Accepted Article
Article Type: Original Article
Phenotypic changes contributing to Enterobacter gergoviae biocide resistance
M. Periame1, N.Philippe13, O. Condell2, S. Fanning2, J-M. Pagès1 and A. Davin-Regli1
1
UMR-MD1, Aix-Marseille Université, IRBA, Transporteurs Membranaires,
Chimiorésistance et Drug Design, Marseille, France 2
UCD-Centre for Food Safety, School of Public Health, Physiotherapy & Population
Science, University College Dublin, Belfield, Dublin 4, Ireland. 3
Laboratoire Information Génomique et Structurale (IGS), UMR 7256 (IMM FR 3479)
CNRS Aix-Marseille Université, Luminy campus, Marseille, France
Corresponding author: Dr. Anne Davin-Regli, UMR-MD1, Transporteurs Membranaires, Chimiorésistance et Drug Design, Facultés de Médecine et Pharmacie, 27 Bd Jean Moulin, 13385 Marseille cedex 05, France ([email protected]).
Keywords: Enterobacter gergoviae, preservatives, cosmetics, triclosan, biofilm, flagellum, fimbriae
Running headline: Mechanisms involved in E. gergoviae biocides adaptation
This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process which may lead to differences between this version and the Version of Record. Please cite this article as an 'Accepted Article', doi: 10.1111/lam.12435 This article is protected by copyright. All rights reserved.
Accepted Article
Significance and impact of the study Recurrent contaminations of cosmetics products by Enterobacter gergoviae, needed a better understanding concerning the bacterial adaptation to preservative agents, with particular concern to triclosan and MIT-CMIT. We demonstrated that bacteria response is associated to various mechanisms represented by expression of external appendages (pili or fimbriae) that control cell motility and biofilm formation and evolving as the concentration of biocides adaptation increased. Such mechanisms which are not chemical specific can also promote a cross-resistance to other biocidal agents. The characterization of E gergoviae adaptability to biocides allows industry to adjust the ranges of concentrations and composition of preservatives in formula.
Abstract Enterobacter gergoviae is a recurrent contaminant of cosmetic and hygiene products. To understand how this bacterium adapts to biocides, we studied E. gergoviae CIP 76.01 and its triclosan and MIT-CMIT tolerant isogenic mutants. They were compared with others also isolated from contaminated cosmetics. Phenotypic differences were noted and these included changes in the bacterial envelope and flagella along with differences in motility, and biofilm growth rates. Triclosan and MIT-CMIT derivatives expressed flagella and other MIT-CMIT derivatives exhibited some external appendages. Those bacteria expressing a high-level MIC to MIT-CMIT, expressed a strong biofilm formation. No differential phenotypes were noted for carbon source utilisation. E. gergoviae demonstrated a diverse response to both of these preservatives contained in cosmetic preparations, depending on their concentrations. Interestingly, this adaptive response is associated with modifications of filament structure-related proteins contributing to increase the organism motility and the production of biofilm.
This article is protected by copyright. All rights reserved.
Accepted Article
Introduction Enterobacter gergoviae is often identified in contaminated manufactured cosmetic products. Bacteriostatic action of preservatives, genetic diversity of strains causing contaminations and the emergence of well-adapted bacteria from these matrices, support E. gergoviae increased tolerance to preservatives (Davin-Regli et al. 2006; Périamé et al. 2014). Diverse mechanisms have been associated with biocide tolerance including alterations of membrane permeability, target gene alterations and expression of efflux systems (Ortega Morente et al. 2013). It has been shown that E. gergoviae presented a natural resistance to parabens by producing an esterase (Davin-Regli et al. 2006).
These threats to public health are a constant challenge for the cosmetics industry and this feature required novel and appropriate solutions to manage the risk. It was reported that the capacity of bacteria to form biofilms appeared to be a key in the bacterial adaptation to these compounds (Lubarsky et al. 2012). It has been demonstrated previously that the biofilm phenotype described in Escherichia coli and its triclosan tolerant mutant were distinct (Sheridan et al. 2013; Lenahan et al. 2014). In the latter case, the tolerant strain exhibited strong biofilm formation associated with the expression of curli fimbriae with the over-expression of FliC protein that contributes to the filament structure of flagella. This structure has been implicated in the biofilm formation by having an important mechanosensory role during the early stages of surface adhesion leading to the biofilm organisation (Belas 2014).
This article is protected by copyright. All rights reserved.
Accepted Article
Triclosan is a broad-spectrum antibacterial compound widely used in personal care products. It inhibits the enoyl-acyl carrier protein reductase FabI and mutations in the fabI gene and/or modifications of membrane permeability by over-expressing efflux pumps contribute to a resistant phenotype (Bailey et al. 2009). Triclosan exposure did not impair bacterial adhesion nonetheless biofilm development was inhibited at high concentrations (Lubarsky et al. 2012). In a triclosan-tolerant E. coli mutant, some 33 of 38 genes belonging to the flagellar assembly pathway were upregulated (Lenahan et al. 2014).
Methylisothiazolinone-chloromethylisothiazolinone (MIT-CMIT) and triclosan compounds have been widely used in cosmetics, toiletries and in various industrial applications (Collier et al. 1990; Fewing and Menné 1999; Maillard et al. 2013). We have previously selected resistant isogenic mutants by using increasing MIT-CMIT and triclosan concentrations. These resistant strains exhibited an overexpression of peroxiredoxin detoxifying enzymes and/or a flagellin overproduction (Périamé et al. 2015). The phenotypic characteristics of these adapted strains were studied to identify metabolic mechanisms and/or envelope changes involved in the biocide stress.
Results and discussion Growth in the presence of cysteine or preservatives Four different concentration ranges were tested and the following results obtained: (i) at very low concentrations (0.2 to 4 µg.ml-1 for triclosan and 0.2 to 1.5. 10-3 % for MIT/CMIT), no inhibitory effects were observed on bacterial growth, -(ii) at intermediate concentrations (ranging from 9 to 15.6 µg.ml-1 for triclosan, and from 2.25.10-5 to 3.1.10-3 % for MIT/CMIT) growth inhibition was observed but bacteria
This article is protected by copyright. All rights reserved.
Accepted Article
were able to restart their growth when the preservative were removed, -(iii) at higher concentrations (125 µg.ml-1 for triclosan and 7.5.10-3 % for MIT-CMIT), growth inhibition was observed but bacteria were able to slowly re-grow when the preservative were removed and -(iv) above the latter concentrations there was no growth observed. As triclosan was sparingly soluble in water, the same concentration of this inhibitor was tested using two solvents, pure ethanol or DMSO, on the following strains eg1, eg14, and eg23 (Table 1). When DMSO was used, bacterial growth was not inhibited with the concentrations of triclosan previously tested. With ethanol (2.20% final concentration), bacteria were more susceptible, due to a probable synergy between the solvent and the biocide. Isogenic resistant strains (including M1-M4, T1-T3) could grow in the presence of the respective concentrations of preservatives. However, when the concentration of the biocide/preservative was high an increase in the lag phase was recorded. It has been reported previously that thiols containing SH-functional groups, like cysteine, are able to negate the inhibitory activity of MIT-CMIT on E. coli (Collier et al. 1990). When E. gergoviae eg1 was cultivated in the presence of MIT-CMIT plus cysteine (final concentration 12.5 µg.ml-1) the lag phase disappeared and the strains were able to grow normally. These data were associated with the peroxiredoxin overexpression in E. gergoviae adapted strains (Périamé et al. 2015). The latter activity is characterized by a peroxidatic cysteine that is oxidized to a sulfenic acid by hydroperoxides (Poynton and Hampton 2014). The inhibitory effects of the MIT-CMIT on the E. gergoviae growth are efficiently quenched by the addition of cystein. These data confirmed the possible involvement of peroxiredoxin previously described as being overexpressed in some E. gergoviae MIT-CMIT resistant derivatives (Périamé et al. 2015).
This article is protected by copyright. All rights reserved.
Accepted Article
Phenotypic analysis E. gergoviae eg1 strain (CIP 76.01), the parental strain of all isogenic derivatives, was used as a reference for comparison of carbon metabolism profiles. Substrate utilizations were measured as differences in the strains ability to carry out normal respiration according to different compound tested (190 tests) (Martins et al. 2013). The derivatives strains exhibited no significant metabolic differences compared to the wild-type strains, however the pentose phosphate metabolic pathways was preferentially used (data not shown). When strain eg1 was compared with eg21 and eg23, metabolic differences for two carbon substrates were observed, nevertheless when eg1 was compared to eg34, 64 different carbons substrates could be obtained (Table 1). The most important metabolic differences were noted among those E. gergoviae strains isolated from cosmetics along with the transformed strain overexpressing the marA gene involved in the MDR phenotype (Davin-Regli et al. 2006). Metabolic divergences were recorded in the three isolates obtained from cosmetic products (Table 2). These differences could reflect the metabolic versatility of each isolate regarding its adaptive response to the environmental stress.
