doi: 10.1111/joim.12004 Biofilm infections, their resilience to therapy and innovative treatment strategies U. Ro ¨ mling 1 & C. Balsalobre 2 From the 1 Department of Microbiology, Tumor and Cell Biology, Karolinska Institutet, Stockholm, Sweden, and 2 Department of Microbiology, University of Barcelona, Barcelona, Spain Abstract. Ro ¨mling U, Balsalobre C (Karolinska Institutet, Stockholm, Sweden; and University of Barcelona, Barcelona, Spain). Biofilm infections, their resilience to therapy and innovative treatment strategies (Review). J Intern Med 2012; 272: 541– 561. Biofilm formation of microorganisms causes per- sistent tissue and foreign body infections resistant to treatment with antimicrobial agents. Up to 80% of human bacterial infections are biofilm associ- ated; such infections are most frequently caused by Staphylococcus epidermidis, Pseudomonas aeru- ginosa, Staphylococcus aureus and Enterobacteria such as Escherichia coli. The accurate diagnosis of biofilm infections is often difficult, which prevents the appropriate choice of treatment. As biofilm infections significantly contribute to patient mor- bidity and substantial healthcare costs, novel strategies to treat these infections are urgently required. Nucleotide second messengers, c-di- GMP, (p)ppGpp and potentially c-di-AMP, are major regulators of biofilm formation and associ- ated antibiotic tolerance. Consequently, different components of these signalling networks might be appropriate targets for antibiofilm therapy in com- bination with antibiotic treatment strategies. In addition, cyclic di-nucleotides are microbial-asso- ciated molecular patterns with an almost universal presence. Their conserved structures sensed by the eukaryotic host have a widespread effect on the immune system. Thus, cyclic di-nucleotides are also potential immunotherapeutic agents to treat antibiotic-resistant bacterial infections. Keywords: biofilm infections, cAMP, c-di-AMP, c-di- GMP, (p)ppGpp, second messenger. Introduction Microbial communities, commonly termed bio- films, are the most ancient multicellular life forms on earth as evidenced by billion-year-old fossils. The preference of natural microbial communities for a sessile lifestyle was observed almost 80 years ago [1]. However, it was not until the 1980s that it was recognized that biofilm formation plays a pathogenic role during the infection process [2]. Today, the almost ubiquitous involvement of biofilm formation during chronic and to some extent acute infection has been realized [3]. Thus, biofilm formation of microbes leads to persistent infections resistant to conventional antimicrobial treatment and is today a major cause of treatment failure. Increased knowledge about the molecular mecha- nisms of biofilm formation is important for the development and analysis of in vivo biofilm models and to establish innovative treatment strate- gies for biofilm infections. Such knowledge has accumulated in recent years leading to the recog- nition that despite some specific variations, there are common structural and regulatory mecha- nisms involved in bacterial biofilm formation. Here, we describe the basis of treatment resis- tance of biofilm infections and discuss the global role of nucleotide second messenger signalling pathways in the regulation of biofilm formation and biofilm-related characteristics, such as anti- biotic resistance and formation of persister cells. Next, the different approaches to interfere with these signalling pathways to develop novel anti- biofilm strategies are presented. The role of cyclic di-nucleotide second messengers in inhibition of biofilm formation as extracellular signalling mol- ecules as well as their role in the communication of microbes with the host resulting in immune stimulation is also discussed. Finally, we consider how these characteristics of cyclic di-nucleotide second messengers can be used to develop com- plementary approaches to treat biofilm-associated infections. ª 2012 The Association for the Publication of the Journal of Internal Medicine 541 Review
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
doi: 10.1111/joim.12004
Biofilm infections, their resilience to therapy and innovativetreatment strategiesU. Romling1 & C. Balsalobre2
From the 1Department of Microbiology, Tumor and Cell Biology, Karolinska Institutet, Stockholm, Sweden, and 2Department of Microbiology,University of Barcelona, Barcelona, Spain
Abstract. Romling U, Balsalobre C (KarolinskaInstitutet, Stockholm, Sweden; and University ofBarcelona, Barcelona, Spain). Biofilm infections,their resilience to therapy and innovative treatmentstrategies (Review). J Intern Med 2012; 272: 541–561.
Biofilm formation of microorganisms causes per-sistent tissue and foreign body infections resistantto treatment with antimicrobial agents. Up to 80%of human bacterial infections are biofilm associ-ated; such infections are most frequently causedby Staphylococcus epidermidis, Pseudomonas aeru-ginosa, Staphylococcus aureus and Enterobacteriasuch as Escherichia coli. The accurate diagnosis ofbiofilm infections is often difficult, which preventsthe appropriate choice of treatment. As biofilminfections significantly contribute to patient mor-bidity and substantial healthcare costs, novel
strategies to treat these infections are urgentlyrequired. Nucleotide second messengers, c-di-GMP, (p)ppGpp and potentially c-di-AMP, aremajor regulators of biofilm formation and associ-ated antibiotic tolerance. Consequently, differentcomponents of these signalling networks might beappropriate targets for antibiofilm therapy in com-bination with antibiotic treatment strategies. Inaddition, cyclic di-nucleotides are microbial-asso-ciated molecular patterns with an almost universalpresence. Their conserved structures sensed by theeukaryotic host have a widespread effect on theimmune system. Thus, cyclic di-nucleotides arealso potential immunotherapeutic agents to treatantibiotic-resistant bacterial infections.
Keywords: biofilm infections, cAMP, c-di-AMP, c-di-GMP, (p)ppGpp, second messenger.
Introduction
Microbial communities, commonly termed bio-films, are the most ancient multicellular life formson earth as evidenced by billion-year-old fossils.The preference of natural microbial communitiesfor a sessile lifestyle was observed almost 80 yearsago [1]. However, it was not until the 1980s that itwas recognized that biofilm formation plays apathogenic role during the infection process [2].Today, the almost ubiquitous involvement ofbiofilm formation during chronic and to someextent acute infection has been realized [3]. Thus,biofilm formation of microbes leads to persistentinfections resistant to conventional antimicrobialtreatment and is today a major cause of treatmentfailure.
Increased knowledge about the molecular mecha-nisms of biofilm formation is important for thedevelopment and analysis of in vivo biofilm modelsand to establish innovative treatment strate-gies for biofilm infections. Such knowledge has
accumulated in recent years leading to the recog-nition that despite some specific variations, thereare common structural and regulatory mecha-nisms involved in bacterial biofilm formation.
Here, we describe the basis of treatment resis-tance of biofilm infections and discuss the globalrole of nucleotide second messenger signallingpathways in the regulation of biofilm formationand biofilm-related characteristics, such as anti-biotic resistance and formation of persister cells.Next, the different approaches to interfere withthese signalling pathways to develop novel anti-biofilm strategies are presented. The role of cyclicdi-nucleotide second messengers in inhibition ofbiofilm formation as extracellular signalling mol-ecules as well as their role in the communicationof microbes with the host resulting in immunestimulation is also discussed. Finally, we considerhow these characteristics of cyclic di-nucleotidesecond messengers can be used to develop com-plementary approaches to treat biofilm-associatedinfections.
ª 2012 The Association for the Publication of the Journal of Internal Medicine 541
Review
The problem: biofilm-related infections are refractory toantimicrobial treatment
According to the National Institutes of Health, up to80% of human bacterial infections involve biofilm-associated microorganisms. Common humandiseases such as dental caries and periodontitisare caused by biofilm-forming bacteria. Biofilmformation has been implicated in persistent tissueinfections such as chronic wound infection, chronicotitis media, chronic osteomyelitis, chronic rhino-sinositis, recurrent urinary tract infection, endo-carditis and cystic fibrosis-associated lung infection[3]. Biofilm-forming bacteria are also associatedwith chronic inflammatory diseases such asCrohn’s disease [4]. In addition, recent experimen-tal evidence indicates a role of biofilm formationin acute infections [5, 6]. As the adherence ofmicroorganisms to tissue is part of the process ofacute infection, the impact of biofilm formation in
infection might in fact be underestimated. There isalso a high incidence of biofilm formation on artifi-cial devices such as catheters, stents, orthopaedicimplants, contact lenses and implantable electronicdevices [3, 7]. Examples of in vitro and in vivo biofilmformation are shown in Fig. 1.
