CHAPTER FIVE

The Challenging World of Biofilm

Physiology
Joke Donné, Sylvia Dewilde1
Protein Chemistry, Proteomics and Epigenetic Signalling (PPES), Department of Biomedical Sciences,
University of Antwerp, Antwerp, Belgium
1
Corresponding author: e-mail address: sylvia.dewilde@uantwerpen.be

Contents
1. Biofilm Structure and Formation 236
1.1 Biofilm Structure 237
1.2 Biofilm Formation 239
2. Biofilm Resistance 245
2.1 An Advantageous Way of Life 245
3. Biofilm-Associated Infections 255
4. Biofilm Research and Its Challenges 261
4.1 In Vitro and In Vivo Biofilm Models 261
4.2 Biofilm Detection, Identification and Quantification Methods 264
4.3 Comparative -Omics Studies 267
4.4 Experimental Design 273
4.5 Multispecies Biofilms 278
References 282

Abstract
Worldwide, infectious diseases are one of the leading causes of death among children.
At least 65% of all infections are caused by the biofilm mode of bacterial growth. Bac-
teria colonise surfaces and grow as multicellular biofilm communities surrounded by a
polymeric matrix as a common survival strategy. These sessile communities endow bac-
teria with high tolerance to antimicrobial agents and hence cause persistent and
chronic bacterial infections, such as dental caries, periodontitis, otitis media, cystic fibro-
sis and pneumonia. The highly complex nature and the rapid adaptability of the biofilm
population impede our understanding of the process of biofilm formation, but an
important role for oxygen-binding proteins herein is clear. Much research on this bac-
terial lifestyle is already performed, from genome/proteome analysis to in vivo antibiotic
susceptibility testing, but without significant progress in biofilm treatment or eradica-
tion. This review will present the multiple challenges of biofilm research and discuss
possibilities to cross these barriers in future experimental studies.

Advances in Microbial Physiology, Volume 67 # 2015 Elsevier Ltd 235

ISSN 0065-2911 All rights reserved.
http://dx.doi.org/10.1016/bs.ampbs.2015.09.003
236 Joke Donné and Sylvia Dewilde

ABBREVIATION
AB antibiotic
AHL acyl-homoserine lactone
AI auto-inducer
A-site active site
BAI biofilm-associated infections
BHI brain heart infusion
c-di-GMP cyclic di-guanosine monophosphate
CE capillary electrophoresis
CFU colony forming units
CLSM confocal laser scanner microscopy
CV crystal violet
CVCs central venous catheters
DGC diguanylate cyclase
DMMB dimethylmethylene blue
EAL motif Glutamic acid-Alanine-Leucine motif
EPS extracellular polymeric substance
FDA fluorescein diacetate
FISH fluorescence in situ hybridisation
GCS globin-coupled sensor
GGDEF motif Glycine-Glycine-Asparticacid-Glutamicacid-Phenylalanine motif
GGEEF motif Glycine-Glycine-Glutamicacid-Glutamicacid-Phenylalanine motif
GMP guanosine monophosphate
ICAT isotope-coded affinity tag
I-site inhibitory binding site
iTRAQ isobaric tag for relative and absolute quantitation
LB Luria–Bertani
MOPS 3-(N-morpholino)propanesulphonic acid
MRD modified Robbins device
MS mass spectrometry
PDE phosphodiesterase
PGA poly-β-1,6-N-acetyl-glucosamine
PMA propidium monoazide
ppGpp guanosine tetraphosphate
qPCR quantitative real-time PCR
QS quorum sensing
TA toxin–antitoxin pair
TOF-MS time-of-flight mass spectrometry
TSB tryptic soy broth
XTT 2,3-bis-(2-methoxy-4-nitro-5-sulphophenyl)-2H-tetrazolium-5-carboxanilide

1. BIOFILM STRUCTURE AND FORMATION

The investigation of natural microbial habitats verifies that most of the
bacteria, perhaps 99%, are sessile and attached as biofilms to a surface. In such
The Challenging World of Biofilm Physiology 237

Dispersal

Maturation
Attachment

Figure 1 Steps of biofilm formation. Aadapted from Abed, Ibnsouda, Latrache, and
Hamadi (2012).

a biofilm, the bacteria behave very differently than in a free-floating plank-

tonic state. Furthermore, according to the U.S. National Institutes of
Health, more than 80% of all microbial infections are associated with biofilm
growth. The development of such a microbial biofilm is a complex and
dynamic process involving several steps: approaching a surface, attachment,
maturation and dispersal. All these steps are tightly regulated by complex sig-
nalling networks connecting a variety of physiological pathways such as
motility, adhesion, communication by quorum sensing (QS), stress
responses, virulence, matrix production and responses to antibiotic treat-
ment. Furthermore, the ability of the biofilm to adapt continuously to
changing environments and conditions, such as oxygen supply, is of utmost
importance. In every stage of the biofilm formation (Fig. 1), the bacteria
need to change their behaviour: from swimming to swarming, from
attaching to growing and from defending to detaching.

1.1 Biofilm Structure

The three basic components of a biofilm culture are (a) microorganisms,
(b) an extracellular matrix and (c) a surface. It must be noted that because
of the wide range of environments where biofilms are found, it is very hard
to generalise the structure and characteristics of biofilms. Many factors con-
stantly affect the architecture of the biofilm population, from external influ-
ences such as the chemical composition of the environment, to internal
factors such as the genetic profile of the microorganisms. Most species act
differently in different environments, which makes biofilm formation diffi-
cult to cure and reproduce in a research setting.
238 Joke Donné and Sylvia Dewilde

1.1.1 Microorganisms
A biofilm can be formed by a single bacterial species, but in most environ-
ments, they are composed of a rich variety of species. Given the right con-
ditions, almost all bacteria are able to form a biofilm, with different growth
and stability characteristics. Also fungi, algae, yeasts and protozoa have
already been observed in a biofilm mode (Cos, Tote, Horemans, & Maes,
2010). For example, most nosocomial Candida albicans bloodstream infec-
tions are caused by polymicrobial biofilms. Bacteria such as Staphylococcus
aureus, Staphylococcus epidermidis and Enterococcus species are commonly
co-isolated with C. albicans. The fungal hyphae enhance S. epidermidis
growth, and the presence of C. albicans increases the vancomycin antibiotic
(AB) resistance of S. epidermidis and S. aureus (Harriott & Noverr, 2011). In a
polymicrobial biofilm, there is competition for nutrients and oxygen from
the environment, effecting mostly the cells in the deeper layers, that are in
turn longer protected from toxic agents. Nevertheless, metabolised products
by one coloniser may stimulate the growth of the others present. These
metabolites can even be ligands allowing the attachment of other species
(Cos et al., 2010; Dunne, 2002). The variance in multispecies biofilms is dis-
cussed further in Section 4.5.

1.1.2 Matrix
The maintenance of the biofilm architecture depends mainly on the presence
of the extracellular matrix. Biofilms are actually composed of more matrix
(80–85%) than microorganisms (15–20%) (Dufour, Leung, & Lévesque,
2012). This extracellular slime layer surrounds and protects the microbial cells
against harmful factors in the surrounding environment. The production of the
polymeric matrix is a social activity wherein each bacterium of the biofilm
community participates to create a protective biofilm society. There is a wide
variety in the composition of the matrix, which depends especially on the bac-
terial species present and the surface (Pamp, Gjermansen, & Tolker-Nielsen,
2007). The biofilm extracellular polymeric substance (EPS) is a complex of
secreted polysaccharides, proteins and nucleic acids from lysed cells and
absorbed nutrients and ions from the surrounding area. The production of
the polysaccharides and proteins of the extracellular matrix, which determine
the viscosity, occurs by the bacteria themselves and is species dependent.
Table 1 shows the basic matrix components of an Escherichia coli biofilm pop-
ulation (Beloin, Roux, & Ghigo, 2008) as an example. To conclude, the poly-
meric matrix is a very important feature in biofilm development and, for that
reason, an interesting research topic. Nevertheless, its very sticky characteristic
makes it difficult to analyse and separate from the microbial cells.
The Challenging World of Biofilm Physiology 239

Table 1 E. coli Biofilm Matrix Components

1. Polysaccharides
Cellulose
Colonic acid
Poly-β-1,6-N-acetyl-glucosamine(PGA)
Lipopolysaccharides
2. Proteins
Multimeric cell appendages Flagellin protein (flagellum)
Curlin protein (curli)
Fimbrin protein (fimbriae)
Surface proteins Ag43 adhesin
AidA and TibA proteins
3. Extracellular DNA
4. Signalling molecules Acyl-homoserine lactones (AHL)
Others
5. Water, nutrients, antimicrobial molecules, ions, etc., from growth environment

1.1.3 Surfaces
Biofilms can form on nearly every surface (Fig. 2), on natural surfaces like
fruit and vegetables, on rocks and in soil. Biofilms also appear in the medical
field where they grow on medical devices such as catheters, tubes and valves,
contact lenses, in wounds or on teeth. Besides health care, biofilms also cre-
ate problems in industry. They can cause food and water contamination,
metal surface corrosion and clogging. Moreover, biofilm formation on
for instance pipelines and fishing nets decreases the efficiency of this equip-
ment which gives problems in every industrial setting. Consequently, bio-
film development has an economical effect (Dufour et al., 2012).

1.2 Biofilm Formation

1.2.1 Different Steps in Biofilm Formation and Maturation
When free-floating planktonic bacteria experience a surface, they are able to
swim towards it and attach to it. Besides passive movements due to
Brownian and gravitational forces, they are also able to migrate in the direc-
tion of favourable conditions. This kind of active motility is dependent on
flagellum machinery. With this active swimming mechanism, the bacterial
240 Joke Donné and Sylvia Dewilde

Figure 2 Biofilm accumulation in wound. An infected wound on a human leg, with a

grey, slimy biofilm. Photograph is a kindly gift from Jennifer Hurlow.

cells are able to overcome repulsive electrostatic and hydrodynamic forces

when approaching surfaces, which makes them able to interact and attach
(Beloin et al., 2008). Thus, flagella are not only required for the
propeller-based swimming capacity of the bacteria but also play a surface-
sensing role in the process of attachment (Belas, 2014). This process involves
(i) a decrease in the rotation speed of the flagella which will make the bacteria
swim slower but keep them sufficiently energetic to exceed the repulsive
forces of the surface, and (ii) an induction of the synthesis of cell appendages
or adhesins which are necessary for adherence. Several additional factors will
affect the efficiency of attachment: (i) the physicochemical properties of the
substratum; (ii) the presence or absence of a conditioning film that can be
formed as a coating on the substratum; (iii) the hydrodynamics of the
medium which is defined by the flow velocity; (iv) the physicochemical
characteristics of the medium like pH, nutrient concentration, temperature,
ionic strength and viscosity; and (v) the surface properties (hydrophobicity,
the presence and number of appendages, the composition of the outer mem-
brane layer) of the bacterial cell that wants to attach (Donlan, 2002).
There are three types of cell-surface components expressed by bacteria
that play a role in strengthening the bacteria-to-surface interactions: fim-
briae, curli and conjugative pili. Fimbriae and curli are filamentous protein-
aceous tubular structures which are present in large numbers (e.g. between
100 and 500 fimbriae) at the bacterial surface (Beloin et al., 2008). Mutants
in both fimA and fimH, the genes encoding the major fimbriae subunit and
The Challenging World of Biofilm Physiology 241

the adhesin at the tip of the filament, respectively, have been reported to
reduce initial attachment of E. coli to abiotic surfaces (Beloin et al., 2004).
The conjugative pili are mainly responsible for cell–cell interactions that sta-
bilise the structure of the biofilm. Furthermore, they are also involved in
horizontal transfer of DNA and plasmids between interacting bacteria. It
is suggested that these plasmids can carry determinants for biofilm initiation
and possibly also for maturation, virulence and resistance (Roberts & Kreth,
2014). The attached bacteria will subsequently proliferate, grow and pro-
duce a slimy matrix. The resulting 3D biofilm community colonises the sur-
face and forms what is assumed to be a mature dense mushroom-like
population which will resist harmful conditions. The mechanisms that are
involved in this protective lifestyle are discussed in more detail in
Section 2. The process of maturation encompasses several pathways such
as cell-to-cell adhesion, secretion of toxins and signal molecules, sensing
microenvironments and environmental factors, quorum-sensing signalling,
regulation of metabolism, stress responses, virulence and resistance. Finally,
biofilms can spread through the detachment of cells, which can attach to
downstream sites at the surface. Dispersal is a result of ageing and environ-
mental shifts, when for instance the population density is too high, nutrient
and oxygen limitations appear or when there is an increased temperature
variation in the surrounding (Romeo, 2006). It is clear that this biofilm life
cycle is a highly regulated process based on a complex regulatory network in
order to tightly control every step of biofilm formation. The nucleotide-
based second messenger, cyclic di-guanosine monophosphate (c-di-
GMP), is a key molecule in this signalling network. Its concentration is con-
trolled by oxygen-binding proteins, among others.

1.2.2 Role of c-di-GMP and Oxygen-Binding Proteins

c-di-GMP interferes in the entire spectrum of cellular processes such as
adhesion, motility, virulence, QS and persistence and plays a key role in
the transition between a motile single-cell planktonic lifestyle and a sessile
multicellular biofilm lifestyle. The cellular c-di-GMP levels are controlled
by two groups of enzymes, diguanylate cyclases (DGCs) and phosphodies-
terases (PDEs). c-di-GMP is synthesised from two molecules GTP by DGCs
and requires dimerisation of two intact GGDEF (Gly-Gly-Asp-Glu-Phe) or
GGEEF (Gly-Gly-Glu-Glu-Phe) motifs in the active site (A-site) of the
functional domain. An inhibitory binding site (I-site) in close proximity
to the A-site is responsible for allosteric product inhibition in these DGCs
(Kalia et al., 2012; Pesavento & Hengge, 2009) (Fig. 3A). This I-site is
242 Joke Donné and Sylvia Dewilde

Figure 3 Structures of the c-di-GMP-metabolising enzymes, DGCs and PDEs.

(A) Synthesis of c-di-GMP (red; dark grey in the print version) by dimerisation of DGC
(light blue; light grey in the print version) and product inhibition via binding in the
I-site(RXXD motif); (B) c-di-GMP hydrolysis by PDE (blue; grey in the print version)
requires two Mg2+ ions (yellow; white in the print version).

characterised by a conserved RXXD motif. Notably, not all DGCs have an

I-site for allosteric feedback control. The degradation of c-di-GMP is con-
trolled by PDEs. Two classes of PDEs exist. One class contains an EAL (Glu-
Ala-Leu) domain which cleaves c-di-GMP into a linear pGpG which can
be broken into two GMP molecules, while others have a HD-GYP domain
which hydrolyses c-di-GMP directly into GMP. Both classes require
Mg2+ or Mn2+ ions for the phosphodiester hydrolysis. A detailed mecha-
nism of the involvement of the Mg2+ or Mn2+ cations in the active site
of PDEs has not been unravelled, but it is assumed that they play a role
in the Lewis acid activation of the phosphate centre of c-di-GMP as well
as activating the bound water molecule, which are both necessary for cleav-
age (Kalia et al., 2012) (Fig. 3B). Other divalent cations (Ca2+, Zn2+, etc.)
will inhibit their activity.
Many bacteria contain several genes that code for these c-di-GMP-
metabolising proteins. E. coli genes encode 29 c-di-GMP-metabolising
enzymes, 12 DGCs (containing a GGDEF domain), 10 PDEs (containing
an EAL domain) and 7 hybrid proteins containing both EAL and GGDEF
domains. Remarkably, no PDE with a HD-GYP domain is expressed in
E. coli. c-di-GMP signalling displays a high diversity concerning the
The Challenging World of Biofilm Physiology 243

¡ Binding gaseous
molecules (O2, CO,...)
¡ Phosphorylation
¡ Protein/peptide binding
¡ lon binding

INPUT

Diguanylate cyclase Phosphodiesterase

Haem GGDEF Haem EAL
2 GTP c-di-GMP pGpG
EcDosC N
O
EcDosP
NH
O
O N
HO P O N NH2
HO O
O OH
H2N N N O P OH
O
O
HN N
O

Quorum sensing Output Persistence

Virulence Motility

Adhesion Matrix production

Sessility and
biofilm formation
Figure 4 c-di-GMP metabolism and targeted pathways.

