Here, shear structures within large Some methodological problems that may have an
cell clusters and the EPS component of the matrix impact on the success of tagged antibody or lectin
were revealed, with displacement direction parallel labelling protocols include target site accessibility
to the fluid flow direction (Brydie, unpublished and specificity and sensitivity. Some biofilms inter-
data). fere with the free diffusion of fluorescent probes,
Lectins may be employed as single or multi- especially where the concentration of matrix associa-
ple probes in complex natural systems (Neu & ted cations is high and copious amounts of cell EPS
Lawrence, 1999). In addition, double lectin labelling are present. Reporter genes, such as green fluo-
can be combined with nucleic acid stains, such as the rescent protein (GFP) provide an alternative,
SYTO series, to demonstrate the localization of molecular means of identifying specific cell types
glycoconjugates relative to cell and microcolonies in in model complex biofilm communities (Skillman
the biofilm structure. However, it is important to et al., 1998). Linking the expression of attachment-
assess beforehand the potential for lectin interactions inducible or biofilm-specific target genes to fluor-
prior to any multiple-labelling experiments, since escent reporter gene constructs will provide an
lectin interactions may preclude their combined indirect method whereby the regulation of these
application (Neu et al., 2001). genes may be analysed quantitatively and, more
However, a note of caution should be exercised importantly, over time. It is important to ensure,
when analysing such images since the lectins may however, that the reporter gene constructs used do
bind to non-EPS targets or adhere non-specifically to not alter the physiological properties of the recipient
other components of the biofilm matrix (Johnsen cells.
et al., 2000). Moreover, in a recent comprehensive Nevertheless, these methods, along with the use of
analysis of lectin-binding proficiency, it was shown dissolved oxygen and pH microsensors, have led to
that the order of addition, lectin specificity and the the understanding that most biofilms, irrespective of
nature of the fluor-conjugate were found to influence environment, comprise aggregates of cells within an
the binding pattern of the lectins (Neu et al., 2001). EPS containing matrix, separated by interstitial voids
Flourescein isothiocyanate (FITC)-conjugated lectins and channels (Sutherland, 2001). Information collec-
possessed more specific binding characteristics than ted by these methods along with computer simu-
either trtramethyl rhodamine isothiocyanate lation has been used to propose three distinct biofilm
(TRITC)- or cyanine dye (CY5)-labelled lectins. This models, depicted schematically in Figure 2. These
may be due to the fact that these two molecules are comprise simple stalked or irregular branching
charged and that TRITC is zwitterionic in nature, structures, dense confluent structures and pene-
thereby affecting lectin-binding interactions. These trated water channel or mushroom-shaped biofilms
general binding effects with little apparent locali- (Wimpenny & Colsanti, 1997). All three models
zation are likely to be more extensive in multi- possess water channels which permeate not only
species biofilms compared to pure culture biofilms. established laboratory monoculture biofilms, but
As such, it is essential that the selection of a panel of have also been found in real-life situations (Coster-
lectins for investigating the EPS component of a ton et al., 1994; DeBeer et al., 1994a; Wimpenny &
biofilm matrix must be based on a full evaluation of Colsanti, 1997; Stoodley et al., 1999a). Even in the
their behaviour in the biofilm system to be studied dense, confluent structure of dental plaque, water
(Neu et al., 2001). channels and fluid filled voids have been detected
144 D G ALLISON

FIGURE 1 Example of a biofilm population examined by CLSM.

A P. aeruginosa biofilm grown in a glass flow cell using an artificial
groundwater medium was probed with the fluorescent stains
SYTO 59 (red) for nucleic acids and FITCA (gold) for b1–3 and
b1 – 4 EPS linkages. The image clearly demonstrates the
distribution of both cells and EPS within the biofilm matrix
(Brydie J et al., unpublished data).

(Lawrence et al., 1998). In all of these systems, the

water channels may be considered as primitive
circulatory systems, providing a mechanism for the
flow of nutrients, enzymes, metabolites and waste
products throughout the biofilm community. The
FIGURE 2 Schematic diagram of three distinct biofilm models,
presence of relatively large channels and pores comprising (A) stalked, (B) penetrated water channel and (C)
within the matrix structure might also allow entry of dense confluent structures. The arrow indicates water channels.
secondary colonizers and their establishment within
the biofilm. The degree and distribution of channels
varies and is dependent upon the rate of EPS
synthesis and turnover. In a dynamic biofilm,
solubilization of EPS occurring at the periphery
will be compensated by increased polysaccharide
production within the depths (Evans et al., 1994),
thereby affecting ultrastructure.
Although these models represent three distinct
forms, in reality biofilms are a combination of all
three depending on many extrinsic factors. Structure
is largely dependent on resource concentration.
What is basically a heterogeneous biofilm one day
may be more of a dense confluent biofilm the next
day following subtle changes in the bulk aqueous
phase. In a recent study by Møller et al. (1998),
biofilms grown on a poorly used substrate produced
mounds of cells, but when switched to a rich
medium produced a much more uniform biofilm.
Similarly, Stoodley et al. (1999b) demonstrated that FIGURE 3 Reaction-diffusion-limitation effects of the biofilm
raising the glucose concentration changed a mixed matrix towards antimicrobial agents. Gradients of biocide occur
species biofilm from a series of cell clusters with (arrow) due to penetration failure through (i) binding and
chemical reaction with the EPS and (ii) enzyme-assisted
ripples to a thick biofilm without ripples. Subsequent inactivation. There, hydrolytic enzymes present in the matrix
lowering of the concentration to initial levels caused will attack and inactivate slowly diffusing biocides.
