B I O F I L M SUSCEPTIBILITY TO ANTIMICROBIALS

P . GI L B E R T
J. DA S
1
I . FOLEY
Department of Pharmacy
University of Manchester
Oxford Road
Manchester M13 9PL, England
1
Present address:
Unilever Research
Port Sunlight Laboratories
Quarry Road East
Bebington, Wirral L63 3JW, England
Adv Dent Res U(l):160-167, April, 1997
AbstractM icrobial biofilms, where organisms are
intimately associated with each other and a solid substratum
through binding and inclusion within an exopolymer matrix,
are widely distributed in nature and disease. In the mouth,
multispecies biofilms are associated not only with dental
plaque and tooth decay but also with soft tissues of the
buccal cavity and with most forms of periodontal disease.
Organization of micro-organisms within biofilms confers, on
the component species, properties which are not evident with
the individual species grown independently or as planktonic
populations in liquid media. While many of these properties
relate to the establishment of functional, mixed-species
consortia within the exopolymeric matrices, others relate to
the establishment of physico-chemical gradients, within the
biofilm, that modify the metabolism of the component cells.
A consequence of biofilm growth that has profound
implications for their control in the environment and in
medicine is a markedly enhanced resistance to chemical
antimicrobial agents and antibiotics. Mechanisms associated
with such resistance in biofilms will form the substance of
the present review. While some aspects of biofilm resistance
are yet only poorly understood, the dominant mechanisms are
thought to be related to: (i) modified nutrient environments
and suppression of growth rate within the biofilm; (ii) direct
interactions between the exopolymer matrices, and their
constituents, and antimicrobials, affecting diffusion and
availability; and (iii) the development of biofilm/attachment-
specific phenotypes.
Key words: B iofilm, antimicrobials, growth rate,
glycocalyx, attachment to surfaces.
Presented at the 14th International Conference on Oral
Biology, "Biofilms on Oral Surfaces: Implications for Health
and Disease", held March 18-20, 1996, in Monterey,
California, organized by the International Association for
Dental Research and supported by Unilever Dental Research
T
he development and application, post-1940, of a
wide variety of antibiotics were generally heralded
as the advent of an 'A ntibiotic E ra' . C ommon
infectious diseases, such as tuberculosis and
pneumonia, became treatable and were essentially eradicated
from the developed world. Since then, major developments
of chemical antimicrobial agents such as bisbiguanides,
isothiazolones, and peroxygens, for use in antisepsis,
disinfection, and preservation, have also been made. In recent
years, however, we have been forced to re-evaluate such
antimicrobial strategies. Widespread, indiscriminate use of
antibiotics has led to the development and emergence of
antibiotic-resistant strains. Similarly, the widespread use, and
dissemination within the environment, of chemical
antimicrobials is leading to reductions in their effectiveness.
This is coupled to increasing demand being placed by man
on the need to control the presence and metabolism of
bacteria in an ever-widening sphere of applications.
T hus, the development and use of a broad range of
medical devices have led to the emergence and recognition of
a variety of infections caused by organisms that were
regarded previously as 'harmless'. In this respect, infections
relate not only to biofilms associated with the surfaces of
implanted medical devices such as prostheses, endocardial
pacemakers, and catheters (Marrie and C osterton, 1983;
Holmes and Evans, 1986), but also to our failure to disinfect
adequately the organisms attached to medical equipment
such as fiber-optic endoscopes. Biofilm infections, associated
with indwelling medical devices, are often chronic and act as
sources for bacteremia. While the latter respond readily to
antibiotic treatment dictated by the results of conventional
sensitivity testing (Thomson et al., 1995), the biofilms from
which they derive display a greatly enhanced resistance and
often fail to respond to even the most aggressive antibiotic
prescribing (Kunim and S teel, 1985; Nickel et al., 1985;
Gristina et al., 1987; Costerton et al, 1993). If the device is
not surgically removed prior to antibiotic treatment, then the
infection will generally recur. Resistance of biofilms is not
restricted to antibiotics but is also shown with respect to a
wide range of chemical biocides. T hese include
isothiazolones (C osterton and L ashen, 1984), quaternary
ammonium compounds (Costerton and Lashen, 1984; Evans
et al., 1990b), and halogens and halogen-release agents
(Favero et al., 1983). Failure of available antimicrobials to
contend adequately with microbial biofilms, together with an
increasing dependence of modern medicine upon the
implantation of devices, has stimulated the search for
antimicrobials which have activity directed primarily toward
the biofilm phenotype (Gilbert and B rown, 1995). T his
process includes not only the development of our knowledge
of biofilm physiology, in the search for novel antimicrobial
targets, but also an examination of the various mechanisms
associated with resistance of biofilms toward current
antibiotic agents.
160
VOL.1 1(1)
BIOFILM SUSCEPTIBILITY TO ANTIMICROBIALS
161
Failure of micro-organisms to succumb to antimicrobial
treatment may arise through: (i) an inherent insusceptibility
to the agents used; (ii) the acquisition of resistance, by
previously susceptible strains, either by genetic mutation or
by transfer of genetic material from another species or genus;
and (iii) the emergence of pre-existing but unexpressed
resistance phenotypes. While it is unequivocal that biofilms
resist conventional treatments, the extent and mechanisms
through which adaptation toward a less-susceptible
phenotype is influenced by growth as a biofilm, on soft tissue
or hard surfaces, will remain a matter for debate until
common in vitro susceptibility-testing methodologies are
adopted that adequately replicate the in vivo complexities
(Anwar and Costerton, 1990; Gilbert and Brown, 1995).