N-terminal sensory input domains fused to the majority of GGDEF, EAL

and HD-GYP domains. Two c-di-GMP-metabolising enzymes from
E. coli, EcDosC and EcDosP, possess a haem-based sensor domain and
are characterised as globin-coupled sensors (GCSs) (Fig. 4).
The DGC EcDosC is transcribed together with the PDE EcDosP, con-
trolling together the intracellular c-di-GMP concentration in an oxygen-
dependent way. When the haem group is in its oxy (Fe(II)-O2) or oxidised
(Fe(III)) form, the EcDosC is active, and when the unligated (Fe(II)) haem is
present, the protein is inactive. The oxygen-binding affinity for EcDosC
(K ¼ 0.07 μM) is extremely low compared to Sperm whale Mb
(K ¼ 1.27 μM) and other GCSs, e.g. the GCS from Bordetella pertussis
(K ¼ 1.8 μM) and the haem-based aerotaxis transducer from Bacillus subtilis
(K ¼ 1.39 μM) (Nakajima et al., 2012). Together with the relative fast
autoxidation rate of the EcDosC (Kitanishi et al., 2010), this DGC enzyme
is regulated in a very narrow oxygen range, living in equilibrium between its
active oxygenated or oxidised form and the inactive ferrous Fe(II) form,
which results in a slow c-di-GMP synthesis. A reducing system that
244 Joke Donné and Sylvia Dewilde

regenerates the reduced (Fe(II)) form is however not reported yet. Under
anaerobic conditions, the synthesis will stop and switch to a controlled
EcDosP-dependentc-di-GMP hydrolysis (Kitanishi et al., 2010). Indeed,
the deoxy form of the PDE EcDosP is an active form, although a very slow
hydrolysis of c-di-GMP (17-fold slower) is also observed when the sensor
domain is saturated with oxygen. However, EcDosP requires high oxygen
concentrations (75–256 μM oxygen) for activation (30–100% saturation).
Moreover, CO saturation enhanced the EcDosP activity in the same way
as oxygen saturation (Tuckerman et al., 2009).
There is less information known about the c-di-GMP effectors. Mainly
based on sequence studies, there appear to be the following types of
c-di-GMP effectors: (a) PilZ-domain effector proteins that bind
c-di-GMP to regulate other proteins/enzymes via protein–protein interac-
tions, (b) proteins with inactive GGDEF or EAL domains to modulate the
neighbouring catalytic or regulatory domains, (c) proteins with an intact
I-site to regulate directly their activity, (d) transcription factors that bind
c-di-GMP to regulate gene expression and (e) riboswitches that regulate
transcription or translation upon c-di-GMP binding (Kalia et al., 2012;
R€ omling, Galperin, & Gomelsky, 2013; Tamayo, Pratt, & Camilli,
2007). An example of c-di-GMP effectors in E. coli are BcsA and YcgR,
the only two PilZ-domain proteins in E. coli. They are the key components
in the motility-to-sessility transition and biofilm formation processes. The
switch from a motile to an adhesive lifestyle when entering the stationary
growth phase requires not only an accumulation of curli fimbriae but also
a precisely timed shutdown of flagella expression. Hence, two inversely
coordinated signalling cascades, dependent on separate c-di-GMP modules,
are controlling this lifestyle switch. At the top of the protein network, there
is competition for the RNA polymerase between sigma factors RpoD
(housekeeping sigma factor), RpoS (the starvation/stationary phase sigma
factor) and FliA (flagellar sigma factor). When cells enter the stationary
growth phase, flagella expression stops because the RpoS-controlled path-
way is induced, which stimulates DGCs YegE and YedQ. The resulting
c-di-GMP production stimulates on one hand the interference of the YciR
effector protein with the flagellar complex that slows down the flagellar rota-
tion and on the other hand the transcription of the adhesive curli fimbriae
activator gene csgD. Also, DGC/PDE pairs YdaM/YciR and YaiC/YoaD
are induced, which do not contribute to motility. YdaM/YciR strongly and
specifically control csgD transcription that results in curli fimbriae expression,
necessary for attachment. The induced YaiC/YoaD pair regulates matrix
The Challenging World of Biofilm Physiology 245

production via the activity of the cellulose synthase BcsA. Moreover, also
the RpoS-induced EcDosC/EcDosP pair regulates curli fimbriae expres-
sion, but in a CsgD-independent way. It also promotes the production of
the matrix component PGA (Tagliabue, Antoniani, et al., 2010;
Tagliabue, MacIg, Antoniani, & Landini, 2010). A brief overview of this
transition network is presented in Fig. 5.

2. BIOFILM RESISTANCE
2.1 An Advantageous Way of Life
Since biofilms are the bacterial dominant lifestyle, there has to be significant
advantages when living in sessile communities compared to free-floating sin-
gle planktonic cells. Bacteria use several mechanisms to create a protected
environment in the biofilm population to tolerate harmful external condi-
tions and host defences (Fig. 6). Their high tolerance to AB is their most
exceptional characteristic. The AB resistance of planktonic bacteria involves
usually (i) inactivation or destruction of the drug by enzymes like
β-lactamases; (ii) target modification via mutations; and (iii) exclusion of
the drug through efflux pumps. These mechanisms require specific regula-
tions of gene expression. However, a biofilm population not only uses the
above factors to be tolerant to antimicrobial agents; a main plank in their
defence is the structural nature of the biofilm itself ( Jolivet-Gougeon &
Bonnaure-Mallet, 2014; Lewis, 2013; Paraje, 2011). Since the biofilm archi-
tecture itself creates such a protective environment, it can be seen as an
‘innate’ resistance, which will be discussed below in more detail.

2.1.1 Biofilm Matrix as Diffusion Barrier

The biofilm matrix can act as a physical diffusion barrier that prevents anti-
microbial agents from reaching their target. The slime layer slows down the
diffusion rate which results in an increased resistance. The high viscosity and
the size exclusion of the polymeric matrix prevent obtaining effective con-
centration of AB being reached in the deeper layers of the biofilm commu-
nity. Bacteria in the outer layer will die, while bacteria buried in the deeper
layers may react with adaptive responses. Furthermore, penetration through
the biofilm can also be inhibited by the formation of ionic interactions
between the negatively charged exopolysaccharide matrix components,
such as colanic acid, and positively charged ABs like aminoglycosides.
Finally, enzymes such as catalases and β-lactamases within the slime layer
can inactivate the AB before they can even reach the bacterial population
Figure 5 Schematic overview of the signalling network which regulates the switch from a motile planktonic to a sessile biofilm lifestyle. Red
(dark grey in the print version) arrows indicate the signalling network of free-floating planktonic cells, while the green (grey in the print
version) arrows show the switch to a sessile lifestyle. Blue (grey in the print version): PDE, light blue (light grey in the print version): DGC,
brown (light grey in the print version): sigma factors, yellow (light grey in the print version): c-di-GMP effectors.
The Challenging World of Biofilm Physiology 247

Figure 6 Biofilm physiology and the effect of the microenvironment. The physical,
chemical and biological heterogeneity that can develop in a biofilm population is
shown. (A) Wild-type P. aeruginosa biofilms, grown in in vitro flow cells, visualised by
confocal laser scanning microscopy (Bjarnsholt et al., 2013). (B) The same wild-type
P. aeruginosa biofilm as (A) treated with 100 mg/ml tobramycin for 24 h. Green bacteria
are alive and yellow/red bacteria are dead. The AB cannot diffuse to the interior of
the biofilm population because of mechanisms of tolerance (Bjarnsholt et al., 2013).
(C) Micrograph of 18-h-old multispecies biofilm grown in flow cells. Fusobacterium
nucleatum (red), Aggregatibacter actinomycetemcomitans (green), Veillonella sp. (dark
blue) and Streptococcus oralis (light blue) (Häussler & Parsek, 2010). (D) E. coli PHL628
biofilm confocal and tomographic pH imaging, with the PH scale shown at the right.
The biofilm population is composed of microenvironments of different pHs (Hidalgo
et al., 2009).

(Cos et al., 2010; Paraje, 2011). These β-lactamase genes are highly induced
in response to AB sensed in the environment. However, this diffusion bar-
rier is species dependent and antibiotic dependent and is not sufficient to
explain the higher tolerance of biofilms against antimicrobial agents,
248 Joke Donné and Sylvia Dewilde

compared to planktonic cultures. A low rate of vancomycin AB penetration

was observed through the mucoid matrix of S. aureus biofilms using
fluorescent-labelled vancomycin ( Jefferson, Goldmann, & Pier, 2005).
A study analysing the penetration of five different antibiotics in S. aureus
and S. epidermidis biofilms, published by Singh, Ray, Das, and Sharma
(2010), demonstrated a reduced penetration of AB oxacillin, cefotaxime
and vancomycin, while the diffusion of amikacin and ciprofloxacin was
unaffected (Singh et al., 2010). Moreover, Anderl and co-workers demon-
strated that ciprofloxacin and chloride ion quickly penetrated the
48-h-oldKlebsiella pneumoniae biofilm, while ampicillin was unable to pen-
etrate when using 4-h treatment periods (Anderl, Franklin, &
Stewart, 2000).

2.1.2 Microenvironments and Stress Responses

The spatial organisation and the physiological function of the whole biofilm
community are vulnerable to environmental signals. The responses of the
biofilm bacteria to switching environmental cues are not homogenous
throughout the biofilm population. This can be explained by the spatial
organisation in the stacked multilayered biofilm structure: the chemical
and biological characteristics vary across the layers of the biofilm population.
Cells in the deeper layers do not experience the same conditions as cells in
the outer layers as mentioned before. Indeed, a mature biofilm consists of
concentration gradients of several solutes, such as nutrients, oxygen and
metabolites, creating different local microscale environments in the different
layers of the population. These result from the diffusion process through the
matrix and the extent of metabolic activity of the present bacteria (Stewart &
Franklin, 2008). Oxygen is able to diffuse through the aqueous biofilm
matrix, but it fails to penetrate to the deepest part of the biofilm population,
creating hypoxic zones, because the bacterial cells in the upper layers actively
consume oxygen. For example, a Pseudomonas aeruginosa biofilm, grown on a
membrane filter on tryptic soy agar for 48 h, showed an oxygen-defined-
two-layer pattern based on oxygen microelectrode measurements. This
study combined oxygen diffusion experiments with protein synthetic activ-
ity assays based on GFP reporter gene constructs (Werner et al., 2004). They
reported that the top layers of cells, where oxygen can diffuse in, exhibit
active protein synthesis, while the cells in the thick bottom layer experience
an anaerobic nature and appear to be metabolically inactive. They concluded
that in the P. aeruginosa colony biofilm, the oxygen is consumed faster than it
can diffuse into the deeper layers of the population (Serra & Hengge, 2014).
The Challenging World of Biofilm Physiology 249

These oxygen gradients will also have an effect on the c-di-GMP con-
centrations within the entire biofilm because of the oxygen-dependent
activity of the c-di-GMP-metabolising DGCs and PDEs, as discussed earlier
in Section 1.2.2. In contrast, metabolites and waste products are available in
higher concentration in the interior compared to the outside of the biofilm.
The methane production of a sewage outlet biofilm of 3.5 mm thick was
analysed, showing high concentrations of methane, building up to the depth
of the biofilm structure (Damgaard, Nielsen, & Revsbech, 2001). The
methane concentration at the surface of the biofilm was approximately
10% of the concentration at 2 mm depth, which is explained by aerobic res-
piration, which consumes a small part of the produced methane.
Due to these microenvironments within the biofilm, bacteria change
their gene expression and metabolic activity which leads to a physiological
heterogeneity and phenotypic diversity within the community (Parker &
Sperandio, 2009). Regions in the interior of the biofilm are limited in
not only oxygen but also nutrients, creating a starved environment. During
this period of starvation, the bacteria slow down metabolism and conse-
quently lower their growth and multiplication rates as an induced stress
response. Because ABs are more effective against fast growing cells, these
slower growing bacteria are co-responsible for the high biofilm resistance.
Furthermore, the diversity of the bacterial phenotypes within a biofilm is
caused by regulation of gene expression due to adaptation to the local envi-
ronment. Also random mutations, genetic rearrangements and horizontal
gene transfer may generate variants that are favoured by natural selection
which results in the creation of more heterogeneity (Fig. 7) (Stewart &
Franklin, 2008).
Next to the already mentioned hypoxia and nutrient starvation, bacteria
encounter a variety of stresses in their natural environments, such as expo-
sure to reactive oxygen and nitrogen species, membrane damage, high tem-
peratures and ribosome disruption (Poole, 2012). Different stresses can
induce several distinct stress responses and the different stress responses
are also correlated with each other (Table 2). The initiation of such bacterial
stress responses positively impacts the antimicrobial resistance mechanisms
of the pathogen in a direct or indirect way. When a stress-induced growth
termination or dormancy occurs, the stress response has an indirect effect on
the AB susceptibility. Direct effects of stress responses are for instance the
stimulation of a biofilm growth mode, increase in antimicrobial efflux
pumps, changes to antimicrobial targets, modifications to the membrane
barrier functions and the induction of resistance mutations (Poole, 2012).
250 Joke Donné and Sylvia Dewilde

A B

Figure 7 Biofilm heterogeneity. (A) Physiological heterogeneity. Cells respond to local

microenvironmental conditions by turning on or off certain genes, (B) genotypic vari-
ation and natural selection. Random mutations or chromosomal rearrangements result
in a variant (purple; dark grey in the print version) that multiplies according to its fitness
in the biofilm. Adapted from Stewart and Franklin (2008).

Table 2 Bacterial Stress Responses as Influencing Factors of Antimicrobial

Resistance (Poole, 2012b)
Types of Stress
Responses References
Stringent response Chatterji and Kumar Ojha (2001)
SOS response Baharoglu and Mazel (2014) and Simmons, Foti, Cohen,
and Walker (2008)
Stress-induced Joly et al. (2010)
mutagenesis
General stress response Hengge-Aronis (2002) and Battesti, Majdalani, and
Gottesman (2011)
Induction of a persister Boehm et al. (2009)
phenotype
References are mostly based on E. coli.
The Challenging World of Biofilm Physiology 251

The general stress response is controlled by the sigma factor S, encoded

by the rpoS gene, which regulates many pathways. RpoS was shown to be
required for a heat shock-promoted increase in carbapenem resistance in
P. aeruginosa, which clearly demonstrates a stress response-induced antimi-
crobial resistance (Murakami et al., 2005). In E. coli, the general stress
response is activated by oxygen and nutrient limitation, hyperosmolality,
pH decrease, non-optimal temperatures, reduced growth rate and high cell
density (Hengge-Aronis, 2002). This response allows cells to become more
resistant not only to the stress that they are first exposed to but also to other
stressful conditions. Such a cross-protection phenomenon is typical of the
general stress response and contrasts with other specific stress responses that
deal only with consequences of the inducing stress (Battesti et al., 2011). The
SOS response is a set of physiological responses induced by DNA damage or
replication fork blockages which can be caused by, e.g., oxidative stress and
exposure to certain AB ( Jolivet-Gougeon & Bonnaure-Mallet, 2014;
Simmons et al., 2008). The produced DNA damage and mutations activate
a set of SOS response-induced genes which express high fidelity DNA repair
proteins and proteins that arrest the cell cycle of the bacteria. This mecha-
nism allows bacteria to repair DNA before damaged DNA is segregated to
daughter cells. Other tolerated mutations may lead to phenotypic variants
within the microbial colony and be favourable allowing better adaptation
to the host environment (Simmons et al., 2008). Consequently, increased
mutation frequencies can result in AB resistance and latent pathogens
( Jolivet-Gougeon & Bonnaure-Mallet, 2014). Moreover, the SOS response
also contributes to AB resistance development by enhancing horizontal gene
transfer and promoting the capture and expression of integrons, which are
mobile genetic elements that carry multiple AB resistance genes and
integrases that catalyse recombination (Poole, 2012). An increased mutation
rate enlarges the chance to adaptation to a specific stress signal, but stress-
induced mutations are rather random.
When bacteria experience nutritional limitations, or starvation, they
adapt by changing to a slow-growing state which is characterised by an accu-
mulation of guanosine tetraphosphates (ppGpps), called the stringent
response. ppGpp negatively affects nucleic acid synthesis and positively reg-
ulates RpoS-dependent genes (Chatterji & Kumar Ojha, 2001). This posi-
tive regulation of RpoS leads to a broad effect of ppGpp because stationary
genes and also genes controlling biofilm formation are RpoS dependent
(Fig. 5). In E. coli, ribosome damage, for example, results in a high
biofilm-forming capacity as a consequence of a decrease in the ppGpp levels
252 Joke Donné and Sylvia Dewilde

and an increase of the c-di-GMP concentrations (Boehm et al., 2009). Fur-

thermore, there are some doubts whether stress can also promote a dormant
cell population, called persister cells (see Section 2.1.4).

2.1.3 Quorum Sensing

Bacteria also communicate with each other to respond to the environment.
Such a cell-to-cell communication system is called QS. It is a cell density-
dependent signal transduction process, which regulates a variety of physio-
logical behaviours in bacteria. The close proximity of the cells in a biofilm
allows rapid coordination of the behaviour of the entire community on the
QS signals, which is beneficial for the biofilm collective responsiveness to
the environment. Necessary collective actions of the biofilm population
include adapting to different levels of nutrients in the environment as men-
tioned earlier in Section 2.1.2, defending against other competitive micro-
organisms, avoiding toxic materials and invading or infecting a host
(Galloway, Hodgkinson, Bowden, Welch, & Spring, 2011).
During QS, bacteria secrete small signal molecules, called auto-
inducers(AIs), in the extracellular environment, which can be recognised
by specific receptors on the surface or even in the cytoplasm of neighbours.
When the concentration of the signal molecules reaches a threshold level, a
signal transduction cascade activates the expression of certain genes of the
responding cells, influencing many regulatory pathways among which bio-
film formation and biofilm-related processes. Gram-negative bacteria pos-
sess three different cell-to-cell communication systems (Parker &
Sperandio, 2009). The main QS mechanism used by these bacteria is based
on acyl-homoserine lactone (AHL) molecules, commonly referred to as
AI-1 molecules. AHL is synthesised by AHL synthase (LuxI and its homo-
logues) and sensed by LuxR-like receptor proteins. E. coli however lack
AHL synthase encoding genes but are able to respond to such AI-1 signals
because they do express SdiA, a LuxR-type AHL receptor (Van Houdt,
Aertsen, Moons, Vanoirbeek, & Michiels, 2006). Thus, AHL-dependent
QS signalling is a communication system between only Gram-negative bac-
teria, in response to secreted AHL molecules (Van Houdt et al., 2006;
Whitehead, Barnard, Slater, Simpson, & Salmond, 2001). Another QS path-
way, which is found in both Gram-positive and Gram-negative bacteria, is
based on AI-2 molecules that are derived from 4,5-dihydroxy-2,3-
pentadione. In E. coli, AI-2 molecules are synthesised by LuxS and are bound
to the ABC-type transporter Lsr. However, there are several different
The Challenging World of Biofilm Physiology 253

bacterial receptors for the detection of QS system 2 molecules, which allows

communication among all bacterial species. Since these receptors can vary in
structure and function, AI-2 signalling can induce different responses
(Rezzonico, Smits, & Duffy, 2012). The third QS circuit used by E. coli uses
the receptor QseC, which has a histidine kinase activity. Upon binding of
AI-3 signals, auto-phosphorylation takes place in order to activate the
response regulator QseB. Less is known about this QS pathway, but Kendall
et al. showed that LuxS is also involved in the production of AI-3 (Kendall,
Rasko, & Sperandio, 2007; Winzer & Williams, 2003).