 THE BIOFILM MATRIX 145

a reversion of the biofilm to the initial phenotype. biofilm contaminant of a food manufacturing sur-
What has become evident from these and other face, it was shown that both EPS possessed a high
studies is that irrespective of growth and physico- proportion of 1,3 linked residues, similar in some of
chemical conditions, all forms of biofilm comprise a their physical properties to the 1,3 linked D -glucans
heterogeneous microenvironment that includes pos- of mutan or curdlan. These polysaccharides possess
session of water channels. poor water solubility and, as such, provide a very
The presence of polysaccharide-synthesizing and effective matrix for bacterial adhesion. This led to the
-degrading enzymes in the biofilm means that matrix suggestion that water-soluble molecules such as
composition will be constantly changing. New disinfectants may be effectively excluded from the
surfaces are created whilst others are masked, interior of such biofilm matrices. Most EPS in
thereby allowing new species to attach to the biofilm. solution undergo a change from order to disorder
Enzymes altering macromolecules within the matrix on heating or on removal of ions. Increasing
will also have a marked influence on its physical solubility and eventual dissolution or sloughing
properties (Sutherland, 2001). Under conditions of from a biofilm (Sutherland, 1997; Willcock et al., 1997)
nutrient starvation, EPS degrading enzymes may be may reflect such changes. Also of great significance
produced which will cause local destruction of the will be the conformation adopted by the EPS. Whilst
matrix and possible release of cells (Allison et al., polymers with charged groups on the outside of the
1998). This may also cause weakening of the tertiary structure will bind ions they are unlikely to
community structure, especially if such enzymes form tight bonds between adjacent strands. Alter-
are carried through the channels to sites remote from natively, polymers such as the alginates produced by
their synthesis. Azotobacter vinelandii which possess block sequences
Many microorganisms are also capable of synthe- of poly-L -guluronic acid can bind cations very
sizing biosurfactants. These clearly might play a effectively when deacetylated to form salt bridges
significant role in localized dissolution of matrix between carboxyl groups on adjacent strands,
material. One possible role for surfactants is to thereby forming a strong gel. In alginates, gel
enhance desorption of the microbial cells from strength can be influenced by both carbohydrate
hydrophobic surfaces which no longer contain composition and the presence or absence of O-acetyl
usable carbon sources. groups. The alginates produced by A. vinelandii most
Numerous studies have shown that the majority of closely resemble the algal products, producing poly-
EPS can exist in ordered or disordered forms L -guluronosyl sequences with acetylation occurring
(Sutherland, 1997). Many bacterial EPS adopt a on many of the D -mannuronosyl residues. These
double helical configuration in the ordered form, polymers provide much greater effective gel strength
with association between double helices being than do the corresponding poly-D -mannuronic acid
facilitated by the ionic content of the surrounding blocks found in Pseudomonas aeruginosa alginate.
bulk aqueous phase and by water molecules. At a Moreover, the presence of O-acetyl groups in
molecular level, the physical properties of EPS are bacterial alginates have been shown to have a strong
dependent on these interactions and may be inhibitory effect on cation binding (Geddie &
influenced by the presence or absence of free anionic Sutherland, 1994), thereby reducing gel strength
groups derived from uronic acid, phosphate groups, further. Nevertheless, non-acetylated alginate does
pyruvate ketals or succinyl half esters (Sutherland, bind cations, particularly Ca2þ ions and will form a
1997). In most natural environments where relatively weak gel. Differences in gel strength, brought about
high ionic concentrations are found, polysaccharides through differences in polymer composition are,
will invariably be found in an ordered configuration therefore, likely to effect the physical properties of
and as gelled and highly hydrated polymers. EPS the biofilm matrix.
also vary in their water solubility, some being highly Some polysaccharides such as the galactoglucans
soluble in water or dilute salt solutions, whilst others found in some Rhizobium sp. lack intra-chain
are virtually insoluble in water or form very rigid hydrogen bonds with the anionic portion of the
gels when in the ordered form (Sutherland, 1997). polymer, provided by the pyruvate ketals, being
Polysaccharides of this type are likely to be present along with acetyl groups on the periphery of
extremely effective molecules at holding biofilms the helix. Such polymers will remain in a disordered
and their components in place. Among these are conformation (Cesáro et al., 1992) which may be
several polymers which are commonly found in characterized by flexibility and will probably cause
biofilms, and include mutan produced by Strepto- loosely attached cells to be very easily removed from
coccus mutans. Mutan is well characterized as being the biofilm.
predominantly water-insoluble due to its 1,3-a- Changes in hydrodynamic conditions can also
linked glucan structure. In a study by Hughes affect matrix structure. Shear rate will influence rates
(1997) on the EPS extracted from two species of of erosion of cells and regions of the matrix from the
Enterobacter agglomerans originally isolated from a biofilm (Stoodley et al., 1999c; Willcock et al., 2000).