Study of biofilm formation The biofilm formation was studied by comparing the isogenic strains adapted to MITCMIT, with eg23, eg1 and E. aerogenes ATCC™13048. Quantification of biofilm was determined following the measurement of the biomass that adhered to the polystyrene surface (O’Toole and Kolter 1998; Peeters et al. 2008). Independent of temperature (30 or 37°C), there were no differences in the production of biomass for either eg23 or eg1 in the presence of the MIT-CMIT or the disinfectant peroxyacid. At 37°C the derivative strains M1, M2, M3, and M4 produced an increase in cell
This article is protected by copyright. All rights reserved.
Accepted Article
biomass in the presence of peroxyacid (0.003 %), and a loss of biomass in the presence of MIT-CMIT (Fig. 1). The biomass of strain M4 was increased in the presence of the disinfectant peroxyacid. In contrast eg23, which was originally isolated from a cosmetic product, produced less biomass, although the ability of biofilm formation was more important in the presence of MIT-CMIT. At 30°C, with MIT-CMIT an increase in biomass production was observed for derivative strain M1 compared to eg1. It was similarly noted for M2 and M3 in presence of MIT-CMIT and peroxyacid. For these three derivates, the biofilm formation was reduced in presence of ADD/PHMB. For all conditions tested (with or without disinfectants or preservative) M4 exhibited a significant augmentation of biomass formation (more than a factor 2 compared to the eg1). This suggested that M4 exhibited diverse regulation of biofilm formation due to its adaptive change.
Motility assay
To test the swimming motility, selected strains including E. gergoviae eg1, M1 and M2, were stabbed into 0.3% agar containing medium (Fig. 2) and their growth was observed. These latter strains produced visible halos, however the growth diameter was reduced for the MIT-CMIT resistant mutants as the concentration of MIT-CMIT increased (from 2.2 cm>1.9 cm>1.5 cm>1.2 cm). These mutants may have an expression of flagella associated to surexpression of FliC, in particular in eg23 and M1, as previously observed (Périamé et al. 2015). In the case of eg23 a reduction in diameter (1.2 cm) was observed compared to that obtained for eg1 (2.2 cm). This may be caused by the overexpression of fliC that can attenuate the swimming phenotype.
This article is protected by copyright. All rights reserved.
Accepted Article
Swarming of bacteria is reflected by colonial outgrowths at the peripheral edges of the bacterial colony when using 0.5% agar. In this experiment eg1 and M1 presented smooth colonies with no roughened and irregular edges. Changes in the growth profile were recorded for M2-M4 and eg23, with these strains forming an irregular halo with limited outgrowths compared to the eg1 strain cultured under similar conditions (Fig. 2).
Twitching motility is a type of bacterial translocation over moist surfaces mediated by the extension-attachment-retraction of pili. Bacteria with no twitching phenotype developed smooth-margined colonies, with no visible bacterial extensions radiating from the central bacterial colony. On 1% agar plates, colonial borders with irregular appearance for all strains were observed with the exception of eg23 exhibiting changes in growth, forming a web-like structure. Eg23 expressed a specific migration profile resulting most likely from the accumulation of different adaptive mechanisms against antimicrobials present in the cosmetic product in which it was originally isolated.
Transmission Electron Microscopy (TEM) data The overexpression of flagellin-type protein has been reported in M1, T1 to T3 and in 2 strains isolated from cosmetics product eg23 and eg26 (Périamé et al. 2015). TEM studies were carried out to investigate whether or not differences in external appendices could be detected comparing eg1 with its resistant derivatives and other cosmetic isolates. TEM images indicated the presence of extended flagella in the isogenic mutants M1, T1, and T2 and in cosmetics strains eg23 and eg26. No similar changes were observed in eg1 strain (Fig. 3). The T3 colonies appeared as aggregated bacteria and it was not possible to observe the flagella in this case.
This article is protected by copyright. All rights reserved.
Accepted Article
Others external appendages similar to pili seemed to be involved in the adaptive profile in the M2, M3 and M4 mutants. The presence of pili in these isolates fitted in with the swimming profiles observed in the motility essays (Fig. 2). The numbers of external appendages appear to be reduced as the concentration of MIT-CMIT increases. E. gergoviae can be recovered from cosmetics products containing preservatives and packaged in conditions requiring disinfectants or preservatives. This work indicates that it can adapt itself by modulating its mobility and/or its ability to form a biofilm through the expression of pili and flagella. The flagellum is used by bacteria to swim in liquids and swarm over surfaces. In Pseudomonas aeruginosa, pili mediate two surface motility mechanisms: horizontally oriented crawling, by which bacterium moves lengthwise with directional persistence and vertically oriented walking (Conrad et al. 2011). The flagellum mediates near surface swimming and surface-anchored spinning. It has been demonstrated that flagella and motility are required for biofilm formation: flagella has a mechanosensory role in surface sensing at the initial stages of surface adhesion (Belas 2014). Nevertheless, studies in E. coli have demonstrated cell motility and flagella synthesis by the way of proteins Fli, need to be repressed in a second time, to promote the switch of biofilm formation and pili synthesis (Ogasawara et al. 2001). In E. coli, pili and flagella inhibit each other in a reciprocal fashion during biofilm formation (Guttenplan and Kearns 2013). Moreover, bacteria, which express flagella and present a swarming motility, are known to be more resistant in a planktonic phase (Lai et al. 2009). Regarding biocides tolerance study, up-regulation of flagellar-middle genes associated to an increase of swarm motility phenotype was characterized in a chlorhexidine-tolerant Salmonella mutant (Condell et al. 2014).
This article is protected by copyright. All rights reserved.
Accepted Article
These data could explain why derivatives mutants to MIT-CMIT can evolve more easily from a free-planktonic state to a surface-associated state and form microcolonies (Conrad et al. 2011). These modifications could be associated with a versatility concerning biofilm formation in presence on biocide, which can modulate their morphology (Dynes et al. 2009). These changes correlated with the resistance levels to biocides obtained in corresponding strains. The E. gergoviae response to triclosan and MIT-CMIT involves different regulations concerning several metabolism pathways: the production of an enzymatic detoxification barrier, an envelope modification with the production of appendages (pilli, flagellum) and a switch from planktonic to biofilm mode of growth for some resistant strains.
Materials and methods Bacterial strains and growth conditions Twenty-two Gram-negative bacteria were studied including eighteen E. gergoviae isolated between 1996 and 2011 from diverse origins (Table 1). They were previously studied for their genetic variability and biocides susceptibility (Périamé et al. 2014, 2015). The Eg34 E. gergoviae strain was transformed with plasmid p9, containing the marA-encoding gene from E. coli (Gambino et al. 1993). Control strains included susceptible E. aerogenes, ATCC™13048, resistant EA27 and CM64 (Ghisalberti et al. 2005). Bacteria were grown on Luria-Bertani (LB) agar 24 h prior to any assay at 30° C. Mueller-Hinton (MH) medium was used for susceptibility tests including selected antimicrobial agents and preservatives.
This article is protected by copyright. All rights reserved.
Accepted Article
Growth curves in presence of cysteine or in presence of preservatives E. gergoviae CIP 76.01 was cultivated at 30°C with and without MIT-CMIT in the presence or the absence of cysteine 12.5 µg.ml-1. The corresponding growth curve was monitored over a 23 h period at 600 nm using a microplate reader Saphas MP96 (Saphas Monaco). For each strain, one 96 wells plate was prepared with 100 μl of solutions containing serial dilutions of the methylisothiazolinone-chloromethylisothiazolinone (MIT-CMIT) and triclosan. DMSO and ethanol were also used as solvents for triclosan, to obtain a better solubilization at high biocide concentrations. The final concentrations did not exceed half of their respective MICs previously described (Periamé et al. 2014). Bacterial colonies obtained on MH agar plates were cultivated overnight in MH to obtain a cell suspension of 109 CFU ml-1. One μl of overnight solution was inoculated in 10 ml in fresh media in order to prepare a suspension of approximately 105 CFU.ml-1 in MH broth twice concentrated. One hundred μl of this bacterial suspension was added per well. Plates were incubated at 37°C for 24 h in an OmnilogTM microplate’s reader (BioLog Inc., Hayward, California) and growth curves were measured at OD 600 nm every minute.