Bacteria frequently involved in biofilm-associatedinfections include the Gram-positive pathogensStaphylococcus epidermidis, Staphylococcus aur-eus and Streptococcus species and Gram-negativePseudomonas aeruginosa and Enterobacteriaceaesuch as Escherichia coli (Table 1). Biofilm infec-tions challenge the ‘one disease, one infectiousagent’ paradigm; more than one bacterial speciescan cause a biofilm infection at the same site, andthese infections are often polymicrobial, with inter-action between microbes increasing persistence.Consequently, personalized antimicrobial treat-ment strategies are likely to emerge.
(b)(a)
(c)
Fig. 1 Examples of bacterial biofilm formation. (a) Biofilm formation of Salmonella typhimurium clinical isolates grown onan agar plate. Left, biofilm-forming colony; right, colony without biofilm. Biofilm colonies by Escherichia coli andPseudomonas spp. have a similar appearance due to the production of extracellular matrix components which bind the dyeCongo red. (b) Electron microscopic analysis of biofilm formation. Left, biofilm-forming S. typhimurium are surrounded by anextracellular matrix; right, non-biofilm-forming S. typhimurium cells. Pictures taken by Manfred Rohde, Helmholtz Center forInfection Research, Braunschweig, Germany. (c) Biofilm formation on an urinary catheter. Left, catheter biofilm of a patientwith urinary tract infection caused by E. coli; right, in vitro catheter biofilm by the same E. coli isolate. Pictures taken byHeinrich Lunsdorf, Helmholtz Center for Infection Research, Brauschweig, Germany. Reproduced from [158] withpermission of the publisher Springer Science+Business Media.
U. Romling & C. Balsalobre Review: Treatment of biofilm infections
542 ª 2012 The Association for the Publication of the Journal of Internal Medicine
Journal of Internal Medicine, 2012, 272; 541–561
Biofilm infections are chronic with a low-gradeimmune response and thus contribute to patientmorbidity. Substantial healthcare costs are causedby biofilm infections due to their high frequency,their resistance to antibiotic treatment and theneed to remove the infected foreign body to cure theinfection [7, 8]. In addition, biofilms contributeto the emergence and spread of antimicrobialresistance [7]. From a wider perspective, biofilmformation favours the colonization of the humanhost by potential pathogens, as well as the trans-mission and persistence of these pathogens in theenvironment [9, 10].
Biofilm infections are difficult to diagnose, as con-ventional culture methods often fail to reliablydetect the infectious agent(s), which prevents theuse of adequate treatment strategies. A number ofcriteria, first proposed by Parsek and Singh [11],have been developed to support accurate diagnosisof biofilm infections [8, 11–13]. These criteriainclude (i) the presence of a localized infection withaggregated bacteria at the infection site, (ii) resis-tance to antibiotic treatment and (iii) ongoing inef-fective host immune responses. To overcome theproblem of poor diagnosis, reliable and accuratedetection of biofilm pathogens is supported byimproved sample preparation and the use ofadvanced molecular and visualization techniques.Of note, there is an urgent need to improve earlydiagnosis of biofilm infections to enhance the effi-cacy of therapy. The development of early diagnos-tic techniques is especially valuable for individualsat high risk of developing biofilm-associated infec-tions including patients with orthopaedic implantswho are affected by S. aureus bacteraemia [14].
Biofilm characteristics
What makes biofilm infections so difficult to treat?Biofilm formation by microbes is a developmentalprocess (Fig. 2) [15]. Initially a (motile) cellapproaches a surface, and the bacterium adheresreversibly to the surface. In the next step, irrevers-ible attachment occurs with the development ofmicrocolonies that produce an extracellular matrix(Fig. 1b). The subsequent development of themature three-dimensional biofilm architectureincludes regulated motility. Upon biofilm dispersal,the cells undergo controlled lysis and escape fromthe microbial community. In the human host,bacteria form a biofilm on a biotic (e.g. an epithelialcell lining) or an abiotic (foreign body) surface. Theabiotic surface is usually coated with proteins orother host molecules forming a conditioning filmthat alters the adhesion capabilities of microbes. Inaddition, biofilm formation can even occur withouta surface. This is the case, for example, in cysticfibrosis-related lung infection where P. aeruginosaforms dense matrix-enclosed cell aggregates in theviscous mucus that are not attached to the epithe-lial cell lining. In addition, host cells can become anintegral part of the biofilm and host componentscan be incorporated in the biofilm matrix.
Mature biofilms are highly resistant not only to theaction of the innate and adaptive immune defencesystems, but also to the action of antimicrobialagents and disinfectants. There are several possi-ble mechanisms underlying this phenotypic resis-tance (Fig. 2), which may depend on the type ofantibiotic treatment and the organism: slow rate ofgrowth in the biofilm, altered metabolism, titration
Table 1 Major pathogens involved in biofilm-associated disease
Bacterial species Biofilm infection References
Escherichia coli Acute and recurrent urinary tract infection,
Streptococcus pneumoniae Colonization of nasopharynx, chronic rhinosinositis, chronic otitis media,
chronic obstructive pulmonary disease
[10, 12]
Streptococcus pyogenes Colonization of oral cavity and nasopharynx, recurrent tonsilitis [3]
U. Romling & C. Balsalobre Review: Treatment of biofilm infections
ª 2012 The Association for the Publication of the Journal of Internal Medicine 543
Journal of Internal Medicine, 2012, 272; 541–561
and inactivation of antimicrobial agents by theextracellular matrix and the presence of an existingoxygen gradient that prevents the action of someantibiotics [7, 12]. In addition, biofilms contain alarge subpopulation of so-called persister cells,that is, dormant cells, which survive antimicrobialtreatment [16] and adapt to a slow growth ratethrough the emergence of small colony variants [7].A limited diffusion of antimicrobials into biofilmshas been suggested, but in most instances, nodirect evidence has been provided [17, 18].
The challenge: defining suitable biofilm models
In vitro biofilm models are indispensable to eluci-date the molecular mechanisms of biofilm forma-tion. They have also been highly valuable fordetermining the role of biofilm formation in theinfectious process. However, the results of in vitroinvestigation of biofilm formation in clinical iso-lates have not been entirely consistent with thefindings from in vivo studies [19]. This might bedue to the poor correlation between in vitro and invivo biofilm formation, the undefined role of biofilmformation in the infection process or insufficientknowledge about the role of biofilms in health and
disease settings. For example, almost all P. aeru-ginosa strains, irrespectively of their origin, formbiofilms to a large extent in in vitro models.However, alginate, a major biofilm matrix compo-nent that supports persistence of P. aeruginosa inthe cystic fibrosis lung, plays a minor role inbiofilm formation in vitro. In addition, correlationbetween biofilm formation and the ability to causeinvasive disease could not be demonstrated forclinical isolates of Streptococcus pneumoniae. Asbiofilm formation has been experimentally shownto negatively affect acute invasive disease ofS. pneumoniae and Streptococcus pyogenes [20,21], biofilm formation might initially aid coloniza-tion of the human host by Streptococcus species[10]. Also, the intercellular adhesion (ica) polysac-charide was considered the major determinantmediating biofilm formation and consequentlyvirulence in the nosocomial pathogen S. epidermi-dis; this was supported by data that invasivestrains carry the ica locus more frequently thancommensal strains [19]. The higher incidence of icaoccurrence in strains recovered from infections canbe related in part to the fact that it inhibitscolonization of the healthy human skin; however,individuals who have contact with the healthcare
Biofilm prevention strategiesDevelopment of antimicrobial surfacesPrevention of attachment
Extracellular matrixProteins (e.g.adhesive pilli)ExopolysaccharidesExtracellular DNA (eDNA)Enzymes
Biofilm treatment strategies
Degradation of biofilm matrixDetachment inductionIntroduction of signal blockersNovel cell-killing strategiesInterference with biofilm regulation
Fig. 2 Biofilm formation is a developmental process. The different stages of biofilm formation include the free-swimming cell(a), reversible attachment to the surface (b), irreversible attachment to the surface (c), formation of microcolonies throughcell division and extracellular matrix production (d) and formation of a mature three-dimensional biofilm architecture (e).Cells can actively disintegrate from the biofilm (f) or passively be shed through mechanical disruption (g).
