International Journal of Antimicrobial Agents 40 (2012) 260 263

Contents lists available at SciVerse ScienceDirect

International Journal of Antimicrobial Agents

j our na l ho me p age: ht tp : / /www.e lsev ier .co

Short communication

Antibio c microo

Susana P oc,a IBBInstitute do Minb Department o N 1N4c LEPAE, Depar rto Fr

a r t i c l

Article history:Received 10 JaAccepted 30 A

Keywords:Cystic brosisAntibioticsPolymicrobial Pseudomonas a

rder bactd the nvenopula

bioispla

osa, tsortiay sure sui

2012 Elsevier B.V. and the International Society of Chemotherapy. All rights reserved.

1. Introduction

Cystic production to frequentthe dominaadopting a brecent studfail to identpresent butniques [4]. Tisolated frolum pigrumis scarce, thventional plung [5].

Antibiotinfections aselective prwhich CF ming to suchCF only reli

CorresponE-mail add

organism, P. aeruginosa. However, the wealth of species of differ-

0924-8579/$ http://dx.doi.obrosis (CF) is a common genetic disease involving theof thick and sticky mucus that predisposes CF patients

pulmonary infections [1,2]. Pseudomonas aeruginosa isnt pathogen colonising older patients in CF [3], ofteniolm mode of growth as a survival strategy. However,

ies have shown that standard laboratory methods mayify or may misidentify other isolates that are actually

are usually only detected using molecular biology tech-wo novel microorganisms that have been increasingly

m CF specimens are Inquilinus limosus and Dolosigranu-. Although available information about these organismseir presence is very likely to occur together with con-athogens, creating a diverse mosaic of bacteria in the CF

ic resistance is a well-known phenomenon in chronicnd is an increasing concern in CF [6]. The intensiveessure provided by the large amount of antibiotics toicrobial populations are exposed is one factor contribut-

resistance. Typically, the choice of antimicrobials ines on antibiotic susceptibility testing of the traditional

ding author. Tel.: +351 253 601 961; fax: +351 253 604 429.ress: supat@deb.uminho.pt (S.P. Lopes).

ent phenotypes and sensitivities prevailing in the airways of thesepatients is clearly more complex than a monomicrobial diseaseand it has been suggested that they can undermine the effective-ness of the treatment commonly practiced, thus resulting in a lessthan optimum treatment outcome for affected individuals. To theauthors knowledge, this is the rst study evaluating the spectrumof antimicrobial resistance of dual-species biolms involving classi-cal and emerging microorganisms related to CF. The ability of thesespecies to develop single and mixed biolms was also assessedthrough biomass and culturable cells analysis and the obtainedresults were correlated with the susceptibility proles found forthe species involved.

2. Materials and methods

2.1. Bacterial strains, growth media and buffers

Three CF-related bacterial species were used in this study, thetraditional pathogen P. aeruginosa PA14 as well as two emergingmicroorganisms, I. limosus M53 and D. pigrum CIP 104051 (Insti-tute Pasteur Collection, Paris, France). Pseudomonas aeruginosa andI. limosus pure cultures were grown in tryptic soy broth (TSB)(EMD Chemicals Inc., Gibbstown, NJ), whereas brainheart infusion(BHI) (EMD Chemicals Inc.) was used to culture D. pigrum. Since P.aeruginosa and D. pigrum require different culture media to grow,

see front matter 2012 Elsevier B.V. and the International Society of Chemotherapy. All rights reserved.rg/10.1016/j.ijantimicag.2012.04.020tic resistance of mixed biolms in cystirganisms on treatment of infection

atrcia Lopesa,, Howard Cerib, Nuno Filipe Azevedfor Biotechnology and Bioengineering, Centre for Biological Engineering, Universidade f Biological Sciences, University of Calgary, 2500 University Dr NW, Calgary, Alberta T2tment of Chemical Engineering, Faculty of Engineering, University of Porto, Rua Dr Robe