Phenotypic microarray assays Bacterial strains and mutants were examined for phenotypic differences using OmnilogTM phenotypic microarrays. A sub-set of the full array was included, using plates denoted PM1 & PM2 which tested 190 carbon metabolism different conditions (BioLog Inc., Hayward, California) (Bochner 2009; Mackie et al. 2014). PM technology used the irreversible reduction of tetrazolium violet to formazan as a reporter of active metabolism. The dye reduction causes formation of a purple color
This article is protected by copyright. All rights reserved.
Accepted Article
that is recorded by a change coupled-device camera providing quantitative and kinetic informations about the growth curve. Colonies onto fresh MH agar plates were picked and re-suspended in 15 ml of IF-0 (Biolog) until a cell density of 42% transmittance was reached (Biolog turbidimeter). Four ml of this suspension were then added to 20 ml of inoculation fluid (IF-0 Biolog) containing dye. Wells of the PM1 and PM2 plates were then inoculated with 100 μl of this final bacterial suspension. Plates were incubated at 37°C for 48 h in OmnilogTM microplate reader. This instrument monitors changes in the bacterial culture respiration in the individual wells with an optical density reading every 15 min over the incubation. The reference and mutant strains were analysed using OmnilogTM software and negative controls (wells inoculated with growth medium alone) were subtracted. PM data were analyzed in comparing gain or lost in respiration area for each condition for E. gergoviae eg1 strain (used as a reference).
Biofilm formation The biofilm formation was tested for the strains eg1, eg23, along with the MIT-CMIT derivatives strains. E. aerogenes ATCC™13048 was included as a control. Attached cells were quantified using the described protocol (O’Toole and Kolter 1998). Cell suspensions were prepared to obtain about 105 colony-forming units (CFU).ml-1. Bacteria were cultivated 48 h in MH broth at 30°C (optimal temperature of growth for E. gergoviae) and 37°C (classical temperature used for growth of Enterobacteriacae) without agitation in 96 wells plates (CORNING incorporated COSTAR, polystyrene serie 3598). The influence on biofilm formation was investigated using a preservative (MIT-CMIT) or a disinfectant (peroxyacid). These compounds were tested at their minimal inhibitory concentration (MIC). Unattached cells were removed by rinsing the
This article is protected by copyright. All rights reserved.
Accepted Article
96-well plates three times, and attached cells were incubated with 1% of crystal violet (CV) for 20 min, allowing the staining of cell biomass for quantification. CV was then solubilized by adding 200 µl of 90% (v/v) ethanol and the absorbance was monitored at 570 nm. Comparisons between strains of their ability to form biofilm, was determined by calculating the ratio of biomass formed for E. gergoviae eg1 (positive control) under biomass formed by other strain. The ability to form biofilm was considered to increase when ratio is superior to 1.
Motility assays Changes in motility were assayed by using different growth conditions on agar plates to respectively evaluate swarming, twitching and swimming: the first medium used was M8 Medium and 0.5% bactoagar (Kohler et al. 2000 ). The second medium was 1% (w/v) Tryptone, 0.5% (w/v) NaCl, 1% (w/v) bactoagar and 0.5% (w/v) yeast extract; the final medium included 1% Tryptone, 0.5% NaCl, and 1% bactoagar, (O' Toole and Kolter 1998). For the first and the second media, bacteria were cultivated to OD600 0.5 at 30°C and 10 µl of suspension was loaded on the plate middle. For the last medium, isolated colonies were picked and deposited on plates. Experiments were repeated three times. Bacterial motility was analyzed after 24 h at 37°C and 48 h at room temperature for the first medium and after 24 h of growth at 37°C for the second medium and the third medium (Rashid and Kornberg 2000). Transmission electron microscopy Strains were streaked on LB agar and grown at 37°C overnight. Several colonies were re-suspended in 0.1 mol phosphate buffer (pH 7.4) to reach approximately 1 unit of OD 600 nm. Copper grids were incubated for 5 min with one drop of cell suspension, and negatively stained with phosphotungstic acid 1% for 20-30 s. The
This article is protected by copyright. All rights reserved.
Accepted Article
morphology of each strain was analyzed under transmission electron microscope (Zeiss EM 912).
Acknowledgments We thank Jean-Paul Chauvin, Fabrice Richard, Aïcha Aouane, (plateforme de microscopie de l’IBDM, AMU) for their assistance in electron microscopy. This work was supported by Aix-Marseille Université. M.P. is recipient from a PACAEntreprise thesis fellowship.
Conflict of interest No conflict of interest declared
References Bailey, A.M., Constantinidou, C., Ivens, A., Garvey, M.I., Webber, M.A., Coldham, N., Hobman, J.L., Wain, J., Woodward, M.J. and Piddock, L.J. (2009) Exposure of Escherichia coli and Salmonella enterica serovar Typhimurium to triclosan induces a species-specific response, including drug detoxification. J Antimicrob Chemother 64, 973-85.
Belas, R. (2014) Biofilms, flagella, and mechanosensing of surfaces by bacteria. Trends in Microbiol 22, 517-27.
Bochner, B.R. (2009) Global phenotypic characterization of bacteria. FEMS Microbiol Rev 33, 191–205. Brenner, D.J., Richard, C., Steigerwalt, A.G., Asbury, M.A. and Mandel M. (1980).
This article is protected by copyright. All rights reserved.
Accepted Article
Enterobacter gergoviae sp. Nov.: a new species of Enterobacteriaceae found in clinical specimens and the environment. Int J Syst Bacteriol 30, 1-6.
Collier, P.J., Ramsey, A.J., Austin, P. and Gilbert, P. (1990) Growth inhibitory and biocidal activity of some isothiazolone biocides. J Appl Microbiol 69, 569–77.
Condell, O., Power, K.A., Händler, K., Finn, S., Sheridan, A., Sergeant, K., Renaut, J., Burgess, C.M. et al. (2014) Comparative analysis of Salmonella susceptibility and tolerance to the biocide chlorhexidine identifies a complex cellular defense network. Front Microbiol 5, 373.
Conrad, J.C., Gibiansky, M.L., Jin, F., Gordon, V.D., Motto, D.A., Mathewson, M.A., Stopka, W.G., Zelasko, D.C., Shrout, J.D. and Wong, G.C. (2011) Flagella and pili-mediated near-surface single-cell motility mechanisms in P. aeruginosa. Biophys J 100,1608-16.
Davin-Regli, A., Chollet, R., Bredin, J., Chevalier, J., Lepine, F. and Pagès, J.M. (2006) Enterobacter gergoviae and the Prevalence of Efflux in Parabens Resistance. J Antimicrob Chemother 57, 757–760.
Dynes, J.J., Lawrence, J.R., Korber, D.R., Swerhone, G.D., Leppard, G.G. and Hitchcock, A.P. (2009) Morphological and biochemical changes in Pseudomonas fluorescens biofilms induced by sub-inhibitory exposure to antimicrobial agents. Can J Microbiol 55,163-78.
Fewings, J. and Menné, T. (1999) An update of the risk assessment for methylchloroisothiazolinone/methylisothiazolinone (MCI/MI) with focus on rinse-off products. Contact Dermatitis 41, 1-13.
This article is protected by copyright. All rights reserved.
Accepted Article
Gambino, L., Gracheck, S.J. and Miller, P.F. (1993) Overexpression of the MarA positive regulator is sufficient to confer multiple antibiotic resistance in Escherichia coli. J Bacteriol 175, 2888-94.
Ghisalberti, D., Masi, M., Pagès, J.M. and Chevalier, J. (2005) Chloramphenicol and expression of multidrug efflux pump in Enterobacter aerogenes. Biochem Biophys Res Commun 328, 1113-8.
Guttenplan, S.B. and Kearns, D.B. (2013) Regulation of flagellar motility during biofilm formation.FEMS Microbiol Rev 37, 849-71.
Köhler, T., Curty, L.K., Barja, F., van Delden, C. and Pechère, J.-C. (2000) Swarming of Pseudomonas aeruginosa is dependent on cell-to-cell signaling and requires flagella and Pili. J Bacteriol 182, 5990–6.
Lai, S., Tremblay, J. and Déziel, E. (2009) Swarming motility: a multicellular behaviour conferring antimicrobial resistance. Environ Microbiol.11, 126-36.