146 D G ALLISON

TABLE I Proposed generalised functions of the biofilm matrix

Function Process
Structural element of biofilm Provides mechanical stability and three dimensional
heterogeneous structure of biofilm; contains water
channels
Attachment to surfaces Some EPS involved in initial adhesion
events; main role is maintenance and
persistence of attached community
Protective barrier i Reactive diffusion limitation to antimicrobial agents
ii Steric resistance to specific and non-specific
host defence mechanisms
iii Resistance to protozoal grazing
iv Water content helps prevent dessication
Adsorption of organic/inorganic nutrients Acts as an ion exchange resin
Concentration of key enzymes and other i Enzymic digestion of trapped macromolecules
biologically active molecules
ii Degradation of component EPS for cell
release and/or attachment of secondary colonizer
iii Allow mediators of cell–cell communication to accumulate to threshold
concentration

Recently, it has been shown that changes in achieved by the occurrence of non-covalent inter-
detachment from a steady-state biofilm due to a actions either directly between EPS chains or via
rapid change in shear force are independent of the multivalent cation bridges. Lectin-like proteins may
initial shear rate and can cause temporary, elastic also contribute to the formation of the 3-dimensional
deformation of biofilm structures (Stoodley et al., network of the biofilm matrix by cross-linking EPS
1999c). Shear stresses applied to the biofilm will directly or indirectly through multivalent cations.
affect EPS solutions, causing flow and elastic A number of other functions have been ascribed to
recovery to occur. Such changes will consequently the biofilm matrix (Table I). Among these, protection
affect the shape of the matrix. Under turbulent flow, of the enveloped cells from antimicrobial treatments
the EPS endowed matrix will flow like a viscous ranks as one of the most important. Indeed, most, but
fluid. Turbulent flow and increased carbon source not all of the resistance characteristics of biofilm
alters the appearance of an established biofilm from communities are lost when those communities are
patches of roughly circular cell clusters to ripples resuspended and separated from their extracellular
and streamers which oscillated in the flow (Stoodley products. The presence of a charged, hydrated EPS
et al., 1999b). matrix around individual cells and microcolonies
In mixed species biofilms, different polymers are profoundly affects the access of solutes. Restricted
likely to be produced by each component species. diffusion from the surrounding medium, by a
These are likely to blend together to generate combination of ionic interaction and molecular
heterogeneous regions of polymer within the biofilm sieving events, will occur for appropriate classes of
matrix and, in so doing, possibly confer some molecule. The EPS of the extracellular matrix thereby
structure to the biofilm (Cooksey, 1992). The act as would an ion exchange resin and actively
physico-chemical properties of such blended EPS remove strongly charged molecules from solution. It
will differ significantly from those of purified is, therefore, not surprising that many groups of
components and will also be substantially affected workers have suggested that the matrix acts as a
by the ionic strength of the surrounding medium and protective umbrella that physically prevents the
the nature of the cationic species (Allison & access of antimicrobials to the cell surface (Gordon
Matthews, 1992). Furthermore, biofilm EPS from et al., 1988; Nichols et al., 1988). Many of the early
two or more species may interact synergistically, studies of antibiotic action on biofilms attributed
leading to the enhancement of bacterial adhesion their recalcitrance to exclusion (Slack & Nichols,
(Skillman et al., 1997). Biofilm EPS may also interact 1981; Suci et al., 1994). Such universal explanations
with other macromolecules such as proteins, lipids have been refuted (Nichols et al., 1989; Allison et al.,
and even nucleic acids. 2000), since reductions in the diffusion coefficients of
antibiotics such as tobramycin and cefsulodin,
within biofilms or microcolonies, are insufficient to
MATRIX FUNCTIONS account for the observed change in susceptibility.
The results from collected studies demonstrate that,
Undoubtedly, one of the most important functions of whilst the possession of a mucoid phenotype may be
the matrix is to provide the structural complexity associated with decreases in susceptibility (Allison
and mechanical stability of the biofilm. This may be et al., 2000), reductions in the diffusion coefficient
 THE BIOFILM MATRIX 147

across polymeric matrices relative to liquid media, relate to the nature of the agent, the binding capacity
are insufficient to account solely for it. This is of the matrix towards it, the levels of agent used
because the equilibration of antibiotic from the bulk therapeutically (Nichols, 1993), the distribution of
phase across the biofilm will be rapid, and at biomass (DeBeer et al., 1994b), the rate of turnover of
equilibrium the concentrations at the cell surfaces the microcolony relative to antibiotic diffusion rate
and in the bulk aqueous phase would be equal (Kumon et al., 1994) and the production and
(Gilbert et al., 2002). Such ready access of antimicro- retention of extracellular products (Allison et al.,
bial agent throughout the community will be further 2000).
