Phenotypic Changes in a Laboratory-Derived Ertapenem-Resistant Escherichia coli Strain

Journal of Chemotherapy Vol. 23 - n. 3 (135-139) - 2011 REVIEW phenotypic Changes in a laboratory-Derived ertapenem-Resistant EscherichiacoliStrain K.V. sAnTOs 1 - M.A.R. CARVALHO 2 - W.A. MARTins 3 - H.M. AnDRADE 3 L.C. VELOsO 2 - s.C. COUTinHO 2 - J.L. BAHiA 2 - J.P.L. AnDRADE 2 - A.C.M. APOLôniO 2 C.G. DiniZ 4 - J.R. niCOLi 2 - L.M. FARiAs 2 1 Departamento de Patologia, Centro de Ciências da saúde, Universidade Federal do Espírito santo, Vitória, Es, Brazil. Departamento de Microbiologia, instituto de Ciências Biológicas, Universidade Federal de Minas Gerais, Belo Horizonte, MG, Brazil. 3 Departamento de Parasitologia, instituto de Ciências Biológicas, Universidade Federal de Minas Gerais, Belo Horizonte, MG, Brazil. 4 Departamento de Parasitologia, Microbiologia e imunologia da Universidade Federal de Juiz de Fora, MG, Brazil. 2 Correspondence: L.M. Farias, Departamento de Microbiologia, instituto de Ciências Biológicas, Universidade Federal de Minas Gerais, Caixa Postal 486, 30.161-970 Belo Horizonte, MG, Brazil. Phone: 55.31.3499.27.59. Fax: 55.31.3499.27.30. E-mail: macedo@icb.ufmg.br and keniavaleria@gmail.com Summary The aim of this study was to identify phenotypic changes in a laboratory-derived strain of ertapenem-resistant Escherichia coli (Ec-ERT) when compared to its susceptible parent strain (Ec-WT). In both strains, we assessed both the effects of ertapenem via time-kill curves and the occurrence of cross resistance with other beta-lactams. The strains were compared based on growth pattern, biochemical-physiological profile and changes in the subproteome using 2D-DIGE followed by MALDI-TOF/TOF MS. To assess virulence, we employed a murine model of intraperitoneal infection in which we investigated the invasiveness of both strains. Growth persistence of the laboratory-derived resistant strain was observed via the time-kill curve assay, but cross resistance was not observed for other beta-lactams. We also observed a slower growth rate and changes in the biochemical and physiological characteristics of the drug-resistant bacteria. In the resistant strain, a total of 51 protein spots were increased in abundance relative to the wild-type strain, including an outer membrane protein A, which is related to bacterial virulence. The mouse infection assay showed a higher invasiveness of the Ec-ERT strain in relation to the Ec-WT strain. In conclusion, the alterations driven by ertapenem in E. coli reinforce the idea that antimicrobial agents may interfere in several aspects of bacterial cell biology, with possible implications for host-bacteria interactions. Key words: Antibiotics, resistance, virulence. inTRODUCTiOn With rare exceptions, once a new antibiotic is introduced into the market it is only a matter of time before resistant strains of bacteria emerge.1 From the standpoint of the bacterium, to efficiently produce an infection, especially in an antibiotic-saturated environment such as a hospital, it must be virulent and resistant to antimicrobials. in fact, there are reports of microorganisms that, when exposed to antibiotics, acquire molecular alterations with resultant implications for their pathogenic proprieties.2,3,4,5 indeed, antimicrobial resistance should be considered an ecological factor for colonization during antimicrobial therapy (i.e., a virulence factor).6 When susceptible bacterial populations are eliminated during antimicrobial therapy, a transient “ecological vacuum” is created in the treated individuals. This vacuum facilitates the recolonization of bacterial clones from the surrounding environment and/or the survival and predominance of resistant populations from the treated site.7,8 This phenomenon is particularly common after treatment with broad-spectrum antibiotics.9-12 Ertapenem is a carbapenem antibiotic that is active against most common pathogens, with the exception of enterococci, nonfermenters and methicillin-resistant staphylococci.13 Because ertapenem is a broad-spectrum antibiotic, there is a potential for the development of bacterial resistance in normal microflora during the antimicrobial therapy.14,15,16 Recently, we showed that a strain of Bacteroides fragilis resistant to ertapenem maintained a pathogenicity pattern similar to that of the susceptible parental strain, although a piperacillin/tazobactam-resistant strain has demonstrated enhanced pathogenic properties.17 To investigate whether these findings