T he present article considers those mechanisms of
resistance that are considered likely to be associated with
attachment of micro-organisms to surfaces and growth into
microcolonies/communities entrapped within extracellular
polymers. We will consider (i) the extent to which resistance
is a reflection of the nutrient environment generated within
the biofilm, (ii) direct modification of antibiotic action
through the presence of extracellular polymers and antibiotic-
modifying enzymes, and (iii) the development of
attachment/biofilm-specific phenotypes.
ANTIBIOTIC RESISTANCE, NUTRIENT
ENVIRONMENT, AND BIOFILMS
Often critical to the long-term survival of micro-organisms is
their ability to attach to surfaces and form adherent biofilms.
Biofilms are functional consortia of microbial cells enveloped
within sometimes-extensive matrices of extracellular
polymers (glycocalyx) and concentrated products of their
own metabolism, together with ions and nutrients sequestered
from the environment. Soft-tissue infection or infections
associated with indwelling medical devices are often the
result of the growth of mono-species biofilms. In the majority
of situations in the human body (i.e., gastro-intestinal tract
and oral cavity), and in the general environment (i.e.,
freshwater and marine ecosystems), biofilm consortia are
composed of a variety of species and genera.
The structural organization of the glycocalyx, which forms
the intercellular matrix, varies according to the prevailing
physico-chemical environment. T hus, in situations of high
shear (i.e., tooth surface during mastication, gastro-intestinal
tract, etc.), the biofilm population is organized within
stratified compacts of exopolymeric material (i.e., dental
plaque; Newman and Barber, 1995). Under low to moderate
shear, with nutrients accessed from the bathing fluids, the
biofilms then appear as attached floccules. These anchor the
micro-colony to the substratum yet maximize diffusive
interactions with a nutrient-bearing environment (Costerton
et al, 1994a). If nutrients are derived from the substratum,
rather than the bathing fluids, then diffuse layers of biofilm
cells which completely coat the available surface are favored,
such as in the colonization of the nasopharynx. With the
exception of those cells that are located at the periphery of
the biofilm, access and availability of nutrients and the
elimination of metabolic by-products are restricted to a
greater extent than it would be for the same cells growing
individually in liquid culture. T hus, cells deep within the
biofilm matrix have available to them only those materials
from the bathing fluids that have failed to be sequestered by
more outlying cells. Conversely, these micro-organisms have
greater access to the secreted metabolic products of the
neighboring cells. This may lead to spatial organization of
species within mixed-species biofilm communities, with
associated cross-feeding, and the development of a functional
inter-species dependence. In both mono- and multi-species
biofilms, nutrient and gaseous gradients, generated by
metabolism, will cause nutrient availability and growth rates
of the enveloped cells to vary with location relative to the
substratum and biofilm/liquid phase interface.
Nutrient limitation and growth rate
The plasticity in structure and physiology of bacterial cells
allows them to make rapid phenotypic responses not only to
changes in their nutrient status and growth rate, but also to
changes in temperature and pH, and following exposure to
subeffective concentrations of antibiotics (B rown and
Williams, 1985; Williams, 1988; Brown et al., 1990; Gilbert
et al., 1990). Such responses may include changes in a wide
variety of cellular componentsincluding proteins, fatty
acids, and phospholipids associated with the cell envelope
and production of extracellular enzymes and polysaccharides
(B rown et al., 1990; Gilbert et al., 1990). S ince all
antimicrobial substances must interact with the cell envelope
either as the primary target or as a means of accessing this
target, then such changes in phenotype inevitably affect
susceptibility toward a wide range of antibiotics,
disinfectants, antiseptics, and preservatives (Brown et al.,
1990; Gilbert etal, 1990).
In well-mixed suspension cultures, all members of the
community experience a common environment at any
particular time. I n batch culture, these environmental
conditions, and hence the phenotype, change with time, while
in open environments, such as chemostat culture, steady-state
conditions prevail. Single phenotypes dominate such cultures
which, as a consequence, tend to demonstrate singular
responses toward antimicrobial treatments. At any given time
within biofilm communities, however, a plethora of
phenotypes is represented for each component species. The
breadth of phenotypes represented reflects the extent of the
chemical heterogeneity within the biofilm and the presence of
various concentration gradients. T hus, the outcome of any
attempt to eliminate a biofilm community by antimicrobial
treatment will often reflect only the susceptibility of the most
resistant phenotype represented.
A distinction can be made between those effects related to
the nature of the least available nutrient (nutrient
limitation/depletion) and the cellular growth rate. Within the
depths of a biofilm, growth rates will generally be suppressed
relative to planktonic cells growing in the same environment.
In this respect, Ashby et al. (1994) used biofilmrplanktonic
ratios of isoeffective concentration (growth inhibition and
bactericidal activity), determined for a wide range of
162
GILBERT E T A L .
ADV DENT RES APRIL 1997
TABLE 1
ACTIVITY INDICES FOR VARIOUS ANTIBIOTICS
AGAINST Escherichia coli (from Ashby et al, 1994)
Antibiotic Minimum Effective
Concentration Activity
Ratio*, Non-growing/
Growing Cells
Cephamycins
cefminox
cefoxitin
cefotetan
cefmetazole
Cephalosporins
cefotaxime
ceftazidime
cefoperazone
cefpirome
Carbapenems
imipenem
meropenem
Miscellaneous
gentamicin
ciprofloxacin
1.0
2.0
32
1.0
16
16
8.0
>32
0.5
8.3
1.0
33.3
Sessile Planktonic
Index Activity Ratio#,
B iofilm/Planktonic
Cells
1.8
1.9
2.5
1.5
5.4
3.2
1.7
3.2
1.2
1.6
1.1
1.2
MIC (mg/L) vs. growing cells/concentration causing 50%
reduction in O D
600
nm for non-growing cells.