aided by the presence of water channels. Given the high water content of the matrix in the
Whilst thickness of the matrix will not affect immediate vicinity of the microbial cells, it can help
diffusion properties per se, it will, however, greatly prevent desiccation under water-deficient conditions
increase the extent to which access of the agent, to the (Ophir & Gutnick, 1994). Such conditions would be
underlying cells, is retarded. This retardation might found in those surface biofilms which are exposed to
possibly be to such an extent that, where treatments frequent cycles of dehydration and rehydration. The
with e.g. biocides are periodic and short-lived, the EPS can also act, in a limited extent, as a diffusion
deeper lying cells might escape lethal exposure barrier to nutrients and cellular products, depending
(Nichols, 1993; Stewart, 1996). In addition, matrix upon the nature of both the EPS and specific
polymers and cellular materials at the periphery of a nutrients (Jensen & Revsbech, 1989). Similarly, it
biofilm community may react chemically with, and has often been suggested that EPS in a biofilm may
neutralize, the treatment agents. Such reaction- bind nutrients essential for growth, thereby creating
diffusion limiting effects (Figure 3) will be most a nutrient rich micro-environment in an otherwise
pronounced with the oxidative biocides such as nutrient poor macro-environment (Costerton et al.,
iodine and iodinepolyvinylpyrollidione complexes 1994). Bacterial polysaccharides have indeed been
(Favero et al., 1983) and for chlorine and peroxides shown to have the ability to bind cations, ion uptake
(Huang et al., 1995), which react non-specifically in a and selectivity for some EPS being influenced by the
consumptive manner with general exopolysaccha- level of acetylation (Geddie & Sutherland, 1993).
rides and cellular materials. Isothiazolone, and other These properties are potentially of great importance
thiol interactive agents might equally be quenched in sewage treatment processes for the removal of
by the presence of thiol-containing amino acids toxic heavy metal pollutants such as lead and
within the matrix. In all instances, the greater the cadmium. The matrix can also act directly as a
activity of the antimicrobial agent the greater will be carbon and energy source. Normally, bacterial cells
the susceptibility to such neutralization (Gilbert et al., are unable to utilise their own EPS. However, in
2001). Furthermore, the reaction-diffusion limitation mixed species biofilms, some bacteria can utilise the
of the matrix could be significantly enhanced if it EPS of other bacteria, providing an opportunity for
contained extracellular enzymes that were capable of cross-feeding (Sutherland, 2001). As a consequence,
degrading the diffusing substrate. A catalytic matrix composition and ultrastructure will be
reaction could lead to severe antimicrobial pene- altered. In pathogenic situations, the presence of
tration failure (Stewart, 1996), provided that the EPS can inhibit macrophage binding and also
turnover of substrate by the enzyme was sufficiently antibody coating, thereby blocking the immunologi-
rapid. Such mechanisms have been attributed to the cal determinants required for opsonic phagocytosis
sequestration of b-lactamase enzymes (Giwercman (Hoiby, 1982). The full effect of EPS on the host
et al., 1991) or formaldehyde lyase and dehydrogen- immune system, however, is still not completely
ase (Sondossi et al., 1985), which would cause the understood.
degradation of b-lactam antibiotics and formal- The biofilm matrix also provides an ideal
dehyde, respectively, thereby enhancing signifi- environment for the accumulation, to effective
cantly the protection afforded. concentrations, of molecules required for cell-to-cell
It is also worth noting that sub-lethal concen- communication. In Gram-negative bacteria, these
trations of some antimicrobial compounds promote signal substances are in the form of N-acyl
EPS synthesis. Although bacterial growth can substituted homoserine lactones (HSL) (Beck von
be inhibited, considerably more EPS can be formed Bodman & Farrand, 1995). These are global
than in the absence of the antimicrobial agents regulators of transcriptional activation in bacteria
(McKenney et al., 1994). Of particular significance is (Williams et al., 1992; Passador et al., 1993; Salmond
the observation that this phenomenon can occur with et al., 1995), responsible for cell – cell signalling and
a range of agents which differ considerably in their implicated in cell-density mediated events. Pro-
mode of antimicrobial action (Hughes et al., 1998). duction of such small, diffusible signals enables the
Clearly, whether or not the matrix constitutes a bacterial population to monitor its own density and
physical barrier to antimicrobial penetration to trigger a co-ordinate and unified response at the
depends greatly on a number of features. These population level (Swift et al., 1994). Once a threshold
148 D G ALLISON

level of HSL has been achieved, the bacteria respond CONCLUSION

by activating the expression of a series of specific, cell
density-dependent genes. In this manner, biofilms Direct knowledge of the structure and composition
can adopt phenotypic properties that are unique and of the biofilm matrix is still rather limited. For
unlikely to be mirrored in natural planktonic example, it is still not known which are the key
populations where HSL accumulation will not polymers, if any, in maintaining matrix structure and
reach such critical levels. for how long cells embedded within the matrix
Exopolysaccharides may also promote adherence produce EPS. In addition, it would be reasonable to
consider each matrix as distinct, with its own unique
of mucoid microorganisms to various substrata. In
microenvironment. How then are these different
cystic fibrosis, there is some evidence to suggest that
matrices ordered? Nevertheless, it is clear that the
the alginate-like EPS produced by infecting strains of
different components of the matrix interact and
P. aeruginosa promotes adherence of mucoid strains provide a wide range of localized environments
to epithelial cells of the pulmonary tract (Doig et al., within the matrix, and that these microenvironments
1987). Moreover, EPS production may be under the are subject to extrinsic forces and are, therefore,
control of HSLs (Beck von Bodman & Farrand, 1995; constantly changing. Consequently, for any biofilm
Davies, 1999). A general assumption is that the EPS in any environment, these changes will contribute to
produced in biofilm matrices are all high molecular the heterogeneous nature of the matrix.