U. Romling & C. Balsalobre Review: Treatment of biofilm infections
544 ª 2012 The Association for the Publication of the Journal of Internal Medicine
Journal of Internal Medicine, 2012, 272; 541–561
system are more frequently colonized with ica-positive strains. Thus, ica-negative strains can stillbe pathogenic harbouring alternative factors thatcontribute to biofilm formation in defined clinicalsettings.
Novel approaches to prevent biofilm formation
Novel approaches to prevent biofilm formation andto treat infections by biofilm-forming bacteria arecurrently in development [22]. Antiadhesivesurfaces with altered physical, chemical and topo-graphical properties that prevent adhesion andthereby biofilm formation are also being sought[23]. Other strategies to prevent biofilm formationhave focused on compounds that inhibit theproduction of functional bacterial adhesins [24].Approaches to aid the dissolution of already estab-lished biofilms include physical treatment of thebiofilm, photodynamic therapy, targeting of thebiofilm matrix for degradation, delivery of signalblockers, interference with biofilm regulation,induction of biofilm detachment and developmentof cytotoxic strategies to treat biofilm-forming bac-teria [22, 25, 26]. Although numerous in vitrostudies have demonstrated effective antibiofilmtreatment, only a few in vivo (preclinical) or clinicalstudies have demonstrated improved treatment ofbiofilm infections (Table 2). For example, interfer-ence with homoserine-lactone (HSL) quorum sens-ing (QS) signalling has been successfully used incombination with antibiotics to treat P. aeruginosabiofilms in vivo [27]. HSL signalling, however,negatively controls biofilm formation in pathogenicbacteria other than P. aeruginosa [28]. It is gener-ally agreed that effective treatment of biofilmsrequires a combination therapy of an antibiofilmcompound with an effective antibiotic, but noantibiofilm therapies are in current clinical use.
There are several challenges to be met in thedevelopment of novel antibiofilm therapies. Screen-ing for effective antibiofilm compounds requiresmodels relevant to the clinical situation [29].Although in vitro investigation of biofilm formationhas made significant progress in the last decade,the in vivo molecular mechanisms remain poorlyunderstood [12]. In addition, the complexity ofbiofilm formation makes it difficult to develop acompound that will affect this process in more thanone species. However, conserved extracellularmatrix components and regulatory mechanismsof biofilm formation have been discovered [30, 31].Almost ubiquitous regulatory mechanisms of
biofilm formation in many Gram-positive andGram-negative pathogens include second messen-ger signalling by nucleotides. These signallingsystems, which are described below, might there-fore provide targets for the development of antibio-film compounds and/or the treatment of biofilminfections through their signalling and immuno-stimulatory properties.
The c-di-GMP signalling network
In recent years, the second messenger cyclicdimeric guanosine monophosphate (c-di-GMP)has evolved as a key activator of biofilm formationin bacteria from all branches of the phylogenetictree [32]. Originally, c-di-GMP was discovered asan allosteric regulator of cellulose synthase in thefruit-degrading bacterium Gluconacetobacter xyli-nus about 20 years ago [33]. c-di-GMP is synthe-sized by so-called GGDEF domain proteins anddegraded by the unrelated EAL and HD-GYPdomain proteins (Fig. 3) [34]. The c-di-GMP signal-ling network is the most complex secondarysignalling system discovered in bacteria with morethan 100 c-di-GMP-metabolizing proteins in somespecies. However, the network can vary signifi-cantly in complexity and c-di-GMP signalling iscompletely absent in some bacteria. This signallingnetwork is especially prominent in c-proteobacte-ria, including the major human pathogensP. aeruginosa, Salmonella typhimurium, E. coliand Vibrio cholerae, which possess numerousc-di-GMP-metabolizing proteins; but also Gram-positive Clostridia and Mycobacteria species canharbour the c-di-GMP signalling pathway.
c-di-GMP signalling has a wide range of effects onbacterial physiology, from the regulation of antibi-otic production in Streptomyces coelicolor to secre-tion of toxins in human pathogens such asV. cholerae [30, 35, 36]. The most prominent roleof c-di-GMP signalling is, however, the regulationof bacterial behaviour with the activation of biofilmformation and suppression of motility [37, 38]. Thislifestyle regulation by c-di-GMP is conserved in allbacterial species that have been investigated.Consequently, as biofilm formation is a hallmarkof persistent infections, c-di-GMP signalling islikely to play a significant role in the regulation ofprocesses associated with persistence. The role ofc-di-GMP signalling in acute infections is lessclear, but emerging evidence indicates that tightregulation is required for a full acute virulencephenotype [35].
U. Romling & C. Balsalobre Review: Treatment of biofilm infections
ª 2012 The Association for the Publication of the Journal of Internal Medicine 545
Journal of Internal Medicine, 2012, 272; 541–561
Table2
Preclinicalandclinicalantibiofilm
treatm
entstrategies
Treatm
ent
Rationale
Biofilm
-associated
infectionmodel
Effect
Reference
QSinhibitora
+
tobra
mycin
QSinhibitors
affectbiofilm
architectu
re,
enhancingsusceptibilityto
antibioticsanddisinfectants,
andexpressionofvirulencefactors
Mouseintraperitonealforeign
bodyinfectionmodel;
Pseudomonasaeru
ginosa
Significantreduction
ofcolony-form
ingunit
counts
whencompared
tomonoth
era
pywith
eithercompound
[27]
Fura
nonederivatives,
garlic
extract,
ajoene
Fura
nonederivativesinhibit
QS;
garlic
extractcontainsQS
inhibitors
(e.g.ajoene)
Mousebiofilm
-associated
lunginfectionmodel;
P.aeru
ginosa
Significantbacterial
cleara
nceb
[159,160]
Mannitol+gentamicin
Metabolitesfeedingin
theupperstepsof
theglycolysis
path
waycreate
NADH,
whichcontributesto
theestablishment
ofaprotonmotiveforce(PMF)in
persistercells.ThePMFfacilitates
uptakeofaminoglycosides
Mouseurinary
cath
eter
infectionmodel;
Esch
erich
iaco
li
Reductionofbiofilm
viabilityby101.5
versusnoeffectof
treatm
entwith
gentamicin
alone;
inhibitionofinfection
spreadto
kidney
[101]
Pilicide,curlicide
Inhibitionofbiogenesis
ofadhesins
requiredforbiofilm
form
ationand
establishmentofinfectionbyadhesion
toepithelialcells
Mouseurinary
tract
infectionmodel;E.co
li
Reductionofbacterial
loadin
thebladder
andintracellular
biofilm
communities
[24]
EspA
proteaseof
Staphyloco
ccusepiderm
idis
EspA
proteasedestroysbiofilm
sof
Staphyloco
ccusaureus;enhances
susceptibilityto
antimicrobialpeptide
humanbeta-d
efensin
2
Nasalcolonizationby
S.aureusin
human
volunteers
Decreasein
nasal
colonizationby
S.aureusupon
co-colonizationwith
Espproducing
S.epiderm
idis
[136]
Biophage-P
Acocktail
(differentphageseffective
againstP.aeru
ginosa)
Biofilm
bacteriaare
susceptible
to
lyticbacteriophages
Humanchronic
otitis;
multi-resistant
P.aeru
ginosa
Significantlylower
bacterialcounts;
improvedclinical
scores
[161]
Bacteriophagetreatm
ent
Biofilm
bacteriaare
susceptible
to
lyticbacteriophages
Mousebiofilm
-associatedlung
infectionmodel;P.aeru
ginosa
Significantbacterial
cleara
nce
[162]
aQuoru
m-sensing(Q
S)inhibitors:fura
noneC-3
0,ajoeneandhorsera
dishjuiceextract.
bGarlic
extracthadnostatisticallysignificanteffectin
aclinicaltrialofcysticfibrosis
patients
infectedwithP.aeru
ginosa[163].
U. Romling & C. Balsalobre Review: Treatment of biofilm infections
546 ª 2012 The Association for the Publication of the Journal of Internal Medicine
Journal of Internal Medicine, 2012, 272; 541–561
The activation of biofilm formation by c-di-GMPsignalling includes the regulation of several indi-vidual physiological processes. The expression ofextracellular matrix components such as exopoly-saccharides, adhesive pili, surface-anchored adhe-sins, secretion systems and extracellular DNA, andcell death and motility which cumulatively lead to amature three-dimensional biofilm structure areregulated by c-di-GMP signalling [32]. Cellularmetabolism, nutrient usage and cell cyclicprogression are also occasionally regulated byc-di-GMP.