e i n f o

nuary 2012pril 2012

biolmseruginosa

a b s t r a c t

Cystic brosis (CF) is a genetic disobetween classical and newly identiedtion in CF lungs. This study investigateand Dolosigranulum pigrum, and the coability to develop biolms of mixed pdifferent antimicrobials. Monospeciesless biomass than P. aeruginosa and ddual-species biolms with P. aeruginincreasing tolerance of the overall conbiolm-encased cells. These results mmight enable the development of mom/locate / i jant imicag

brosis: impact of emerging

Maria Olvia Pereiraa

ho, Campus de Gualtar 4710-057 Braga, Portugal, Canadaias, 4200-465 Porto, Portugal

associated with multispecies infections where interactionseria might be crucial to understanding the persistent colonisa-interactions between two emerging species, Inquilinus limosustional CF pathogen Pseudomonas aeruginosa by evaluating thetions and then studying their susceptibility patterns to eightlms formed by I. limosus and D. pigrum produced signicantlyyed greater sensitivity to antimicrobials. However, when inhe emerging species I. limosus and D. pigrum were crucial ina to most antibiotics, even without a change in the number ofggest that revising these and other species interactions in CFtable and effective therapies in the future.

S.P. Lopes et al. / International Journal of Antimicrobial Agents 40 (2012) 260 263 261

Table 1Bacterial cultures and growth conditions used in this study.

Bacterial culture Strain Growth medium Incubationtemperature (C)

Incubationperiod (h)

PseudomonaInquilinus limDolosigranulP. aeruginosaP. aeruginosa

TSB, tryptic so

for experimisms, P. aergrown in Tperiod of bused for sinTable 1.

2.2. In vitro

Single aDevice (CBDviously des

2.3. Analys

Single anlysed in ter

2.3.1. BiomPeg lids

to dry for 1stained witCanada) forwater. Pegsthe optical sured usingCanada). At

2.3.2. Cell cThe num

mined by bAn Aquasonnational, Edon the brokmented witbiolms wetryptic soy aSelective agisolation agagar (BCSA100 mg/L ticI. limosus, TSwere incubforming un

2.4. Antibio

Eight ctobramycincefotaximefrom Sigmwere prepastored at of use at 1

iagnotic c

tibio

r wed action challhat u

weren theeria tic; hoccuinim

eterm

atisti

a wePad s wand Tre pant

ults

this -relaing lly e

aft6 CF

iolm conMIC)) arebitins; hon higonic y a ely ltly gshedf ants aeruginosa PA14 TSB/TSA osus M53 TSB/TSA

um pigrum CIP 107041 BHI/TSA + I. limosus TSB/TSA + D. pigrum TSB or BHI/TSA

y broth; TSA, tryptic soy agar; BHI, brainheart infusion.

ents of mixed biolms involving these two microorgan-uginosa was also grown in BHI and D. pigrum was alsoSB. The temperature of incubation was 37 C and theiolm formation varied for each strain. All conditionsgle and mixed biolm experiments are summarised in

biolm formation

nd mixed biolms were grown on a Calgary Biolm) (MBEC Biolm Technologies, Calgary, Alberta) as pre-

cribed by Ceri et al. [7].

is of pre-formed biolms

d mixed biolms formed on the CBD were further ana-ms of biomass and number of culturable cells.

asswere rinsed with double-distilled water and were left0 min. Biolm-growing bacteria on the pegs were thenh 1% (v/v) crystal violet (Sigma-Aldrich, Oakville, ON,

1 min and were washed twice with double-distilled were then decolorised with pure methanol (Sigma) anddensity at 550 nm of the obtained solution was mea-

a microtitre plate reader (Labequip Ltd., Markham, ON, least eight replicates were run for each condition.