2.1.4 Differentiation into Persister Cells

In the deepest interior of the biofilm, a specific non-growing biofilm phe-
notype, called persister cells, exists. These cells are significantly different
from other bacteria in the biofilm population because they tolerate excessive
antibiotic concentrations. Persister cells gain their resistance property by
entering a state of dormancy. A specific and extraordinary characteristic
of these dormant persister cells is recalcitrance, which means that after anti-
biotic treatment, the survived persister cells can relocate and form a new bio-
film again with the same sensitivity as the original population. This unique
property explains why persister cells are the major concern in biofilm treat-
ment (Lewis, 2007).
Several mechanisms are hypothesised to explain the formation of this
persistent subpopulation of a biofilm (Fig. 8) (Williams & Hergenrother,
2012). It is reported that dormant cells arise when the population enters
the mid-exponential growth phase and they reach a maximum in the sta-
tionary phase (Lewis, 2007). It is assumed that a phenotypic switch takes
place due to stochastic gene expression changes.
Another hypothesis states that persister formation can be induced by
stresses such as exposure to AB. A specific level of SOS response induction
is required for persister formation based on a study where all actively grow-
ing cells were exposed to ciprofloxacin, but where not all cells turned into
persister cells, suggesting that there are different mechanisms of persister for-
mation to different antibiotics (D€ orr, Lewis, & Vulić, 2009).
A transcriptome study showed that genes coding for toxin–antitoxin(TA)
module proteins were highly expressed in E. coli persisters, compared to cells
in the stationary phase (Shah et al., 2006). This indicates a role for TA systems
in the formation of persister cells. TA systems typically consist of a stable toxin
protein that disturbs an essential cellular process like, e.g., translation, and an
antitoxin (RNA or a protein) that prevents the toxin from being poisonous.
254 Joke Donné and Sylvia Dewilde

Figure 8 Four different models (A–D) to explain persister formation. Copied from Kint,
Verstraeten, Fauvart, and Michiels (2012).

Type I antitoxins are RNA antitoxins that inhibit translation by acting as anti-
sense RNA. When an RNA antitoxin binds and inhibits the toxin protein, it
is known as a type III antitoxin. Type II TA systems are based on protein–
protein interactions between the toxin and the antitoxin, which hinder toxin
activity. Type IV protein antitoxins prevent the toxin from binding its target
instead of inhibiting the toxin directly, and type V protein antitoxins cleave
specifically the toxin mRNA (Wood, Knabel, & Kwan, 2013). The E. coli
chromosome encodes several TA systems such as the relBE, mazEF, chpBIK,
hipBA, dinJ–yafQ, tisAB–istR and yefM–yoeB systems (Fiedoruk, Daniluk,
Swiecicka, Sciepuk, & Leszczynska, 2014; Karimi, Ghafourian, Kalani, &
Jalilian, 2015; Tsilibaris, Maenhaut-Michel, Mine, & Van Melderen,
2007). In E. coli, the SOS response probably induces persister cell formation
via the type I TA pair tisAB–istR. An SOS-activated cleavage of LexA stim-
ulates the transcription of the tisB gene, which encodes a membrane-acting
toxin. As a consequence, free TisB inserts in the membrane and disturbs the
The Challenging World of Biofilm Physiology 255

proton motive force which leads to decreased ATP concentrations and a state
of dormancy and persistence (Cavalcanti et al., 2015; D€ orr et al., 2009; Kint
et al., 2012).
Although environmental changes seem to positively influence the persister
phenotype, they can also result in the degradation of an antitoxin which is
highly unstable (Cos et al., 2010; Paraje, 2011). Released toxin proteins can
now inhibit cellular processes such as replication, translation, cell division
and ATP synthesis. Additionally, several antitoxins are degraded by Lon pro-
teases, whose cleavage activity is controlled by environmental factors. Both
facts lead to the hypothesis that the amount of persister cells is dependent of
the surrounded conditions (Unterholzner, Poppenberger, & Rozhon, 2013).

3. BIOFILM-ASSOCIATED INFECTIONS
The same microorganism can cause an acute or a chronic infection.
The nature of the pathogenesis is dependent on environmental signals, such
as the route of entry for infection in the host, the immune and nutrition sta-
tus of the host, the tissue integrity and the nutrient and oxygen availability in
the surrounding. Additionally, the type of infection is also correlated with
the growth mode of the bacteria when entering the patient: growing and
proliferating rapidly in the host is mostly associated with an acute infection
and will involve planktonic bacteria, while settling in slow-growing com-
munities, such as biofilms, mostly gives rise to chronic infections
(Bjarnsholt et al., 2013; Furukawa, Kuchma, & O’Toole, 2006). Some com-
mon biofilm-associated infections (BAI) are shown in Table 3.
During an acute infection, planktonic bacteria cause a short-time illness,
and ABs and vaccines are effective treatments of such infectious diseases. On
the contrary, in patients with a chronic infection, the infection lasts longer
but is mostly associated with mild symptoms, sometimes going unnoticed for
a few months to sometimes years. These ‘silent’ infections rarely lead to indi-
cations in the bloodstream which gives difficulties for early diagnosis. There-
fore, some criteria are formulated to evaluate several BAI characteristics
(Hall-Stoodley & Stoodley, 2009; Hall-Stoodley et al., 2012; Parsek &
Singh, 2003) (Table 4). Routine microbial examinations include sample col-
lection, cultivation, identification and antibiotic susceptibility testing. The
first step in this diagnostic protocol is not only the most important one
but also the most challenging one. A correct sample collection is essential
for a reliable diagnosis of the BAI but is associated with invasive procedures
since the microorganisms from the biofilm-attached surfaces have to be
256 Joke Donné and Sylvia Dewilde

Table 3 Examples of Biofilm-Associated Infections in Humans

Infection or Disease Biofilm Species
Tissue-related infections
Dental carries Acidogenic Gram-positive cocci (e.g. S. mutans)
Periodontitis Gram-negative anaerobic oral bacteria (e.g. P. gingivalis)
Otitis media Haemophilus influenzae
Osteomyelitis Various bacterial and fungal species (e.g. S. aureus)
Cystic fibrosis pneumonia P. aeruginosa and Burkholderia cenocepacia
Infective endocarditis Streptococci, Enterococci, Staphylococci
Bacterial prostatitis E. coli and other Gram-negative bacteria
Chronic wounds S. aureus, P. aeruginosa and several others
Urinary tract infections E. coli
Device-related infections
Central venous catheters S. aureus, S. epidermidis, P. aeruginosa, C. albicans,
Enterococcus faecalis
Urinary catheters S. epidermidis, E. coli, E. faecalis, K. pneumoniae, Proteus
mirabilis
Heart valves Streptococci, Enterococci
Orthopaedic implants P. aeruginosa, E. coli, Enterococci, Streptococci
Endotracheal tubes Various bacterial and fungal species
Contact lenses P. aeruginosa and Gram-positive cocci
Adapted from Aparna and Yadav (2008), Lebeaux, Chauhan, Rendueles, and Beloin (2013) and Cos
et al. (2010).

isolated (Wu, Moser, Wang, Høiby, & Song, 2014). When the isolated
colonised surface is removed, microscopic analysis can already determine
the biofilm phenotype based on diagnostic criteria 1 and 2 (Table 4). Fur-
thermore, it has to be noted that the presence of biofilms is not necessarily
associated with a positive blood culture neither when dealing with intravas-
cular device-related biofilm infections. This is illustrated in criteria 3 and 5
(Table 4). Indeed, peripheral blood cultures reflect individual bacteria float-
ing around in the bloodstream, but they do not detect biofilm-attached cells
which can be present. The method used for detection is as such crucial. Post
et al. studied diagnostic assays to detect bacterial pathogens in chronic otitis
media patients (Post et al., 1995). They were able to detect a pathogen using
The Challenging World of Biofilm Physiology 257

Table 4 Diagnostic Criteria for Biofilm-Associated Infections

1. Pathogenic bacteria are associated with a surface.

2. Direct examination of infected tissue or materials demonstrates aggregated cells
encapsulated in a matrix, which may be of bacterial and host origin.
3. Infection is confined to a particular site in the host.
4. Recalcitrance to antibiotic treatment in spite of a demonstrated standard
susceptibility testing of the specific bacterium (since standard tests are performed
on planktonic single cells).
5. Culture-negative result in spite of clinically documented high suspicion of
infection (since localised bacteria in a biofilm may be missed in a conventional
blood sample or aspirate).
6. Evidence of ineffective host clearance with bacterial aggregates (microcolonies)
demonstrated by the co-localisation of host inflammatory cells with discrete
areas of the host tissue.
Adapted from Hall-Stoodley et al. (2012) and Hall-Stoodley and Stoodley (2009).

standard cultural methods in only 25–30% of the patients, while in an acute

otitis media pathogenesis 90% of the patients report a positive culturing
result. On the contrary, a PCR-based detection system and/or fluorescence
in situ hybridisation (FISH) combined with confocal laser scanner micros-
copy (CLSM) identified in 80–100% of the chronic patients a harmful
microorganism (Post et al., 1995; Wilkins, Hall-Stoodley, Allan, & Faust,
2014). Additionally, as mentioned above, biofilms can withstand both anti-
biotics and host defences, since they are persistent (10–100 times more resis-
tant than their planktonic life form), which explains criteria 4 and 6 for the
diagnosis of BAI (Hall-Stoodley et al., 2012).
There is an increasing concern about the role played by microbial bio-
films in infection. Recent surveys clearly indicate that bacteraemia from
catheter-related infections is by far the leading cause of nosocomial infec-
tions in intensive care units. Today, the common hospital-used-strategy
to treat patients with device-related biofilm infections is the removal of
the colonised indwelling device. Medical devices such as central venous
catheters (CVCs), endotracheal tubes and urinary catheters are easy to
remove and can be replaced by new ones, but for pacemakers or prostheses,
it is more difficult and a surgical operation is required (Amalaradjou &
Venkitanarayanan, 2013). In addition, ABs are administered prior to cat-
heterisation in order to kill the incoming pathogens that are at that point still
planktonic.
258 Joke Donné and Sylvia Dewilde

To try to avoid colonisation of the medical devices, a lot of research has

been made on the biofilm susceptibility of the material the devices are made
of and the possible pretreatment of this material. The surface properties,
such as the hydrophobicity, charge and shape of the surface, can be changed
in such a way that attachment will be inefficient. Biocompatibility of the
catheter material, and the ability of the device to perform its function with-
out provoking undesirable side effects, is of crucial importance and also
dependent on the physical and chemical properties of the material and on
the body’s reaction to it. The most modern tubes used today are silicone-
based or polyurethane catheters. They are non-allergenic, show a high
flexibility and have low surface tension, diminishing the chance of biofilm
formation (Hawser & Douglas, 1994). However, silicones are also hydro-
phobic which makes in turn the bacteria eager to adhere (Cohen et al.,
2011; Cox, 1990; Curtis & Colas, 2004). In addition, a conditional film will
be formed on, for example, CVC tubing. Components of the blood flow
will form a natural coating on the surface area. The adsorption of blood pro-
teins, like fibrinogen and fibronectin, or other organic molecules will affect
the adhesion capacity of bacteria in a positive way (Murga, Miller, &
Donlan, 2001; Teughels, Assche, Sliepen, & Quirynen, 2006). That is
why despite the fact that silicone and polyurethane catheters are showing
the highest biocompatibility, additional coatings will be necessary to reduce
the susceptibility to biofilm formation. The search for new synthetic films
which are coated directly on the surface material (with, e.g., metals or anti-
microbials) and which should prevent or diminish natural conditioning
and/or bacterial attachment, is progressing, but to date not entirely success-
ful. That is why a combinational approach is necessary where not only a
coated catheter is used but where the patient is treated with bacterial attach-
ment inhibitors. Several anti-adhesion therapies have been reviewed such as
the use of adhesin analogues (synthetic or recombinant adhesin fragments)
and anti-adhesin antibodies. However, such therapies are adhesin specific
and are not effective enough to prevent bacterial attachment. Another strat-
egy is the administration of receptor-like carbohydrates, which interact with
adhesins before they can attach to the glycoprotein/glycopeptide receptors
on host cells. This is a more attractive therapy because of its broad-spectrum
anti-adhesion activity. The anti-adhesion effects of dietary inhibitors are
questioned. The most studied dietary inhibitor in E. coli adhesion in urinary
tract infections is cranberry juice (Ofek, Hasty, & Sharon, 2003).
If such procedures are not feasible, as for instance when biofilms are
already formed on tissue surfaces or implanted medical devices, other
The Challenging World of Biofilm Physiology 259

approaches must be implicated. A possibility to manage such biofilm infec-

tions is to drain biofilm components out of the body. This will not be suf-
ficient to remove tightly attached cells and surgically removing the biofilm
population will be the last option. However, some studies demonstrate an
enhanced antimicrobial effect when using different AB dosing approaches,
dependent on the bacterial species and the used antimicrobial agent. A pulse-
dosing method, once-daily or thrice-daily, for example, can result in a dif-
ferent response dependent on the pathogen and the type of ABs (Ibrahim,
Gunderson, Hermsen, Hovde, & Rotschafer, 2004). Grant and Bott publi-
shed in 2003 a study investigating the effects of dose concentration, duration
and frequency of dosing of biocides on Pseudomonas fluorescens biofilms
attached to glass tubes (Grant & Bott, 2003). However, more research about
evaluating different dosing approaches is required to achieve an optimum
dosing method for each biofilm. Such a strategy could have promising ben-
eficial effects in future treatments of biofilm-related infections, but complete
eradication cannot be achieved with AB alone. There is a need for specific
anti-biofilm agents that focus on the manipulation of important parameters
in the process of biofilm formation such as attachment, motility, matrix pro-
duction and QS (Table 5).
An interesting target for therapy is the bacterial QS communication sys-
tem (see Section 2.1.3). There are four different ways to attenuate this system
and consequently reduce biofilm virulence: (i) inhibition of the enzymes
responsible for AI synthesis; (ii) targeting the AI receptors;
(iii) preventing AI-binding on receptor by, e.g., AI antibodies; and
(iv) inhibition of downstream proteins that link the QS receptor to the
activation of QS-regulated genes (Sintim, Smith, Wang, Nakayama, &
Yan, 2010). Because AI-1 molecules are produced by species-specific
enzymes, it is hard to discover broad-spectrumAI-1 synthase antagonists.
On the contrary, more than 70 bacterial species synthesise AI-2 molecules
via a LuxS enzyme, which makes LuxS inhibitors superior therapeutic
agents. Some attempts in finding LuxS substrate analogues were published
(Shen, Rajan, Zhu, Bell, & Pei, 2006; Wnuk et al., 2009). Furthermore,
up till now, there are no QS receptors found in humans, which means
anti-AI-receptor molecules will show low cytotoxicity. Mostly, these mol-
ecules are structurally similar to the natural AI and thus are called AI ana-
logues. Using AI-binding macromolecules that prevent the signal
molecule binding to its receptor is another principle for biofilm research.
A protective role for monoclonal antibodies that recognise AHL compounds
against P. aeruginosa infections has been published (Palliyil, Downham,
260 Joke Donné and Sylvia Dewilde

Table 5 Anti-Biofilm Therapy

Point of Interference Specific Inhibitory Effect
Motility Stimulation of flagella expression or activity
EPS biofilm matrix Matrix-degrading enzymes (e.g. dispersin B, cellulase,
DNase)
Inhibition of matrix production
Attachment Anti-adhesion coatings
QS communication Inhibition AI synthesis
system
AI antibodies
Competitive AI receptor-binding molecules
c-di-GMP signalling DGC inhibition
PDE stimulation
Interference with c-di-GMP effectors (e.g. c-di-GMP
analogues)
AB tolerance Inhibitors of efflux pumps
Antimicrobial lock solutions (e.g. chelators, anticoagulants)
Membrane-acting AB to kill also persisters
Antimicrobial peptides
Probiotics
Stimulation of cell death (e.g. TA activation)
Bacteriophage therapy
Different points of interference for biofilm therapy are demonstrated, each with some examples of
possible inhibitory approaches.

Broadbent, Charlton, & Porter, 2014). Also the use of AI-degrading

enzymes, such as lactonases and acylases, is a promising approach. Acylase
enzymes, for example, remove a side chain from a ring-like molecule by
hydrolysing the connecting amide bond (Sio et al., 2006). The enzymes
were listed that have the ability to degrade AHL signal molecules that are
characterised to date (Fetzner, 2014).
Another interesting point of interference is c-di-GMP signalling (see
Section 1.2.2). Controlling the intracellular c-di-GMP concentrations may
suppress biofilm formation. A decreased DGC activity or an increased PDE
activity will lead to lower levels of c-di-GMP and result in the disturbance
The Challenging World of Biofilm Physiology 261

of the biofilm signalling network. A screen for small molecules that inhibit the
c-di-GMP synthesising DGC WspR of P. aeruginosa identified ebselen as an
I-site binding compound (Lieberman, Orr, Wang, & Lee, 2014). Also
c-di-GMP analogues have potential to treat BAI because they can occupy
the active site of c-di-GMP receptors. The two previously discussed possible
therapeutic approaches (QS and c-di-GMP signalling as targets) appear more
beneficial than conventional ABs because they are not directly involved in
bacterial growth and may be less sensitive to the development of resistance.
Finally, compounds that are capable of inducing cell death have potential
as antibacterial agents. Chromosomal TA systems are identified in almost all
bacterial pathogens and are involved in several cellular pathways, including
stress response, biofilm formation, QS, virulence and persistence, making
them an interesting therapeutic target. A direct activation approach is based
on TA disruptors that, e.g., inhibit the antitoxin and prevent it from com-
plex formation with the toxin, or that bind to and activate the toxin. An
indirect toxin activation can take place due to the modulation of TA expres-
sion via interfering with the promotor regions of the TA genes or due to the
enhancement of the proteolytic susceptibility of the antitoxin (Williams &
Hergenrother, 2012). Overproduction of the toxin of the yefM–yoeB TA
system leads to reduced cell growth and cell viability in E. coli (Nieto
et al., 2007).
To conclude, all above discussed therapies could be very powerful anti-
biofilm agents because they also target the dormant persister cells of the pop-
ulation. Additionally, membrane-acting strategies can also be successful in
killing the whole biofilm community.