Ratio of concentrations causing a 50% inhibition of the
incorporation of [3H]-leucine in biofilm and planktonic
populations, respectively.
antibiotics against cells grown either in broth or on urinary
catheter discs, to indicate the extent of biofilm resistance.
They noted that such ratios (Table 1) closely followed those
generated between non-growing and actively growing
cultures. With the exception of ciprofloxacin, antibiotic
agents that were most effective against non-growing cultures
{i.e., imipenem, meropenem) were also the most active
against these biofilms. Other workers have used perfused
biofilm fermenters (Gilbert et al, 1989) directly to control
and study the effects of growth rate within biofilms.
Planktonic controls grown in chemostats can be used for the
evaluation of the separate contributions of growth rate and
association within a biofilm. Decreased susceptibility of
Staphylococcus epidermidis to tobramycin (Duguid et al,
1992a) and of Escherichia coli to tobramycin (Evans et al.,
1990a) and cetrimide (E vans et al., 1990b) could be
explained largely in terms of growth rate. Cells re-suspended
from growth-rate-controlled biofilms and planktonic cells of
the same growth rate possessed virtually identical
susceptibilities to these agents. When intact biofilms were
treated, however, susceptibility was decreased somewhat
from that of planktonic and re-suspended biofilm cells,
indicating some benefit to the cells of organization within a
glycocalyx (see below).
Stewart (1994) developed a mathematical model which
incorporated the concepts of metabolism-driven oxygen
gradients and growth-rate-dependent killing to examine the
susceptibility of S. epidermidis biofilms to various
antibiotics. The model accurately predicted that susceptibility
would be reduced in thicker biofilms due to oxygen
limitation. Oxygen gradients within the biofilm may also
directly influence the activity of some antibacterials
(Shepherd et al., 1988; Zabinski et al., 1995). Since nutrient
and gaseous gradients will increase within maturing biofilms,
then growth-rate effects on susceptibility, such as these, will
become particularly marked in aged biofilms (Anwar et al.,
1989, 1990) and might also lead to an onset of dormancy and
the triggering of stringent-response genes (Zambrano and
Kolter, 1995). Such changes probably contribute to reports
that aged biofilms are more recalcitrant to antibiotic and
biocide treatment than are younger ones (Anwar et al., 1989).
The presence of sub-inhibitory levels of antibiotic agents
within the depths of the biofilm will provide selective
pressures for the development of more resistant phenotypes
and for the selection and expression of resistance plasmids.
S ub-inhibitory concentrations may be either generated
through the failure of antibiotics to penetrate the glycocalyx
adequately or caused through decreases in the susceptibility
of the enveloped cells.
T his view of biofilm resistance being related
predominantly to the low growth rates within them (Prosser
et al, 1987; Brown et al, 1988; Gilbert et al, 1990) does not
offer much hope to those searching for novel anti-
biofilm/anti-plaque agents (Gilbert and Brown, 1995). In the
study by Ashby et al (1994), described above and presented
in Table 1, ciprofloxacin, an agent that does not distinguish
itself particularly against non-growing cells, nevertheless has
good anti-biofilm activity. Similarly, experiments with this
quinolone, utilizing the perfused biofilm fermenter (Gilbert et
al., 1989), suggested that while growth rate played a crucial
role in the ciprofloxacin susceptibility of S. epidermidis and
E. coli, slow-growing (u < 0.15 h
1
) biofilms were especially
sensitive (Evans et al, 1991; Duguid et al, 1992b). These
observations suggest that there might be some physiological
properties, with potential to act as antimicrobial targets, that
are unique to biofilm-grown cells.
MATRIX POLYMERS, GLYCOCALYX, AND
EXTRACELLULAR ENZYMES
Electron microscopic examinations of antibody-stabilized
biofilm preparations reveal ordered arrays of fine fibers that
provide a relatively thick, hydrated, polyanionic matrix
around the cells (Glycocalyx; Costerton et al, 1981). While
the exopolymers that compose the glycocalyx are
predominantly highly hydrated, gelled polysaccharides, other
polymers, particularly globular glycoproteins (Sutherland,
1985), are also represented. Whether the ' footprint
polymers' , which cement the primary colonizers to the
substratum, differ from the matrix polymers, which bind cells
to other cells, and whether these differ from the exopolymers
VOL.U(l)
BIOFILM SUSCEPTIBILITY TO ANTIMICROBIALS
163
found associated with planktonically grown cells is unknown
(Sutherland, 1995). Nevertheless, in mixed-species biofilms,
each component species will produce a different set of
polymers, and these will merge to give heterogenous regions
of polymers within the matrix (C ooksey, 1992). T he
physicochemical properties of the blended exopolymers will
differ significantly from those of the purified components and
will also be substantially affected by the ionic strength of the
surrounding medium and the nature of the cationic species
(Allison and Matthews, 1992).
Regulation of exopolymer synthesis
The synthesis of matrix polymers appears to be regulated by
a variety of factors, of which surface attachment appears to
be of particular importance. T hus, Davies et al. (1993)
showed EP S (alginate) production to be de-repressed in
biofilm cells compared with planktonics, and Evans et al.
(1994) showed exopolysaccharide production to be increased
at low growth rates and substantially higher for biofilm than
planktonic cells. The latter effect would provide for increased
exopolymer production within the slow-growing heart of a
thick micro-colony/biofilm. This would alter the distribution
and density of cells throughout the matrix, and once again
confer some structural organization upon the community to
provide customized micro-niches at various points within the
biofilm (Costerton et al, 1994a). Recently, it has also been
suggested that in some Gram-negative organisms the
production of exopolysaccharides, such as alginate, may be
under the control of signal substances such as homoserine
lactone (HSL). These are global regulators of transcriptional
activation in bacteria (Williams et al, 1992; Gambello et al,
1993; Cooper et al., 1995), responsible for cell-cell signaling,
and implicated in cell-density-mediated events. In biofilms,
signal substances such as HSL would become concentrated
within the geometric center of the micro-colonies/biofilm,
where cell physiology would be altered accordingly. The
extent and nature of exopolymer production are also
dependent upon physiological factors such as the relative
availability of carbon and nitrogen (Sutherland, 1985).