Lenahan, M., Sheridan, Á., Morris, D., Duffy, G., Fanning, S. and Burgess, C.M. (2014) Transcriptomic analysis of triclosan-susceptible and -tolerant Escherichia coli O157:H19 in response to triclosan exposure. Microb Drug Resist 20, 91-103.
Lubarsky, H.V., Gerbersdorf, S.U., Hubas, C., Behrens, S., Ricciardi, F. and Paterson, D.M. (2012) Impairment of the bacterial biofilm stability by triclosan PLoS One 7, e31183.
Mackie, A.M., Hassan, K.A., Paulsen, I.T. and Tetu, S.G. (2014) Biolog Phenotype Microarrays for phenotypic characterization of microbial cells. Methods Mol Biol 1096, 123-30.
This article is protected by copyright. All rights reserved.
Accepted Article
Maillard, J-Y., Bloomfield, S., Coelho, J.R., Collier, P., Cookson, B., Fanning, S., Hill, A., Hartemann, P. et al. (2013) Does microbicide use in consumer products promote antimicrobial resistance ? A critical review and recommendations for a cohesive approach to risk assessment. Microb Drug Resist 19, 344-54.
Martins, M., McCusker, M.P., McCabe, E.M., O'Leary, D., Duffy, G. and Fanning, S. (2013) Evidence of metabolic switching and implications for food safety from the phenome(s) of Salmonella enterica serovar Typhimurium DT104 cultured at selected points across the pork production food chain. Appl Environ Microbiol 79, 5437-49.
Ogasawara, H., Yamamoto, K. and Ishihama, A. (2001) Role of the biofilm master regulator CsgD in cross-regulation between biofilm formation and flagellar synthesis. J Bacteriol 193, 2587-97.
Ortega Morente, E., Fernández-Fuentes, M.A., Grande Burgos, M.J., Abriouel, H., Pérez Pulido, R. and Gálvez, A. (2013) Biocide tolerance in bacteria. Int J Food Microbiol 162,13-25.
O’Toole, G.A. and Kolter, R. (1998) Initiation of biofilm formation in Pseudomonas fluorescens WCS365 proceeds via multiple, convergent signalling pathways: a genetic analysis. Mol Microbiol 28, 449–61.
Peeters, E., Nelis, H.J. and Coenye, T. (2008) Comparison of multiple methods for quantification of microbial biofilms grown in microtiter plates. J Microbiol Methods 72, 157–65.
This article is protected by copyright. All rights reserved.
Accepted Article
Périamé, M., Pagès, J-M. and Davin-Regli, A. (2014) Enterobacter gergoviae adaptation to preservatives commonly used in cosmetic industry. Int J Cosmet Sci 36, 386-95.
Périamé, M., Pagès, J-M. and Davin-Regli A. (2015) Enterobacter gergoviae membrane modifications are involved in the adaptive response to preservatives used in cosmetic industry.J Appl Microbiol 118, 49-61.
Poynton, R.A. and Hampton, M.B. (2014) Peroxiredoxins as biomarkers of oxidative stress. Biochim Biophys Acta 1840, 906-12.
Rashid, M.H. and Kornberg, A. (2000) Inorganic polyphosphate is needed for swimming, swarming, and twiching motilities of Pseudomonas aeruginosa. PNAS 97, 4885-90.
Sheridan, A., Lenahan, M., Condell, O., Bonilla-Santiago, R., Sergeant, K., Renaut, J., Duffy, G., Fanning, S. et al. (2013) Proteomic and phenotypic analysis of triclosan tolerant verocytotoxigenic Escherichia coli O157:H19. J Proteomics 80C, 78-90.
This article is protected by copyright. All rights reserved.
metabolic differences
Accepted Article
Bacterial strains Origin
Name in this study
(number/190 differences total)
References compared to eg1
Enterobacter gergoviae CIP 76.01 (CDC604-77)
Brenner et al. 1980
eg1
-
Eye liner Davin-Regli et al. (2006)
Cosmetic Industry1,1997
eg10
3
CIP 105140 Davin-Regli et al. (2006)
used for challenge test, 1998
eg14
9
Exfoliating gel Davin-Regli et al. (2006)
Cosmetic Industry2
eg18
8
Syrup Davin-Regli et al. (2006)
Food Industry , Canada
eg 21
2
Rose bath Periamé et al. (2014)
Cosmetic Industry3, 2009
eg 23
2
Cream Periamé et al. (2014)
Cosmetic Industry3, 2009
eg 24
6
Cream Periamé et al. (2014)
Cosmetic Industry3, 2009
eg 25
32
Shower gel Periamé et al. (2014)
Cosmetic Industry3, 2009
eg 26
20
Foam Periamé et al. (2014)
Cosmetic Industry3, 2010
eg 38
20
CIP 76.01 (p9) Davin-Regli et al. (2006)
eg1 transformed with plasmid p9 (Apr; 2.2 kb marAB–E.coli insert in pBR322)
eg 34
64
MIT-CMIT derivative Periamé et al. (2015)
eg1 derivated with MIT-CMIT 1.875.10 %; 3.75.10-4%; 4.5.10-4% & 6.10-4%
M1, M2, M3, M4
nd
Triclosan derivative Periamé et al. (2015)
eg1 derivated with Triclosan 5; 10; 20 μg/ml
T1, T2, T3
nd
ATCC 13048
ATCC13048
nd
EA27 Ghisalberti et al. (2005) CM64 Ghisalberti et al. (2005)
EA27
62
CM64
26
-4
Enterobacter aerogenes
Table 1: Bacterial strains and derivatives used in this study and metabolic differences obtained with various E. gergoviae and the multidrug resistant E. aerogenes, compared to the reference strain E. gergoviae eg1, by OmnilogTM phenotypic microarrays. Plates PM1 & PM2, which tested 190 carbon metabolisms in different conditions, were used. Significant results for all strains tested are presented. nd, not determined.
This article is protected by copyright. All rights reserved.
Accepted Article
Compounds involved in:
Metabolic difference compared to eg1 reference strain eg25
eg26
eg34
Metabolic Pathways
yes
yes
yes
Microbial metabolism in diverse environments
yes
yes
yes
Pentose and glucuronate interconversions
no
yes
yes
Starch and sucrose metabolism
no
yes
yes
Amino sugar and nucleotide sugar metabolism
yes
no
yes
Biosynthesis of secondary metabolites
yes
no
yes
Glycoxylate and dicarboxylate metabolism
yes
no
no
ABC Transporters
no
yes
yes
PTS phosphotransferase
no
yes
yes
Table 2: Presentation of the various metabolic pathways which were significatively used by three E. gergoviae isolated from cosmetics compared to the eg1 strain profile (reference strain).
This article is protected by copyright. All rights reserved.
Accepted Article
Figures legend Figure 1: Crystal violet quantification of biofilm formation at 30 and 37°C, with or without disinfectant (MIT-CMIT or peroxyacid). Comparison of ability to form biofilm was measured by calculating the ratio of biomass formed for eg1 under biomass formed by other strain. The ability to form biofilm was considered to increase when ratio is superior to 1. Strains were E. gergoviae eg1, eg23, derivatives M1 to M4 and for reference E. aerogenes ATCC 13048.
Figure 2: Mobility experiments. For the swimming motility (upper line), bacteria were inoculated on a surface of 0.3% agar plates. Swimming haloes were determined after 18 h of incubation at 28°C. Average swimming halo diameters (mm) are then measured. Swarming (middle line) can be objective by outgrowths aureole around the periphery of the bacterial colony in small decreases in agar from 0.5%. Twitching in 1% agar require to analyse the borders of the migration zone. Observation of colonies is made for their irregular appearance on periphery or extensions irradiating from the migration zones. Bacteria with no twitching phenotype developed smoothmargined colonies. Strain eg1 was tested as reference, strain eg23 is of cosmetic origin and derivatives M1 to M4 corresponded to eg1 selected on increased concentration of MIT-CMIT.
Figure 3: Transmission electron microscopy (TEM) Copper grids were incubated for 5 minutes with one drop of cell suspension, and negatively stained with phosphotungstic acid 1% for 20-30 seconds. The morphology of each strain was analyzed under transmission electron microscope (Zeiss EM 912). A: eg1; B1: M1; B2:M2; B3:M3; B4:M4; C1:T1; C2:T2; C3:T3; D1: eg23; D2: eg26.
This article is protected by copyright. All rights reserved.
Accepted Article
Figure 1
This article is protected by copyright. All rights reserved.
Accepted Article
Figure 2
Figure 3
This article is protected by copyright. All rights reserved.