Regulation of biofilm-associated processes occurson the transcriptional, post-transcriptional andpost-translational levels upon c-di-GMP binding
to a protein or RNA receptor (Fig. 3) [39–41].Protein receptors include effectors that changetheir output activity upon c-di-GMP binding. Forexample, diverse transcriptional regulators acti-vate biofilm-related genes following c-di-GMP bind-ing. Other c-di-GMP receptors are enzymes such asribonucleases and glycosyltransferases. PilZdomain proteins and degenerated GGDEF andEAL domain proteins are intermediaries thatundergo a conformational change upon c-di-GMPbinding and act on an effector through protein–protein interactions.
The c-di-GMP pool available for binding to aspecific receptor is frequently regulated by morethan one c-di-GMP-metabolizing protein [42, 43].
GTP c-di-GTP pGpG
Intermediary Effector
Effector
Effector
Targete.g. Cellulose expression, inhibition of cholera toxin expression
Fig. 3 Basic principles of a nucleotide second messenger signalling pathway with c-di-GMP as an example. An extra- orintracellular signal is sensed by a receptor or directly by sensory domains of the c-di-GMP-metabolizing proteins. As aconsequence of signal perception, alteration of the activity of the enzymes involved in c-di-GMP turnover will lead to atemporal and/or spatial alteration of the concentration of c-di-GMP. Second messenger concentrations are sensed byc-di-GMP-binding proteins (receptors) or RNA aptamers. C-di-GMP receptors can function as intermediaries that transmit thesignal either by protein–protein interactions to effector proteins or by RNA–RNA interactions to a downstream functionalplatform. Effector proteins, which can be itself receptors can function as transcription factors or may have enzymaticactivity. The physiological consequences of c-di-GMP interactions with cellular receptors include alterations in majoressential cellular functions such as replication, gene expression, RNA turnover, translation, post-translational enzymaticactivity and protein functionality and degradation. The overall biological target output is stimulation of biofilm formationwith activation of the biogenesis of adhesive extracellular matrix components, restriction of motility, alteration of proteinsecretion and other functions. Alternative global target outputs include inhibition of motility and inhibition or stimulationof virulence.
U. Romling & C. Balsalobre Review: Treatment of biofilm infections
ª 2012 The Association for the Publication of the Journal of Internal Medicine 547
Journal of Internal Medicine, 2012, 272; 541–561
Thereby, the activities of c-di-GMP-metabolizingproteins are tightly regulated from the level of geneexpression to allosteric feedback inhibition inresponse to intra- and extracellular conditions[32]. c-di-GMP-metabolizing proteins can be regu-lated in concert by global transcriptional factors,such as the stress sigma factor RpoS and at thepost-transcriptional level by the RNA-binding pro-tein CsrA which regulates carbon metabolism.Whereas RpoS expression is required for biofilmformation and leads to an overall rise in c-di-GMPlevels, CsrA inhibits biofilm formation and down-regulates the c-di-GMP levels in E. coli.
c-di-GMP-metabolizing proteins are multidomainproteins consisting of N-terminal signalling domain(s) and C-terminal c-di-GMP-metabolizing domains[30]. Phosphotransfer within two-component sig-nalling cascades is a common signal that couplesc-di-GMP turnover activity with extracellular sig-nals sensed by membrane-standing histidinekinases. Oxygen, nitrogen oxide and light aredirectly sensed by the sensor domains ofc-di-GMP-metabolizing proteins. Furthermore,c-di-GMP synthesizing activity is allosterically reg-ulated by binding of c-di-GMP to an inhibitory sitein the GGDEF domain. Signals that cumulativelydownregulate c-di-GMP concentrations in the cellcan be used as potential therapies for biofilm-associated infections (see below). Bacterial biofilmformation is a cause of persistent infection [3].A role for c-di-GMP is implied by the fact that c-di-GMP-regulated components such as the exopoly-saccharide alginate contribute to persistence inP. aeruginosa infection of the cystic fibrosis lung[44]. Indeed, small colony variants, highly adhesiveantibiotic-resistant variants of P. aeruginosa,which emerge after long-term colonization of thecystic fibrosis lung, have elevated c-di-GMP levels[45, 46]. Nevertheless, using a chinchilla model ofotitis media, overexpression of c-di-GMP hasrecently been shown to increase the persistenceof the infection [47]. Of note, the regulation ofbiofilm components by c-di-GMP-metabolizingprotein in vivo and in vitro can be significantlydifferent [48].
As cells in biofilms show resistance to antimicro-bial treatment and c-di-GMP targets are involvedin resistance to antimicrobial agents anddisinfectants, it is likely that the c-di-GMP signal-ling network contributes to antimicrobial resis-tance. However, only one c-di-GMP-metabolizingprotein, the phosphodiesterase PvrR, has been
demonstrated to control the switch between anti-microbial-susceptible and -resistant forms ofP. aeruginosa [49].
c-di-GMP signalling also contributes to the acuteinfection process. The formation of bacterial aggre-gates (morulae) that resembles biofilm formationby the obligate intracellular pathogens Anaplasmaphagocytophilum and Ehrlichia chaffeensis in whiteblood cells is dependent on c-di-GMP signalling [5,50]. This signalling is also required for pathoge-nicity of P. aeruginosa in a murine model of ther-mal injury [51]. Frequently, however, c-di-GMPsignalling inhibits the acute virulence phenotype.Loss of virulence has been attributed to the inhi-bition of flagella-related functions, suppression oftoxin secretion and/or the deregulated expressionof biofilm extracellular matrix components [52–54].In these instances, the c-di-GMP signalling systemmight also provide a potential target for interfer-ence with the process of acute infection.
The c-di-AMP signalling network
Recently it was discovered serendipitously, whilstdetermining its crystal structure, that the secondmessenger molecule cyclic dimeric adenosinemonophosphate (c-di-AMP) is associated with theDNA integrity scanning protein DisA from Bacillussubtilis [55]. This finding immediately led to thesuggestion that c-di-AMP might be yet anotherbacterial second messenger [56].
Is c-di-AMP signalling as widespread as othersecond messenger pathways? The identification ofthe basic components of c-di-AMP turnover,di-adenylate cyclase activity in the DAC domainof the DNA integrity scanning protein DisA [55] andphysiologically relevant c-di-AMP phosphodiester-ase activity in a subgroup domain of the DHHfamily [57] provides a clue to the phylogeneticdistribution of the c-di-AMP signalling pathway. Itis interesting that c-di-AMP signalling seems tohave a fundamentally different phylogenetic distri-bution compared with c-di-GMP signalling. Forexample, c-di-AMP-metabolizing proteins are notfound in most branches of the proteobacteria; theyare absent in important Gram-negative pathogenssuch as P. aeruginosa and E. coli. However,c-di-AMP signalling components can be found inmajor Gram-positive pathogens such as Staphylo-cocci, Enterococci and Clostridia, as well as inTreponema pallidum and Borrelia burgdorferi.In addition, the c-di-AMP pathway is present in
U. Romling & C. Balsalobre Review: Treatment of biofilm infections
548 ª 2012 The Association for the Publication of the Journal of Internal Medicine
Journal of Internal Medicine, 2012, 272; 541–561
intracellular pathogens such as Chlamydia andMycoplasma spp. It is worth noting that thec-di-AMP signalling network appears to be far lesscomplex than the c-di-GMP system.