Exopolymers and antimicrobial susceptibility
The exopolymers serve two major functions within biofilm
communities. First, while not adhesins in their own right,
exopolysaccharides are overproduced following the initial
attachment of cells to surfaces. As such, they have been
suggested to act as cements, which may reinforce the primary
adhesion mechanisms (A llison and S utherland, 1987).
Second, the glycocalyx protects the cells contained within it
against desiccation and against predatory phagocytic entities,
such as white blood cells and protozoa, and may restrict the
diffusion of agents from the surrounding medium through a
combination of ionic-interaction and molecular-sieving
events. Thus, in analogy to the penetration of agents through
peptidoglycan (M arquis, 1968), the polymers of the
glycocalyx may act as an ion-exchange resin where strongly
charged molecules are actively removed from solution as
they pass through. I ncoming molecules would have to
saturate all available binding sites, just as gentamicin must
saturate all of the binding sites on the polysaccharide fibers
before it can pass through a cellulose filter (Wagman et al.,
1975). While it has been suggested that the glycocalyx
reduces access of antibiotics to the biofilm population in this
manner (Gordon et al., 1988; N ichols et al., 1989), such
effects are most pronounced for the activity of chemically,
highly reactive biocides such as iodine and
iodinepolyvinylpyrollidone complexes (Favero et al., 1983).
A ccess of such agents is substantially reduced by the
presence of exopolymers which, in addition to acting as
adsorption sites, will react chemically with, and neutralize,
biocides.
The presence of high cell numbers within an exopolymer
matrix will undoubtedly profoundly influence their access to
molecules and ions, including protons (C osterton et al.,
1981). It is not surprising, therefore, that many groups of
workers have suggested that the glycocalyx physically
prevents access of antimicrobials to the cell surface, and that
the recalcitrance of biofilms is purely and simply a matter of
exclusion (Costerton et al., 1987; Slack and Nichols, 1981;
Suci et al., 1994). Such universal explanations have been
refuted (Gordon et al., 1988; Nichols et al, 1988, 1989),
since reductions in the diffusion coefficients of antibiotics
such as tobramycin and cefsulodin, within biofilms or
microcolonies, are insufficient to account for the observed
change in susceptibility. In this light, Gristina et al (1989)
compared the susceptibility of biofilms formed by slime-
producing and non-slime-producing strains of S. epidermidis.
Lack of any change suggested that the slime was not of great
significance in antibiotic penetration. While a variety of
antibiotic agents has been shown readily to perfuse biofilms
(Dunne et al, 1993) and to attain concentrations within the
matrix that exceed the MIC/MB C observed for planktonic
organisms (Darouchie et al, 1994), there are also reports that
sub-inhibitory concentrations of 6-lactams are selective of
mucoid phenotypes (Govan, 1976). A solution to these
apparent contradictions might be a reinforcement of the
barrier properties of the glycocalyx through a local
concentration, within the glycocalyx, of drug-inactivating
enzymes such as beta-lactamases (Giwercman et al, 1991).
T his will create marked concentration gradients of the
antibiotic across them and protect, to some extent, the
underlying cells.
C learly, whether or not the glycocalyx constitutes a
physical barrier to antibiotic penetration depends greatly
upon the nature of the antibiotic, the binding capacity of the
glycocalyx toward it, the levels of agent used therapeutically,
and the rate of growth of the microcolony relative to the
antibiotic diffusion rate (Kumon et al, 1994). For antibiotics
such as tobramycin and cefsoludin, therefore, such effects are
suggested to be minimal (Nichols et al, 1988, 1989), but for
positively charged antibiotics such as the aminoglycosides,
binding to the polyanionic matrix polymers is high and then-
access to cells impeded (Nichols et al, 1988). Curiously,
macrolide antibiotics, which are also positively charged but
also very hydrophobic, are relatively unaffected by the
presence of the exopolymers (Ichimiya et al, 1994). Poor
penetration through anionic matrices might therefore be a
164
GILBERT E T A L . ADV DENT RES APRIL 1997
TABLE 2
RATIOS OF THE MINIMUM INHIBITORY
CONCENTRATIONS (MIC)
OF ANTIMICROBIAL AGENTS TOWARD BIOFILM
AND PLANKTONIC POPULATIONS
OF Staphylococcus epidermidis AND Escherichia coli
AT VARIOUS TIMES AFTER INOCULATION OF THE
PLANKTONIC PHASE (from Das et al, 1995)
Micro-
organism
S. epidermidis
E. coli
Time
(hrs)
1
0
6
14
20
0
6
14
20
Biofilm:Planktonic MI C
2
Phenoxy-
ethanol
1.00
1.96
1.92
1.96
1.00
2.04
2.00
2.04
Cetrimide PHMB
3
1.14
1.14
1.14
1.47
0.94
0.94
0.94
0.94
2.86
8.33
7.14
10.00
2.03
5.26
6.25
6.25
Ratios
Chloro-
xylenol
1.00
1.00
0.97
1.00
1.00
1.00
1.00
1.00
r
Time between inoculation of the planktonic phase and
addition of biocide.
2
A ratio of greater than 1 indicates a higher MIC toward
attached organisms and biofilms than toward planktonic
cells.
3
PHMB, polyhexamethylene biguanide (Vantocil).
phenomenon restricted to the more hydrophilic, positively
charged agents.