weight polymers. This may not be the case, and again
knowledge is dependent upon employment of
different EPS isolation and analysis procedures that Acknowledgements
will allow detection of materials other than high
molecular weight polymers. To this extent, there is a This article was written in honour of Professor Ian W
growing body of evidence to suggest that high Sutherland on the occasion of his retirement to
molecular weight exopolysaccharides do not act as acknowledge the significant contribution he has
adhesins during the process of irreversible adhesion. made to the development, understanding and
awareness of microbial physiology, particularly in
Instead, some microorganisms such as Rhizobium sp.,
the fields of bacterial exopolysaccharides and biofilm
Acetobacter sp., Agrobacterium sp. and Alkaligenes sp.
physiology.
have been shown to possess the ability to co-
synthesize two or more chemically distinct exopoly-
saccharides (Sutherland, 1985). These include low References
molecular weight exopolysaccharides (LMW.EPS),
Allison D G, Sutherland I W (1987) The role of exopolysaccha-
such as the b1 –2 glucans. It is possible, therefore, rides in adhesion of freshwater bacteria. J Gen Microbiol 133:
that these low molecular weight polymers are 1319–1327
synthesized and secreted in response to a nearby Allison D G, Matthews M J (1992) Effect of polysaccharide
interactions on antibiotic susceptibility of Pseudomonas
surface. Consequently, high molecular weight EPS aeruginosa. J Appl Bacteriol 73: 484–488
formation is manifested in subsequent events such as Allison D G, McBain A J, Gilbert P (2000) Biofilms: problems of
control. In: Allison D G, Gilbert P, Lappin-Scott H, Wilson M
stabilization and persistence of the colony (Allison & (eds) Community Structure and Co-operation in Biofilms, SGM
Sutherland, 1987; Neu & Marshall, 1991; Davies, et al., Vol 59. Cambridge University Press, Cambridge, pp 309–327
1998). Thus, although non-mucoid mutants of a Allison D G, Ruiz B, San Jose C, Jaspe A, Gilbert P (1998)
Extracellular products as mediators of the formation and
freshwater bacterium were shown to be able to attach detachment of Pseudomonas fluorescens biofilms. FEMS Micro-
to a glass surface in equally high numbers to the biol Lett 167: 179– 184
Angles M L, Goodman A E (2000) Plasmid transfer between
mucoid wild type, they were unable to form bacteria in biofilms. In: Evans L V (ed) Biofilms: Recent
microcolonies (Allison & Sutherland, 1987). Some Advances in their Study and Control. Harwood Academic
Publishers, Amsterdam, pp 81 –99
other cellular factor(s) are thought to be involved in Angles M L, Marshall K C, Goodman A E (1993) Plasmid transfer
the initial adhesive interaction. Studies on the between marine bacteria in the aqueous phase and biofilms in
production and interaction of LMW.EPS with reactor microcosms. Appl Environ Microbiol 59: 843–850
Battisti L, Lara J C, Leigh J A (1992) Specific oligosaccharide form
surfaces suggest that they may fulfil such a role. In of the Rhizobium meliloti exopolysaccharide promotes nodule
the soil ecosystem, the successful invasion of alfalfa invasion in alfalfa. Proc Natl Acad Sci USA 89: 5625–5629
Beck von Bodman S, Farrand S K (1995) Capsular polysaccharide
by Rhizobium meliloti is dependent upon both a high biosynthesis and pathogenicity in Erwinia stewartii require
and low molecular weight EPS being produced induction by an N-acyl homoserine lactone autoinducer.
(Battisti et al., 1992). Whereas the LMW-succinogly- J Bacteriol 177: 5000–5008
Boyd A, Chakrabarty A M (1994) Role of alginate lyase in
can acts as a signal in Rhizobium symbiosis and is cell detachment of Pseudomonas aeruginosa. J Bacteriol 60:
required to initiate alfalfa root nodule invasion, the 2355–2359
Brown M R W, Williams P (1985) The influence of the environment
high molecular weight polymer stabilizes the on envelope properties affecting survival of bacteria in
symbiotic interaction. infections. Ann Rev Microbiol 39: 527 –556
 THE BIOFILM MATRIX 149

Brown M R W, Gilbert P (1993) Sensitivity of biofilms to Giwercman B, Jensen E T, Høiby N, Kharazmi A, Costerton J W
antimicrobial agents. J Appl Bacteriol 74 (Suppl): 87S–97S (1991) Induction of b-lactamase production in Pseudomonas
Burns R G (1989) Microbial and enzymic activities in soil biofilms. aeruginosa biofilms. Antimicrob Agents Chemother 35:
In: Characklis W G, Wilderer P A (eds) Structure and Function 1008–1010
of Biofilms. Wiley, Chichester, pp 339 –349 Goodwin J A S, Foster C F A N (1989) Examination of the
Cesáro A, Tomasi G, Gamini A, Vidotto S, Navarini L (1992) extracellular polymers produced by activated-sludge. Micro-
Solution conformation and properties of the galactoglucan bios 57: 2179–2185
from Rhizobium meliloti strain (S1). Carbohyd Res 231: 117– 135 Gordon C A, Hodges N A, Marriot C (1988) Antibiotic interaction
Cooksey K E (1992) Extracellular polymers in biofilms. In: Melo L F, and diffusion through alginate and exopolysaccharide of
Bott T R, Fletcher M, Capdeville B (eds) Biofilms: Science and cystic fibrosis derived Pseudomonas aeruginosa. J Antimicrob
Technology. Kluwer Academic Press, Dordrecht, pp 137–147 Chemother 22: 667–674
Costerton J W, Irwin R T, Cheng K J (1981) The bacterial glycocalyx Henrici A T (1932) Studies of freshwater bacteria. J Bacteriol 25:
in nature and disease. Ann Rev Microbiol 35: 399 –424 277 –288
Costerton J W, Lewandowski Z, deBeer D, Caldwell D, Korber D, Hoiby N (1982) Microbiology of lung infections in cystic fibrosis
James G (1994) Biofilms, the customised microniche. J Bacteriol patients. Acta Paediatrica Scand 301 (Suppl): 33– 54
176: 2137–2142 Huang C T, Yu F P, McFeters G A, Stewart P S (1995) Non-uniform
Dalton H M, Poulsen L K, Halasz P, Angles M L, Goodman A E, spatial patterns of respiratory activity within biofilms during
Marshall K C (1994) Substratum-induced morphological disinfection. Appl Env Microbiol 61: 2252– 2256
changes in a marine bacterium and their relevance to biofilm Hughes K A (1997) Bacterial biofilms and their exopolysacchar-
structure. J Bacteriol 176: 6900–6906 ides. PhD Thesis, University of Edinburgh
Davies D G (1999) Regulation of matrix polymer in biofilm Hughes K A, Sutherland I W, Jones M V (1998) The function of
formation and dispersion. In: Wingender J, Neu T R, biofilm exopolysaccharides: increased exopolysaccharide
Flemming H-C (eds) Microbial Extracellular Polymeric Sub- production following exposure to antimicrobial agents and
stances. Springer, Berlin, pp 91 –117 enhanced desiccation resistance. Microbiology 144: 3039–3047
Davies D G, Geesey G G (1995) Regulation of the alginate Jahn A, Nielsen P H (1998) Cell biomass and exopolymer
biosynthesis gene algC in Pseudomonas aeruginosa during composition in sewer biofilms. Water Sci Technol 37: 17–24
biofilm development in continuous culture. Appl Env Jensen J, Revsbech N P (1989) Photosynthesis and respiration of a
Microbiol 61: 860 –867 diatom biofilm cultured in a new gradient growth chamber.
Davies D G, Parsek M R, Pearson J P, Iglewski B H, Costerton J FEMS Microbiol Ecol 62: 29–38
W, Greenberg E P (1998) The involvement of cell-to-cell Johnsen A R, Hausner M, Schnell A, Wuertz S (2000) Evaluation of
signals in the development of a bacterial biofilm. Science 280: fluorescently labeled lectins for noninvasive localization of
295– 298 extracellular polymeric substances in Sphingomonas biofilms.
DeBeer D, Stoodley P, Lewandowski Z (1994a) Liquid flow in Appl Env Microbiol 66: 3487–3491
heterogeneous biofilms. Biotechnol Bioeng 44: 636–641 Kolari M, Matilla K, Mikkola R, Salkinoja-Salonen M S (1998)
DeBeer D, Srinivasan R, Stewart P S (1994b) Direct measurement of Community structure of biofilms on ennobled stainless steel
chlorine penetration into biofilms during disinfection. Appl in Baltic sea water. J Ind Microbiol Biotechnol 21: 261–274
Environ Microbiol 60: 4339–4344 Kumon H, Tomochika K-I, Matunga T, Ogawa M, Ohmori H (1994)
Doig P, Smith N R, Todd T, Irvin R T (1987) Characterisation of the A sandwich cup method for the penetration assay of
binding of Pseudomonas aeruginosa to human epithelial cells. antimicrobial agents through Pseudomonas exopolysaccharide.
Infect Immun 55: 1517–1522 Microbiol Immunol 38: 615 –619
Ellwood D E, Tempest D W (1972) Effects of environment on Lawrence J R, Neu T R, Swerhome G D W (1998) Multiple
bacterial wall content and composition. Adv Microb Physiol parameter imaging, and microbial biofilms. Application of
7: 83 –117 multiple parameter imaging for the quantification of algal,
Evans E, Brown M R W, Gilbert P (1994) Iron chelator, bacterial and exopolymer components of microbial biofilms.
exopolysaccharide and protease production of Staphylococcus J Microbiol Meth 32: 253 –261
epidermidis: a comparative study of the effects of specific Lewandowski Z, Altobelli S A, Fukushima E (1993) NMR and
growth rate in biofilm and planktonic culture. Microbiology microelectrode studies of hydrodynamics and kinetics in
140: 153–157 biofilms. Biotechnol Prog 9: 40 –45
Favero M S, Bond W W, Peterson N J, Cook E H (1983) Scanning Lisle J T, Rose J B (1995) Gene exchange in drinking water and
electron microscopic observations of bacteria resistant to biofilms by natural transformation. Water Sci Technol 31: 41 –46
iodophor solutions. In: Proc Int Symp Povidone, University of Lorenz M G, Aardema B W, Wackernagel W (1988) Highly efficient
Kentucky, Lexington, USA, pp 158–166 genetic transformation in the environment. Microbiol Rev
Flemming H-C, Wingender J (2001) Relevance of microbial 58: 563–602
extracellular polymeric substances (EPSs)– Part I: Structural McKenney D, Willcock L, Trueman P, Allison D G (1994) Effect of
and ecological aspects. Water Sci Technol 43: 1–8 sub-MIC antibiotics on the cell surface and extracellular
Flemming H-C, Wingender J, Mayer C, Korstgens V, Borchard W virulence determinants of Pseudomonas cepacia. J Appl Bacteriol
(2000) Cohesiveness in biofilm matrix polymers. In: Allison 76: 190–195
D G, Gilbert P, Lappin-Scott H, Wilson M (eds) Community Møller S, Sternberg C, Andersen J B, Christensen B B, Ramos J L,
Structure and Co-operation in Biofilms. SGM Volume 59, Givskov M, Molin S (1998) In situ gene expression in mixed-
Cambridge University Press, Cambridge, pp 87– 105 culture biofilms: evidence of metabolic interactions between
Frølund B, Griebe T, Nielsen P H (1995) Enzymic activity in the community members. Appl Env Microbiol 64: 721 –732
activated sludge-floc matrix. Appl Microbiol Biotechnol 43: Neu T R, Lawrence J R (1999) In situ characterization of
755– 761 extracellular polymeric substances (EPS) in biofilm systems.