Although the physiological functions of c-di-AMPsignalling have not been extensively studied,emerging data indicate that this pathway couldbe a potential target for antimicrobial and antibio-film therapy. First, cumulative evidence suggeststhat c-di-AMP signalling is essential as deletion ofdi-adenylate cyclase activity was found to bedeleterious in a number of investigated pathogensand B. subtilis [58–61]. Second, a general role ofc-di-AMP signalling in the increased resistance tovarious stress factors such as acid and oxidativestress has been suggested in Gram-positive bacte-ria [57, 62, 63]. In S. aureus, c-di-AMP has a role inresistance to extreme cell wall stress [64]. A mutantof the c-di-AMP phosphodiesterase (i.e. mutation inthe gdpP gene) suppressed the lethal effect causedby the absence of lipotechoic acid in the cell wall,enhanced the cross-linking of the peptidoglycanand altered the expression of autolysin. Third,elevated c-di-AMP levels are associated withincreased resistance to some cell wall-active anti-microbials in S. aureus and B. subtilis [64, 65].Fourth, c-di-AMP is secreted by virulent Listeriamonocytogenes to activate a cytosolic surveillancepathway (CSP) in host immune cells such asmacrophages, which stimulates type 1 interferonproduction [66]. Balanced c-di-AMP production islikely to have a role in virulence as it is attenuatedby deregulation of c-di-AMP delivery [67]. Finally,there are indications that biofilm formation isstimulated by c-di-AMP signalling [64].
In summary, c-di-AMP seems to have an essentialrole in intracellular pathogens and plays a role inantibiotic and stress resistance, virulence andbiofilm formation. In addition, as c-di-AMP isrequired for efficient sporulation [68, 69], it mightalso be a target for prevention of sporulation infood-contaminating pathogens such as Bacilluscereus and Clostridia spp. In future studies,c-di-AMP receptors need to be identified and thetargets of c-di-AMP signalling elucidated.
The (p)ppGpp signalling network
The term (p)ppGpp is used for the two modifiednucleotides ppGpp and pppGpp, which are tetra-and penta-phosphate guanosines. These compoundswere initially described in E. coli as mediators of the
control of ribosomal content in response to aminoacid availability, the so-called stringent response[70]. Today, (p)ppGpp is considered a global stressresponse regulator that responds to additionalenvironmental inputs including phosphorus, fattyacid and iron starvation. In E. coli and mostc-proteobacteria, (p)ppGpp turnover depends onthe activity of two enzymes: RelA, a (p)ppGppsynthetase and SpoT, a bifunctional enzyme withsynthetase and hydrolase activities. The mecha-nisms by which E. coli senses amino acid and fattyacid starvation by RelA and SpoT, respectively,have been demonstrated [70]. In the absence ofcertain amino acids, RelA is activated after inter-action with stalled ribosomes. During fatty acidstarvation, the acyl carrier protein interacts withSpoT promoting (p)ppGpp synthesis under condi-tions of limited fatty acid synthesis. However, thesemechanisms of sensing nutrient starvation maynot be universal amongst bacteria [71, 72]. More-over, non-homologous enzymes are involved in (p)ppGpp turnover in different bacteria. Although themajority of bacteria encode for a RelA/SpoT homo-logue (RSH), several Gram-positive species andV. cholerae also code for small (p)ppGpp syntheta-ses and/or hydrolases unrelated to RSH. Thedescribed diversity in (p)ppGpp turnover enzymessuggests substantial complexity of the (p)ppGpp-mediated sensing of environmental changesamongst bacteria [73]; however, this requires fur-ther investigation.
(p)ppGpp acts via the following mechanisms. (i) (p)ppGpp directly affects transcription by interactingwith RNA polymerase (RNAP), thereby stimulatingor repressing expression of the specific promoter[74]. (ii) Interaction of RNAP with (p)ppGppdecreases the affinity for RpoD (major sigma sub-unit during exponential growth), promoting inter-action with other sigma factors such as RpoS(stationary phase sigma subunit). This mechanismis known as sigma factor competition. (iii) It seemsthat (p)ppGpp binds to proteins other than RNAPand thereby affects translation, replication andRNA turnover [75]. (iv) In addition, there is cros-stalk between (p)ppGpp and other second messen-gers, such as c-di-GMP, c-di-AMP and cAMP [76].
Alteration in the level of (p)ppGpp above a certainthreshold leads to deviation of resources fromactive growth to promote rapid adaptation to stressand survival under harsh conditions [77, 78]. Thisreprogramming of the cell reduces basic cellularfunctions, such as protein synthesis, cell division
U. Romling & C. Balsalobre Review: Treatment of biofilm infections
ª 2012 The Association for the Publication of the Journal of Internal Medicine 549
Journal of Internal Medicine, 2012, 272; 541–561
and cell wall synthesis, and induces protectiveresponses against oxidative and osmotic stress.Moreover, (p)ppGpp can coordinate the regulationof cellular processes such as sporulation, compe-tence, antibiotic production and cell-to-cell com-munication [79]. The relevance of (p)ppGpp in thecontrol of virulence has been extensively studied indifferent bacterial pathogens [80]. In vivo andin vitro experiments have shown that the (p)ppGppsignalling pathway coordinates the expression ofvirulence factors required at different steps duringboth acute and persistent infectious processes; (p)ppGpp contributes to initial adherence to hosttissues during infection, host cell invasion andintracellular survival. Moreover, in vivo experi-ments have highlighted the pivotal role of (p)ppGppin pathogenesis.
(p)ppGpp is also involved in the control of biofilmformation and maturation by important Gram-positive and Gram-negative pathogens includingListeria monocytogenes, Streptococcus mutans,Enterococcus faecalis, V. cholerae and uropatho-genic E. coli [81–85]. In the majority of studies, aclear reduction in biofilm formation is detected inthe absence of (p)ppGpp, consistent with a role inprotection against environmental stress. Directregulation of biofilm-promoting type 1 fimbriationby (p)ppGpp has been described for uropatho-genic E. coli strains [85]. In other bacteria, (p)ppGpp acts indirectly by altering the levels ofcentral regulators that control the expression ofbiofilm-promoting genes [84]. In S. mutans, inac-tivation of relA causes a significant reduction inbiofilm formation, presumably by deregulation ofthe luxS gene involved in quorum sensing [82]. InE. coli, (p)ppGpp interferes with the CsrAregulatory pathway that negatively controls bio-film formation by inducing expression of non-coding regulatory RNAs that inhibit the functionof CsrA [75].
(p)ppGpp also contributes to antimicrobial drugresistance [86]. The mechanisms behind the (p)ppGpp-mediated resistance to microcin J25 andpenicillins in E. coli and to vancomycin in Entero-coccus faecalis have been investigated [87, 88].Moreover, it is well known that nutrient starvation,the signal that increases (p)ppGpp levels and acommon condition in biofilm subpopulations,causes an increased tolerance to antibioticexposure. The conventional explanation of thisobservation has been that passive mechanismssuch as slow growth and a low level of metabolic
activity in senescent cells would promote antibioticresistance, as most of the antibiotic targets arebarely active. Recently, however, an activeresponse to starvation that causes an increase intolerance to antibiotic exposure mediated by (p)ppGpp has been shown in P. aeruginosa [89]. Afternutrient depletion, (p)ppGpp promotes a decreasein the synthesis of pro-oxidant molecules such ashydroxy-2-alkyl quinolones and a concomitantincrease in the expression of oxidative stress-protecting enzymes such as catalase and superox-ide dismutase. Of note, during antibiotic exposurein a murine biofilm model, the stringent responsemutant was more efficiently eradicated than thewild type. Thus, (p)ppGpp signalling is ubiquitousamongst pathogens and has a central role inpromoting biofilm formation, drug resistance,stress survival and virulence.
The cAMP signalling network
The first of the second messengers to be describedwas cAMP. It is ubiquitously present amongstdifferent life forms. In prokaryotic cells, the cAMPsignal is commonly relayed via cAMP receptorprotein (CRP)-like transcription factors.
In E. coli, the cAMP–CRP complex is a metabolicsensor that regulates gene expression in responseto carbohydrate availability (commonly known ascatabolite repression) and the energy status of thecell. In addition, cAMP plays a pivotal role in thefine-tuning of a diverse set of cellular processes,including pathogenicity. cAMP deficiency causesattenuation of virulence in different pathogens, viaregulation of the expression of toxins, adhesinsand secretion systems associated with virulence[90]. It is intriguing that the M. tuberculosisgenome contains at least fifteen adenylate cyclasehomologues, suggesting that cAMP signalling hasboth a relevant and complex role in the biology ofthis major pathogen [91].