Modification of exopolymer properties
The presence of adsorbed ions within the biofilm matrix
polymers will affect its net charge and thereby its antibiotic-
exclusion properties. In this respect, Hoyle et al. (1992) have
shown the tobramycin-binding capacity of bacterial
exopolysaccharides to be less important, in terms of reduced
susceptibility, than is the reduction in diffusivity of the
matrix imposed by Ca
2+
condensation of the polymer. Such
effects are dependent upon the nature of the adsorbed species
and are not seen, for example, with Mg
2+
. This might have
profound effects upon the susceptibility of biofilms growing
at different sites within the body or in patients with some
predisposing clinical conditions (i.e., cystic fibrosis, chronic
renal failure, hypercalcemia) which create abnormal body
chemistries which include markedly elevated calcium ion
levels.
Novel treatment strategies
A potential application of these observations might be the use
of adjuvants to alter the charge characteristics of the polymer
matrix (Lee et al., 1995), or the use of novel agents with non-
lethal effects on exopolymer synthesis in combination with
conventional antibiotics (Gagnon et al., 1994; Pascual et al.,
1994).
Observation of the effects of charge on the functionality of
the biofilm matrix in excluding antibiotic agents has led to
various investigations of possible synergy between
bioelectric treatments and antibiotics or biocides. Significant
enhancements in killing action have been reported when low,
direct-current fields (12 V/cm), at a low current density
( 2.1 mA /cm
2
), were applied to biofilms together with
various biocides. In this instance, several of these biocides
were bactericidal against biofilms at concentrations that were
less than the planktonic MI C (B lenkinsopp et al., 1992;
C osterton et al., 1994b). S imilar results have also been
reported for the activity of antibiotics in combination with the
bioelectric effect (Costerton et al., 1994b). Modifications of
implanted devices to facilitate bioelectric treatment are now
in the developmental stages. Similar approaches to reducing
the recalcitrance of biofilms through modification of the
polymer matrix have involved the use of 67-kHz ultrasound
(Pitt et al., 1994) to produce synergy with gentamicin.
ANTIMICROBIAL SUSCEPTIBILITY AND
ATTACHMENT-SPECIFIC PHYSIOLOGY
The possibility remains that bacteria are able to sense the
presence of a surface to which they become attached, and, as
a consequence, transcriptionally activate genes/operons to
confer an attachment-specific phenotype which has a
modified susceptibility toward antimicrobials. Variations of
the 'bottle effect', whereby the metabolic activity of micro-
organisms is stimulated by their attachment to surfaces, have
been reported in the literature since the early 1940s (Zobell,
1943; Fletcher, 1984, 1986). Only recently, however, have
attempts been made to define a genetic and physiological
basis for such phenomena. Dagostino et al. (1991) utilized
transposon mutagenesis to insert randomly into the
chromosome of E. coli a marker gene which lacked its own
promoter element. They then went on to isolate mutant cell
lines which expressed the gene when attached to a
polystyrene surface but not when grown on agar or in liquid
media. T he isolation of mutant cells such as these, with
reporter genes that respond to attachment onto surfaces, has
not diminished the level of debate as to the cause of surface-
induced metabolic stimulation. This may reflect de-repression
or induction of specific operons/genes, or it may be a
physico-chemical manifestation (i.e., localized concentration
of nutrients, viscosity changes, pH effects, etc.) of the
proximity of the surface (Van Loosdrecht et al, 1990).
E vidence in favor of physico-cemical effects includes
work on the regulation of lateral flagella gene transcription in
Vibrio parahaemolyticus. This organism has been shown to
produce a single polar flagellum in liquid, and numerous
lateral, unsheathed flagella on solid culture media (Belas et
al, 1984, 1986; M cC arter et al, 1988). C hanges in
flagellation, in this instance, reflect an increased viscosity at
the surface which restricts the movement of the polar
flagellum and switches the laf genes. In a similar vein, Lee
VOL.ll(l)
BIOFILM SUSCEPTIBILITY TO ANTIMICROBIALS
and Falkow (1990) recognized that reduced oxygen tension,
as experienced by cells enveloped within a biofilm or in
association with a surface, causes the triggering of expression
of Salmonella invasins.
Other examples of surface-induced behavior defy such
physico-chemical explanation. Thus, nitrilotriacetate does not
adsorb to surfaces, but its breakdown is enhanced when the
degradative organisms are attached to inert surfaces
(McFeters et al., 1990). This suggests increased expression of
the degradative enzymes by attached cells. In a similar
fashion, gliding bacteria do not synthesize extracellular
polymers when they are grown in suspension culture
(Humphrey et al, 1979; Abbanat et al, 1988), but do so
rapidly after they become irreversibly bound to a surface.
On balance, it now seems probable that bacteria can sense
the proximity of surfaces and, through cell density
transcriptional activation (Cooper et al, 1995), the proximity
of other cells. Das et al. (1995) developed spectrophotometric
methods which allowed for simultaneous monitoring of the
growth of planktonic and biofilm bacteria within the wells of
a microtiter plate. B acterial cultures were exposed to a
variety of agents at various times (from 0 to 20 hrs) following
inoculation of the wells and the initiation of biofilm
formation (T able 2). M inimum growth inhibitory
concentrations were determined against the planktonic and
attached subpopulations and expressed as activity ratios. The
activities of both cetrimide and polyhexamethylene biguanide
were affected immediately following bacterial attachment. In
this respect, some bacterial cells were able to attach to the
polystyrene surface and grow as a biofilm, while growth of
the suspensions was inhibited (biofilm:planktonic MIC ratio
> 1). Since there had been insufficient time for the formation
of biofilms, then such data relate either to a decreased
disposition of these agents at the colonized surface or to the
expression of an attachment-specific phenotype. If biofilms
were allowed to form for 6-20 hrs before exposure to biocide,
then the MI C 's were further reduced. Again, these incubation
times are insufficient for an extensive biofilm to develop and
might indicate the protective effects of microcolony
formation. While it is likely that similar physiological
responses, invoked as attachment-specific phenotypes, will
affect antibiotic susceptibility, these have not been
demonstrated.