Geddie J L, Sutherland I W (1993) Uptake of metals by bacterial In: Wingender J, Neu T R, Flemming H-C (eds) Microbial
polysaccharides. J Appl Bacteriol 74: 467–472 Extracellular Polymeric Substances. Springer, Berlin, pp 21–47
Geddie J L, Sutherland I W (1994) The effect of acetylation on Neu T R, Marshall K C (1991) Microbial ‘footprints’– a new
cation binding by algal and bacterial alginates. Biotech Appl approach to adhesive polymers. Biofouling 3: 101–112
Biochem 20: 117 –129 Neu T R, Poralla K (1990) Emulsifying agents from bacteria
Gilbert P, Das J, Foley I (1997) Biofilm susceptibility to isolated during screening for cells with hydrophobic surfaces.
antimicrobials. Adv Dent Res 11: 160–167 Appl Microbiol Biotechnol 32: 521–525
Gilbert P, Das J R, Jones M, Allison D G (2001) Assessment of the Neu T R, Swerhone G D W, Lawrence J R (2001) Assessment of
biocide activity upon various bacteria following their lectin-binding analysis for in situ detection of glycoconjugates
attachment to and growth on surfaces. J Appl Microbiol 91: in biofilm systems. Microbiology 147: 299– 313
248– 255 Nichols W W (1993) Biofilm permeability to antibacterial agents.
Gilbert P, Maira-Litran T, McBain A J, Rickard A H, Whyte F (2002) In: Wimpenny J, Nichols W, Stickler D, Lappin-Scott H (eds)
The physiology and collective recalcitrance of microbial Bacterial Biofilms and their Control in Medicine and Industry.
biofilm communities. Adv Microb Physiol 46: 203–255 Bioline Press, Cardiff, pp 141 –149
150 D G ALLISON

Nichols W W, Dorrington S M, Slack M P E, Walmsley H L (1988) Sutherland I W (1985) Biosynthesis and composition of Gram-
Inhibition of tobramycin diffusion by binding to alginate. negative bacterial extracellular and wall polysaccharides.
Antimicrob Agents Chemother 32: 518–523 Ann Rev Microbiol 10: 243–270
Nichols W W, Evans M J, Slack M P E, Walmsley H L (1989) The Sutherland I W (1990) Biotechnology of Microbial Exopolysaccharides.
penetration of antibiotics into aggregates of mucoid and Cambridge University Press, Cambridge, 163 pp
non-mucoid Pseudomonas aeruginosa. J Gen Microbiol 135: Sutherland I W (1995) Biofilm specific polysaccharides–do they
1291–1303 exist? In: Wimpenny J, Handley P, Gilbert P, Lappin-Scott H
Nivens D E, Palmer R J Jr, White D C (1995) Continuous non- (eds) The Life and Death of Biofilm. Bioline Press, Cardiff,
destructive monitoring of microbial biofilms: a review of pp 103–106
analytical techniques. J Indust Microbiol 15: 263–276 Sutherland I W (1997) Microbial biofilm exopolysaccharides–
Ophir T, Gutnick D L (1994) A role for exopolysaccharides in the superglues or velcro? In: Wimpenny J, Handley P, Gilbert P,
protection of microorganisms from desiccation. Appl Env Lappin-Scott H, Jones M (eds) Biofilms: Community Interactions
Microbiol 147: 195 –202 and Control. Bioline Press, Cardiff, pp 33– 39
Passador L, Cook J M, Gambello M J, Rust L, Iglewski B H (1993) Sutherland I W (1999a) Biofilm exopolysaccharides. In: Wingender
Expression of Pseudomonas aeruginosa virulence genes requires J, Neu T R, Flemming H-C (eds) Microbial Extracellular
cell-to-cell communication. Science 260: 1127 –1130 Polymeric Substances. Springer, Berlin, pp 73– 92
Salmond G P C, Bycroft B W, Stewart G S A B, Williams P (1995) Sutherland I W (1999b) Polysaccharases in biofilms–source–
The bacterial “enigma” cracking the code of cell – cell action–consequences!. In: Wingender J, Neu T R, Flemming
communication. Mol Microbiol 16: 615–624 H-C (eds) Microbial Extracellular Polymeric Substances.