4. BIOFILM RESEARCH AND ITS CHALLENGES

4.1 In Vitro and In Vivo Biofilm Models
To get a better understanding of the underlying mechanisms of biofilm for-
mation and its resistant characteristic, a first essential step is to develop appro-
priate biofilm model systems to produce, quantify and evaluate the biofilm
phenotype. Several different in vitro and in vivo models have been developed
over the last decade, aiming at a general model that approximates the natural
biofilm condition the most (Coenye & Nelis, 2010; Cos et al., 2010). Both
static and dynamic in vitro biofilm production model systems are available to
grow adherent biofilm populations. Static test systems contain diverse flat
surfaces such as microtitre plates, petri dishes or plastic tubes. The most pop-
ular static biofilm production model is a 96-well plate, where the bacterial
262 Joke Donné and Sylvia Dewilde

suspension is added to every well of the plate and incubated. Such static sys-
tems cannot mimic the reality of a natural biofilm environment, where there
is a free flow of nutrients and waste products and where there are influences
of, e.g., the host immune system. A dynamic in vitro system is developed to
create biofilms under a continuous flow of nutrients. Centre for Disease
Control biofilm reactors and modified Robbins devices (MRDs) (Fig. 9)
are commercially available flow displacement model systems (Coenye &
Nelis, 2010; Lebeaux et al., 2013). The MRD is a linear tank reactor with
several ports that accept plug-holding discs where biofilms can be formed.
After an adhesion phase where a microorganism suspension is flowed
through the device, a controlled medium flow can be applied to mimic
the in vivo biofilm formation process as in, for instance, water pipelines or
medical tubing. Numerous custom-made dynamic models are appearing
in the literature. A simple example is a catheter which is connected at
one end to a pump which provides the medium and/or cell suspension flow
and at the other end to a waste reservoir. Dynamic methods generally allow
better control of the biofilm growth environment (e.g. shear forces, nutrient
concentrations, waste product drainage) and for that reason they have been
frequently used to study the physical and chemical resistance capacity of bio-
film populations (Lebeaux et al., 2013).

Figure 9 Schematic presentation of the modified Robbins device. Via a pump system,
medium is pumped automatically through the device, containing several discs for bio-
film formation. Adapted from Oosterhof, Buijssen, Busscher, Van Der Laan, and Van Der Mei
(2006).
The Challenging World of Biofilm Physiology 263

Both static and dynamic models are extensively used to characterise bio-
film physiology in, for example, AB screens because of high-throughput
capability, and in genetic studies to investigate the roles of different genes
in the process and regulation of biofilm formation. However, a major draw-
back of these in vitro models is that they do not take into account some impor-
tant environmental factors, like host factors and other signalling molecules
that are present in the surrounding of the adherent biofilm. These shortcom-
ings can be partly addressed by using an in vivo model system. Many animal
models have been developed to study BAI because they are better suited for
observing the biological behaviour of both the host and the biofilm popula-
tion. Most biofilm model systems use rats, mice or rabbits, not only for testing
AB survival but also for studying factors important for biofilm formation. For
example, rats are commonly used as an in vivo CVC model where catheters
are inserted in the external jugular vein, followed by contamination with bac-
teria. This model gives the opportunity to study biofilm adhesion to CVC
and evaluate the effect of antimicrobial agents, which can be injected. In for-
eign body infection models, a foreign body is implanted subcutaneously and
allow biofilms to grow on the inserted material. The animals can be infected
pre- or post-implantation. This model is well suited to compare the biofilm
formation capacity on different substrates. Additionally, inserting foreign
materials in the peritoneal cavity is performed to study chronic infections.
Other model systems involve inserting polyethylene tubes in the bladder
or endotracheal tube to develop, respectively, urinary tract and respiratory
tract infections. For a more comprehensive overview of the used in vivo
models, see Bjarnsholt et al. (2013) and Coenye and Nelis (2010). There
are also ex vivo tissue/organ models such as teeth, kidneys and human cell
lines, which may be used to study biofilm attachment to the tissues or cells.
These models are, for example, used to study wound infections, where bio-
films are formed on the wound surface (Rhoads, Wolcott, & Percival, 2008).
Although these extracted tissue cells are placed in an artificial environment,
they provide a good alternative to laboratory animals.
Because of the high biofilm diversity/complexity, today, there is no ideal
laboratory biofilm model that is representative for all biofilms and suitable for
biofilm susceptibility testing in order to develop anti-biofilm strategies. The
choice of a suitable model system is mainly dependent on the objective of the
biofilm study. Every model has strengths and weaknesses, which makes it
suitable for different applications (McBain, 2009). For an extensive over-
view of all in vitro and in vivo biofilm models, discussing their advantages
and disadvantages and their choice of application, see Lebeaux et al. (2013).
264 Joke Donné and Sylvia Dewilde

4.2 Biofilm Detection, Identification and Quantification

Methods
To analyse biofilm growth in an in vitro or in vivo model system, it is necessary
to measure the dimension/proportion of biofilm formation. Different
approaches have been developed to visualise and quantify attached biofilm
populations. They can be classified into biofilm biomass assays (quantifica-
tion of matrix and both living and dead cells), viability assays (quantification
of viable cells) and matrix assays (quantification of matrix). The most com-
monly used quantification techniques for microbial biofilms are listed in
Table 6 (Coenye & Nelis, 2010; Peeters, Nelis, & Coenye, 2008). These
staining methods are biofilm-destructive, but non-destructive visualisation
of biofilms can be achieved by, for instance, CLSM. Besides microscopy,
analytical techniques such as Raman spectroscopy, FTIR spectroscopy
and magnetic resonance imaging have increased our knowledge of biofilm
structure (Høiby et al., 2015).
The most popular biofilm biomass assay is crystal violet (CV) staining.
CV is a basic cationic dye and will stain both the biofilm bacteria and the

Table 6 Mostly Used Quantification Assays for Microbial Biofilms

Quantification Type of
Technique Quantification Based on
Crystal violet Biomass Binding of negatively charged surface molecules
(CV) stain and polysaccharides in the extracellular matrix
Syto9 stain Biomass Binding of intracellular and extracellular DNA
XTT dye Viability Reduction to formazan by metabolically active
cells
Resazurin Viability Reduction to pink, fluorescent resorufin by
metabolically active cells
FDA Reduction to yellow, fluorescent fluorescein by
metabolically active cells
DMMB Matrix Binding of sulphated polysaccharides in matrix
Conventional Viability Counting colony-forming units after plating
plate count biofilm cells
Life/dead PCR Viability Inhibition of PMA-modified DNA PCR
amplification
Flow cytometry Viability Difference in fluorescence intensity using
membrane-impermeable dyes
The Challenging World of Biofilm Physiology 265

extracellular matrix. The protocol involves fixation of the biofilms grown in

microtitre plates using methanol. After drying and washing, the stain is
solubilised using 33% glacial acetic acid and absorbance is measured at
590 nm (Peeters et al., 2008). The most widely used methods to analyse
the viability of a biofilm population are conventional plate counting and
the resazurin assay. Resazurin is a blue non-fluorescent compound that
can be reduced to pink resorufin, which is fluorescent. Resazurin is only
reduced by metabolically active, viable cells. This compound is generally
added to the biofilms in the wells, in combination with fresh medium,
and incubated for 2–4 h before measuring the fluorescent signal (Peeters
et al., 2008). The resazurin reduction kinetics are different for each strain,
require a certain bacterial density and can be influenced by the cultured
medium or testing compounds (Riss et al., 2004). Plate counting is a
time-consuming,labour-intensive and subjective method that only allows
determination of organisms that easily grow on solid medium and is as such
not a good technique for high-throughput analysis (Kerstens et al., 2014).
Flow cytometry, combined with the usage of membrane-impermeable dyes,
is a powerful and faster alternative for cell counts. A TO-PRO-3iodide-
based flow cytometry approach to isolate and quantitate viable and dead
S. aureus, E. coli or Bacillus subtilis cells has been described (Kerstens et al.,
2014). It is clear that parameters such as high-throughput ability, sensitivity
and rapid performance need to be taken into account when choosing the
assay to use.
Another alternative for viability assays is a PCR-based assay that discrim-
inates between live and dead bacteria in a community. Because propidium
monoazide (PMA), which can only enter cells with a damaged membrane,
strongly intercalates with DNA and inhibits PCR amplification, only DNA
from living cells will be amplified. Based on standard curves, made from
analysis of serial dilutions of bacterial suspensions, the viability can be deter-
mined (Clais et al., 2015; Nocker, Richter-Heitmann, Montijn, Schuren, &
Kort, 2010). Sheridan and co-workers noted that mRNA has a half-life of
only a few minutes, indicating that the detection of mRNA levels by quan-
titative RT-PCR could also be a powerful tool to determine the amount of
living cells (Sheridan, Masters, Shallcross, & Mackey, 1998).
Before applying any of the above-mentioned biofilm analysis methods,
there is the need for a good sample preparation. This includes washing and
collecting the cells. The adherent multilayered biofilms are usually washed
with PBS to remove non-adherent cells and to keep only biofilm cells for
detection and phenotyping. In order to maintain the integrity of the biofilm
266 Joke Donné and Sylvia Dewilde

structure, two aspects of the washing step are relevant to incorporate in every
protocol: (i) the number of washings and (ii) the methodology used for
washing (Stepanović et al., 2007). Two washing steps appear to be sufficient
to get rid of the planktonic cells and minimise biofilm impairment (Toté,
Vanden Berghe, Maes, & Cos, 2008). Concerning the techniques used
for washing, a variety of methods are applied, but the most simple and effec-
tive technique is pipetting carefully the medium out of the wells, while
slightly lifting the plate (Deighton, Capstick, Domalewski, & Van
Nguyen, 2001; Stepanović et al., 2007). However, monitoring the biofilm
integrity during washing is very important and in order to be accurate, the
wells that lose visible clusters of biofilm cells should be excluded from further
calculations (Deighton et al., 2001). Collecting biofilm cells from well plates
can be done using cell scrapers or sonication (Bjerkan, Witsø, & Bergh,
2009). However, when biofilm bacteria are separated from the surface they
were attached to, they change their physiology and become planktonic cells
again when back in suspension (Kaplan, 2010; Kostakioti,
Hadjifrangiskou, & Hultgren, 2013). As a consequence, analysis of the iso-
lated biofilm cells has to be done immediately, before gene expression
variations arise.
A last comment about biofilm analysis refers to several applied micro-
scopic techniques, from electron microscopy to fluorescence microscopy,
which are used for the detection and visualisation of biofilms. During such
applications, it is important to standardise the position in the well where the
structure and thickness of the biofilm population will be analysed since the
biofilm morphology at the centre (mid-point) of the well will differ signif-
icantly from the edges. The flow characteristics in well plates were analysed
in order to link hydrodynamics to the behaviour of cell cultures using com-
putational simulations (Salek, Sattari, & Martinuzzi, 2012). They concluded
that the fluid dynamics and wall shear stress distribution within a cylindrical
well undergoing orbital shaking are controlled by the liquid medium vol-
ume, the orbital radius of gyration and angular speed. The influence of fluc-
tuating shear stresses on the S. aureus biofilm morphology and tolerance to
ABs, using a six-well plate agitated by an orbital shaker, has been reported
(Kostenko, Salek, Sattari, & Martinuzzi, 2010). They observed that the bio-
film cell density increases significantly and proportionally to the radial dis-
tance from the centre, when the microtitre plates were agitated at 100 rpm.
Also the 3D structures of the biofilms became thicker at the exterior of the
edges. Resistance to ABs was correlated with the morphology of the bacteria
and was mostly higher in regions with high shear stress.
The Challenging World of Biofilm Physiology 267

4.3 Comparative -Omics Studies

Numerous genomic and proteomic studies have been performed to unravel
the molecular pathways involved in the different stages of the biofilm devel-
opment. Important studies are comparative set-ups which evaluate the func-
tional difference between a bacteria living in a biofilm and a planktonic
bacterium. Such comparative approaches have taught us that the biofilm
phenotype is extremely different from their planktonic counterparts (De
Angelis, Siragusa, Campanella, Di Cagno, & Gobbetti, 2015; Mukherjee,
Ow, Noirel, & Biggs, 2011; Post et al., 2014; Ren, Bedzyk, Thomas,
Ye, & Wood, 2004; Tremoulet, Duche, Namane, Martinie, &
Labadie, 2002).
Genomic studies are mostly based on DNA microarrays and have
resulted in differentially expressed gene sets between biofilm and planktonic
organisms. These cDNA-based studies revealed that there is no common
gene expression pattern for ‘biofilms’ (Sauer, 2003). However, some general
differences between the two bacterial lifestyles are revealed (e.g. higher cell
density, lower oxygen levels and higher osmolarity in biofilms than in a
planktonic lifestyle (Prigent-Combaret, Vidal, Dorel, & Lejeune, 1999)).
Differential genes in mRNAs of planktonic and 6-h biofilms of Pseudomonas
putida were identified. Some genes encode for structural components of fla-
gella and type IV pili, and other differential genes are involved in polysac-
charide biosynthesis. With immunoblot analysis of FliC, they validated the
presence of flagella in planktonic cultures, but as expected flagella were not
expressed in 12 or 24 h biofilms (Sauer & Camper, 2001). Other studies con-
firmed flagellum repression in biofilm populations (Pratt & Kolter, 1998;
Prigent-Combaret et al., 1999). Furthermore, Beloin et al. made an inter-
esting comparison between biofilm cells, cells in exponential planktonic
phase and cells in stationary planktonic phase with the aim of identifying true
biofilm-specific genes using DNA microarrays (Beloin et al., 2004). They
demonstrated that among the 64 genes found to be the most induced
(twofold ratio) in biofilm versus exponential phase, 39 of them were
not induced in the biofilm condition when compared with stationary phase.
These 39 genes (for instance, genes involved in cell division and ion trans-
port and genes coding for fimbriae components) are not biofilm specific but
rather related to a stationary growth phenotype within the mature E. coli
biofilm. The remaining 25 genes were also overexpressed in biofilm versus
stationary growth phase, defining a set of biofilm-specific gene expressions.
These genes encode mainly membrane proteins, regulators involved in
envelope stress and carbohydrate transporters. Other studies confirm the fact
268 Joke Donné and Sylvia Dewilde

that bacteria develop stress responses within biofilms (Ferrières & Clarke,
2003; Otto & Silhavy, 2002; Wang et al., 2011). To validate the differen-
tially expressed genes functionally, analysing the biofilm phenotype of spe-
cific mutants is a widely used approach. Most genes overexpressed in mature
E. coliK-12 biofilms, such as recA, cpxAPR, pspF and msrA, were not involved
in the early steps of the biofilm formation process (Beloin & Ghigo, 2005).
The mature biofilm-specific gene mutants were not able to form a mature
biofilm, but their adhesion capacities could not be distinguished from the
wild-type strain (Beloin et al., 2004). On the other hand, pgaABCD muta-
genesis experiments in E. coli showed that synthesis of the polymeric sub-
stance PGA promotes adhesion and consequently biofilm formation
(Wang, Preston, & Romeo, 2004).
Many studies comparing different ages of biofilms have been performed
(Domka, Lee, Bansal, & Wood, 2007; Park et al., 2014; Resch et al., 2006).
The gene expression in E. coliK-12 biofilms of different ages (4, 7, 15 and
24 h), grown in Luria–Bertani(LB) medium on glass wool was analysed
(Domka et al., 2007). Colonic acid genes are only induced in mature
24-h-old biofilms, while fimbriae-related genes were induced in biofilms
of all ages. These transient gene expression profiles can be considered as
snapshots of the biofilm developmental process, which is a better method
than analysing only the mature biofilms, but still less realistic than in vivo
approaches. Besides transient alterations in the process of biofilm formation,
there is also a spatial phenotypic variation within the biofilm population
itself (discussed in Section 2.1.2). Studying localised expression could give
more realistic insights into the composition of the multilayered community.
Using fluorescent reporter genes or fluorescent staining methods coupled to
fluorescence microscopy are appropriate techniques for visualising and
mapping biofilm heterogeneity in vivo. An overview of some
fluorescence-based methods to measure respiratory activity, membrane per-
meability, species diversity and gene-specific expressions within biofilms
was published (Stewart & Franklin, 2008). As an example, Finelli et al. used
in vivo expression technology to screen a reporter gene library and to select
for those clones in which gene expression is upregulated in P. aeruginosa
grown to a 5-day-old biofilm (Finelli, Gallant, Jarvi, Lori, & Burrows,
2003). They used promoterless gene-fusion constructs that complement
an adenine auxotrophic mutation (purEK deletion) to identify promoters
that are induced in vivo in the biofilm by plating out biofilm cells on
adenine-containing medium (Angelichio & Camilli, 2002). An interesting
local transcriptomics analysis (Williamson et al., 2012) reported a
The Challenging World of Biofilm Physiology 269

microarray and quantitative real-time PCR (qPCR) experiment on two

separate layers of a 72-h-oldP. aeruginosa biofilm. Through cryo-based-
micro-dissection, they were able to compare a biofilm layer 30 μm from
the top with a population that grows 30 μm from the bottom from a total
biofilm of 350 μm thick. At the top of the biofilm, the cells were actively
dividing and showed high mRNA levels for genes regulated by the
hypoxia-induced regulator Anr. Slow-growing cells deep in the biofilm
showed little expression of Anr-regulated genes and thus may have sensed
anoxia.
Combined localised and transient gene expression analysis can be per-
formed by using gene-targeted GFP fusions to follow live, via in vivo fluo-
rescence microscopy, specific gene expressions in the subpopulations of the
biofilm architecture (Beloin & Ghigo, 2005). A detailed study of the spatio-
temporal expression of the transcriptional regulator agr, important in biofilm
formation and virulence, in Listeria monocytogenes during non-motile growth
in two different nutritional environments was reported (Rieu et al., 2008).
They revealed that agr was expressed preferentially in cells located outside
the microcolonies, but not in static growth circumstances.
As the high AB tolerance characteristic of biofilm populations is still a
limiting factor, an underutilised approach in this research area is a drug-
induced comparative biofilm set-up. For example, the transcriptional
response of sessile Burkholderia cenocepacia cultures to treatment with the
AB chlorhexidine was analysed (Coenye et al., 2011). Upregulated genes
in the treated biofilms encoded membrane-related proteins, efflux pumps
and chemotaxis- and motility-related proteins. This, together with the
downregulation of an adhesin-encoding gene, they hypothesised that sessile
cells try to escape the biofilm when treated with AB. Also other induction-
based experiments can be advantageous in exposing specific regulatory net-
works in biofilms, for example, induction by AI, acidity, oxygen supply,
reactive oxygen species or even components of the immune system like
cytokines (Kanangat, Postlethwaite, Cholera, Williams, & Schaberg,
2007; McLaughlin & Hoogewerf, 2006; Villa et al., 2012; Welin
et al., 2003).
Remarkably, the small overlap in up- and downregulated genes between
different studies, together with the very small fraction of the total bacterial
genome that experience variations in expression, demonstrates that biofilm
formation is a complex developmental process. In addition, because proteins
are a better representation of the bacterial activity; proteomics is increasing
in popularity. Table 7 lists all proteomic studies based on E. coli biofilms.
270 Joke Donné and Sylvia Dewilde