ACKNOWLEDGMENT
T his work was supported in part by a B B S R C (U K)
studentship to Ian Foley.
REFERENCES
Abbanat DR, Godchaux W, Leadbetter ER (1988). Surface-
induced synthesis of new sulphonolipids in the gliding
bacterium Cytophaga johnsonae. Arch Microbiol 149:358-
364.
Allison DG, Matthews MJ (1992). Effect of polysaccharide
interactions on antibiotic susceptibility of Pseudomonas
aeruginosa. J Appl Bacteriol 73:484-488.
A llison DG, S utherland I W (1987). T he role of
exopolysaccharides in adhesion of freshwater bacteria. /
Gen Microbiol 133:1319-1327.
A nwar H, C osterton JW (1990). E nhanced activity of
combination of tobramycin and piperacillin for eradication
of sessile biofilm cells of Pseudomonas aeruginosa.
Antimicrob Agents Chemother 34:1666-1671.
Anwar H, Dasgupta M, Lam K, Costerton JW (1989).
Tobramycin resistance of mucoid Pseudomonas
aeruginosa biofilm grown under iron limitation. /
Antimicrob Chemother 24:647-655.
Ashby MJ, Neale JE, Knott SJ, Critchley IA (1994). Effect of
antibiotics on non-growing cells and biofilms of
Escherichia coli. J Antimicrob Chemother 33:443-452.
B elas R, M ileham A , S imon M , S ilverman M (1984).
Transposon mutagenesis of marine vibrio spp. / Bacteriol
158:890-896.
Belas R, Simon M, Silverman M (1986). Regulation of lateral
flagellar gene transcription in Vibrio parahaemolyticus. J
Bacteriol 167:210-218.
B lenkinsopp S A , Khoury A E , C osterton JW (1992).
E lectrical enhancement of biocide efficacy against
Pseudomonas aeruginosa biofilms. Appl Environ
Microbiol 58:3770-3773.
B rown M R W, Williams P (1985). T he influence of
environment on envelope properties affecting survival of
bacteria in infections. Ann Rev Microbiol 39:527-556.
Brown MRW, Allison DG, Gilbert P (1988). Resistance of
bacterial biofilms to antibiotics: A growth rate related
effect. J Antimicrob Chemother 22:777-789.
B rown MR W, C ollier P J, Gilbert P (1990). I nfluence of
growth rate on the susceptibility to antimicrobial agents:
M odification of the cell envelope and batch and
continuous culture. Antimicrob Agents Chemother
34:1623-1628.
Cooksey KE (1992). Extracellular polymers in biofilms. In:
Biofilms: science and technology. Melo L F , Bott TR,
F letcher M, C apdeville B , editors. Dordrecht, T he
Netherlands: Kluwer, Academic Press, pp. 137-147.
Cooper M, Batchelor SM, Prosser JI (1995). Is cell density
signalling applicable to biofilms? In: The life and death of
biofilm. Wimpenny J, Handley P, Gilbert P, Lappin-Scott
H, editors. Cardiff: Bioline Press, pp. 93-97.
Costerton JW, Lashen ES (1984). Influence of biofilm on the
efficacy of biocides on corrosion-causing bacteria. Mat
Performance 23:34-37.
Costerton JW, I rvin R T , Cheng KJ (1981). The bacterial
glycocalyx in nature and disease. Ann Rev Microbiol
35:399-424.
Costerton JW, Cheng KJ, Geesey GG, Ladd TI, Nickel JC,
Dasgupta M, et al. (1987). Bacterial biofilms in nature and
disease. Ann Rev Microbiol 41:435-464.
C osterton JW, Khoury A E , Ward KH, Anwar H (1993).
P ractical measures to control device-related bacterial
infections. IntJ ArtifOrg 16:765-770.
C osterton JW, L ewandowski Z, DeB eer D, C aldwell D,
Kober D, James G (1994a). B iofilms, the customized
166
GILBERT E T A L . ADV DENT RES APRIL 1997
microniche. J Bacteriol 176:2137-2142.
C osterton JW, E llis B , L am K, Johnson F , Khoury AE
(1994b). Mechanism of electrical enhancement of efficacy
of antibiotics in killing biofilm bacteria. Antimicrob
Agents Chemother 38:2803-2809.
Dagostino L , Goodman A E , M arshall KC (1991).
Physiological responses induced in bacteria adhering to
surfaces. Biofouling 4:113-119.
Darouchie R O , Dhir A, Miller AJ, L andon GC , Raad I I ,
M uscher DM (1994). Vancomycin penetration into
biofilm covering infected prostheses and effect on
bacteria. J Infect Dis 170:720-723.
Das J, Jones M, Bhakoo M, Gilbert P (1995). An automated
method for the assessment of biofilm formation and
antimicrobial screening (abstract). 95th General Meeting
of the American Society for Microbiology, May 21-25,
1995. Washington, DC: ASM Press.
Davies DG, C hakrabarty A M , Geesey GG (1993).
E xopolysaccharide production in biofilms: S ubstratum
activation of alginate gene expression by Pseudomonas
aeruginosa. Appl Environ Microbiol 59:1181-1186.
Duguid IG, Evans E, Brown MRW, Gilbert P (1992a). Effect
of biofilm culture upon the susceptibility of
Staphylococcus epidermidis to tobramycin. / Antimicrob
Chemother 30:803-810.
Duguid I G, E vans E , B rown M R W, Gilbert P (1992b).