Sand W, Gehrke T (1999) Analysis and function of the EPS from the Springer, Berlin, pp 201 –230
strong acidophile Thiobacillus ferrooxidans. In: Wingender J, Sutherland I W (2001) The biofilm matrix– an immobilized but
Neu T R, Flemming H-C (eds) Bacterial Extracellular Polymeric dynamic environment. Trends Microbiol 9: 222–227
Substances. Springer, Berlin, pp 127 –141 Swift S, Bainton N J, Winson K (1994) Gram-negative bacterial
Schmitt J, Flemming H-C (1999) Water binding in biofilms. Water communication by N-acyl homoserine lactones: a universal
Sci Technol 39: 77–82 language? TIMS 2: 193–198
Schmitt J, Nivens D, White D C, Flemming H-C (1995) Changes of Tait M I, Sutherland I W, Clarke-Sturman A J (1986) Effect of
biofilm properties in response to sorbed substances– an FTIR- growth conditions on the production, composition and
ATR study. Water Sci Technol 32: 149 –155 viscosity of Xanthomonas campestris exopolysaccharide. J Gen
Skillman L C, Sutherland I W, Jones M V (1997) Co-operative Microbiol 132: 1483–1492
biofilm formation between two species of Enterobacteriaceae. Whitfield C (1988) Bacterial extracellular polysaccharides. Can J
In: Wimpenny J, Handley P, Gilbert P, Lappin-Scott H, Jones M Microbiol 34: 415–420
(eds) Biofilms:Community Interactions and Control. Bioline Willcock L, Holah J, Allison D G, Gilbert P (1997) Population
Press, Cardiff, pp 119–127 dynamics in steady-state biofilms: effects of growth environ-
Skillman L C, Sutherland I W, Jones M V, Goulsbra A (1998) Green ment upon dispersal. In: Wimpenny J, Handley P, Gilbert P,
fluorescent protein as a novel species-specific marker in Lappin-Scott H, Jones M (eds) Biofilms: Community Interactions
enteric dual-species biofilms. Microbiology 144: 2095–2101 and Control. Bioline Press, Cardiff, pp 23– 31
Slack M P E, Nichols W W (1981) The penetration of antibiotics Willcock L, Gilbert P, Holah J, Wirtanen G, Allison D G (2000) A
through sodium alginate and through the exopolysaccharide new technique for the performance evaluation of clean-in-
of a mucoid strain of Pseudomonas aeruginosa. Lancet 11: place disinfection of biofilms. J Ind Microbiol 25: 235–241
502 –503 Williams P, Bainton N J, Swift S, Chhabra S R, Winson M K, Stewart
Sondossi M, Rossmore H W, Wireman J W (1985) Observation of G S A B, Salmond G P C, Bycroft B W (1992) Small molecule-
resistance and cross-resistance to formaldehyde and a mediated density-dependent control of gene expression in
formaldehyde condensate biocide in Pseudomonas aeruginosa. prokaryotes: bioluminescence and the biosynthesis of
Int Biodeterior Biodegr 21: 105– 106 carbapenem antibiotics. FEMS Microbiol Lett 100: 161–168
Stewart P S (1996) Theoretical aspects of antibiotic diffusion Wimpenny J W T (2000) An overview of biofilms as functional
into microbial biofilms. Antimicrob Agents Chemother 40: communities. In: Allison D G, Gilbert P, Lappin-Scott H,
2517–2522 Wilson M (eds) Community Structure and Co-operation in
Stoodley P, Boyle J D, DeBeer D, Lappin-Scott H M (1999a) Biofilms. SGM Volume 59, Cambridge University Press,
Evolving perspectives of biofilm structure. Biofouling 14: Cambridge, pp 1–24
75– 90 Wimpenny J W T, Colsanti R (1997) A unifying hypothesis for the
Stoodley P, Dodds I, Boyle J D, Lappin-Scott H (1999b) Influence of structure of microbial biofilms. FEMS Microbiol Ecol 22: 1 –16
hydrodynamics and nutrients on biofilm structure. J Appl Wingender J, Neu T R, Flemming H-C (1999a) What are bacterial
Microbiol 85 (Suppl): 19S–28S extracellular polymeric substances? In: Wingender J, Neu T R,
Stoodley P, Jørgensen F, Williams P, Lappin-Scott H (1999c) The Flemming H-C (eds) Microbial Extracellular Polymeric Sub-
role of hydrodynamics and AHL signalling molecules as stances. Springer, Berlin, pp 1–19
determinants of the structure of Pseudomonas aeruginosa Wingender J, Jaeger K-E, Flemming H-C (1999b) Interaction
biofilms. In: Wimpenny J, Gilbert P, Walker J, Brading M, between extracellular polysaccharides and enzymes. In:
Bayston R (eds) Biofilms–The Good, the Bad and the Ugly. Bioline Wingender J, Neu T R, Flemming H-C (eds) Microbial
Press, Cardiff, pp 223 –230 Extracellular Polymeric Substances. Springer, Berlin,
Suci P A, Mittleman M W, Yu F U, Geesey G G (1994) Investigation pp 231– 251
of ciprofloxacin penetration into Pseudomonas aeruginosa Zobell C E, Allen E C (1935) The significance of marine bacteria in
biofilms. Antimicrob Agents Chemother 38: 2125–2133 the fouling of submerged surfaces. J Bacteriol 29: 230–251