Table 7 Studies of E. coli Biofilm Proteome

Biofilm
Growth Techniques
Research Goal Conditions Employed Result References
Spatial UPEC Imaging MS Type 1 pili Floyd et al.
proteome of cystitis localised to the air- (2015)
biofilm isolate exposed region,
UTI89, curli fibres to the
ITO- air–liquid interface
coated glass
slides,
YESCA
broth, 48 h
Effect cell EHEC 2D gel, 7 differential Kim, Lee,
extract of O157:H7, Coomassie Blue proteins including Kang, Oh,
Bifidobacterium microtitre staining, formation of iron– and Griffiths
spp. on plate, LB, MALDI-TOF/ sulphur protein (2012)
biofilm 24 h MS (NifU), thiol:
proteome disulphide
interchange
protein (DsbA) and
flagellar P-ring
protein (FlgI)
Outer E. coli 1D gel, Unique to the Mukherjee,
membrane MG1655, Coomassie Blue biofilm growth are Karunakaran,
proteins glass wool, staining, LC– proteins involved and Biggs
biofilm versus LB MS/MS in stress resistance, (2012)
planktonic medium, metabolism of
24 h carbohydrates and
amino acids and
ATP synthesis
Biofilm versus E. coli Gel free, Differential Mukherjee
planktonic MG1655, iTRAQ, LC– proteins involved et al. (2011)
glass wool, MS/MS in acid resistance,
LB DNA protection
medium, and binding and
24 h ABC transporters
Impact of rpoS E. coli 2D gel, silver 35 rpoS-regulated Collet et al.
deletion on ZK126, staining, proteins, some of (2008)
proteome glass wool, MALDI- them dependent on
biofilms versus MSM, TOF/MS and growth condition
planktonic 3 days LC–MS/MS
The Challenging World of Biofilm Physiology 271

Table 7 Studies of E. coli Biofilm Proteome—cont'd

Biofilm
Growth Techniques
Research Goal Conditions Employed Result References
Impact indole E. coliS17- 2D gel, silver 14 proteins Collet et al.
on biofilm 1,6-well staining, overexpressed, (2007)
versus plate, LB, MALDI-TOF/ 37 underexpressed
planktonic 20 h MS,N-terminal in biofilms in the
sequencing presence of indole
compared to
planktonic
Intracellular STEC 2D gel, silver Role of periplasmic Kim et al.
proteome of O157:H7, staining, antioxidant systems (2006)
planktonic glass fibre MALDI-TOF/ (SodC and Tpx) in
versus biofilm filter, M9 MS biofilm growth
agar, 5 days
Biofilms on 3 E. coli XL1 2D gel, silver Upregulation of Orme,
types of blue and staining, outer membrane Douglas,
hydrogel- E. coli LC-MS/MS protein A (OmpA) Rimmer, and
coated petri U125544, in biofilms Webb (2006)
dishes versus hydrogel- compared to
planktonic coated planktonic
petri dish,
YT, 16 h
Biofilm versus E. coli 2D gel, 17 differential Tremoulet
planktonic O157:H7 Coomassie Blue proteins, including et al. (2002)
NCTC staining, general metabolism
12900, MALDI-TOF/ proteins, sugar and
glass fibre MS amino acid
filter, M9 transporters and
agar, 7 days regulator proteins
Based on Pubmed key word searches ‘Proteomics and Escherichia coli and Biofilm’ and ‘Proteome and
Escherichia coli and Biofilm’.

Gel-based (e.g. two-dimensional difference gel electrophoresis) as well as

gel-free (e.g. isobaric tag for relative and absolute quantitation (iTRAQ),
stable isotope labelling of amino acids in cell culture, isotope-coded affinity
tag (ICAT)) approaches are employed (Mukherjee et al., 2011; Petrova &
Sauer, 2009; Phillips et al., 2012; Post et al., 2014; Rathsam, Eaton,
Simpson, Browne, Valova, et al., 2005). To improve the protein coverage
of proteomics in bacterial biofilms, pre-fractionation is a commonly applied
technique. Proteins extracted from different cellular compartments are
272 Joke Donné and Sylvia Dewilde

analysed separately, which gives a more detailed view of the protein distri-
bution in the biofilm structure. A technical paper compares five different
fractionation protocols for isolating outer membrane, cytoplasmic mem-
brane, periplasmic and cytosolic proteins and evaluated the separation qual-
ity and the suitability of the samples for mass spectrometric analysis (Thein,
Sauer, Paramasivam, Grin, & Linke, 2010). Washing steps with chaotropic
reagents increased the purity and the amount of isolated outer membrane
proteins (36 proteins) as would be expected. This method showed, for
example, a contamination of the isolated outer membrane fraction of 37%
with soluble cytoplasmic proteins, which were mostly ribosomal proteins.
Other methods resulted in 42%, 43%, 55% and 57% cytoplasmic contami-
nants. Another study used a pre-fractionation of cell wall-,membrane- and
cytoplasmic proteins of both planktonic and biofilm Streptococcus mutans cells
to compare both phenotypes by gel-based mass spectrometry (MS) analysis.
Their method could hardly separate the cytoplasmic fraction from the mem-
brane fraction, resulting in very few cytoplasmic protein spots (<100), while
between 200 and 600 membrane protein spots (Rathsam, Eaton, Simpson,
Browne, Berg, et al., 2005).
Nonetheless, only a small proportion (<10%) (Mukherjee et al., 2011;
Schembri, Kjaergaard, & Klemm, 2003) of the genome and proteome
undergoes changes in expression levels when switching from non-biofilm
mode of growth to biofilm mode of growth. It is assumed that the key proteins
are regulated in quality instead of quantity. That is why post-translational
modifications (phosphorylation, glycosylation and nitrosylation) of selected
or all proteins should be enclosed in future research. Petrova and Sauer
published in 2009 a phosphoproteomic analyses during P. aeruginosa biofilm
development (Petrova & Sauer, 2009). Phosphorylated proteins of planktonic
cells and of five different ages of biofilms were enriched by metal oxide affinity
chromatography and detected by both 2D gel-based immunoblot analysis
with anti-Phospho-(Ser/Thr)Phe antibodies and LC–MS/MS analysis with
ICAT labelling. They identified three undescribed two-component systems
(BfiSR harbouring an RpoD-like domain, an OmpR-like BfmSR and
MifSR) which were sequentially phosphorylated during biofilm formation.
Moreover, this phosphoproteomic study is also a good example of a transient
biofilm analysis. When biofilm development proceeds, the local chemical
environments and thus also the bacterial behaviours and/or genetic profiles
change. Consequently, it is important to identify transient gene/protein
expression profiles in order to completely unravel the process of biofilm
formation.
The Challenging World of Biofilm Physiology 273

Next to genomic, transcriptomic and proteomic analyses, metabolome

analysis or metabolomics maps the final output of the biological function
and is directly related to pathogenesis. Measuring the produced metabolites
can tell us which pathways are truly active in the cellular population. Cap-
illary electrophoresis (CE) combined with time-of-flight mass spectrometry
(TOF-MS), or CE–MS, is excellent in separating ionised small molecules
and has been proven suitable to divide, identify and quantify metabolic
intermediates (Takahashi, Washio, & Mayanagi, 2012). This was proven
by a study where the metabolites from the carbohydrate pathways (central
carbon metabolism, EMP pathway, pentose phosphate pathway and Krebs
cycle) from supragingival plaque microflora from teeth were examined by
CE–MS (Takahashi, Washio, & Mayanagi, 2010). Junka and co-workers
searched differences in intracellular metabolite patterns of S. aureus biofilm
versus planktonic forms ( Junka et al., 2013). They used nuclear magnetic
resonance spectroscopy to measure four significantly different metabolites
between these conditions, namely elevated levels of isoleucine, 2,3-
butanediol and alanine in biofilm cells and a higher glycine-betaine concen-
tration in planktonic cells. The higher levels of the anaerobic fermentation
product 2,3-butanediol in biofilm cells for instance support the hypothesis
that biofilms experience anoxic conditions. This is also confirmed by the
study where the comparison of planktonic and biofilm cells from
Acinetobacter baumannii (Yeom, Shin, Yang, Kim, & Hwang, 2013) revealed
increased levels of intracellular metabolites acetate, pyruvate and succinate in
the sessile fraction that may explain pyruvate fermentation in the mature
biofilm population. Also UDP-glucose, AMP, glutamate and lysine were
increased in the biofilm condition compared to planktonic cells. A recent
review includes information on designing a metabolomics experiment
(Zhang & Powers, 2012).

4.4 Experimental Design

It is difficult to standardise the process of biofilm formation in a laboratory
set-up. Therefore, it is important to define the different parameters that
affect the biofilm formation capacity in the models being used. The exper-
imental set-up is crucial to the outcome of the biofilm experiment and its
interpretation.

4.4.1 Growth Conditions

First of all, the growth conditions chosen to produce biofilms, such as the
surface used to grow on, the growth medium and the type of bacterial/yeast
274 Joke Donné and Sylvia Dewilde

strain(s), are crucial in the standardisation of the biofilm formation process in

a laboratory set-up.

4.4.1.1 Surface
As discussed earlier, the properties of the surface, where the biofilm is
attached to, play an important role. The chemical composition of the mate-
rial, but also the surface area where the biofilm adheres, will determine the
biofilm growth. Working with, for instance, a 96- or a 24-well plate as an
in vitro model to produce biofilms in a laboratory environment determines
the final biofilm phenotype, even when growth is started from identical
inocula. Additionally, the crude material of the surface influences the
biofilm-forming capacity as the roughness and hydrophobicity of the surface
affect the bacterial attachment phase. A comparison of stainless steel (hydro-
philic), glass (hydrophilic) and polyvinyl chloride (hydrophobic) surfaces to
grow Salmonella spp. biofilms, concluded that glass was the best surface mate-
rial to diminish biofilm production of that species (De Oliveira et al., 2014).
Additionally, a surface exposed in a liquid medium will inevitably and rather
immediately (within seconds) become conditioned or coated by compo-
nents from that surrounded medium, which changes the physicochemical
properties (roughness, hydrophobicity, chemical composition) of the sur-
face, affecting bacterial attachment. Such conditioning films facilitate bacte-
rial attachment due to their polar character, which is a stimulus for biofilm
formation (Garrett, Bhakoo, & Zhang, 2008). Additional oxidations and
hydrations due to cell appendage interactions even induce irreversible adhe-
sion (Garrett et al., 2008). As an example, fibrinogen-coated polymeth-
ylmethacrylate coverslips promoted adherence for several clinical isolated
S. aureus strains to a higher extent than laminin-coated surfaces
(Herrmann et al., 1988). Addition of bovine serum albumin to periwinkle
wilt culture medium will also enhance the formation of a conditioning film
on a glass surface (Lorite et al., 2011).

4.4.1.2 Effect of Nutrients and Oxygen

Environmental conditions such as the available nutrients, oxygen and vita-
mins, the pH and the temperature of the surrounding in which the bacteria
are living influence the efficacy of biofilm growth (Garrett et al., 2008). For
example, our own experiments comparing 1640 and tryptic soy broth (TSB)
indicated that E. coli ATCC10536 produce abundantly more biofilm in 1640
medium than in TSB medium (Fig. 10). We suppose that this is due to the
The Challenging World of Biofilm Physiology 275

3.5

3
OD570 nm 2.5

1.5 TSB
RPMI 1640
1

0.5

0
24 48 72
Incubation time (h)
Figure 10 The effect of medium on biofilm formation. E. coli ATCC10536 biofilms were
grown for 24, 48 and 72 h in 96-well plates (N ¼ 24) in two different media, TSB and 1640.
They are inoculated with 105 CFU/ml. Error bars represent the standard deviation. Bio-
film mass was determined using crystal violet staining (unpublished data).

presence of high vitamin concentrations and a reducing agent, glutathione,

in the 1640 medium.
The effect of growth media was confirmed by Labrie et al. where
Actinobacillus pleuropneumoniae biofilm formation, grown in five different cul-
ture media, was analysed by comparing the biofilm mass and the transcrip-
tional profile (Labrie et al., 2010). They observed no biofilm formation for
the A. pleuropneumoniae reference strain S4074 after 24 h of growth in a
96-well plate in LB medium, while there was a slight biofilm formation
in wells with TSB, Mueller Hinton and brain heart infusion-A(BHI-A)
medium. They tested BHI broth from two different suppliers, called
BHI-A and BHI-B, with the latter showing a much higher biofilm forma-
tion than the former, which can be explained by the higher concentrations
of zinc in BHI-A, which inhibits biofilm formation. Transcriptional profiles
comparing A. pleuropneumoniae biofilms in these media demonstrated an
upregulation of autotransporter adhesin genes and genes involved in matrix
biosynthesis and zinc transport after growth in BHI-B. A study evaluating
the effects of growth medium, temperature and incubation time on biofilm
formation by Enterobacter cloacae strains using a microtitre plate model
system (Nyenje, Green, & Ndip, 2013) showed stronger biofilm formation
when incubated at 37 °C compared to 25 °C, which is expected due to
the faster growth rate of bacteria at higher temperatures. However,
for Stenotrophomonas maltophilia, a higher biofilm mass was measured at
32 °C compared to 37 °C (Di Bonaventura, Stepanović, Picciani,
276 Joke Donné and Sylvia Dewilde

Pompilio, & Piccolomini, 2007). It is postulated that the expression of bac-

terial surface appendages as pili and flagella that has an impact on biofilm for-
mation is temperature dependent (Herald & Zottola, 1988). The attachment
of L. monocytogenes isolates to stainless steel at three different temperatures,
10, 21 and 35 °C, resulted in cells with fibrils only at 21 °C (Herald &
Zottola, 1988). Thus, depending on the strain and species, an optimum
growing temperature will favour the bacterial growth efficiency and conse-
quently biofilm formation (Garrett et al., 2008).
Protein expression and bacterial growth are also influenced by the intra-
cellular and extracellular pH of the bacteria. pH changes disturb several met-
abolic processes, leading to an alteration of the biofilm development (Garrett
et al., 2008). Many papers report the effect of temperature and pH on biofilm
formation (Del Carpio-Perochena et al., 2015; Di Bonaventura et al., 2007;
Hostacká, Ciznár, & Stefkovicová, 2010; Marsh, 2009; Nostro et al., 2012),
but no general conclusion can be drawn, since the effect is again species
dependent.
Changing the agitation velocity can also change the lifestyle of the bac-
teria, from free-living cells at a very high rotation speed to sticky biofilm cells
when no or slow shaking is used. For producing mature biofilms in micro-
titre plates, we have demonstrated that a slow horizontal shaking at 25 rpm
promotes more biofilm formation than no agitation at all (unpublished data)
(Fig. 11). We suppose that biofilm formation will be enhanced by low-speed

0.600

0.500

0.400
OD570 nm

0.300

0.200

0.100

0.000
Shaking 25 rpm No shaking
Figure 11 Effect of agitation on biofilm formation. E. coli BW25113 biofilms were grown
in 1640, in 24-well plates (N ¼ 20) for 72 h at 37 °C with or without shaking at 25 rpm.
The biofilm mass is quantitated with 0.1% CV staining. The error bars represent the stan-
dard deviation (unpublished data).
The Challenging World of Biofilm Physiology 277

shaking because it supports a uniform dispersion of nutrients and oxygen.

However, some researchers showed that mechanical agitation at high speed
can partly remove biofilm populations and thus enhances dispersal ( Jiang,
Pei, & Hu, 2010). The effect of agitation in in vitroset-ups can be extrapo-
lated to the effect of flow rates in in vivo situations. The effects of these flow
rates on the biofilm-forming capacity can be investigated in a laboratory sys-
tem with dynamic models. It is reported that E. coli biofilm formation was
favoured at low flow rates in a flow cell system. It is hypothesised that the
shear stress effects were playing a more important role than mass transfer
(Moreira et al., 2013).

4.4.1.3 Strain
The above discussed factors affecting biofilm development are species
dependent. Every bacterial species exhibits a different biofilm phenotype,
probably due to morphology and metabolism. Some microorganisms pro-
duce a more ‘slimy’ matrix than others (Silva et al., 2009), and some bacterial
biofilms grow more in height, some more in width. But even within the
same species group, there are differences. For example, the ATCC10536
E. coli show a much stronger biofilm phenotype than E. coli BW25113, based
on its total biofilm mass (unpublished data), even when using the same inoc-
ulation size. In addition, the use of inocula derived from a culture medium, a
nutrient agar or a cryogenic stock can influence the biofilm production. All
tested E. cloacae isolates were able to produce a biofilm, cultured for 24 h in
BHI medium in 96-well plates at 37 °C, when starting from a nutrient agar
inoculum, while only 93% of the isolates from culture medium were able to
form a biofilm under the same growth conditions (Nyenje et al., 2013). It is
also important to inoculate single cells and thus avoid cell clusters to diminish
false-positive results. Such bacterial clusters will experience a small advan-
tage over single cells to form biofilms. Consequently, thoroughly mixing
the cell suspension is necessary before inoculation (Stepanović et al., 2007).