Growth-rate-independent killing by ciprofloxacin of
biofilm-derived Staphylococcus epidermidis: evidence for
cell-cycle dependency. / Antimicrob Chemother 30:791-
802.
Dunne WM Jr, Mason EO, Kaplan SL (1993). Diffusion of
rifampicin and vancomycin through a Staphylococcus
epidermidis biofilm. Antimicrob Agents Chemother
37:2522-2526.
E vans DJ, Brown MRW, Allison DG, Gilbert P (1990a).
Susceptibility of bacterial biofilms to tobramycin: role of
specific growth rate and phase in the division cycle. /
Antimicrob Chemother 25:585-591.
E vans DJ, Brown MRW, Allison DG, Gilbert P (1990b).
Growth rate and resistance of Gram-negative biofilms
towards Cetrimide USP. / Antimicrob Chemother 26:473-
478.
E vans DJ, B rown M R W, A llison DG, Gilbert P (1991).
S usceptibility of Escherichia coli and Pseudomonas
aeruginosa biofilms to ciprofloxacin: effect of specific
growth rate. / Antimicrob Chemother 27:177-184.
E vans E, B rown M R W, Gilbert P (1994). I ron chelator,
exopolysaccharide and protease production of
Staphylococcus epidermidis: a comparative study of the
effects of specific growth rate in biofilm and planktonic
culture. Microbiol 140:153-157.
F avero M S , B ond WW, P eterson N J, C ook EH (1983).
Scanning electron microscopic observations of bacteria
resistant to iodophor solutions. P roceedings of the
International Symposium on Povidone, April 17-29, 1983.
Lexington, KY: University of Kentucky.
Fletcher M (1984). Comparative physiology of attached and
free-living bacteria. I n: M icrobial adhesion and
aggregation. Marshall KC, editor. Berlin: Springer Verlag,
pp. 223-232.
Fletcher M (1986). Measurement of glucose utilisation by
Pseudomonas fluorescens that are free living and that are
attached to surfaces. Appl Environ Microbiol 52:672-676.
Gagnon R F , Richards GK, Kostiner GB (1994). Time-kill
efficacy of antibiotics in combination with rifampicin
against Staphylococcus epidermidis biofilms. Adv Perit
Dial 10:189-192.
Gambello M J, Kaye S, I nglewski BH (1993). LasR of
Pseudomonas aeruginosa is a transcriptional activator of
the line protease gene (apr) and an enhancer of exotoxin A
expression. Infect Immun 61:1180-1184.
Gilbert P , B rown M R W (1995). S creening for novel
antimicrobial activity/compounds in the pharmaceutical
industry. In: Microbial quality assurance: a guide towards
relevance and reproducibility of inocula. Brown MRW,
Gilbert P, editors. Boca Raton, FL: CRC Press, pp. 247-
260.
Gilbert P, Allison DG, Evans DJ, Handley P S , Brown MRW
(1989). Growth rate control of adherent bacterial
populations. Appl Environ Microbiol 55:1308-1311.
Gilbert P , C ollier P J, B rown MRW (1990). I nfluence of
growth rate on susceptibility to antimicrobial agents:
biofilms, cell cycle and dormancy. Antimicrob Agents
Chemother 34:1865-1868.
Giwercman B , Jensen ET, Hoiby N, Kharazmi A, Costerton
JW (1991). I nduction of B -lactamase production in
Pseudomonas aeruginosa biofilms. Antimicrob Agents
Chemother 35:1008-1010.
Gordon C A , Hodges N A , M arriot C (1988). A ntibiotic
interaction and diffusion through alginate and
exopolysaccharide of cystic fibrosis derived Pseudomonas
aeruginosa. J Antimicrob Chemother 22:661-61 A.
Govan JRW (1976). Antibiotic therapy and cystic fibrosis:
increased resistance of mucoid Pseudomonas aeruginosa
to carbenicillin. J Antimicrob Chemother 2:215-217.
Gristina AG, Hobgood CD, Webb LX, Myrvik QN (1987).
A dhesive colonisation of biomaterials and antibiotic
resistance. Biomaterials 8:423-426.
Gristina AG, Jennings RA, Naylor PT, Myrvik QN, Barth E,
Webb L X (1989). C omparative in-vitro antibiotic
resistance of surface colonising coagulase negative
staphylococci. Antimicrob Agents Chemother 33:813-824.
Holmes C F , E vans R (1986). B iofilm and foreign body
infectionthe significance to CAPD peritonitis. Periton
Dialysis Bull 6:168-177.
Hoyle B D, Wong C K, C osterton JW (1992). Disparate
efficacy of tobramycin on Ca(2+) Mg(2+)-, and HEPES-
treated Pseudomonas aeruginosa biofilms. Can J
Microbiol 38:1214-1218.
Humphrey B A , Dixon M R , M arshall KC (1979).
Physiological and in-situ observations on the adhesion of
gliding bacteria to surfaces. Arch Microbiol 120:231-238.
Ichimiya T, Yamasaki T, Nasu M (1994). In-vitro effects of
antimicrobial agents on Pseudomonas aeruginosa biofilm
formation. / Antimicrob Chemother 34:331-341.
Kumon H, Tomochika K-I, Matsunaga T, Ogawa M, Ohmori
VoUl(l) BIOFILM SUSCEPTIBILITY TO ANTIMICROBIALS
167
H (1994). A sandwich cup method for the penetration
assay of antimicrobial agents through Pseudomonas
exopolysaccharides. Microbiol Immunol 38:615-619.
Kunim CM, Steel C (1985). Culture of the surfaces of urinary
catheters to sample urethral flora and study the effect of
antimicrobial therapy. / Clin Microbiol 21:902-908.