Article Type: Original Article
Phenotypic changes contributing to Enterobacter gergoviae biocide resistance
M. Periame1, N.Philippe13, O. Condell2, S. Fanning2, J-M. Pagès1 and A. Davin-Regli1
1
UMR-MD1, Aix-Marseille Université, IRBA, Transporteurs Membranaires,
Chimiorésistance et Drug Design, Marseille, France 2
UCD-Centre for Food Safety, School of Public Health, Physiotherapy & Population
Science, University College Dublin, Belfield, Dublin 4, Ireland. 3
Laboratoire Information Génomique et Structurale (IGS), UMR 7256 (IMM FR 3479)
CNRS Aix-Marseille Université, Luminy campus, Marseille, France
Corresponding author: Dr. Anne Davin-Regli, UMR-MD1, Transporteurs Membranaires, Chimiorésistance et Drug Design, Facultés de Médecine et Pharmacie, 27 Bd Jean Moulin, 13385 Marseille cedex 05, France ([email protected]).
Keywords: Enterobacter gergoviae, preservatives, cosmetics, triclosan, biofilm, flagellum, fimbriae
Running headline: Mechanisms involved in E. gergoviae biocides adaptation
This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process which may lead to differences between this version and the Version of Record. Please cite this article as an 'Accepted Article', doi: 10.1111/lam.12435 This article is protected by copyright. All rights reserved.
Accepted Article
Significance and impact of the study Recurrent contaminations of cosmetics products by Enterobacter gergoviae, needed a better understanding concerning the bacterial adaptation to preservative agents, with particular concern to triclosan and MIT-CMIT. We demonstrated that bacteria response is associated to various mechanisms represented by expression of external appendages (pili or fimbriae) that control cell motility and biofilm formation and evolving as the concentration of biocides adaptation increased. Such mechanisms which are not chemical specific can also promote a cross-resistance to other biocidal agents. The characterization of E gergoviae adaptability to biocides allows industry to adjust the ranges of concentrations and composition of preservatives in formula.
Abstract Enterobacter gergoviae is a recurrent contaminant of cosmetic and hygiene products. To understand how this bacterium adapts to biocides, we studied E. gergoviae CIP 76.01 and its triclosan and MIT-CMIT tolerant isogenic mutants. They were compared with others also isolated from contaminated cosmetics. Phenotypic differences were noted and these included changes in the bacterial envelope and flagella along with differences in motility, and biofilm growth rates. Triclosan and MIT-CMIT derivatives expressed flagella and other MIT-CMIT derivatives exhibited some external appendages. Those bacteria expressing a high-level MIC to MIT-CMIT, expressed a strong biofilm formation. No differential phenotypes were noted for carbon source utilisation. E. gergoviae demonstrated a diverse response to both of these preservatives contained in cosmetic preparations, depending on their concentrations. Interestingly, this adaptive response is associated with modifications of filament structure-related proteins contributing to increase the organism motility and the production of biofilm.
This article is protected by copyright. All rights reserved.
Accepted Article
Introduction Enterobacter gergoviae is often identified in contaminated manufactured cosmetic products. Bacteriostatic action of preservatives, genetic diversity of strains causing contaminations and the emergence of well-adapted bacteria from these matrices, support E. gergoviae increased tolerance to preservatives (Davin-Regli et al. 2006; Périamé et al. 2014). Diverse mechanisms have been associated with biocide tolerance including alterations of membrane permeability, target gene alterations and expression of efflux systems (Ortega Morente et al. 2013). It has been shown that E. gergoviae presented a natural resistance to parabens by producing an esterase (Davin-Regli et al. 2006).
These threats to public health are a constant challenge for the cosmetics industry and this feature required novel and appropriate solutions to manage the risk. It was reported that the capacity of bacteria to form biofilms appeared to be a key in the bacterial adaptation to these compounds (Lubarsky et al. 2012). It has been demonstrated previously that the biofilm phenotype described in Escherichia coli and its triclosan tolerant mutant were distinct (Sheridan et al. 2013; Lenahan et al. 2014). In the latter case, the tolerant strain exhibited strong biofilm formation associated with the expression of curli fimbriae with the over-expression of FliC protein that contributes to the filament structure of flagella. This structure has been implicated in the biofilm formation by having an important mechanosensory role during the early stages of surface adhesion leading to the biofilm organisation (Belas 2014).
This article is protected by copyright. All rights reserved.
Accepted Article
Triclosan is a broad-spectrum antibacterial compound widely used in personal care products. It inhibits the enoyl-acyl carrier protein reductase FabI and mutations in the fabI gene and/or modifications of membrane permeability by over-expressing efflux pumps contribute to a resistant phenotype (Bailey et al. 2009). Triclosan exposure did not impair bacterial adhesion nonetheless biofilm development was inhibited at high concentrations (Lubarsky et al. 2012). In a triclosan-tolerant E. coli mutant, some 33 of 38 genes belonging to the flagellar assembly pathway were upregulated (Lenahan et al. 2014).
Methylisothiazolinone-chloromethylisothiazolinone (MIT-CMIT) and triclosan compounds have been widely used in cosmetics, toiletries and in various industrial applications (Collier et al. 1990; Fewing and Menné 1999; Maillard et al. 2013). We have previously selected resistant isogenic mutants by using increasing MIT-CMIT and triclosan concentrations. These resistant strains exhibited an overexpression of peroxiredoxin detoxifying enzymes and/or a flagellin overproduction (Périamé et al. 2015). The phenotypic characteristics of these adapted strains were studied to identify metabolic mechanisms and/or envelope changes involved in the biocide stress.
Results and discussion Growth in the presence of cysteine or preservatives Four different concentration ranges were tested and the following results obtained: (i) at very low concentrations (0.2 to 4 µg.ml-1 for triclosan and 0.2 to 1.5. 10-3 % for MIT/CMIT), no inhibitory effects were observed on bacterial growth, -(ii) at intermediate concentrations (ranging from 9 to 15.6 µg.ml-1 for triclosan, and from 2.25.10-5 to 3.1.10-3 % for MIT/CMIT) growth inhibition was observed but bacteria
This article is protected by copyright. All rights reserved.
Accepted Article
were able to restart their growth when the preservative were removed, -(iii) at higher concentrations (125 µg.ml-1 for triclosan and 7.5.10-3 % for MIT-CMIT), growth inhibition was observed but bacteria were able to slowly re-grow when the preservative were removed and -(iv) above the latter concentrations there was no growth observed. As triclosan was sparingly soluble in water, the same concentration of this inhibitor was tested using two solvents, pure ethanol or DMSO, on the following strains eg1, eg14, and eg23 (Table 1). When DMSO was used, bacterial growth was not inhibited with the concentrations of triclosan previously tested. With ethanol (2.20% final concentration), bacteria were more susceptible, due to a probable synergy between the solvent and the biocide. Isogenic resistant strains (including M1-M4, T1-T3) could grow in the presence of the respective concentrations of preservatives. However, when the concentration of the biocide/preservative was high an increase in the lag phase was recorded. It has been reported previously that thiols containing SH-functional groups, like cysteine, are able to negate the inhibitory activity of MIT-CMIT on E. coli (Collier et al. 1990). When E. gergoviae eg1 was cultivated in the presence of MIT-CMIT plus cysteine (final concentration 12.5 µg.ml-1) the lag phase disappeared and the strains were able to grow normally. These data were associated with the peroxiredoxin overexpression in E. gergoviae adapted strains (Périamé et al. 2015). The latter activity is characterized by a peroxidatic cysteine that is oxidized to a sulfenic acid by hydroperoxides (Poynton and Hampton 2014). The inhibitory effects of the MIT-CMIT on the E. gergoviae growth are efficiently quenched by the addition of cystein. These data confirmed the possible involvement of peroxiredoxin previously described as being overexpressed in some E. gergoviae MIT-CMIT resistant derivatives (Périamé et al. 2015).
This article is protected by copyright. All rights reserved.
Accepted Article
Phenotypic analysis E. gergoviae eg1 strain (CIP 76.01), the parental strain of all isogenic derivatives, was used as a reference for comparison of carbon metabolism profiles. Substrate utilizations were measured as differences in the strains ability to carry out normal respiration according to different compound tested (190 tests) (Martins et al. 2013). The derivatives strains exhibited no significant metabolic differences compared to the wild-type strains, however the pentose phosphate metabolic pathways was preferentially used (data not shown). When strain eg1 was compared with eg21 and eg23, metabolic differences for two carbon substrates were observed, nevertheless when eg1 was compared to eg34, 64 different carbons substrates could be obtained (Table 1). The most important metabolic differences were noted among those E. gergoviae strains isolated from cosmetics along with the transformed strain overexpressing the marA gene involved in the MDR phenotype (Davin-Regli et al. 2006). Metabolic divergences were recorded in the three isolates obtained from cosmetic products (Table 2). These differences could reflect the metabolic versatility of each isolate regarding its adaptive response to the environmental stress.