ulturabilityber of adhering bacteria within the biolm was deter-reaking four pegs of the CBD under aseptic conditions.ic Water-table Sonicator (model 250 HT; VWR Inter-monton, AB, Canada) was used to disrupt the biolmen pegs submerged in sterile 0.9% (v/v) saline comple-h 1% (v/v) Tween 20 for 10 min (Sigma). The disruptedre subsequently serially diluted in saline and plated ongar (TSA) (EMD Chemicals Inc.) for viable cell counting.ar media were used to plate P. aeruginosa [Pseudomonasar (PIA)] and I. limosus [Burkholderia cepacia selective) supplemented with 300 000 IU/L polymyxin B andarcillin]. TSA and PIA plates were incubated for 24 h (forA plates were incubated for 40 h), whereas BCSA platesated for 48 h at 37 C before enumeration of colony-its (CFU). All samples were run in quadruplicate.

tic stock solutions

(BD Dantibio

2.5. An

Aftepreparcentraof the tures tvaluesdient oof bactantibiobition [7,8]. Mwere d

2.6. St

Dat(Graphbiolmtests, acompasignic

3. Res

In ing CFprovidclinicaformed10410each b

Theteria ((MBECin inhitrationor eveplankt

Onlrelativnicanestablidoses olinically relevant antibiotics were used, including, gentamicin, levooxacin, ciprooxacin, clindamycin,, chloramphenicol and rifampicin. All antibiotics werea-Aldrich. Stock solutions of antimicrobial agentsred at 5120 g/mL and then 500 L aliquots were70 C. Working solutions were prepared on the day024 g/mL in cation-adjusted MuellerHinton broth

biolms invtant to moorganisms.

The minteria in mixneeded to kspecies, wh37 637 4837 7237 2437 24

stics, Franklin Lakes, NJ). For susceptibility testing,oncentrations ranged from 2 g/mL to 1024 g/mL.

tic susceptibility testing

ashing biolms with saline, a challenge plate wascording to Ceri et al. [7]. The minimum inhibitory con-(MIC) was determined by reading the optical densityenge plate at 650 nm, with the exception of those cul-sed sheep blood in the susceptibility testing, where MIC

determined by visual observation of the turbidity gra- challenge plate. This turbidity demonstrates the abilityto grow as a planktonic population in the presence ofence, the minimum concentration where growth inhi-rs is equivalent to the MIC value for most organismsum biolm eradication concentration (MBEC) valuesined by enumerating spot plates for bacterial growth.

cal analysis

re analysed using Prism software v.4.0 for MacintoshSoftware Inc., La Jolla, CA). The ability of strains to forms assessed by one-way analysis of variance (ANOVA)ukeys post hoc test was performed to subsequently

airs of columns. Results were considered statisticallyat P < 0.05.

study, bacterial biolms of traditional and emerg-ted microorganisms were readily formed on the CBD,a valuable and reliable technology for selection offfective antibiotics. Susceptibility assays were per-er all biolms achieved a threshold concentration ofU/peg, which required different incubation times for

(Table 1).centrations of antibiotic able to inhibit planktonic bac-

and those required to kill biolm-encased bacteria summarised in Table 2. Most antibiotics were effectiveg planktonic growth of single species at low concen-wever, mixed planktonic populations required equalher concentrations than those applied to inhibit thegrowth of single populations.few antibiotics were able to kill biolm bacteria atow concentrations. Generally, MBEC values were sig-reater compared with MIC data, suggesting that once, biolms are notoriously difcult to eradicate and highimicrobials are needed to eliminate them. Monospecies

olving only P. aeruginosa were considerably more resis-st antibiotics tested than those developed by other

imum bactericidal concentration required to kill bac-ed biolms was generally equal to the concentrationill the more resistant single biolm of the encompassedich was predominately P. aeruginosa. Results obtained

262 S.P. Lopes et al. / International Journal of Antimicrobial Agents 40 (2012) 260 263

Table 2In vitro susceptibility patterns of single-species and dual-species cultures of cystic brosis-related organisms to eight clinically relevant antibioticsa.