extend to other bacterial species, we examined changes in the biology of a laboratory-derived strain of Escherichia coli with resistance to ertapenem. E. coli was chosen as a model organism not only due to its ecological importance in the human intestinal microflora but also due to the extensive knowledge already accumulated regarding this species. © E.s.i.F.T. srl - Firenze MATERiALs AnD METHODs Bacterial strains: Two E. coli strains (E. coli ATCC 25922 and Ec-ERT) with different susceptibility patterns to ertapenem were assayed. The parent E. coli ATCC 25922 is a susceptible strain designated as wild-type (Ec-WT) (ertapenem minimal inhibitory concentrations [MiC] = 0.016 mg mL-1). Ec-ERT (ertapenem MiC = 8 mg mL-1) was the laboratory-derived drug-resistant strain, selected by in vitro exposure to increasing concentrations of ertapenem as previously described.17 Briefly, the selection of resistant bacteria was carried out by serially sub culturing the parental strain onto agar plates containing a linear gradient of the antibiotics. Gradients were prepared in Petri dishes, which were poured with two layers of agar. The bottom layer consisted of Brain Heart infusion agar (BHi) (Difco), allowed to harden with the plate slanted sufficiently to cover the entire bottom. The top layer, added to the dish in the normal position, contained antibiotics at concentrations of 2 x MiC. An inoculum of 109 colony forming units (CFU) (sufficiently high to be representative of bacterial strain) was homogeneously spread on each plate, and incubated for 24h, at 37°C. Colonies growing at the highest antibiotic concentration were sampled, checked for purity, grown overnight into antibiotic-free broth, and plated again on new antibiotic gradient plates. This procedure was repeated until the bacterial growth issn 1120-009x 136 K. V. sAnTOs - M.A. R. CARVALHO - W. A. MARTins - H. M. AnDRADE - L. C. VELOsO - s.C. COUTinHO - J. L. BAHiA - J.P. L. AnDRADE - A.C. M. APOLôniO - et al. occurred at concentrations above the breakpoint of each drug. MiCs were determined before and after each passage by the agar dilution technique as described bellow. in order to test the stability of the resistant phenotype the MiC was retest after five passages on antibiotic-free agar. The bacterial strains were cultured and maintained on BHi, at 37°C in a bacteriological chamber during the course of all experiments. Antibiotics: standard powder of ertapenem sodium (Merck, sharp & Dohme), piperacillin-tazobactam (Tazocin®, Lederle Piperacillin), ceftazidime (GlaxosmithKline), imipenem (Merck, sharp & Dohme), meropenem (AstraZeneca) and aztreonam (Bristol Myers squibb) were used in this study. Antibiotic solutions were freshly prepared according to manufacturer instructions. Time- kill curves: To confirm the ertapenem-resistant phenotype, time-kill curves under antimicrobial selective pressure were compared between wild-type and the laboratory-derived drug-resistant strain. Time-kill curves were obtained using peak concentrations of the indicated antibiotics in serum (8.0 mg mL1 ). Prior to the time-kill curve experiments, 3 to 5 colonies of the wild-type and derived drug-resistant strain were incubated overnight in 10 mL of BHi broth. Each overnight culture was then adjusted to a 0.05 optical density (550 nm) with fresh BHi broth to yield a starting inoculum of approximately 106 CFU mL-1. This concentration was confirmed by ten-fold serial dilution. Cultures were incubated at 37°C in the bacteriological chamber. When the bacterial population reached the logarithmic growth phase, ertapenem solution was added to each test vial to produce antibiotic concentrations of 8.0 mg mL-1. no antibiotic was added to the control vial. Cultures were then incubated at 37°C. At time points of 0, 1, 2, 3, 4, 5 and 6 h after antibiotic exposure, 1 mL samples of the cultures were aseptically withdrawn for bacterial quantification. Bacterial counts were determined by 1:10 serial dilution in sterile saline, and a 0.1 mL sample of each dilution was plated onto BHi Agar. Antibiotic carryover was addressed by using saline dilution techniques. The lower limit accuracy (LLA) of bacterial counting was 300 CFU mL-1. Cross-resistance investigation: Cross-resistance was evaluated for the derived ertapenem-resistant strain against piperacillin-tazobactam, ceftazidime, imipenem, meropenem and aztreonam by MiC determination. MiC determinations were performed by the standard microdilution method using MuellerHinton broth (Oxoid