The involvement of cAMP in the modulation ofbiofilm formation has not been extensively studied.The available data, however, point to a negativerole of cAMP in biofilm formation through repres-sion of the cdgA di-guanylate cyclase in V. choleraeand transcriptional repression of the biofilm-promoting factor type 1 fimbriae in uropathogenicE. coli strains and Serratia marcescens [92, 93].The potential use of cAMP signalling as a targetto prevent biofilm formation warrants furtherinvestigation.
U. Romling & C. Balsalobre Review: Treatment of biofilm infections
550 ª 2012 The Association for the Publication of the Journal of Internal Medicine
Journal of Internal Medicine, 2012, 272; 541–561
Antibiotic resistance phenotypes related to nucleotide secondmessenger signalling and biofilm formation
Several antibiotic resistance phenotypes affectbiofilm formation and/or nucleotide second mes-senger signalling, or vice versa. However, theseinter-relationships have not been fully explored.
Persister cells
A minor fraction of a bacterial population consistsof persister cells, which are highly drug tolerant[16]. There are high levels of persister cells inrecalcitrant chronic bacterial infections such asP. aeruginosa infection of the lung in patientswith cystic fibrosis and possibly tuberculosisinfection [16]. These cells are particularly abun-dant in biofilms. The mechanism of dormancy ofpersister cells is not fully understood, but may bedue to the expression of toxin–antitoxin (TA)systems [94]. Shifting the TA balance throughdegradation of the antitoxin activates the toxinand induces a bacteriostatic-like effect leadingto dormancy. TA systems are also intimatelyinvolved in the regulation of biofilm formationand affect the c-di-GMP network [95]. In E. coli,the antitoxin MqsA, which is also a transcriptionalregulator, represses expression of the biofilmregulatory cascade including the stress sigmafactor rpoS and the biofilm regulator csgD, whichreduces cellular levels of c-di-GMP [96]. In addi-tion, the level of persister cells is reduced in a (p)ppGpp-deficient strain. It has been suggested thatthe toxin HipA might elicit an increase in (p)ppGpp level causing a concomitant increase inantibiotic tolerance [97]. According to the pro-posed model, inhibition of translation and expres-sion of the cell stasis-associated HipA toxin willinduce the activation of ribosome-associated (p)ppGpp synthetases [98, 99].
Toxin–antitoxin systems are highly redundant inbacterial genomes, as E. coli and M. tuberculosispossess at least 37 and 88 TA systems, respec-tively. Consequently, global mechanisms of TAsystem regulation and persister formation [94,100] are potential targets for successful elimina-tion of persister cells. Further studies will revealwhether the c-di-GMP and (p)ppGpp signallingpathways have a global impact on the regulationof persister cell formation and TA systems and maythus provide alternative targets for persister cellelimination. Reversion of dormancy by stimulationof metabolism is another approach to target the
antibiotic tolerance of persister cells. It hasrecently been demonstrated in a biofilm-associatedanimal infection model that this approach canenable aminoglycoside-specific killing of persistercells (Table 2) [101].
Antioxidant defence
A common cytotoxic mechanism of bactericidalantibiotics, independently of their specific molecu-lar target, is the induction of oxidative damage as aresult of hydroxyl radical production [102]. Fur-thermore, bacteria have general strategies to coun-teract the deleterious effects of antimicrobial agentssuch as endogenousH2S production which protectsagainst exposure to a wide range of antibiotics inboth Gram-positive and Gram-negative bacteria[103]. H2S acts as an antioxidant by chelating iron,which prevents the Fenton reaction (a hallmark ofantibiotic-induced oxidative stress) and conse-quently suppresses DNA damage. Moreover, H2Sstimulates catalase and superoxide dismutaseactivities in the cell. This H2S-mediated protectionagainst antibiotics is reminiscent of the molecularmechanism of (p)ppGpp-mediated antibiotic toler-ance induced by nutrient starvation as discussedabove [89]. The production of H2S is ubiquitousamongst bacteria, although different metabolicpathways have been described. To promote bacter-ial killing by blocking H2S-mediated antibioticresistance in biofilms will therefore require theinhibition of diverse H2S-producing enzymes asthis molecule, which easily diffuses through biolog-ical membranes, might otherwise exert a protectiveeffect on the bacterial community.
Antibiotic–biofilm interaction
In addition to a growth inhibiting effect, anti-biotics are important as signalling molecules.Exposure of bacteria to a subminimum inhibitoryconcentration (MIC) of different classes of antibi-otics with diverse cellular targets can globallyaffect gene expression regulating not only biofilmformation, but also stress response, virulenceand motility [104, 105]. Importantly, sub-MICantibiotic levels can affect biofilm formation pos-itively and negatively which potentially determinestreatment outcome. For example, the beneficialeffect of low-dose chemotherapy with the macro-lide antibiotic azithromycin for the treatment oflung infection with P. aeruginosa might be par-tially due to its inhibition of biofilm formation[106].
U. Romling & C. Balsalobre Review: Treatment of biofilm infections
ª 2012 The Association for the Publication of the Journal of Internal Medicine 551
Journal of Internal Medicine, 2012, 272; 541–561
Antibiotic exposure can stimulate biofilm forma-tion. For example, the commonly used aminogly-coside tobramycin induced biofilm formation ofP. aeruginosa and E. coli at sub-MIC levels. It isinteresting that this effect is mediated directly orindirectly through the c-di-GMP signalling system[107]. In S. epidermidis, expression of the icaADBCoperon, encoding the biofilm matrix componentpoly-N-acetylglucosamine (PNAG), is induced afterexposure to a wide variety of antibiotics. In addi-tion, antibiotic-mediated induction of biofilmformation is observed in ica-negative strains. Sim-ilarly, in E. coli, antibiotics targeting translationinduce biofilm formation by upregulation of PNAGsynthesis [76]. Biofilm activation occurs through(p)ppGpp signalling, which acts synergisticallywith c-di-GMP. The mechanisms by which antibi-otic exposure affects biofilm formation are targetsto improve the efficacy of antimicrobial therapy.
Another aspect of antibiotic–biofilm interaction isthe alteration of the mechanism of biofilm forma-tion upon acquisition of antibiotic resistance. Forexample, methicillin-resistant S. aureus isolatesacquire PNAG-independent biofilm formation,which is related to c-di-AMP signalling [108].Expression of b-lactamases type A and D, but notother types, decreased biofilm formation in P. aeru-ginosa and E. coli, most probably through interfer-ence with normal cell wall turnover [109].
Approaches to interfering with second messenger signallingpathways
There are several rational approaches to interferewith second messenger signalling pathways thatalter the activities of the second messenger metab-olizing network: manipulation of metabolizingactivities, interference with second messengerperception and direct inactivation of the secondmessenger molecules. Most experimental studieshave targeted the c-di-GMP signalling network;however, the same principles of interference can beapplied to the manipulation of other second mes-senger pathways.
Manipulation of enzymatic activities
Intrinsic expression of c-di-GMP-specific phospho-diesterases prevents biofilm formation and/orleads to biofilm dispersal which can be mimickedby ectopic expression of certain phosphodiesteras-es in biofilm-forming bacteria such as S. typhimu-rium, E. coli, P. aeruginosa and Clostridium difficile
[37, 110–113]. Similarly, the general stimulatoryeffect of (p)ppGpp on biofilm formation has beendemonstrated in a wide range of pathogens eitherby blocking or by promoting ectopic expression of(p)ppGpp synthetases [81, 83–85]. These examplesshow that in principle, biofilm formation can beeffectively prevented by manipulation of nucleotidesecond messenger metabolizing activities, either byinhibition of the synthesizing activity or by stimu-lation of phosphodiesterase activity, or both, whichcumulatively leads to reduction in the secondmessenger concentration. Phage therapy, whichhad already been used successfully in the pre-antibiotic era to treat bacterial infections, hasrecently been proven to be effective in treatingbiofilm infections (Table 2). It is possible thatdelivery of nucleotide second messenger degradingproteins can be combined with phage therapy toincrease efficacy against biofilm infections [114].