Antibiotic Pseudomonasaeruginosa

Inquilinuslimosus

Dolosigranulumpigrum

P. aeruginosa + I.limosus

P. aeruginosa + D.pigrum (TSB)

P. aeruginosa + D.pigrum (BHI)

TobramycinMIC 64 MBEC 256

GentamicinMIC 1024 MBEC 1024

LevooxacinMIC 128 512 >1024 >

n1024 32 512 >

>1024 512 16 >>1024 >1024 >1024 >

16 1024 >1024 >

nicol128 256 4 >

>1024 1024 32 >

>1024 1024 8 32 >

inhibitory concentration; MBEC, minimum biolm eradication concentration; TSBBEC values expressed in mg/L.

ve media enabled the observation that P. aeruginosa ism prevailing within dual-species biolms with I. limo-ing ca. 70% of the total cell number. Because there is no

about specic selective media available for D. pigrum,dy was performed for this microorganism. By com-susceptibility patterns of P. aeruginosa and D. pigrums biolms growing in different media (Table 2, lasts), it was observed that antibiotics easily inhibited thelanktonic suspensions when TSB was used to culture

ulations. Susceptibility proles of these dual-speciesTSB were similar to those obtained with P. aeruginosa,duced susceptibility to antibiotics as biolms. BiolmsBHI did not follow any similar pattern of susceptibility,riations in both MIC and MBEC values.onas aeruginosa was clearly the organism that pro-reatest amount of biomass (Fig. 1a) and also showed

4. Dis

Althber ofinjury of the this wdifcuknownbe mumetabprovidwithinnumbe[10].

Sin

counts (Fig. 1b). Inquilinus limosus and D. pigrum pre-istically signicantly less biomass than P. aeruginosaThese species showed similar numbers of CFU (ca.g), however 2 log below the cell number produced bysa (P < 0.001) (Fig. 1b). Dual-species biolms formed

aeruginosa and each of the unconventional speciesl numbers similar to biolms of P. aeruginosa alone (ca.g), however biomass was markedly reduced in mixed-lms. To conrm that the observed biomass decreaseted with the amount of extracellular matrix produced, the protein and polysaccharide content in the matrixined using established methods [9]. It was observed

pecies biolms of P. aeruginosa with I. limosus or D. a reduction in the matrix of ca. 30% and 35%, respec-ared with P. aeruginosa biolms alone (Supplementary

entary material related to this article can be found,ne version, at http://dx.doi.org/10.1016/j.ijantimicag..tically signicant difference was found between dual-lms of P. aeruginosa and D. pigrum growing in TSB and

in biomass formation (Fig. 1a) or in the number of pro-(Fig. 1b).

otics than thto the highbiolm forknown intrP. aeruginoslarge and padaptabilityment, meantreat [12].

Conversform biolmdo not formorganisms tIndeed, thebiolms waThese organcells than Pcrobial agenincubation sequently rof producecharacter ogrowth on t16 8>1024 >1024

16 8>1024 >1024

128 8>1024 >1024

1024>1024 >1024

32 256>1024 >1024

1024 128>1024 >1024

>1024 >1024>1024 >1024

tic soy broth; BHI, brainheart infusion.

on

h it was originally thought that only a limited num-nisms could cause symptomatic infection and lung