Ltd., Basingstoke, Hampshire, United Kingdom) and interpreted according to CLsi guidelines.18 Growth curves: Overnight cultures grown in BHi broth of each bacterial strain were adjusted to a 0.05 optical density (550 nm) with fresh BHi broth to yield a starting inoculum of approximately 106 CFU mL-1. Cultures were then incubated at 37°C in the bacteriological chamber, and an aliquot was removed to measure the optical density every 20 min, until an endpoint of 350 min of incubation. Biochemical assays: The biochemical characteristics of the wild-type and the laboratory-derived drug-resistant strain were comparatively investigated using the APi 20E identification system (BioMérieux, Marcy – i’Etoile, France). 2D differential in-gel electrophoresis (DIGE): 2D-DiGE technology was used to investigate the protein composition of both the Ec-ERT and Ec-WT strains. By labeling the protein samples prior to the electrophoretic processing, 2D-DiGE allows an accurate analysis of differentially expressed proteins in independent samples. Total protein was extracted from mid-log phase cultures as previously described5 and then resuspended in a buffer containing 7 M urea, 2 M thiourea, 4% CHAPs, 40 mM DTT, 2% iPG Buffer (pH 3-10), 40 mM Tris base, and protease inhibitor mix (GE Healthcare, Upsala, sweden). The protein content was measured with the 2-D Quant Kit (GE Healthcare). samples containing 160 mg protein from Ec-WT, Ec-ERT or Ec-WT/Ec-ERT 1:1 (internal standard) were labeled with 320 pMol Cy3, Cy5 and Cy2, respectively, for 30 min on ice. The reaction was then quenched with 10 mM lysine. samples were then pooled and mixed 1:1 with 2x sample buffer [7 M urea, 2 M thiourea, 2% CHAPs, 2% pharmalyte and 40 mM DTT (GE Healthcare)]. immobilized pH-gradient strips (iPG) (pH 3-10 nL, 18 cm) were rehydrated in Destreak plus iPGBUFFER (3-10 pH) (all GE Healthcare), and 480 mg total protein was applied per strip. They were then run in Ettan iPGphor system (GE Healthcare) for 65.000 Vh. strips were incubated for 15 min in 10 mL of rehydration buffer [50 mM Tris HCl, pH 8.8, 6 M urea, 30% (v/v) glycerol, 2% (w/v) sDs, 0.002% BPB, and 125 mM DTT], followed by a second incubation step in the same buffer solution, but with DTT replaced by 125 mM iodoacetamide. Proteins were separated in a 12% polyacrylamide gel using a Tris/glycine/sDs buffer at 200 W (constant). Gels were scanned on a Typhoon scanner, and differential analysis was performed using the imageMaster 2D Platinum software (both GE Healthcare). The protein spots of interest were excised from the gel manually. The excised protein spots were digested in gel with porcine trypsin (Promega), as previously described by santos et al.5 Briefly, the digested and dried peptide samples were dissolved in 30 ml of 5% formic acid in 50% acetonitrile. The dissolved samples were loaded on a Zip-Tip (C18 resin, Millipore Corporation), and the elution from the column was mixed with matrix (5 mg/mL recrystallized a-cyano-4-hydroxycinnamic acid) in a 1:1 proportions and then spotted on the target (AnchorChipTM 600/384, Bruker Daltonics) for MALDi-TOF-TOF (Autoflex iii, Bruker Daltonics, Billerica, UsA) analysis. Ms and tandem Ms analysis were performed as described by santos and colleagues5 with the instrument in positive/reflector mode controlled by FlexControlTM software. The results from the Ms/Ms were used to search the nCBinr protein database using MAsCOT® software. The statistical analyses of the sequences were determined by the probability-based MOWsE score offered by MAsCOT® software. A p-value of less than 0.05 was considered significant and was used to generate the results. Mouse infection: A murine model of intraperitoneal infection was used to access the pathogenicity of the resistant strain. All experimental procedures with animals were approved by the ethics committee in animal experimentation of the Universidade Federal de Minas Gerais (CETEA/UFMG). Three-week-old female swiss mice (Animal Care Center, Universidade Federal de Minas Gerais, Brazil) weighing 15 to 18 g were inoculated intraperitoneally with 0.1 mL of a culture containing approximately 107 CFU mL-1 of Ec-WT or Ec-ERT, according to dos santos et al.19 The bacterial inoculum was prepared by suspending overnight cultures in 3% porcine gastric mucin (sigmaAldrich). A control group was inoculated with the bacterial suspension vehicle. All experimental animals were sacrificed by cervical displacement after 24 h of infection. For microbiological analysis, samples