Signals that target second messenger pathways
External signals that either inhibit nucleotidesynthases and/or activate phosphodiesterasescan be used to control biofilm formation. It hasbeen demonstrated that NO, which is also intrin-sically produced by prokaryotes, inhibits biofilmformation and/or triggers biofilm dispersal in avariety of microorganisms when applied exoge-nously at levels below the MIC [115] and is thusan example of a global biofilm inhibitor. At themolecular level, reduction in biofilm formation bythe freely diffusible NO occurs through direct orindirect interference with the c-di-GMP or c-di-AMP-metabolizing components, which ultimatelyleads to a decrease in second messenger concen-tration [116, 117]. In P. aeruginosa, exposure toNO stimulates c-di-GMP-specific phosphodiester-ase activity that promotes biofilm dispersal by apathway which also responds to other dispersionsignals [112, 118, 119]. In E. coli, NO directly bindsto the PAS sensory domain of a phosphodiesterase,which stimulates cyclic di-nucleotide degradingactivity and eventually leads to biofilm dispersal[120, 121]. The level of NO must be regulated ashigh concentrations were found to induce biofilmformation, probably as part of a protectiveresponse [122]. Of interest, biofilm upregulationis also mediated by the c-di-GMP signalling net-work [123]. Very high concentrations of NO subse-quently exhibit a bactericidal effect.
Other commonly used dispersion signals that canaid the dissolution of medically important biofilms
U. Romling & C. Balsalobre Review: Treatment of biofilm infections
552 ª 2012 The Association for the Publication of the Journal of Internal Medicine
Journal of Internal Medicine, 2012, 272; 541–561
have been identified in studies of biofilm dispersion[28]. In addition, cell death and lysis, whichprecede the dispersal of biofilm microcolonies, arephenotypes that can be targeted in c-di-GMP-mediated biofilm eradication approaches.
Interference with signal perception
Although c-di-GMP and other nucleotides are inprinciple freely diffusible molecules, data suggestthat they are localized by receptor binding (Fig. 3).Thus, the sequestration of c-di-GMP by high-affin-ity receptors removes the nucleotide from thegeneral signalling system and promotes biofilmdispersal [124, 125]. Consequently, overexpressionof c-di-GMP receptors mimics the phenotype ofphosphodiesterase expression. To decrease c-di-GMP levels in the cells by sequestration throughreceptors could be an effective method to preventbiofilm formation or to dissolve existing biofilms.
Synthetic chemistry
In addition to the genetic approaches to controlsecond messenger pathways by manipulating theexpression of turnover proteins and receptors,synthetic chemistry approaches can be used tointerfere with the signalling function. In the mostobvious approach, synthetic analogues of secondmessengers can be designed to interfere with thesynthesis, degradation or receptor perception ofthe messenger. A (p)ppGpp analogue has beendescribed which effectively inhibits RelA synthaseactivity of Gram-positive and Gram-negative bac-teria in vitro [126]. In the obligate intracellularpathogens A. phagocytophilum and E. chaffeensis,2′-O-(tert-butyldimethylsilyl) (TBDMS)-c-di-GMP, ahydrophobic analogue of c-di-GMP, has beenshown to specifically interfere with the biofilmphenotype required for growth in host cells [5, 50,127]. A. phagocytophilum and E. chaffeensis har-bour only one di-guanylate cyclase making thetarget choice straightforward. In general, however,it is not clear whether targeting of a single turnoverprotein or all components of the c-di-GMP meta-bolic network is the most useful strategy.
c-di-GMP-binding sites in proteins are diverse andbind different conformations of the nucleotide oreven dimeric c-di-GMP aggregates [32]. Thisdiversity of binding sites allows the developmentof class-specific c-di-GMP pathway inhibitors.It has been demonstrated experimentally thata non-hydrolysable analogue of c-di-GMP is
conformationally locked and shows binding spec-ificity for a phosphodiesterase, but not for adi-guanylate cyclase and a c-di-GMP receptor[128]. Selective inhibition of phosphodiesteraseswill prevent virulence and/or promote biofilmformation.
It will not be trivial to design synthetic analogues ofnucleotide second messenger molecules or smallmolecules that globally interfere with componentsof these signalling pathways due to substantialsequence diversity, for example in GGDEF domaindi-guanylate cyclases. As an alternative approach,it has been suggested that the signalling moleculecould be targeted directly by small molecules,which interact with c-di-GMP, to form biologicallyinactive higher-order complexes [129]. Due tohighly altered structures, these complexes are notexpected to bind to c-di-GMP-binding sites andthus remain biologically inactive. Future work willshow whether these aromatic intercalators, previ-ously shown to bind to DNA, indeed have thepotential to function as biofilm inhibitors in vivo[129, 130]. Formation of higher-order aggregateshas not been reported for the other nucleotidesecond messengers, c-di-AMP and (p)ppGpp.
On the other hand, global screens have detectedbiofilm inhibitory compounds that interfere withnucleotide secondary signalling pathways. Forexample, a molecule with S. mutans antibiofilmactivity, described after screening of a library of506 compounds, causes significant downregula-tion of relA, highlighting the role of (p)ppGpp inbiofilm formation [131].
Inherent microbial components that inhibit biofilm formation
Clear understanding of the mechanisms of disag-gregation to liberate planktonic cells from thebiofilm (Fig. 2) and competitive inhibitory mecha-nisms in multispecies biofilms provides new strat-egies to design efficient antibiofilm drugs [28, 132].
Studies of the mechanisms of biofilm disaggrega-tion have demonstrated that mature biofilms ofB. subtilis secrete several non-proteinaceous cel-lular factors that not only induce disaggregation ofexisting biofilms but also prevent biofilm forma-tion. The first cellular factor identified was amixture of D-amino acids (D-Tyr, D-Leu, D-Metand D-Trp), which causes the mislocation of cru-cial components of the extracellular matrix bysubstitution of the D-Ala present in the peptide
U. Romling & C. Balsalobre Review: Treatment of biofilm infections
ª 2012 The Association for the Publication of the Journal of Internal Medicine 553
Journal of Internal Medicine, 2012, 272; 541–561
side chain of the peptidoglycan [106, 133]. AsD-amino acids are produced by many bacteria andalso prevent biofilm formation in S. aureus andP. aeruginosa, they have emerged as global anti-biofilm compounds. Another biofilm-disassemblyfactor of B. subtilis is norspermidine; this poly-amine interacts with exopolysaccharides, whichare major component of the extracellular matrix.The disassembling activity of norspermidine relieson a basic motif of three methylene groups flankedby two positively charged amino acids [134]. Syn-thetically produced compounds carrying this motifwere found to be active as biofilm disassemblingagents. It is remarkable that norspermidine inhib-its biofilm formation in both major Gram-positiveand Gram-negative pathogens. Although the com-position of the extracellular matrix varies betweenspecies, these findings suggest that approachesbased on D-amino acids and norspermidine mightbe exploited as general antibiofilm strategies.
Many bacteria secrete factors that competitivelyinhibit biofilm formation in other species. Forexample, although certain polysaccharides arepart of the extracellular matrix of biofilms (Fig. 2),the group II capsular polysaccharide produced bycertain strains of E. coli was shown to preventbiofilm formation [132]. Broad spectrum activitywas demonstrated as biofilm formation was notonly prevented in E. coli strains, but also inP. aeruginosa, S. aureus and other pathogens.Recently, additional antibiofilm polysaccharideshave been discovered [135].
The presence of the commensal flora prevents theintrusion of pathogens. Recently, it was demon-strated that commensal strains of S. epidermidisthat secrete the EspA protease prevent biofilmformation and nasal colonization by S. aureus bya unique, non-bactericidal mechanism (Table 2)[136]. This finding might lead to novel mechanismsof interference as the nasal cavity is the primaryreservoir of S. aureus and a major risk factor fortransmission and infection.
Application of extracellular c-di-GMP
In addition to its role as an intracellular secondmessenger, extracellular c-di-GMP has beenreported to paradoxically act as a biofilm inhibitor.Exposure to c-di-GMP of different clinically rele-vant S. aureus strains including a human methi-cillin-resistant S. aureus isolate inhibited not onlybiofilm formation but also cell-to-cell adhesion and
bacterial adherence to human epithelial cells [137].In a mouse model of mastitis, bacterial load wasreduced upon co-administration of c-di-GMP withbacteria [138]. It is noteworthy that S. aureus doesnot possess a functional c-di-GMP signalling path-way. A membrane-standing receptor that sensesextracellular c-di-GMP has not been identified. Inaddition, the broader significance of this findingremains to be demonstrated as the repressive effectof extracellular c-di-GMP on biofilm formationremains restricted to S. aureus.