F, it has now been shown that the microbial ecologyung is far more complex than originally thought. Inbacteria growing in biolms were notoriously moreeradicate than when growing planktonically. The well-eased resistance associated with biolms is likely totorial, depending for instance on the alteration of the

of bacterial cells, on the extra barrier of protectiony the extracellular matrix that so often encases cellslm, or even by the spatial arrangement, as well as thethose bacterial cells that form the layers of the biolm

aeruginosa biolms were more resistant to most antibi-

ose formed by emerging species. This may be attributed

er biomass and cell numbers obtained for P. aeruginosamation compared with the other biolms. The well-insic resistance and acquired tolerance to antibiotics ofa biolms [11] may also support the previous results. Itslastic genome favours the species in providing greater

to most hostile environments and to antibiotic treat-ing that infection with this species is more arduous to

ely, emerging bacteria did not show a great ability tos on the CBD. Nevertheless, the fact that some isolates

in vitro biolms does not impair the ability of theseo survive in the patient lung, as recently suggested [13].

reduced capability of I. limosus and D. pigrum to forms reected in their sensitivity to most tested antibiotics.isms showed signicantly less biomass and number of. aeruginosa, making them more vulnerable to antimi-ts. The fact that these organisms require an extendedtime could help explain the slow growth, which con-esults in the lowest amount of biomass and numberd cells by their single biolms. In addition, the slimyf I. limosus bacterial colonies may contribute to the slowhe peg surface of the CBD.

S.P. Lopes et al. / International Journal of Antimicrobial Agents 40 (2012) 260 263 263

Fig. 1. (a) Biom fter thdeviation for a broth

Most stuclassical paCo-infectionlungs remathe importawas demonity against dspecies. Gencentrationsin monospeshowing thcould be ththe antibiotrecognised consortiumpopulation.overall bioganisms, ima decreasedwe have shoformed andeven the hibiolms wasortia. This other unusucome and tnon-convenbetween allstrategies t

In conclto be morecal pathogepopulationsinteractionsthe complefor the increin CF will pthe diseasein those casment occurstrain of P. colonisation

e: Tics.o

ding:ationto SPpeti

ical a

nces

ley CDbial petic bjen FAe 200ler DEar F, leculapatienley CDl naturh cystenza Gce andatienass and (b) culturable cells obtained for single-species and dual-species biolms at least six replicates is illustrated. OD550, optical density at 550 nm; TSB, tryptic soy

dies involving mixed biolms in CF have only includedthogens such as P. aeruginosa and B. cepacia [14].s of traditional pathogens with rare species in CFin largely unexplored, limiting the understanding ofnce of these interspecies interactions. In this study itstrated that most antibiotics presented a poor activ-ual-species biolms of P. aeruginosa with an emergingerally, these biolms required effective antibiotic con-

at least equal to that used to kill the same species whencies biolms. The results obtained by selective media,e predominance of P. aeruginosa within the consortia,e basis for a higher contribution of P. aeruginosa toic resistances presented by dual-species biolms. It isthat mixed biolms alter the metabolic activity of the

and hence may alter the susceptibility patterns of the This can reect itself for instance in an alteration in thelm structure and extracellular matrix by both microor-pairing access of antibiotics into the consortium, or by

antibiotic uptake rate through the cell membrane. Herewn that independently of the reduced biolm biomass

a decrease in matrix content, the arrangement andgh number of biolm-encased cells in mixed-species

Notbiofomposes.

FunFound2008 (

ComEth

Refere

[1] Sibcrocys

[2] RatCar

[3] Gel[4] Bitt

Mosis

[5] Sibbiawit

[6] Vallenof ps enough to imply an increased resistance on those con-strongly suggests that these organisms and eventuallyal species might have a great importance in the out-

reatment of infection in CF. Inaccurate identication oftional pathogens and the disregard for the interactions

bacteria may lead to ineffective antibiotic therapeutichat could select for antibiotic-resistant pathogens.usion, although the novel species found in CF appear

easily treated by antibiotic therapy than the classi-ns, they can enhance the antibiotic resistance of mixed

where they are involved. It is clear that the complex between bacteria in the host play an important role in

x pathology of the disease and may be often responsiblease in antibiotic tolerance. Thus, treatment of infectionrobably be more effective in the future by categorising

as polymicrobial. It remains to be understood whetheres where resistance of P. aeruginosa CF biolms to treat-s, we might in fact not be in the presence of a resistantaeruginosa in the patient but rather of a polymicrobial.