4.4.2 Expression Profiles

The gene expression profiles are dependent on the experimental set-up. As a
comprehensive example, two separate DNA microarray-based studies (Ren
et al., 2004; Schembri et al., 2003), which both compare the gene expression
profiles of E. coli as a biofilm phenotype and as planktonic cells, will be dis-
cussed. Ren and co-workers identified only 19 altered genes (induced and
repressed) in E. coliK-12 biofilms, compared to planktonic cells, while
Schembri et al. reported hundreds of differential genes. Remarkably, both
278 Joke Donné and Sylvia Dewilde

researchers used the same statistical t-test analysis, reporting genes as signif-
icantly differential when fold changes were greater than 2.5 and p-values less
than 0.05. There were less than 10 differentially expressed genes in common,
based on their published data. Such incomparability is likely due to differ-
ences in growth conditions, the type of strain and the age of the biofilm. Ren
et al. grew E. coli K-12 ATCC25404 biofilms in LB medium on metal plates
in continuous reactors for 5 days, while Schembri et al. used a four-channel
flow system to form E. coli K-12 MG1655 biofilms of 42 h old in 3-(N-
morpholino)propanesulphonic acid (MOPS) minimal media supplemented
with 0.2% glucose. The influence of the experimental design on the genetic
character of the biofilm is also relevant in differential proteomic studies.
Twelve identified and quantified proteins were determined in an iTRAQ
study that were significantly increased or decreased in abundance between
biofilm and planktonically grown E. coli cells (Mukherjee et al., 2011).
There were only a few differential proteins in common with the experi-
ments of Collet et al., which can be partly explained by their different bio-
film growth conditions (E. coliS17-1 biofilms grown in LB in a six-well plate
for 20 h versus E. coli MG1655 biofilms grown in LB on glass wool for 24 h).
These two studies also differ in their used proteomic approach: a gel-based
method versus a study based on peptide labelling (Table 7) (Collet et al.,
2007; Mukherjee et al., 2011).

4.5 Multispecies Biofilms

Biofilms in nature consist of multiple species. This characteristic is a serious
impediment in the progress of biofilm research. Limited experiments have
been performed on multispecies biofilms since many methodologies are not
able to distinguish between different types of cells within the biofilm pop-
ulation. It is crucial to identify and localise the different species in poly-
microbial biofilms. Viable plate count on selective and differential agars
(Table 8) can select for certain bacteria, but this is mostly a species-unspecific
detection. The most popular methods to identify the diversity of microor-
ganisms are based on ribosomal amplification, e.g., denaturing gradient gel
electrophoresis of PCR-amplified 16S rRNA or qPCR analysis with a
Taqman probe. FISH in combination with CLSM is another method to
identify and visualise microbial species in the multispecies biofilms using
species-dependent probes (Yang et al., 2011). A five-colour multiplex FISH
analysis is an interesting example (Al-Ahmad et al., 2007). Furthermore, it
was demonstrated that both qPCR levels and fluorescence microscopic
The Challenging World of Biofilm Physiology 279

Table 8 Some Selective and Differential Agars

Selective and Selected
Medium Differential Agents Organisms Differentiation
Blood agar Sheep blood cells Most Haemolytic bacteria, e.g.
bacteria – S. aureus: yellow colonies
– S. epidermidis: white pigment
Chocolate Haemoglobin and Haemophilus
agar supplements and Neisseria
species
Eaton’s agar Penicillin Mycoplasma
pneumoniae
Eosin Lactose, eosin and Gram- – Enterobacter aerogenes: large,
methylene methylene blue negative pink colonies with dark centre
blue agar enteric – E. coli: small, dark colonies
species with a green sheen
– Pseudomonas, Proteus,
Salmonella and Shigella sp.:
colourless colonies
Hektoen Lactose, sucrose, Gram-
Enteric agar salicin, bile salts, negative
sodium enteric
thiosulphate, ferric species,
ammonium especially
citrate, Salmonella
bromothymol and Shigella
blue, acid fuchsin
MacConkey’s Lactose, bile salts, Gram- – Lactose fermenting bacteria
agar neutral red, crystal negative (E. coli, Enterobacter, Klebsiella):
violet bacteria red/pink colonies
– Non-lactose fermenting
bacteria (P. aeruginosa,
Salmonella, Shigella, Proteus
sp.): white/colourless colonies
MRS agar Sodium acetate Lactobacillus
species
Mannitol salt 7.5% NaCl, Staphylococci
agar mannitol, phenol – S. aureus: yellow colonies
red – S. epidermidis: red colonies
Phenylethyl Phenylethyl Gram-
alcohol agar alcohol positive
bacteria
280 Joke Donné and Sylvia Dewilde

counts based on species-specific staining by FISH or immunofluorescence

yielded in higher values than colony forming units (CFU) counts on selec-
tive agars when quantifying species in polymicrobial biofilms (Ammann,
Bostanci, Belibasakis, & Thurnheer, 2013).
Interactions between different species (inter-species interactions) often
change the physiology of individual bacteria as well as the whole biofilm
community. Mostly these interactions are competitions for nutrients to
stimulate their own growth (Yang et al., 2011). In multispecies communi-
ties, the different bacteria can stimulate or inhibit each other’s biofilm-
forming ability comparing to the biofilm production by any of the single
species (Varposhti, Entezari, & Feizabadi, 2014; Wen, Yates, Ahn, &
Burne, 2010). If the cooperation between the species is synergistic or
non-synergistic depends mostly on their secreted molecules, which can
be for instance signal molecules, metabolic products or toxins (see
Sections 2.1.4). Aggregatibacter actinomycetemcomitans inhibits C. albicans bio-
film formation by means of the signalling molecule AI-2 that was secreted
during growth. C. albicans biofilm formation increased when they were
co-cultured with the Gram-negative bacteria that were deficient in AI-2
production (Bachtiar et al., 2014). Besides the biofilm-forming capacity,
other physiological functions can alter in polymicrobial biofilms, e.g., viru-
lence and drug resistance. Some interesting examples are discussed below.
A study showed an increased biofilm phenotype when growing S. aureus
in the presence of C. albicans, compared to the mono-bacterial biofilm.
The S. aureus bacteria seemed to become coated in the secreted matrix of
the underlying yeast cells. In addition, a higher vancomycin resistance of
S. aureus was demonstrated when living in a dual-species biofilm
(Harriott & Noverr, 2009, 2010). Furthermore, a mixed-biofilm study of
S. mutans and Veillonella parvula showed no significant difference in biofilm
growth between single- and dual-species biofilms, grown for 72 h, based on
the viable count levels (Kara, Luppens, & Ten Cate, 2006). However, a
much higher lactic acid concentration was detected in the medium of
72-h-oldsingle-species biofilms of S. mutans compared to the dual-species
community, although the pH of the medium was similar. This could be
explained by the significant correlation between decreasing concentrations
of lactic acid and increasing concentrations of acetic acid. Another study
evaluated the phenazine toxin production of P. aeruginosa in the presence
of C. albicans. The signalling molecule farnesol produced by C. albicans stim-
ulates the production of phenazine toxins in P. aeruginosa clinical isolates.
Even bacterial mutants that lack the QS regulator responsible for phenazine
The Challenging World of Biofilm Physiology 281

expression regained their toxin-producing ability when cultured with

C. albicans (Cugini, Morales, & Hogan, 2010). To conclude, many dual-
species biofilm research studies can be found in the literature, but multi-
species biofilm experiments are less frequently reported (Cavalcanti et al.,
2015; Da Silva Fernandes, Kabuki, & Kuaye, 2015; Kim & Izadjoo,
2015). It is important in biofilm research to have a better understanding
of multispecies biofilms and find ways to manipulate the biofilm properties,
especially its virulence. Effective biofilm inhibitory compounds, based on
single-species experiments, are probably less antibacterial to multispecies
biofilms because many more factors are involved. Some attempts to
approach a more realistic scenario have been made by using spent growth
medium during biofilm formation. In this way, secreted molecules from
other bacteria or yeasts are incorporated in the biofilm development. For
example, Kim et al. screened the spent media from 4104 separate actinomy-
cetes strains in order to identify anti-biofilm strains against P. aeruginosa (Kim
et al., 2012). The spent media of two strains, namely Streptomyces sp. BFI 230
and Kribbella sp. BFI 1562, inhibited P. aeruginosa biofilm formation by 90%
without affecting the growth of planktonic P. aeruginosa cells. Transcriptome
analysis revealed an interference with the iron acquisition by P. aeruginosa.
Another study (Bandara, Cheung, Watt, Jin, & Samaranayake, 2013) inves-
tigated the effects of E. coli biofilms supernatants of different ages on the
biofilm-forming capacity of Candida species and demonstrated a significant
inhibitory effect due to the E. coli secretory elements. Another study
reported future research ideas concerning spent medium analysis.
Seventy-nine cell-free culture supernatants from a variety of single oral
streptococci was screened to identify extracellular compounds that inhibit
biofilm formation by the oral anaerobe Porphyromonas gingivalis strain 381
(Christopher, Arndt, Cugini, & Davey, 2010). These authors discovered
a 48-kDa arginine deiminase protein that was responsible for inhibition.
By using qPCR, it was concluded that P. gingivalis can sense this extracellular
protein, produced by an oral Streptococcus intermedius, and respond by down-
regulating the expression of cell-surface appendages.
Before producing multispecies biofilms in a laboratory for experimental
research, we want to firstly note that the order in which the organisation of a
multispecies biofilm happens in nature is unknown. That is why simulta-
neous inoculation of the several species in a lab model system is a good alter-
native. For example, S. aureus bacteria nestle in the secreted matrix and the
hyphae of the underlying C. albicans yeast cells, when culturing them
together (Harriott & Noverr, 2009, 2010). Secondly, distinguishing
282 Joke Donné and Sylvia Dewilde

between the identified genes or proteins derived from the different species is
extremely hard to accomplish, as well as discovering the cause of the differ-
entially expressed genes/proteins by each species. An outstanding technical
paper demonstrated a powerful technique to investigate the complexity of
transcriptional changes in mixed-species biofilms, which is based on
multiplex-labelled cDNA hybridisations with a NimbleGen DNA array
(Redanz, Standar, Podbielski, & Kreikemeyer, 2011). With a 385 KTM
chip, 385,000 gene probes could be tested allowing coverage of several bac-
terial genomes on one array.
To conclude, these concerns of species identification, inter-species inter-
actions, species-specific expression profiling and inoculation order hinder
researchers to venture to competitive polymicrobial biofilm research.

REFERENCES
Abed, S. E., Ibnsouda, S. K., Latrache, H., & Hamadi, F. (2012). Scanning electron micros-
copy (SEM) and environmental SEM: Suitable tools for study of adhesion stage and bio-
film formation. In V. Kazmiruk (Ed.), Scanning electron microscopy (pp. 717–730).
INTECH Open Access Publisher.
Al-Ahmad, A., Wunder, A., Auschill, T. M., Follo, M., Braun, G., Hellwig, E., et al. (2007).
The in vivo dynamics of Streptococcus spp., Actinomyces naeslundii, Fusobacterium nucleatum
and Veillonella spp. in dental plaque biofilm as analysed by five-colour multiplex fluores-
cence in situ hybridization. Journal of Medical Microbiology, 56, 681–687.
Amalaradjou, M., & Venkitanarayanan, K. (2013). Role of bacterial biofilms in catheter-
associated urinary tract infections (CAUTI) and strategies for their control.
In T. Nelius (Ed.), Recent advances in the field of urinary tract infections (pp. 1–15).
INTECH Open Access Publisher.
Ammann, T. W., Bostanci, N., Belibasakis, G. N., & Thurnheer, T. (2013). Validation of a
quantitative real-time PCR assay and comparison with fluorescence microscopy and
selective agar plate counting for species-specific quantification of an in vitro subgingival
biofilm model. Journal of Periodontal Research, 48, 517–526.
Anderl, J. N., Franklin, M. J., & Stewart, P. S. (2000). Role of antibiotic penetration lim-
itation in Klebsiella pneumoniae biofilm resistance to ampicillin and ciprofloxacin. Antimi-
crobial Agents and Chemotherapy, 44(7), 1818–1824.
Angelichio, M. J., & Camilli, A. (2002). In vivo expression technology. Society, 70(12),
6518–6523.
Aparna, M. S., & Yadav, S. (2008). Biofilms: Microbes and disease. The Brazilian Journal of
Infectious Diseases: An Official Publication of the Brazilian Society of Infectious Diseases, 12,
526–530.
Bachtiar, E. W., Bachtiar, B. M., Jarosz, L. M., Amir, L. R., Sunarto, H., Ganin, H.,
et al. (2014). AI-2 of Aggregatibacter actinomycetemcomitans inhibits Candida albicans biofilm
formation. Frontiers in Cellular and Infection Microbiology, 4, 94.
Baharoglu, Z., & Mazel, D. (2014). SOS, the formidable strategy of bacteria against aggres-
sions. FEMS Microbiology Reviews, 38, 1126–1145.
Bandara, H. M. H. N., Cheung, B. P. K., Watt, R. M., Jin, L. J., & Samaranayake, L. P.
(2013). Secretory products of Escherichia coli biofilm modulate Candida biofilm forma-
tion and hyphal development. Journal of Investigative and Clinical Dentistry, 4, 186–199.
The Challenging World of Biofilm Physiology 283

Battesti, A., Majdalani, N., & Gottesman, S. (2011). The RpoS-mediated general stress
response in Escherichia coli. Annual Review of Microbiology, 65, 189–213.
Belas, R. (2014). Biofilms, flagella, and mechanosensing of surfaces by bacteria. Trends in
Microbiology, 22, 517–527.
Beloin, C., & Ghigo, J. M. (2005). Finding gene-expression patterns in bacterial biofilms.
Trends in Microbiology, 13(1), 16–19.
Beloin, C., Roux, A., & Ghigo, J.-M. (2008). Escherichia coli biofilms. Current Topics in Micro-
biology and Immunology, 322, 249–289.
Beloin, C., Valle, J., Latour-Lambert, P., Faure, P., Kzreminski, M., Balestrino, D.,
et al. (2004). Global impact of mature biofilm lifestyle on Escherichia coliK-12 gene
expression. Molecular Microbiology, 51, 659–674.
Bjarnsholt, T., Alhede, M., Alhede, M., Eickhardt-Sørensen, S. R., Moser, C., Kühl, M.,
et al. (2013). The in vivo biofilm. Trends in Microbiology, 21(9), 466–474.
Bjerkan, G., Witsø, E., & Bergh, K. (2009). Sonication is superior to scraping for retrieval of
bacteria in biofilm on titanium and steel surfaces in vitro. Acta Orthopaedica, 80(2),
245–250.
Boehm, A., Steiner, S., Zaehringer, F., Casanova, A., Hamburger, F., Ritz, D., et al. (2009).
Second messenger signalling governs Escherichia coli biofilm induction upon ribosomal
stress. Molecular Microbiology, 72, 1500–1516.
Cavalcanti, Y. W., Morse, D. J., da Silva, W. J., Del-Bel-Cury, A. A., Wei, X., Wilson, M.,
et al. (2015). Virulence and pathogenicity of Candida albicans is enhanced in biofilms con-
taining oral bacteria. Biofouling, 31, 27–38.
Chatterji, D., & Kumar Ojha, A. (2001). Revisiting the stringent response, ppGpp and star-
vation signaling. Current Opinion in Microbiology, 4, 160–165.
Christopher, A. B., Arndt, A., Cugini, C., & Davey, M. E. (2010). A streptococcal effector
protein that inhibits Porphyromonas gingivalis biofilm development. Microbiology, 156,
3469–3477.
Clais, S., Boulet, G., Van kerckhoven, M., Lanckacker, E., Delputte, P., Maes, L.,
et al. (2015). Comparison of viable plate count, turbidity measurement and real-time
PCR for quantification of Porphyromonas gingivalis. Letters in Applied Microbiology, 60,
79–84.
Coenye, T., & Nelis, H. J. (2010). In vitro and in vivo model systems to study microbial
biofilm formation. Journal of Microbiological Methods, 83(2), 89–105.
Coenye, T., Van Acker, H., Peeters, E., Sass, A., Buroni, S., Riccardi, G., et al. (2011).
Molecular mechanisms of chlorhexidine tolerance in Burkholderia cenocepacia biofilms.
Antimicrobial Agents and Chemotherapy, 55(5), 1912–1919.
Cohen, A., Dagli, M., Stavropoulos, S. J., Mondschein, J., Soulen, M., Shlansky-Goldberg, R.,
et al. (2011). Silicone and polyurethane tunneled infusion catheters: A comparison of
durability and breakage rates. Journal of Vascular and Interventional Radiology, 22(5), 638–641.
Collet, A., Cosette, P., Beloln, C., Ghlgo, J. M., Rlhouey, C., Lerouge, P., et al. (2008).
Impact of rpoS deletion on the proteome of Escherichia coli grown planktonically and
as biofilm. Journal of Proteome Research, 7, 4659–4669.
Collet, A., Vilain, S., Cosette, P., Junter, G. A., Jouenne, T., Phillips, R. S., et al. (2007).
Protein expression in Escherichia coliS17-1 biofilms: Impact of indole. Antonie van Leeu-
wenhoek International Journal of General and Molecular Microbiology, 91, 71–85.
Cos, P., Tote, K., Horemans, T., & Maes, L. (2010). Biofilms: An extra hurdle for effective
antimicrobial therapy. Current Pharmaceutical Design, 16(20), 2279–2295.
Cox, A. J. (1990). Comparison of catheter surface morphologies. British Journal of Urology, 65,
55–60.
Cugini, C., Morales, D. K., & Hogan, D. A. (2010). Candida albicans-produced farnesol stim-
ulates Pseudomonas quinolone signal production in LasR-defectivePseudomonas
aeruginosa strains. Microbiology, 156, 3096–3107.
284 Joke Donné and Sylvia Dewilde

Curtis, J., & Colas, A. (2004). Medical applications of silicones. In B. D. Ratner,