Study of biofilm formation The biofilm formation was studied by comparing the isogenic strains adapted to MITCMIT, with eg23, eg1 and E. aerogenes ATCC™13048. Quantification of biofilm was determined following the measurement of the biomass that adhered to the polystyrene surface (O’Toole and Kolter 1998; Peeters et al. 2008). Independent of temperature (30 or 37°C), there were no differences in the production of biomass for either eg23 or eg1 in the presence of the MIT-CMIT or the disinfectant peroxyacid. At 37°C the derivative strains M1, M2, M3, and M4 produced an increase in cell
This article is protected by copyright. All rights reserved.
Accepted Article
biomass in the presence of peroxyacid (0.003 %), and a loss of biomass in the presence of MIT-CMIT (Fig. 1). The biomass of strain M4 was increased in the presence of the disinfectant peroxyacid. In contrast eg23, which was originally isolated from a cosmetic product, produced less biomass, although the ability of biofilm formation was more important in the presence of MIT-CMIT. At 30°C, with MIT-CMIT an increase in biomass production was observed for derivative strain M1 compared to eg1. It was similarly noted for M2 and M3 in presence of MIT-CMIT and peroxyacid. For these three derivates, the biofilm formation was reduced in presence of ADD/PHMB. For all conditions tested (with or without disinfectants or preservative) M4 exhibited a significant augmentation of biomass formation (more than a factor 2 compared to the eg1). This suggested that M4 exhibited diverse regulation of biofilm formation due to its adaptive change.
Motility assay
To test the swimming motility, selected strains including E. gergoviae eg1, M1 and M2, were stabbed into 0.3% agar containing medium (Fig. 2) and their growth was observed. These latter strains produced visible halos, however the growth diameter was reduced for the MIT-CMIT resistant mutants as the concentration of MIT-CMIT increased (from 2.2 cm>1.9 cm>1.5 cm>1.2 cm). These mutants may have an expression of flagella associated to surexpression of FliC, in particular in eg23 and M1, as previously observed (Périamé et al. 2015). In the case of eg23 a reduction in diameter (1.2 cm) was observed compared to that obtained for eg1 (2.2 cm). This may be caused by the overexpression of fliC that can attenuate the swimming phenotype.
This article is protected by copyright. All rights reserved.
Accepted Article
Swarming of bacteria is reflected by colonial outgrowths at the peripheral edges of the bacterial colony when using 0.5% agar. In this experiment eg1 and M1 presented smooth colonies with no roughened and irregular edges. Changes in the growth profile were recorded for M2-M4 and eg23, with these strains forming an irregular halo with limited outgrowths compared to the eg1 strain cultured under similar conditions (Fig. 2).
Twitching motility is a type of bacterial translocation over moist surfaces mediated by the extension-attachment-retraction of pili. Bacteria with no twitching phenotype developed smooth-margined colonies, with no visible bacterial extensions radiating from the central bacterial colony. On 1% agar plates, colonial borders with irregular appearance for all strains were observed with the exception of eg23 exhibiting changes in growth, forming a web-like structure. Eg23 expressed a specific migration profile resulting most likely from the accumulation of different adaptive mechanisms against antimicrobials present in the cosmetic product in which it was originally isolated.
Transmission Electron Microscopy (TEM) data The overexpression of flagellin-type protein has been reported in M1, T1 to T3 and in 2 strains isolated from cosmetics product eg23 and eg26 (Périamé et al. 2015). TEM studies were carried out to investigate whether or not differences in external appendices could be detected comparing eg1 with its resistant derivatives and other cosmetic isolates. TEM images indicated the presence of extended flagella in the isogenic mutants M1, T1, and T2 and in cosmetics strains eg23 and eg26. No similar changes were observed in eg1 strain (Fig. 3). The T3 colonies appeared as aggregated bacteria and it was not possible to observe the flagella in this case.
This article is protected by copyright. All rights reserved.
Accepted Article
Others external appendages similar to pili seemed to be involved in the adaptive profile in the M2, M3 and M4 mutants. The presence of pili in these isolates fitted in with the swimming profiles observed in the motility essays (Fig. 2). The numbers of external appendages appear to be reduced as the concentration of MIT-CMIT increases. E. gergoviae can be recovered from cosmetics products containing preservatives and packaged in conditions requiring disinfectants or preservatives. This work indicates that it can adapt itself by modulating its mobility and/or its ability to form a biofilm through the expression of pili and flagella. The flagellum is used by bacteria to swim in liquids and swarm over surfaces. In Pseudomonas aeruginosa, pili mediate two surface motility mechanisms: horizontally oriented crawling, by which bacterium moves lengthwise with directional persistence and vertically oriented walking (Conrad et al. 2011). The flagellum mediates near surface swimming and surface-anchored spinning. It has been demonstrated that flagella and motility are required for biofilm formation: flagella has a mechanosensory role in surface sensing at the initial stages of surface adhesion (Belas 2014). Nevertheless, studies in E. coli have demonstrated cell motility and flagella synthesis by the way of proteins Fli, need to be repressed in a second time, to promote the switch of biofilm formation and pili synthesis (Ogasawara et al. 2001). In E. coli, pili and flagella inhibit each other in a reciprocal fashion during biofilm formation (Guttenplan and Kearns 2013). Moreover, bacteria, which express flagella and present a swarming motility, are known to be more resistant in a planktonic phase (Lai et al. 2009). Regarding biocides tolerance study, up-regulation of flagellar-middle genes associated to an increase of swarm motility phenotype was characterized in a chlorhexidine-tolerant Salmonella mutant (Condell et al. 2014).
This article is protected by copyright. All rights reserved.
Accepted Article
These data could explain why derivatives mutants to MIT-CMIT can evolve more easily from a free-planktonic state to a surface-associated state and form microcolonies (Conrad et al. 2011). These modifications could be associated with a versatility concerning biofilm formation in presence on biocide, which can modulate their morphology (Dynes et al. 2009). These changes correlated with the resistance levels to biocides obtained in corresponding strains. The E. gergoviae response to triclosan and MIT-CMIT involves different regulations concerning several metabolism pathways: the production of an enzymatic detoxification barrier, an envelope modification with the production of appendages (pilli, flagellum) and a switch from planktonic to biofilm mode of growth for some resistant strains.
Materials and methods Bacterial strains and growth conditions Twenty-two Gram-negative bacteria were studied including eighteen E. gergoviae isolated between 1996 and 2011 from diverse origins (Table 1). They were previously studied for their genetic variability and biocides susceptibility (Périamé et al. 2014, 2015). The Eg34 E. gergoviae strain was transformed with plasmid p9, containing the marA-encoding gene from E. coli (Gambino et al. 1993). Control strains included susceptible E. aerogenes, ATCC™13048, resistant EA27 and CM64 (Ghisalberti et al. 2005). Bacteria were grown on Luria-Bertani (LB) agar 24 h prior to any assay at 30° C. Mueller-Hinton (MH) medium was used for susceptibility tests including selected antimicrobial agents and preservatives.
This article is protected by copyright. All rights reserved.
Accepted Article
Growth curves in presence of cysteine or in presence of preservatives E. gergoviae CIP 76.01 was cultivated at 30°C with and without MIT-CMIT in the presence or the absence of cysteine 12.5 µg.ml-1. The corresponding growth curve was monitored over a 23 h period at 600 nm using a microplate reader Saphas MP96 (Saphas Monaco). For each strain, one 96 wells plate was prepared with 100 μl of solutions containing serial dilutions of the methylisothiazolinone-chloromethylisothiazolinone (MIT-CMIT) and triclosan. DMSO and ethanol were also used as solvents for triclosan, to obtain a better solubilization at high biocide concentrations. The final concentrations did not exceed half of their respective MICs previously described (Periamé et al. 2014). Bacterial colonies obtained on MH agar plates were cultivated overnight in MH to obtain a cell suspension of 109 CFU ml-1. One μl of overnight solution was inoculated in 10 ml in fresh media in order to prepare a suspension of approximately 105 CFU.ml-1 in MH broth twice concentrated. One hundred μl of this bacterial suspension was added per well. Plates were incubated at 37°C for 24 h in an OmnilogTM microplate’s reader (BioLog Inc., Hayward, California) and growth curves were measured at OD 600 nm every minute.