Lee CA, Falkow S (1990). The ability of Salmonella to enter
mammalian cells is affected by bacterial growth state.
ProcNatlAcadSci USA 87:4304-4308.
Lee CK, Rubin LG, Moldwin RM (1995). Synergy between
protamine and vancomycin in the treatment of
Staphylococcus epidermidis biofilms. Urology 45:720-
724.
Marquis RE (1968). Salt-induced contraction of bacterial cell
walls. / Bacteriol 95:775-781.
M arrie T J, C osterton JW (1983). A scanning and
transmission electron microscopic study of the surface of
interuterine devices. Am J Obstet Gynecol 14:384-393.
M cC arter L, Hilmen M, S ilverman M (1988). F lagellar
dynamometer controls swarmer cell differentiation of
Vibrio parahaemolyticus. Cell 54:345-351.
McF eters GA, Egil T, Wilberg E , Adler A, S chneider R,
Snozzy M, Geiger W (1990). Activity and adaptation of
nitrilo(N T A )-degrading bacteria: F ield and laboratory
studies. Water Res 24:875-881.
Newman HN, Barber PM (1995). Dental plaque structure in-
vivo. I n: T he life and death of biofilm. Wimpenny J,
Handley P, Gilbert P, Lappin-Scott H, editors. Cardiff,
UK: Bioline, pp. 27-32.
Nichols WW, Dorrington SM, Slack MP E , Walmsley HL
(1988). Inhibition of tobramycin diffusion by binding to
alginate. Antimicrob Agents Chemother 32:518-523.
Nichols WW, Evans MJ, Slack MPE, Walmsley HL (1989).
The penetration of antibiotics into aggregates of mucoid
and non-mucoid Pseudomonas aeruginosa. J Gen
Microbiol 135:1291-1303.
N ickel JC , R useska I , Wright JB , C osterton JW (1985).
Tobramycin resistance of Pseudomonas aeruginosa cells
growing as a biofilm on urinary catheter materials.
Antimicrob Agents Chemother 27:619-624.
Pascual A, Ramirez de Arellano E, Pera EJ (1994). Activity
of glycopeptides in combination with amikacin or
rifampin against Staphylococcus epidermidis biofilms on
plastic catheters. Eur J Clin Microbiol Infect Dis 13:515-
517.
Pitt WG, McBride MO, Lunceford JK, Roper RJ, Sagers RD
(1994). Ultrasonic enhancement of antibiotic action on
Gram-negative bacteria. Antimicrob Agents Chemother
38:2577-2582.
Prosser BT, Taylor B, Dix BA, Cleeland R (1987). Method of
evaluating effects of antimicrobials upon bacterial
biofilms. Antimicrob Agents Chemother 31:1502-1506.
Shepherd JE , Waigh R D, Gilbert P (1988). Antimicrobial
action of 2-bromo-2-nitropropan-l,3-diol (B ronopol)
against Escherichia coli. Antimicrob Agents Chemother
32:1693-1698.
S lack M P E , N ichols WW (1981). T he penetration of
antibiotics through sodium alginate and through the
exopolysaccharide of a mucoid strain of Pseudomonas
aeruginosa. Lancet 11:502-503.
Stewart PS (1994). Biofilm accumulation model that predicts
antibiotic resistance of Pseudomonas aeruginosa biofilms.
Antimicrob Agents Chemother 38:1052-1058.
Suci PA, Mittelman MW, Yu FU, Geesey GG (1994).
I nvestigation of ciprofloxacin penetration into
Pseudomonas aeruginosa biofilms. Antimicrob Agents
Chemother 38:2125-2133.
S utherland IW (1985). B iosynthesis and composition of
Gram-negative bacterial extracellular and wall
polysaccharides. Ann Rev Microbiol 39:243-270.
Sutherland IW (1995). Biofilm-specific polymers: Do they
exist? I n: The life and death of biofilm. Wimpenny J,
Handley P, Gilbert P, Lappin-Scott H, editors. Cardiff,
UK: Bioline, pp. 97-101.
Thomson KS , Bakken JS , Sanders CC (1995). Antimicrobial
susceptibility testing in the clinic. In: Microbial quality
assurance: a guide towards relevance and reproducibility
of inocula. Brown MRW, Gilbert P, editors. Boca Raton,
FL: CRC Press, pp. 275-289.
Van Loosdrecht MCM, Lyklema J, Norde W, Zehnder AJB
(1990). I nfluence of interfaces on microbial activity.
Microbiol Rev 54:75-87.
Wagman GH, Bailey JV, Weinstein MJ (1975). Binding of
aminoglycoside antibiotics to filtration materials.
Antimicrob Agents Chemother 7:316-319.
Williams P (1988). Role of the cell envelope in bacterial
adaptation to growth in vivo in infections. Biochimie
70:987-1011.
Williams P, Bainton NJ, Swift S, Chabra SR, Winson MK,
Stewart GS A B , et al. (1992). Small molecule-mediated
density dependent control of gene expression in
prokaryotes: bioluminescence and the biosynthesis of
carbapenem antibiotics. FEMS Microbiol Lett 100:161-
168.
Zabinski RA, Walker KJ, Larsson AJ, Moody JA, Kaatz GW,
Rotschafer JC (1995). Effect of aerobic and anaerobic
environments on antistaphylococcal activities of five
fluoroquinolones. Antimicrob Agents Chemother 39:507-
512.
Zambrano MM, Kolter R (1995). Changes in bacterial cell
properties on going from exponential growth to stationary
phase. In: Microbial quality assurance: a guide towards
relevance and reproducibility of inocula. Brown MRW,
Gilbert P, editors. Boca Raton, FL: CRC Press, pp. 21-30.
Zobell CE (1943). The effect of solid surfaces on bacterial
activity. / Bacteriol 46:39-56.