A. S. Hoffman, F. J. Schoen, & J. E. Lemons (Eds.), Biomaterials science: An introduction
to materials in medicine (2nd ed., pp. 697–707). Canada: Elsevier Academic Press.
Damgaard, L. R., Nielsen, L. P., & Revsbech, N. P. (2001). Methane microprofiles in a sew-
age biofilm determined with a microscale biosensor. Water Research, 35(6), 1379–1386.
Da Silva Fernandes, S., Kabuki, D. Y., & Kuaye, A. (2015). Behavior of Listeria monocytogenes
in a multi-species bio film with Enterococcus faecalis and Enterococcus faecium and control
through sanitation procedures. International Journal of Food Microbiology, 200, 5–12.
De Angelis, M., Siragusa, S., Campanella, D., Di Cagno, R., & Gobbetti, M. (2015). Com-
parative proteomic analysis of biofilm and planktonic cells of Lactobacillus plantarum
DB200. Proteomics, (Epub ahead of print).
Deighton, M. A., Capstick, J., Domalewski, E., & Van Nguyen, T. (2001). Methods for
studying biofilms produced by Staphylococcus epidermidis. Methods in Enzymology,
336(1996), 177–195.
Del Carpio-Perochena, A., Bramante, C. M., de Andrade, F. B., Maliza, A. G. A.,
Cavenago, B. C., Marciano, M. A., et al. (2015). Antibacterial and dissolution ability
of sodium hypochlorite in different pHs on multi-species biofilms. Clinical Oral Investi-
gations, (Epub ahead of print).
De Oliveira, D. C. V., Fernandes Júnior, A., Kaneno, R., Silva, M. G., Araújo Júnior, J. P.,
Silva, N. C. C., et al. (2014). Ability of Salmonella spp. to produce biofilm is depen-
dent on temperature and surface material. Foodborne Pathogens and Disease, 11(6),
478–483.
Di Bonaventura, G., Stepanović, S., Picciani, C., Pompilio, A., & Piccolomini, R. (2007).
Effect of environmental factors on biofilm formation by clinical Stenotrophomonas
maltophilia isolates. Folia Microbiologica, 52(1), 86–90.
Domka, J., Lee, J., Bansal, T., & Wood, T. K. (2007). Temporal gene-expression in
Escherichia coliK-12 biofilms. Environmental Microbiology, 9, 332–346.
Donlan, R. M. (2002). Biofilms: Microbial life on surfaces. Emerging Infectious Diseases, 8(9),
881–890.
D€orr, T., Lewis, K., & Vulić, M. (2009). SOS response induces persistence to
fluoroquinolones in Escherichia coli. PLoS Genetics, 5(12), e1000760.
Dufour, D., Leung, V., & Lévesque, C. M. (2012). Bacterial biofilm: Structure, function, and
antimicrobial resistance. Endodontic Topics, 22, 2–16.
Dunne, W. M. (2002). Bacterial adhesion: Seen any good biofilms lately? Society, 15(2),
155–166.
Ferrières, L., & Clarke, D. J. (2003). The RcsC sensor kinase is required for normal biofilm
formation in Escherichia coliK-12 and controls the expression of a regulon in response to
growth on a solid surface. Molecular Microbiology, 50, 1665–1682.
Fetzner, S. (2014). Quorum quenching enzymes. Journal of Biotechnology, 201, 2–14.
Fiedoruk, K., Daniluk, T., Swiecicka, I., Sciepuk, M., & Leszczynska, K. (2014). Type II
toxin-antitoxin systems are unevenly distributed among Escherichia coli phylogroups.
Microbiology, 161, 158–167.
Finelli, A., Gallant, C. V., Jarvi, K., Lori, L., & Burrows, L. L. (2003). Use of in-biofilm
expression technology to identify genes involved in Pseudomonas aeruginosa biofilm devel-
opment. Journal of Bacteriology, 185(9), 2700–2710.
Floyd, K. A., Moore, J. L., Eberly, A. R., Good, J. A. D., Shaffer, C. L., Zaver, H.,
et al. (2015). Adhesive fiber stratification in uropathogenic Escherichia coli biofilms unveils
oxygen-mediated control of type 1 pili. PLoS Pathogens, 11, e1004697.
Furukawa, S., Kuchma, S. L., & O’Toole, G. A. (2006). Keeping their options open: Acute
versus persistent infections. Journal of Bacteriology, 188(4), 1211–1217.
The Challenging World of Biofilm Physiology 285

Galloway, W. R. J., Hodgkinson, J. T., Bowden, S. D., Welch, M., & Spring, D. R. (2011).
Quorum sensing in Gram-negative bacteria: Small-molecule modulation of AHL and
AI-2 quorum sensing pathways. Chemical Reviews, 111, 28–67.
Garrett, T. R., Bhakoo, M., & Zhang, Z. (2008). Bacterial adhesion and biofilms on surfaces.
Progress in Natural Science, 18, 1049–1056.
Grant, D. M., & Bott, T. R. (2003). Biocide dosing strategies for biofilm control. In Heat
exchanger fouling and cleaning: Fundamentals and applications Engineering conference interna-
tional 2003.
Hall-Stoodley, L., & Stoodley, P. (2009). Evolving concepts in biofilm infections. Cellular
Microbiology, 11, 1034–1043.
Hall-Stoodley, L., Stoodley, P., Kathju, S., Høiby, N., Moser, C., William Costerton, J.,
et al. (2012). Towards diagnostic guidelines for biofilm-associated infections. FEMS
Immunology and Medical Microbiology, 65, 127–145.
Harriott, M. M., & Noverr, M. C. (2009). Candida albicans and Staphylococcus aureus form
polymicrobial biofilms: Effects on antimicrobial resistance. Antimicrobial Agents and
Chemotherapy, 53(9), 3914–3922.
Harriott, M. M., & Noverr, M. C. (2010). Ability of Candida albicans mutants to induce
Staphylococcus aureus vancomycin resistance during polymicrobial biofilm formation.
Antimicrobial Agents and Chemotherapy, 54(9), 3746–3755.
Harriott, M. M., & Noverr, M. C. (2011). Importance of Candida—Bacterial polymicrobial
biofilms in disease. Trends in Microbiology, 19(11), 557–563.
Häussler, S., & Parsek, M. R. (2010). Biofilms 2009: New perspectives at the heart of surface-
associated microbial communities. Journal of Bacteriology, 192(12), 2941–2949.
Hawser, S. P., & Douglas, L. J. (1994). Biofilm formation by Candida species on the surface of
catheter materials in vitro. Infection and Immunity, 62(3), 915–921.
Hengge-Aronis, R. (2002). Signal transduction and regulatory mechanisms involved in con-
trol of the sigma(S) (RpoS) subunit of RNA polymerase. Microbiology and Molecular Biol-
ogy Reviews, 66(3), 373–395. table of contents.
Herald, P. J., & Zottola, E. A. (1988). Attachment of Listeria monocytogenes to stainless steel
surfaces at various temperatures and pH values. Journal of Food Science, 53(5), 1549–1562.
Herrmann, M., Vaudaux, P. E., Pittet, D., Auckenthaler, R., Lew, P. D., Schumacher-
Perdreau, F., et al. (1988). Fibronectin, fibrinogen, and laminin act as mediators of
adherence of clinical Staphylococcal isolates to foreign material. The Journal of Infectious
Diseases, 158(4), 693–701.
Hidalgo, G., Burns, A., Herz, E., Hay, A. G., Houston, P. L., Wiesner, U., et al. (2009).
Functional tomographic fluorescence imaging of pH microenvironments in microbial
biofilms by use of silica nanoparticle sensors. Applied and Environmental Microbiology,
75(23), 7426–7435.
Høiby, N., Bjarnsholt, T., Moser, C., Bassi, G. L., Coenye, T., Donelli, G., et al. (2015).
ESCMID guideline for the diagnosis and treatment of biofilm infections 2014. Clinical
Microbiology and Infection, 21, S1–S25.
Hostacká, A., Ciznár, I., & Stefkovicová, M. (2010). Temperature and pH affect the produc-
tion of bacterial biofilm. Folia Microbiologica, 55(1), 75–78.
Ibrahim, K. H., Gunderson, B. W., Hermsen, E. D., Hovde, L. B., & Rotschafer, J. C.
(2004). Pharmacodynamics of pulse dosing versus standard dosing: In vitro metronida-
zole activity against Bacteroides fragilis and Bacteroides thetaiotaomicron. Antimicrobial Agents
and Chemotherapy, 48(11), 4195–4199.
Jefferson, K. K., Goldmann, D., & Pier, G. B. (2005). Use of confocal microscopy to analyze
the rate of vancomycin penetration through Staphylococcus aureus biofilms. Antimicrobial
Agents and Chemotherapy, 49(6), 2467–2473.
286 Joke Donné and Sylvia Dewilde

Jiang, B., Pei, H., & Hu, W. (2010). The effect of mechanical agitation on the stripping of
bio-film from ceramic particles. In 2010 4th international conference on bioinformatics and
biomedical engineering, iCBBE 2010 (pp. 10–13) (2007).
Jolivet-Gougeon, A., & Bonnaure-Mallet, M. (2014). Biofilms as a mechanism of bacterial
resistance. Drug Discovery Today. Technologies, 11, 49–56.
Joly, N., Engl, C., Jovanovic, G., Huvet, M., Toni, T., Sheng, X., et al. (2010). Managing
membrane stress: The phage shock protein (Psp) response, from molecular mechanisms
to physiology. FEMS Microbiology Reviews, 34(1), 797–827.
Junka, A. F., Deja, S., Smutnicka, D., Szymczyk, P., Ziółkowski, G., Bartoszewicz, M.,
et al. (2013). Differences in metabolic profiles of planktonic and biofilm cells in Staph-
ylococcus aureus—1H nuclear magnetic resonance search for candidate biomarkers. Acta
Biochimica Polonica, 60(4), 701–706.
Kalia, D., Merey, G., Nakayama, S., Zheng, Y., Zhou, J., Luo, Y., et al. (2012). Nucleotide,
c-di-GMP, c-di-AMP, cGMP, cAMP, (p)ppGpp signaling in bacteria and implications
in pathogenesis. Chemical Society Reviews, 42, 305–341.
Kanangat, S., Postlethwaite, A., Cholera, S., Williams, L., & Schaberg, D. (2007). Modula-
tion of virulence gene expression in Staphylococcus aureus by interleukin-1β: Novel impli-
cations in bacterial pathogenesis. Microbes and Infection, 9, 408–415.
Kaplan, J. B. (2010). Biofilm dispersal: Mechanisms, clinical implications, and potential ther-
apeutic uses. Journal of Dental Research, 89(3), 205–218.
Kara, D., Luppens, S. B. I., & Ten Cate, J. M. (2006). Differences between single- and dual-
species biofilms of Streptococcus mutans and Veillonella parvula in growth, acidogenicity and
susceptibility to chlorhexidine. European Journal of Oral Sciences, 114(12), 58–63.
Karimi, S., Ghafourian, S., Kalani, M. T., & Jalilian, F. A. (2015). Association between toxin-
antitoxin systems and biofilm formation. Jundishapur Journal of Microbiology, 8(1), 1–6.
Kendall, M. M., Rasko, D. A., & Sperandio, V. (2007). Global effects of the cell-to-cell sig-
naling molecules autoinducer-2,autoinducer-3, and epinephrine in a luxS mutant of
enterohemorrhagic Escherichia coli. Infection and Immunity, 75(10), 4875–4884.
Kerstens, M., Boulet, G., Tritsmans, C., Horemans, T., Hellings, M., Delputte, P.,
et al. (2014). Flow cytometric enumeration of bacteria using TO-PRO(R)-3 iodide
as a single-stain viability dye. Journal of Laboratory Automation, 19, 555–561.
Kim, H., & Izadjoo, M. J. (2015). Antibiofilm efficacy evaluation of a bioelectric dressing in
mono- and multi-species biofilms. Journal of Wound Care, 24(2), 10–15.
Kim, Y., Lee, J. W., Kang, S. G., Oh, S., & Griffiths, M. W. (2012a). Bifidobacterium spp.
influences the production of autoinducer-2 and biofilm formation by Escherichia coli
O157: H7. Anaerobe, 18, 539–545.
Kim, Y. G., Lee, J. H., Kim, C. J., Lee, J. C., Ju, Y. J., Cho, M. H., et al. (2012b). Antibiofilm
activity of Streptomyces sp. BFI 230 and Kribbella sp. BFI 1562 against Pseudomonas
aeruginosa. Applied Microbiology and Biotechnology, 96, 1607–1617.
Kim, Y. H., Lee, Y., Kim, S., Yeom, J., Yeom, S., Kim, B. S., et al. (2006). The role of
periplasmic antioxidant enzymes (superoxide dismutase and thiol peroxidase) of the
Shiga toxin-producingEscherichia coli O157:H7 in the formation of biofilms. Proteomics,
6, 6181–6193.
Kint, C. I., Verstraeten, N., Fauvart, M., & Michiels, J. (2012). New-found fundamentals of
bacterial persistence. Trends in Microbiology, 20(12), 577–585.
Kitanishi, K., Kobayashi, K., Kawamura, Y., Ishigami, I., Ogura, T., Nakajima, K.,
et al. (2010). Important roles of Tyr43 at the putative heme distal side in the oxygen rec-
ognition and stability of the Fe(II)-O2 complex of YddV, a globin-coupledheme-based
oxygen sensor diguanylate cyclase. Biochemistry, 49(49), 10381–10393.
Kostakioti, M., Hadjifrangiskou, M., & Hultgren, S. J. (2013). Bacterial biofilms: Develop-
ment, dispersal, and therapeutic strategies in the dawn of the postantibiotic era. Cold
Spring Harbor Perspectives in Medicine, 3, 1–24.
The Challenging World of Biofilm Physiology 287

Kostenko, V., Salek, M. M., Sattari, P., & Martinuzzi, R. J. (2010). Staphylococcus aureus bio-
film formation and tolerance to antibiotics in response to oscillatory shear stresses of phys-
iological levels. FEMS Immunology and Medical Microbiology, 59, 421–431.
Labrie, J., Pelletier-Jacques, G., Deslandes, V., Ramjeet, M., Auger, E., Nash, J. H. E.,
et al. (2010). Effects of growth conditions on biofilm formation by Actinobacillus
pleuropneumoniae. Veterinary Research, 41, 3.
Lebeaux, D., Chauhan, A., Rendueles, O., & Beloin, C. (2013). From in vitro to in vivo
models of bacterial biofilm-related infections. Pathogens, 2, 288–356.
Lewis, K. (2007). Persister cells, dormancy and infectious disease. Nature Reviews. Microbiology,
5, 48–56.
Lewis, K. (2013). Platforms for antibiotic discovery. Nature Reviews. Drug Discovery, 12(5),
371–387.
Lieberman, O. J., Orr, M. W., Wang, Y., & Lee, V. T. (2014). High-throughput screening
using the differential radial capillary action of ligand assay identifies ebselen as an inhibitor
of diguanylate cyclases. ACS Chemical Biology, 9(1), 183–192.
Lorite, G. S., Rodrigues, C. M., de Souza, A. A., Kranz, C., Mizaikoff, B., & Cotta, M. A.
(2011). The role of conditioning film formation and surface chemical changes on Xylella
fastidiosa adhesion and biofilm evolution. Journal of Colloid and Interface Science, 359(1),
289–295.
Marsh, P. D. (2009). Dental plaque as a biofilm: The significance of pH in health and caries.
Compendium of Continuing Education in Dentistry (Jamesburg, N.J.: 1995), 30, 76–78.
McBain, A. J. (2009). In vitro biofilm models: An overview. In S. Sima & M. G. Geoffrey
(Eds.), Advances in applied microbiology: Vol. 69 (1st ed., pp. 99–132): Academic Press.
McLaughlin, R. A., & Hoogewerf, A. J. (2006). Interleukin-1-induced growth enhancement
of Staphylococcus aureus occurs in biofilm but not planktonic cultures. Microbial Pathogen-
esis, 41, 67–79.
Moreira, J. M. R., Teodósio, J. S., Silva, F. C., Simões, M., Melo, L. F., & Mergulhão, F. J.
(2013). Influence of flow rate variation on the development of Escherichia coli biofilms.
Bioprocess and Biosystems Engineering, 36, 1787–1796.
Mukherjee, J., Karunakaran, E., & Biggs, C. A. (2012). Using a multi-faceted approach to
determine the changes in bacterial cell surface properties influenced by a biofilm lifestyle.
Biofouling, 28, 1–14.
Mukherjee, J., Ow, S. Y., Noirel, J., & Biggs, C. A. (2011). Quantitative protein expression
and cell surface characteristics of Escherichia coli MG1655 biofilms. Proteomics, 11,
339–351.
Murakami, K., Ono, T., Viducic, D., Kayama, S., Mori, M., Hirota, K., et al. (2005). Role
for rpoS gene of Pseudomonas aeruginosa in antibiotic tolerance. FEMS Microbiology Letters,
242, 161–167.
Murga, R., Miller, J. M., & Donlan, R. M. (2001). Biofilm formation by Gram-negative
bacteria on central venous catheter connectors: Effect of conditioning films in a labora-
tory model. Journal of Clinical Microbiology, 39(6), 2294–2297.
Nakajima, K., Kitanishi, K., Kobayashi, K., Kobayashi, N., Igarashi, J., & Shimizu, T. (2012).
Leu65 in the heme distal side is critical for the stability of the Fe(II)-O2 complex of
YddV, a globin-coupled oxygen sensor diguanylate cyclase. Journal of Inorganic Biochem-
istry, 108, 163–170.
Nieto, C., Cherny, I., Seok, K. K., De Lacoba, M. G., Wai, T. C., Chew, C. Y., et al. (2007).
The yefM-yoeBtoxin-antitoxin systems of Escherichia coli and Streptococcus pneumoniae:
Functional and structural correlation. Journal of Bacteriology, 189(4), 1266–1278.
Nocker, A., Richter-Heitmann, T., Montijn, R., Schuren, F., & Kort, R. (2010).
Discrimination between live and dead cells in bacterial communities from environ-
mental water samples analyzed by 454 pyrosequencing. International Microbiology, 13,
59–65.
288 Joke Donné and Sylvia Dewilde