Phenotypic microarray assays Bacterial strains and mutants were examined for phenotypic differences using OmnilogTM phenotypic microarrays. A sub-set of the full array was included, using plates denoted PM1 & PM2 which tested 190 carbon metabolism different conditions (BioLog Inc., Hayward, California) (Bochner 2009; Mackie et al. 2014). PM technology used the irreversible reduction of tetrazolium violet to formazan as a reporter of active metabolism. The dye reduction causes formation of a purple color
This article is protected by copyright. All rights reserved.
Accepted Article
that is recorded by a change coupled-device camera providing quantitative and kinetic informations about the growth curve. Colonies onto fresh MH agar plates were picked and re-suspended in 15 ml of IF-0 (Biolog) until a cell density of 42% transmittance was reached (Biolog turbidimeter). Four ml of this suspension were then added to 20 ml of inoculation fluid (IF-0 Biolog) containing dye. Wells of the PM1 and PM2 plates were then inoculated with 100 μl of this final bacterial suspension. Plates were incubated at 37°C for 48 h in OmnilogTM microplate reader. This instrument monitors changes in the bacterial culture respiration in the individual wells with an optical density reading every 15 min over the incubation. The reference and mutant strains were analysed using OmnilogTM software and negative controls (wells inoculated with growth medium alone) were subtracted. PM data were analyzed in comparing gain or lost in respiration area for each condition for E. gergoviae eg1 strain (used as a reference).
Biofilm formation The biofilm formation was tested for the strains eg1, eg23, along with the MIT-CMIT derivatives strains. E. aerogenes ATCC™13048 was included as a control. Attached cells were quantified using the described protocol (O’Toole and Kolter 1998). Cell suspensions were prepared to obtain about 105 colony-forming units (CFU).ml-1. Bacteria were cultivated 48 h in MH broth at 30°C (optimal temperature of growth for E. gergoviae) and 37°C (classical temperature used for growth of Enterobacteriacae) without agitation in 96 wells plates (CORNING incorporated COSTAR, polystyrene serie 3598). The influence on biofilm formation was investigated using a preservative (MIT-CMIT) or a disinfectant (peroxyacid). These compounds were tested at their minimal inhibitory concentration (MIC). Unattached cells were removed by rinsing the
This article is protected by copyright. All rights reserved.
Accepted Article
96-well plates three times, and attached cells were incubated with 1% of crystal violet (CV) for 20 min, allowing the staining of cell biomass for quantification. CV was then solubilized by adding 200 µl of 90% (v/v) ethanol and the absorbance was monitored at 570 nm. Comparisons between strains of their ability to form biofilm, was determined by calculating the ratio of biomass formed for E. gergoviae eg1 (positive control) under biomass formed by other strain. The ability to form biofilm was considered to increase when ratio is superior to 1.
Motility assays Changes in motility were assayed by using different growth conditions on agar plates to respectively evaluate swarming, twitching and swimming: the first medium used was M8 Medium and 0.5% bactoagar (Kohler et al. 2000 ). The second medium was 1% (w/v) Tryptone, 0.5% (w/v) NaCl, 1% (w/v) bactoagar and 0.5% (w/v) yeast extract; the final medium included 1% Tryptone, 0.5% NaCl, and 1% bactoagar, (O' Toole and Kolter 1998). For the first and the second media, bacteria were cultivated to OD600 0.5 at 30°C and 10 µl of suspension was loaded on the plate middle. For the last medium, isolated colonies were picked and deposited on plates. Experiments were repeated three times. Bacterial motility was analyzed after 24 h at 37°C and 48 h at room temperature for the first medium and after 24 h of growth at 37°C for the second medium and the third medium (Rashid and Kornberg 2000). Transmission electron microscopy Strains were streaked on LB agar and grown at 37°C overnight. Several colonies were re-suspended in 0.1 mol phosphate buffer (pH 7.4) to reach approximately 1 unit of OD 600 nm. Copper grids were incubated for 5 min with one drop of cell suspension, and negatively stained with phosphotungstic acid 1% for 20-30 s. The
This article is protected by copyright. All rights reserved.
Accepted Article
morphology of each strain was analyzed under transmission electron microscope (Zeiss EM 912).
Acknowledgments We thank Jean-Paul Chauvin, Fabrice Richard, Aïcha Aouane, (plateforme de microscopie de l’IBDM, AMU) for their assistance in electron microscopy. This work was supported by Aix-Marseille Université. M.P. is recipient from a PACAEntreprise thesis fellowship.
Conflict of interest No conflict of interest declared
References Bailey, A.M., Constantinidou, C., Ivens, A., Garvey, M.I., Webber, M.A., Coldham, N., Hobman, J.L., Wain, J., Woodward, M.J. and Piddock, L.J. (2009) Exposure of Escherichia coli and Salmonella enterica serovar Typhimurium to triclosan induces a species-specific response, including drug detoxification. J Antimicrob Chemother 64, 973-85.
Belas, R. (2014) Biofilms, flagella, and mechanosensing of surfaces by bacteria. Trends in Microbiol 22, 517-27.
Bochner, B.R. (2009) Global phenotypic characterization of bacteria. FEMS Microbiol Rev 33, 191–205. Brenner, D.J., Richard, C., Steigerwalt, A.G., Asbury, M.A. and Mandel M. (1980).
This article is protected by copyright. All rights reserved.
Accepted Article
Enterobacter gergoviae sp. Nov.: a new species of Enterobacteriaceae found in clinical specimens and the environment. Int J Syst Bacteriol 30, 1-6.
Collier, P.J., Ramsey, A.J., Austin, P. and Gilbert, P. (1990) Growth inhibitory and biocidal activity of some isothiazolone biocides. J Appl Microbiol 69, 569–77.
Condell, O., Power, K.A., Händler, K., Finn, S., Sheridan, A., Sergeant, K., Renaut, J., Burgess, C.M. et al. (2014) Comparative analysis of Salmonella susceptibility and tolerance to the biocide chlorhexidine identifies a complex cellular defense network. Front Microbiol 5, 373.
Conrad, J.C., Gibiansky, M.L., Jin, F., Gordon, V.D., Motto, D.A., Mathewson, M.A., Stopka, W.G., Zelasko, D.C., Shrout, J.D. and Wong, G.C. (2011) Flagella and pili-mediated near-surface single-cell motility mechanisms in P. aeruginosa. Biophys J 100,1608-16.
Davin-Regli, A., Chollet, R., Bredin, J., Chevalier, J., Lepine, F. and Pagès, J.M. (2006) Enterobacter gergoviae and the Prevalence of Efflux in Parabens Resistance. J Antimicrob Chemother 57, 757–760.
Dynes, J.J., Lawrence, J.R., Korber, D.R., Swerhone, G.D., Leppard, G.G. and Hitchcock, A.P. (2009) Morphological and biochemical changes in Pseudomonas fluorescens biofilms induced by sub-inhibitory exposure to antimicrobial agents. Can J Microbiol 55,163-78.
Fewings, J. and Menné, T. (1999) An update of the risk assessment for methylchloroisothiazolinone/methylisothiazolinone (MCI/MI) with focus on rinse-off products. Contact Dermatitis 41, 1-13.
This article is protected by copyright. All rights reserved.
Accepted Article
Gambino, L., Gracheck, S.J. and Miller, P.F. (1993) Overexpression of the MarA positive regulator is sufficient to confer multiple antibiotic resistance in Escherichia coli. J Bacteriol 175, 2888-94.
Ghisalberti, D., Masi, M., Pagès, J.M. and Chevalier, J. (2005) Chloramphenicol and expression of multidrug efflux pump in Enterobacter aerogenes. Biochem Biophys Res Commun 328, 1113-8.
Guttenplan, S.B. and Kearns, D.B. (2013) Regulation of flagellar motility during biofilm formation.FEMS Microbiol Rev 37, 849-71.
Köhler, T., Curty, L.K., Barja, F., van Delden, C. and Pechère, J.-C. (2000) Swarming of Pseudomonas aeruginosa is dependent on cell-to-cell signaling and requires flagella and Pili. J Bacteriol 182, 5990–6.
Lai, S., Tremblay, J. and Déziel, E. (2009) Swarming motility: a multicellular behaviour conferring antimicrobial resistance. Environ Microbiol.11, 126-36.