[7] Ceri H, ODevice: nof bacter

[8] Harrisonsusceptibized bio123654

[9] Silva S, Hof non-Cacomposit

[10] Hiby N,bacterial

[11] Tre-Hardactivity biolm 32936.

[12] Sykes R.tance: a184252

[13] DeligiannPseudomoexhibit d38.

[14] Tomlin KBurkholdee respective period of incubation (see Table 1). The mean standard; BHI, brainheart infusion.

he data in this manuscript are available in www.rg, where it is possible to access them for research pur-

Financial support was provided by the Portuguese for Science and Technology [grant SFRH/BD/47613/L) and project PTDC/SAU-ESA/64609/2006].ng interests: None declared.pproval: Not required.

, Parkins MD, Rabin HR, Duan K, Norgaard JC, Surette MG. A polymi-rspective of pulmonary infections exposes an enigmatic pathogen in

rosis patients. Proc Natl Acad Sci USA 2008;105:150705.. Cystic brosis: pathogenesis and future treatment strategies. Respir9;54:595605.. Aerosol antibiotics in cystic brosis. Respir Care 2009;54:65870.Richet H, Dubus JC, Reynaud-Gaubert M, Stremler N, Sarles J, et al.r detection of multiple emerging pathogens in sputa from cystic bro-ts. PLoS One 2008;3:e2908., Parkins MD, Rabin HR, Surette MG. The relevance of the polymicro-e of airway infection in the acute and chronic management of patientsic brosis. Curr Opin Investig Drugs 2009;10:78794., Tappe D, Turnwald D, Frosch M, Konig C, Hebestreit H, et al. Preva-

antimicrobial susceptibility of microorganisms isolated from sputats with cystic brosis. J Cyst Fibros 2008;7:1237.

lson ME, Stremick C, Read RR, Morck D, Buret A. The Calgary Biolmew technology for rapid determination of antibiotic susceptibilitiesial biolms. J Clin Microbiol 1999;37:17716.

JJ, Stremick CA, Turner RJ, Allan ND, Olson ME, Ceri H. Microtiterility testing of microbes growing on peg lids: a miniatur-lm model for high-throughput screening. Nat Protoc 2010;5:.enriques M, Martins A, Oliveira R, Williams D, Azeredo J. Biolmsndida albicans Candida species: quantication, structure and matrixion. Med Mycol 2009;47:6819.

Bjarnsholt T, Givskov M, Molin S, Ciofu O. Antibiotic resistance of biolms. Int J Antimicrob Agents 2010;35:32232.y M, Vanderbist F, Traore H, Devleeschouwer MJ. In vitroof antibiotic combinations against Pseudomonas aeruginosaand planktonic cultures. Int J Antimicrob Agents 2008;31:

The 2009 Garrod lecture: the evolution of antimicrobial resis- Darwinian perspective. J Antimicrob Chemother 2010;65:.i E, Pattison S, Berrar D, Ternan NG, Haylock RW, Moore JE, et al.nas aeruginosa cystic brosis isolates of similar RAPD genotypeiversity in biolm forming ability in vitro. BMC Microbiol 2010;10:

L, Coll OP, Ceri H. Interspecies biolms of Pseudomonas aeruginosa andria cepacia. Can J Microbiol 2001;47:94954.

Antibiotic resistance of mixed biofilms in cystic fibrosis: impact of emerging microorganisms on treatment of infection1 Introduction2 Materials and methods2.1 Bacterial strains, growth media and buffers2.2 In vitro biofilm formation2.3 Analysis of pre-formed biofilms2.3.1 Biomass2.3.2 Cell culturability

2.4 Antibiotic stock solutions2.5 Antibiotic susceptibility testing2.6 Statistical analysis

3 Results4 DiscussionReferences