Nostro, A., Cellini, L., Di Giulio, M., D’Arrigo, M., Marino, A., Blanco, A. R., et al. (2012).
Effect of alkaline pH on staphylococcal biofilm formation. APMIS, 120(6), 733–742.
Nyenje, M. E., Green, E., & Ndip, R. N. (2013). Evaluation of the effect of different growth
media and temperature on the suitability of biofilm formation by Enterobacter cloacae
strains isolated from food samples in South Africa. Molecules (Basel, Switzerland), 18,
9582–9593.
Ofek, I., Hasty, D. L., & Sharon, N. (2003). Anti-adhesion therapy of bacterial diseases: Pros-
pects and problems. FEMS Immunology and Medical Microbiology, 38, 181–191.
Oosterhof, J. J. H., Buijssen, K. J. D., Busscher, H. J., Van Der Laan, B. F., & Van Der
Mei, H. C. (2006). Effects of quaternary ammonium silane coatings on mixed fungal
and bacterial biofilms on tracheoesophageal shunt prostheses. Applied and Environmental
Microbiology, 72(5), 3673–3677.
Orme, R., Douglas, C. W. I., Rimmer, S., & Webb, M. (2006). Proteomic analysis of
Escherichia coli biofilms reveals the overexpression of the outer membrane protein OmpA.
Proteomics, 6, 4269–4277.
Otto, K., & Silhavy, T. J. (2002). Surface sensing and adhesion of Escherichia coli controlled by
the Cpx-signaling pathway. Proceedings of the National Academy of Sciences of the United
States of America, 99(4), 2287–2292.
Palliyil, S., Downham, C., Broadbent, I., Charlton, K., & Porter, A. J. (2014). High-
sensitivity monoclonal antibodies specific for homoserine lactones protect mice from
lethal Pseudomonas aeruginosa infections. Applied and Environmental Microbiology, 80(2),
462–469.
Pamp, S. J., Gjermansen, M., & Tolker-Nielsen, T. (2007). The biofilm matrix: A sticky
framework. In M. Givskov & S. Kjelleberg (Eds.), The biofilm mode of life: mechanisms
and adaptations (bacterial). Wymondham, UK: Horizon BioScience.
Paraje, M. G. (2011). Antimicrobial resistance in biofilms. In A. Méndez-Vilas (Ed.), Science
against microbial pathogens: Communicating current research and technological advances: Vol. 2
(pp. 736–744). Badajoz: Formatex Research Center.
Park, A. J., Murphy, K., Krieger, J. R., Brewer, D., Taylor, P., Habash, M., et al. (2014).
A temporal examination of the planktonic and biofilm proteome of whole cell Pseudo-
monas aeruginosa PAO1 using quantitative mass spectrometry. Molecular & Cellular
Proteomics, 13, 1095–1105.
Parker, C. T., & Sperandio, V. (2009). Cell-to-cell signalling during pathogenesis. Molecular
Microbiology, 11(3), 363–369.
Parsek, M. R., & Singh, P. K. (2003). Bacterial biofilms: An emerging link to disease path-
ogenesis. Annual Review of Microbiology, 57, 677–701.
Peeters, E., Nelis, H. J., & Coenye, T. (2008). Comparison of multiple methods for quan-
tification of microbial biofilms grown in microtiter plates. Journal of Microbiological
Methods, 72, 157–165.
Pesavento, C., & Hengge, R. (2009). Bacterial nucleotide-based second messengers. Current
Opinion in Microbiology, 12, 170–176.
Petrova, O. E., & Sauer, K. (2009). A novel signaling network essential for regulating Pseu-
domonas aeruginosa biofilm development. PLoS Pathogens, 5(11), e1000668.
Phillips, N. J., Steichen, C. T., Schilling, B., Post, D. M. B., Niles, R. K., Bair, T. B.,
et al. (2012). Proteomic analysis of Neisseria gonorrhoeae biofilms shows shift to anaer-
obic respiration and changes in nutrient transport and outermembrane proteins. PLoS
One, 7(6), e38303.
Poole, K. (2012). Bacterial stress responses as determinants of antimicrobial resistance. Journal
of Antimicrobial Chemotherapy, 67(9), 2069–2089.
Post, D., Held, J. M., Ketterer, M. R., Phillips, N. J., Sahu, A., Apicella, M. A., et al. (2014).
Comparative analyses of proteins from Haemophilus influenzae biofilm and planktonic
The Challenging World of Biofilm Physiology 289

populations using metabolic labeling and mass spectrometry. BMC Microbiology, 14,
1–16.
Post, J., Preston, R., Aul, J., Larkins-Pettigrew, M., Rydquist-White, J., Anderson, K.,
et al. (1995). Molecular analysis of bacterial pathogens in otitis media with effusion.
JAMA, the Journal of the American Medical Association, 273(20), 1598–1604.
Pratt, L. A., & Kolter, R. (1998). Genetic analysis of Escherichia coli biofilm formation: Roles
of flagella, motility, chemotaxis and type I pili. Molecular Microbiology, 30, 285–293.
Prigent-Combaret, C., Vidal, O., Dorel, C., & Lejeune, P. (1999). Abiotic surface sensing
and biofilm-dependent regulation of gene expression in Escherichia coli. Journal of Bacte-
riology, 181(19), 5993–6002.
Rathsam, C., Eaton, R. E., Simpson, C. L., Browne, G. V., Berg, T., Harty, D. W. S.,
et al. (2005a). Up-regulation of competence—But not stress-responsive proteins accom-
panies an altered metabolic phenotype in Streptococcus mutans biofilms. Microbiology, 151,
1823–1837.
Rathsam, C., Eaton, R. E., Simpson, C. L., Browne, G. V., Valova, V. A., Harty, D. W. S.,
et al. (2005b). Two-dimensional fluorescence difference gel electrophoretic analysis of
Streptococcus mutans biofilms. Journal of Proteome Research, 4, 2161–2173.
Redanz, S., Standar, K., Podbielski, A., & Kreikemeyer, B. (2011). A five-species trans-
criptome array for oral mixed-biofilm studies. PLoS One, 6(12), e27827.
Ren, D., Bedzyk, L. A., Thomas, S. M., Ye, R. W., & Wood, T. K. (2004). Gene expression
in Escherichia coli biofilms. Applied Microbiology and Biotechnology, 64, 515–524.
Resch, A., Leicht, S., Saric, M., Pásztor, L., Jakob, A., G€ otz, F., et al. (2006). Comparative
proteome analysis of Staphylococcus aureus biofilm and planktonic cells and correlation
with transcriptome profiling. Proteomics, 6, 1867–1877.
Rezzonico, F., Smits, T. H. M., & Duffy, B. (2012). Detection of AI-2 receptors in genomes
of Enterobacteriaceae suggests a role of type-2 quorum sensing in closed ecosystems. Sen-
sors (Basel, Switzerland), 12, 6645–6665.
Rhoads, D. D., Wolcott, R. D., & Percival, S. L. (2008). Biofilms in wounds: Management
strategies. Journal of Wound Care, 17(11), 502–508.
Rieu, A., Briandet, R., Habimana, O., Garmyn, D., Guzzo, J., & Piveteau, P. (2008). Listeria
monocytogenesEGD-e biofilms: No mushrooms but a network of knitted chains. Applied
and Environmental Microbiology, 74(14), 4491–4497.
Riss, T. L., Moravec, R. A., Niles, A. L., Benink, H. A., Worzella, T. J., Minor, L.,
et al. (2004). Cell viability assays. In G. S. Sittampalam, N. P. Coussens, H. Nelson,
M. Arkin, D. Auld, & C. Austin, et al. Assay guidance manual [Internet]. Bethesda
(MD): Eli Lilly & Company and the National Center for Advancing Translational
Sciences. 2013 May 1 [updated 2015 Jun 29].
Roberts, A. P., & Kreth, J. (2014). The impact of horizontal gene transfer on the adaptive
ability of the human oral microbiome. Frontiers in Cellular and Infection Microbiology, 4,
1–9.
Romeo, T. (2006). When the party is over: A signal for dispersal of Pseudomonas aeruginosa
biofilms. Journal of Bacteriology, 188(21), 7325–7327.
R€omling, U., Galperin, M. Y., & Gomelsky, M. (2013). Cyclic di-GMP: The first 25 years of
a universal bacterial second messenger. Microbiology and Molecular Biology Reviews, 77(1),
1–52.
Salek, M. M., Sattari, P., & Martinuzzi, R. J. (2012). Analysis of fluid flow and wall shear
stress patterns inside partially filled agitated culture well plates. Annals of Biomedical Engi-
neering, 40, 707–728.
Sauer, K. (2003). The genomics and proteomics of biofilm formation. Genome Biology, 4, 219.
Sauer, K., & Camper, A. K. (2001). Characterization of phenotypic changes in Pseudomonas
putida in response to surface-associated growth. Society, 183(22), 6579–6589.
290 Joke Donné and Sylvia Dewilde

Schembri, M. A., Kjaergaard, K., & Klemm, P. (2003). Global gene expression in Escherichia
coli biofilms. Molecular Microbiology, 48, 253–267.
Serra, D. O., & Hengge, R. (2014). Stress responses go three dimensional—The spatial order
of physiological differentiation in bacterial macrocolony biofilms. Environmental Microbi-
ology, 16, 1455–1471.
Shah, D., Zhang, Z., Khodursky, A., Kaldalu, N., Kurg, K., & Lewis, K. (2006). Persisters:
A distinct physiological state of E. coli. BMC Microbiology, 6, 53.
Shen, G., Rajan, R., Zhu, J., Bell, C. E., & Pei, D. (2006). Design and synthesis of substrate
and intermediate analogue inhibitors of S-ribosylhomocysteinase. Journal of Medicinal
Chemistry, 49(10), 3003–3011.
Sheridan, G. E. C., Masters, C. I., Shallcross, J. A., & Mackey, B. M. (1998). Detection of
mRNA by reverse transcription-PCR as indicator of viability in Escherichia coli cells.
Applied and Environmental Microbiology, 64(4), 1313–1318.
Silva, S., Henriques, M., Martins, A., Oliveira, R., Williams, D., & Azeredo, J. (2009). Bio-
films of non-Candida albicans Candida species: Quantification, structure and matrix com-
position. Medical Mycology, 47, 681–689.
Simmons, L. A., Foti, J. J., Cohen, S. E., & Walker, G. C. (2008). The SOS regulatory net-
work. EcoSal Plus, 1–48. http://dx.doi.org/10.1128/ecosalplus.5.4.3.
Singh, R., Ray, P., Das, A., & Sharma, M. (2010). Penetration of antibiotics through Staph-
ylococcus aureus and Staphylococcus epidermidis biofilms. Journal of Antimicrobial Chemotherapy,
65, 1955–1958.
Sintim, H. O., Smith, J. A., Wang, J., Nakayama, S., & Yan, L. (2010). Paradigm shift in
discovering next-generationanti-infective agents: Targeting quorum sensing,
c-di-GMP signaling and biofilm formation in bacteria with small molecules. Future
Medicinal Chemistry, 2(6), 1005–1035.
Sio, C. F., Otten, L. G., Cool, R. H., Stephen, P., Braun, P. G., Bos, R., et al. (2006). Quo-
rum quenching by an N-acyl-homoserine lactone acylase from Pseudomonas aeruginosa
PAO1. Infection and Immunity, 74(3), 1673–1682.
Stepanović, S., Vuković, D., Hola, V., Di Bonaventura, G., Djukić, S., Ćirković, I.,
et al. (2007). Quantification of biofilm in microtiter plates: Overview of testing condi-
tions and practical recommendations for assessment of biofilm production by staphylo-
cocci. APMIS, 115, 891–899.
Stewart, P. S., & Franklin, M. J. (2008). Physiological heterogeneity in biofilms. Nature
Reviews. Microbiology, 6, 199–210.
Tagliabue, L., Antoniani, D., Macia¸g, A., Bocci, P., Raffaelli, N., & Landini, P. (2010). The
diguanylate cyclase YddV controls production of the exopolysaccharide poly-N-
acetylglucosamine(PNAG) through regulation of the PNAG biosynthetic pgaABCD
operon. Microbiology, 156(10), 2901–2911.
Tagliabue, L., MacIg, A., Antoniani, D., & Landini, P. (2010). The yddV-dos operon con-
trols biofilm formation through the regulation of genes encoding curli fibers’ subunits in
aerobically growing Escherichia coli. FEMS Immunology and Medical Microbiology, 59(3),
477–484.
Takahashi, N., Washio, J., & Mayanagi, G. (2010). Metabolomics of supragingival plaque and
oral bacteria. Journal of Dental Research, 89, 1383–1388.
Takahashi, N., Washio, J., & Mayanagi, G. (2012). Metabolomic approach to oral biofilm
characterization—A future direction of biofilm research. Journal of Oral Biosciences,
54(3), 138–143.
Tamayo, R., Pratt, J. T., & Camilli, A. (2007). Roles of cyclic diguanylate in the regulation of
bacterial pathogenesis. Annual Review of Microbiology, 61, 131–148.
Teughels, W., Assche, N., Sliepen, I., & Quirynen, M. (2006). Effect of material character-
istics and/or surface topography on biofilm development. Clinical Oral Implants Research,
17(Suppl. 2), 68–81.
The Challenging World of Biofilm Physiology 291

Thein, M., Sauer, G., Paramasivam, N., Grin, I., & Linke, D. (2010). Efficient sub-
fractionation of Gram-negative bacteria for proteomics studies. Journal of Proteome
Research, 9, 6135–6147.
Toté, K., Vanden Berghe, D., Maes, L., & Cos, P. (2008). A new colorimetric microtitre
model for the detection of Staphylococcus aureus biofilms. Letters in Applied Microbiology,
46, 249–254.
Tremoulet, F., Duche, O., Namane, A., Martinie, B., & Labadie, J. (2002). A proteomic
study of Escherichia coli O157: H7 NCTC 12900 cultivated in bio film or in planktonic
growth mode. FEMS Microbiology Letters, 215, 7–14.
Tsilibaris, V., Maenhaut-Michel, G., Mine, N., & Van Melderen, L. (2007). What is the ben-
efit to Escherichia coli of having multiple toxin-antitoxin systems in its genome? Journal of
Bacteriology, 189(17), 6101–6108.
Tuckerman, J. R., Gonzalez, G., Sousa, E. H. S., Wan, X., Saito, J. A., Alam, M.,
et al. (2009). An oxygen-sensing diguanylate cyclase and phosphodiesterase couple for
c-di-GMP control. Biochemistry, 48(41), 9764–9774.
Unterholzner, S. J., Poppenberger, B., & Rozhon, W. (2013). Toxin-antitoxin systems:
Biology, identification, and application. Mobile Genetic Elements, 3, e26219.
Van Houdt, R., Aertsen, A., Moons, P., Vanoirbeek, K., & Michiels, C. W. (2006). N-Acyl-
L-homoserine lactone signal interception by Escherichia coli. FEMS Microbiology Letters,
256, 83–89.
Varposhti, M., Entezari, F., & Feizabadi, M. M. (2014). Synergistic interactions in mixed-
species biofilms of pathogenic bacteria from the respiratory tract. Revista da Sociedade
Brasileira de Medicina Tropical, 47, 649–652.
Villa, F., Remelli, W., Forlani, F., Gambino, M., Landini, P., & Cappitelli, F. (2012). Effects
of chronic sub-lethal oxidative stress on biofilm formation by Azotobacter vinelandii.
Biofouling, 28, 823–833.
Wang, X., Kim, Y., Hong, S. H., Ma, Q., Brown, B. L., Pu, M., et al. (2011). Antitoxin MqsA
helps mediate the bacterial general stress response. Nature Chemical Biology, 7(6), 359–366.
Wang, X., Preston, J. F., & Romeo, T. (2004). The pgaABCD Locus of Escherichia coli pro-
motes the synthesis of a polysaccharide adhesin required for biofilm formation. Journal of
Bacteriology, 186(9), 2724–2734.
Welin, J., Wilkins, J. C., Beighton, D., Wrzesinski, K., Fey, S. J., Mose-Larsen, P.,
et al. (2003). Effect of acid shock on protein expression by biofilm cells of Streptococcus
mutans. FEMS Microbiology Letters, 227, 287–293.
Wen, Z. T., Yates, D., Ahn, S.-J., & Burne, R. a. (2010). Biofilm formation and virulence
expression by Streptococcus mutans are altered when grown in dual-species model. BMC
Microbiology, 10, 111.
Werner, E., Roe, F., Bugnicourt, A., Michael, J., Heydorn, A., Molin, S., et al. (2004). Strat-
ified growth in Pseudomonas aeruginosa biofilms. Applied and Environmental Microbiology,
70(10), 6188–6196.
Whitehead, N. A., Barnard, A. M. L., Slater, H., Simpson, N. J. L., & Salmond, G. P. C.
(2001). Quorum-sensing in Gram-negative bacteria. FEMS Microbiology Reviews, 25,
365–404.
Wilkins, M., Hall-Stoodley, L., Allan, R. N., & Faust, S. N. (2014). New approaches to the
treatment of biofilm-related infections. Journal of Infection, 69, S47–S52.
Williams, J. J., & Hergenrother, P. J. (2012). Artificial activation of toxin-antitoxin systems as
an antibacterial strategy. Trends in Microbiology, 20(6), 291–298.
Williamson, K. S., Richards, L. A., Perez-Osorio, A. C., Pitts, B., McInnerney, K.,
Stewart, P. S., et al. (2012). Heterogeneity in Pseudomonas aeruginosa biofilms includes
expression of ribosome hibernation factors in the antibiotic-tolerant subpopulation
and hypoxia-induced stress response in the metabolically active population. Journal of Bac-
teriology, 194, 2062–2073.
292 Joke Donné and Sylvia Dewilde

Winzer, K., & Williams, P. (2003). Escherichia coli gets the message. Nature Medicine, 9(9),
1118–1119.
Wnuk, S. F., Robert, J., Sobczak, A. J., Meyers, B. P., Malladi, V. L., Zhu, J., et al. (2009).
Inhibition of S-ribosylhomocysteinase(LuxS) by substrate analogues modified at the
ribosyl C-3 position. Bioorganic and Medicinal Chemistry, 17(18), 6699–6706.
Wood, T. K., Knabel, S. J., & Kwan, B. W. (2013). Bacterial persister cell formation and
dormancy. Applied and Environmental Microbiology, 79, 7116–7121.
Wu, H., Moser, C., Wang, H.-Z., Høiby, N., & Song, Z.-J. (2014). Strategies for combating
bacterial biofilm infections. International Journal of Oral Science, 7, 1–7.
Yang, L., Liu, Y., Wu, H., Hóiby, N., Molin, S., & Song, Z. (2011). Current understanding
of multi-species biofilms. International Journal of Oral Science, 3, 74–81.
Yeom, J., Shin, J. H., Yang, J. Y., Kim, J., & Hwang, G. S. (2013). 1H NMR-based metab-
olite profiling of planktonic and biofilm cells in Acinetobacter baumannii1656-2. PLoS One,
8(3), 1–7.
Zhang, B., & Powers, R. (2012). Analysis of bacterial biofilms using NMR-based met-
abolomics. Future Medicinal Chemistry, 4(10), 1273–1306.