Lenahan, M., Sheridan, Á., Morris, D., Duffy, G., Fanning, S. and Burgess, C.M. (2014) Transcriptomic analysis of triclosan-susceptible and -tolerant Escherichia coli O157:H19 in response to triclosan exposure. Microb Drug Resist 20, 91-103.
Lubarsky, H.V., Gerbersdorf, S.U., Hubas, C., Behrens, S., Ricciardi, F. and Paterson, D.M. (2012) Impairment of the bacterial biofilm stability by triclosan PLoS One 7, e31183.
Mackie, A.M., Hassan, K.A., Paulsen, I.T. and Tetu, S.G. (2014) Biolog Phenotype Microarrays for phenotypic characterization of microbial cells. Methods Mol Biol 1096, 123-30.
This article is protected by copyright. All rights reserved.
Accepted Article
Maillard, J-Y., Bloomfield, S., Coelho, J.R., Collier, P., Cookson, B., Fanning, S., Hill, A., Hartemann, P. et al. (2013) Does microbicide use in consumer products promote antimicrobial resistance ? A critical review and recommendations for a cohesive approach to risk assessment. Microb Drug Resist 19, 344-54.
Martins, M., McCusker, M.P., McCabe, E.M., O'Leary, D., Duffy, G. and Fanning, S. (2013) Evidence of metabolic switching and implications for food safety from the phenome(s) of Salmonella enterica serovar Typhimurium DT104 cultured at selected points across the pork production food chain. Appl Environ Microbiol 79, 5437-49.
Ogasawara, H., Yamamoto, K. and Ishihama, A. (2001) Role of the biofilm master regulator CsgD in cross-regulation between biofilm formation and flagellar synthesis. J Bacteriol 193, 2587-97.
Ortega Morente, E., Fernández-Fuentes, M.A., Grande Burgos, M.J., Abriouel, H., Pérez Pulido, R. and Gálvez, A. (2013) Biocide tolerance in bacteria. Int J Food Microbiol 162,13-25.
O’Toole, G.A. and Kolter, R. (1998) Initiation of biofilm formation in Pseudomonas fluorescens WCS365 proceeds via multiple, convergent signalling pathways: a genetic analysis. Mol Microbiol 28, 449–61.
Peeters, E., Nelis, H.J. and Coenye, T. (2008) Comparison of multiple methods for quantification of microbial biofilms grown in microtiter plates. J Microbiol Methods 72, 157–65.
This article is protected by copyright. All rights reserved.
Accepted Article
Périamé, M., Pagès, J-M. and Davin-Regli, A. (2014) Enterobacter gergoviae adaptation to preservatives commonly used in cosmetic industry. Int J Cosmet Sci 36, 386-95.
Périamé, M., Pagès, J-M. and Davin-Regli A. (2015) Enterobacter gergoviae membrane modifications are involved in the adaptive response to preservatives used in cosmetic industry.J Appl Microbiol 118, 49-61.
Poynton, R.A. and Hampton, M.B. (2014) Peroxiredoxins as biomarkers of oxidative stress. Biochim Biophys Acta 1840, 906-12.
Rashid, M.H. and Kornberg, A. (2000) Inorganic polyphosphate is needed for swimming, swarming, and twiching motilities of Pseudomonas aeruginosa. PNAS 97, 4885-90.
Sheridan, A., Lenahan, M., Condell, O., Bonilla-Santiago, R., Sergeant, K., Renaut, J., Duffy, G., Fanning, S. et al. (2013) Proteomic and phenotypic analysis of triclosan tolerant verocytotoxigenic Escherichia coli O157:H19. J Proteomics 80C, 78-90.
This article is protected by copyright. All rights reserved.
metabolic differences
Accepted Article
Bacterial strains Origin
Name in this study
(number/190 differences total)
References compared to eg1
Enterobacter gergoviae CIP 76.01 (CDC604-77)
Brenner et al. 1980
eg1
-
Eye liner Davin-Regli et al. (2006)
Cosmetic Industry1,1997
eg10
3
CIP 105140 Davin-Regli et al. (2006)
used for challenge test, 1998
eg14
9
Exfoliating gel Davin-Regli et al. (2006)
Cosmetic Industry2
eg18
8
Syrup Davin-Regli et al. (2006)
Food Industry , Canada
eg 21
2
Rose bath Periamé et al. (2014)
Cosmetic Industry3, 2009
eg 23
2
Cream Periamé et al. (2014)
Cosmetic Industry3, 2009
eg 24
6
Cream Periamé et al. (2014)
Cosmetic Industry3, 2009
eg 25
32
Shower gel Periamé et al. (2014)
Cosmetic Industry3, 2009
eg 26
20
Foam Periamé et al. (2014)
Cosmetic Industry3, 2010
eg 38
20
CIP 76.01 (p9) Davin-Regli et al. (2006)
eg1 transformed with plasmid p9 (Apr; 2.2 kb marAB–E.coli insert in pBR322)
eg 34
64
MIT-CMIT derivative Periamé et al. (2015)
eg1 derivated with MIT-CMIT 1.875.10 %; 3.75.10-4%; 4.5.10-4% & 6.10-4%
M1, M2, M3, M4
nd
Triclosan derivative Periamé et al. (2015)
eg1 derivated with Triclosan 5; 10; 20 μg/ml
T1, T2, T3
nd
ATCC 13048
ATCC13048
nd
EA27 Ghisalberti et al. (2005) CM64 Ghisalberti et al. (2005)
EA27
62
CM64
26
-4
Enterobacter aerogenes
Table 1: Bacterial strains and derivatives used in this study and metabolic differences obtained with various E. gergoviae and the multidrug resistant E. aerogenes, compared to the reference strain E. gergoviae eg1, by OmnilogTM phenotypic microarrays. Plates PM1 & PM2, which tested 190 carbon metabolisms in different conditions, were used. Significant results for all strains tested are presented. nd, not determined.
This article is protected by copyright. All rights reserved.
Accepted Article
Compounds involved in:
Metabolic difference compared to eg1 reference strain eg25
eg26
eg34
Metabolic Pathways
yes
yes
yes
Microbial metabolism in diverse environments
yes
yes
yes
Pentose and glucuronate interconversions
no
yes
yes
Starch and sucrose metabolism
no
yes
yes
Amino sugar and nucleotide sugar metabolism
yes
no
yes
Biosynthesis of secondary metabolites
yes
no
yes
Glycoxylate and dicarboxylate metabolism
yes
no
no
ABC Transporters
no
yes
yes
PTS phosphotransferase
no
yes
yes
Table 2: Presentation of the various metabolic pathways which were significatively used by three E. gergoviae isolated from cosmetics compared to the eg1 strain profile (reference strain).
This article is protected by copyright. All rights reserved.
Accepted Article
Figures legend Figure 1: Crystal violet quantification of biofilm formation at 30 and 37°C, with or without disinfectant (MIT-CMIT or peroxyacid). Comparison of ability to form biofilm was measured by calculating the ratio of biomass formed for eg1 under biomass formed by other strain. The ability to form biofilm was considered to increase when ratio is superior to 1. Strains were E. gergoviae eg1, eg23, derivatives M1 to M4 and for reference E. aerogenes ATCC 13048.
Figure 2: Mobility experiments. For the swimming motility (upper line), bacteria were inoculated on a surface of 0.3% agar plates. Swimming haloes were determined after 18 h of incubation at 28°C. Average swimming halo diameters (mm) are then measured. Swarming (middle line) can be objective by outgrowths aureole around the periphery of the bacterial colony in small decreases in agar from 0.5%. Twitching in 1% agar require to analyse the borders of the migration zone. Observation of colonies is made for their irregular appearance on periphery or extensions irradiating from the migration zones. Bacteria with no twitching phenotype developed smoothmargined colonies. Strain eg1 was tested as reference, strain eg23 is of cosmetic origin and derivatives M1 to M4 corresponded to eg1 selected on increased concentration of MIT-CMIT.
Figure 3: Transmission electron microscopy (TEM) Copper grids were incubated for 5 minutes with one drop of cell suspension, and negatively stained with phosphotungstic acid 1% for 20-30 seconds. The morphology of each strain was analyzed under transmission electron microscope (Zeiss EM 912). A: eg1; B1: M1; B2:M2; B3:M3; B4:M4; C1:T1; C2:T2; C3:T3; D1: eg23; D2: eg26.
This article is protected by copyright. All rights reserved.
Accepted Article
Figure 1
This article is protected by copyright. All rights reserved.
Accepted Article
Figure 2
Figure 3
This article is protected by copyright. All rights reserved.