Immunostimulation using cyclic di-nucleotides: a complementarystrategy to inhibit biofilm formation?
Microorganisms that cause biofilm-associatedinfections and cancer cells have evolved analogousredundant strategies to resist drug treatment andto effectively evade the immune system. Immuno-therapy has been successful as part of cancertreatment [139], and this therapeutic approach hasalso been considered for biofilm-related infections.In particular, active or passive immunization strat-egies based on proteins or polysaccharides specif-ically expressed by biofilm cells have beendeveloped [140].
The cyclic di-nucleotides c-di-GMP and c-di-AMPare microbial-associated molecular patterns asthey are present in bacteria, but not in highereukaryotes. The almost exclusive bacterialpresence makes them suitable for interkingdomcommunication (i.e. between microorganisms andtheir hosts) and as targets of immunosurveillance.Indeed, their conserved structures are sensed byboth mouse and human cells with widespreadeffects on the innate and adaptive immuneresponses [141–143]. For example, in dendriticcells exposed to c-di-GMP in vitro, maturation wasinduced and chemokine production and receptorexpression were increased [141]. These cells have acentral role in bridging the innate and adaptiveimmune responses. Consistent with in vitro effects,the production of specific antibodies is enhancedby in vivo application of c-di-GMP alone or incombination with an antigen [141, 144].
Subsequent studies have shown that the receptorsfor cyclic di-nucleotides are located insidehost cells.Cyclic-di-AMP is secreted into the cytoplasm by theintracellular bacterium L. monocytogenes, whichcauses the activation of the CSP [66]. A robust type1 interferon response is also triggered by Legionellapneumophila overexpressing a di-guanylate cyclase
U. Romling & C. Balsalobre Review: Treatment of biofilm infections
554 ª 2012 The Association for the Publication of the Journal of Internal Medicine
Journal of Internal Medicine, 2012, 272; 541–561
in host macrophages or cytosol-delivered c-di-GMPalone [145, 146]. Cyclic di-nucleotides are thereforeprominent non-viral CSP-activating ligands.Recently, STING, an essential signalling adaptor,which links cytosolic detection of DNA to down-streamevents,was identifiedasa c-di-GMPreceptor[147]. Considering the broad effects of cyclicdi-nucleotides on functions of the immune system,it is expected that additional detection and signal-ling mechanisms by the host for immunosurveil-lance of cyclic di-nucleotide secondmessengers willsoon be uncovered [145]. Although the biologicalconsequences are different, cytosolic delivery ofcyclic di-nucleotides resembles the delivery of cAMPby intracellular M. tuberculosis [90].
The immunomodulatory functions described abovein combination with the non-toxicity and stabilityof cyclic di-nucleotides support their use as im-munotherapeutics. Indeed, c-di-GMP, c-di-AMPand cyclic dimeric inosine monophosphate, whichis not a naturally occurring nucleotide, have beenshown to function successfully as an adjuvant invaccination strategies [148, 149], leading to theclearance of microbial infections [150–152].Although cyclic di-nucleotides have not been usedas adjuvants in combination with biofilm-relatedantigens, their higher efficacy to mount an immuneresponse compared to established adjuvantsmakes them safe candidate molecules for activeor passive immunization strategies against biofilminfections. In addition, as cyclic di-nucleotides perse are highly immunomodulatory, local adminis-tration might aid the clearance of these infections
through immunostimulation in combination withcomplementary antibiofilm therapies.
Future perspectives
Biofilm infections are resistant to conventionalantimicrobial treatment and require the develop-ment of innovative combination therapies for suc-cessful eradication. Candidate targets for thedevelopment of antimicrobial treatment strategiesinclude bacterial nucleotide second messengersignalling systems, which integrate and amplifychanges in the intra- and extracellular environ-ment, mediate resistance to environmental stressand monitor the nutrient status leading to adap-tation of the bacterial cell (Fig. 4).
Thus, c-di-GMP is a major regulator of the switchbetween the sessile (biofilm) and motile bacteriallifestyle and between the acute and chronic infec-tion state. c-di-AMP has emerged as a global stressresponse regulator that modulates the cell wall andaffects biofilm formation and antibiotic resistance.(p)ppGpp plays a major role in the adaptation ofbacteria to a broad range of environmental stressfactors and coordinates cellular processes such asvirulence, biofilm formation and drug resistance.A common feature of these nucleotide secondmessengers is their promotion of resistance toantimicrobial agents through effects on biofilmformation, but also independent of such effects.Consequently, nucleotide second messenger sig-nalling systems are suitable targets for antibiofilmand/or antibacterial interference. Because of the
Biofilm
(p)ppGpp
cAMP c-di-AMP
c-di-GMP
Nutrient starvation Environmental changes
Carbon availability
Virulence
Antibioticresistance
Stress conditions
Persistent/chronicInfectious disease
Fig. 4 The second messenger systems in bacteria: major stimulatory signals and their effect on biofilm formation, virulence,antibiotic tolerance and persistent infections.
U. Romling & C. Balsalobre Review: Treatment of biofilm infections
ª 2012 The Association for the Publication of the Journal of Internal Medicine 555
Journal of Internal Medicine, 2012, 272; 541–561
absence of cyclic di-nucleotides in higher eukary-otes combined with their broad effects on the hostimmune response and cell growth, cyclic di-nucle-otides are likely to prove effective as therapeuticagents.
The regulation of biofilm formation by nucleotidesecond messengers is complex. Despite consider-able progress, few molecular components of thesesignalling systems have been identified. Increasedknowledge regarding nucleotide receptors, the tar-gets of the nucleotide signals, target output andother molecular aspects of these signalling systemsin major pathogens will help to identify moreprecisely targets for the development of novelantibiofilm and antimicrobial therapies. In addi-tion, the relationship between persister cell forma-tion and nucleotide second messenger signalling isworth investigating in more detail and may lead tothe discovery of novel global mechanisms to treatthis subpopulation of survivors of antimicrobialtherapies.
As regulation of biofilm components by nucleotidesecond messenger signalling systems might differin vitro and in vivo, the precise role of individualelements and their targets in in vivo infection needsto be defined. In vivo models of biofilm formation incombination with expression technology may shedlight on this regulation [153].
Biofilm populations consist of heterogeneous celltypes, which are bistable with respect to nucleotidesecond messenger signalling pathways [154] andrespond differently to antibiotic treatment [155].Therefore, considering the rapid emergence ofresistance, more than one antibiofilm/antimicro-bial agent might be required to effectively treatbiofilm infections.
Furthermore, the wide impact of nucleotide signal-ling systems on bacterial physiology makes themideal tools to engineer bacteria with defined andcontrollable secretory, adhesive and multicellularproperties, for example for the design of strainswith improved antibiotic production or novelprobiotics.
Because of the pronounced effect of cyclic di-nucleotides on the host immune system, thesemolecules are ideal candidate therapeutic agentsfor the treatment of biofilm infections. To tailorbacteria to accurately and efficiently deliver cyclicdi-nucleotides for therapeutic purposes will require
increased knowledge of the expression of cyclicdi-nucleotide-metabolizing proteins by bacteria inthe host and inside host cells and the deliverymechanisms of these nucleotides for presentationto the antibacterial surveillance systems of thehost. As c-di-GMP has also been reported to affecthost cell growth, cyclic di-nucleotides might evenbe applied for cancer treatment [156, 157].
In summary, global regulators of bacterial physi-ology have emerged not only with important rolesin biofilm formation, virulence and antibiotic resis-tance, but also as immunostimulatory moleculeswith adjuvant function. All these characteristicscan be taken into consideration for the develop-ment of treatment strategies for bacterial infectionsresistant to conventional antimicrobial agents.
Conflict of interest statement
No conflicts of interest to declare.
Acknowledgements
The authors would like to thank Borje Akerlund forcarefully reading the manuscript. Work conductedin the laboratory of U.R. is supported by Vetensk-apsradet, the Petrus and Augusta Hedlund Foun-dation, Karolinska Institutet and the EuropeanCommission. C.B. is supported by grants from theSpanish Ministry of Science and Innovation(BIO2010-15417 and CSD2008-00013) and theGeneralitat de Catalunya (2009SGR66).
References
1 Henrici AT. Studies of freshwater bacteria: I. A direct