of blood, peritoneal fluid and liver macerate were treated to a ten-fold serial dilution in buffered saline, with subsequent plating in Trypic soy Agar (Difco, UsA). The viable colonies were counted after overnight incubation at 37°C. REsULTs To confirm the ertapenem-resistant phenotype, time-kill curves under antimicrobial pressure were compared between the wild-type and the derived drug-resistant strain. Ertapenem exhibited a marked bactericidal effect against the wild-type strain (Ec-WT), whereas Ec-ERT strain was unaffected by this drug (Figure 1). Cross-resistance was not observed in the resistant strain (Ec-ERT) but there were slight variations in the MiC for imipenem, meropenem and ceftazidime and a significant increase in the MiC for aztreonam, as shown in Table 1. Re- 137 PHEnOTYPiC CHAnGEs in A LABORATORY-DERiVED ERTAPEnEM-REsisTAnT ESCHERICHIA COLI sTRAin garding the growth pattern, although the curves profiles were similar, the laboratory-derived drug-resistant strain displayed a slight delay in the growth rate, reaching the mid-log phase 40 minutes after the wild-type E. coli strain. fluid, spleen and liver of animals challenged with Ec-ERT (>105 CFU/g), while microorganisms were not detected in clinical specimens from animals challenged with the parental strain. DisCUssiOn FiGURE 1 - Time–kill curves of ertapenem (ERT) at 8.0 mg mL-1 against the wild-type (Ec-WT) and laboratory-derived drug-resistant E. coli (Ec-ERT). A growth control and the LLA (dashed line) are shown. Differences between the wild-type and laboratory-derived drug-resistant strain were observed in the biochemical-physiological characterization. The altered patterns for Ec-ERT were related to arginine dehydrolase production and melibiose fermentation. These differences, however, were not enough to modify the identity of the strain according to the identification scheme, and they were correctly identified as E. coli (>98% of trust). TABLE 1 - MICs for the susceptible E. coli strain (Ec-WT) and the laboratory-derived resistant strain (Ec-ERT) MIC (mg/mL) Drugs Piperacillin-tazobactam Ceftazidime Ec-WT Ec-ERT 0.5 0.5 <0.12 1.0 Ertapenem 0.016 8.0 imipenem 0.25 0.12 Meropenem 0.016 Aztreonam 0.06 0.06 8 The analysis of DiGE gels revealed changes in the abundance of numerous protein species in the resistant strain in relation to the susceptible wild-type strain. Approximately 666 spots were visualized, and 20 were increased in abundance in Ec-WT and 51 increased in Ec-ERT (Figure 2). One of the identified proteins increased in abundance in the resistant strain was OmpA (outer membrane protein A; gene ompA; molecular weight of 37292 kDa; swiss-Prot identification OMPA_ECOLi; required score of 50, founded score of 239) (Figures 2 and 3). in the model of murine peritoneal infection, after 24 h of infection, high bacteria counts were detected in the peritoneal The management of patients with complicated infections due to Enterobacteriaceae has increased in complexity because of the evolution of antibiotic resistance and the development of multidrug-resistance (MDR), E. coli being the most commonly isolated MDR pathogen.20,21 Although carbapenems are the most active agents against these bacteria, some resistant strains have already been reported.20,21,22 Resistance to carbapenems in Enterobacteriaceae may be related to carbapenemases or to dual mechanisms associated with the outer membrane permeability defect and beta-lactamases, such as AmpCs and EsBLs.13,23 Considering that E. coli ATCC 25922 lacks the ampC and plasmid-mediated beta-lactamase genes, the ertapenem resistance can be associated with deficiency in the expression of outer membrane proteins 24,23,21 and/or increased expression of several efflux systems, as observed in carbapenem resistant clinical isolates of Pseudomonas aeruginosa25 and Enterobacter cloacae.26 skurnik et al.27 reported a EsBLs producing Klebsiella pneumonia selected in vivo by an ertapenem-containing regimen in a patient with mediastinitis. The ribotyping showed that the carbapenem-resistant strain was a derivative of the original mediastinal isolate. As stated by the authors, this observation stresses the risk of selecting for a panpenem resistant strain of enterobacteria when ertapenem is used for the treatment of severe infections caused by EsBL producing enterobacteria. Mutations that confer antibiotic resistance to a microorganism reciprocally and negatively impact their fitness, with the rate of bacterial growth being a factor subject to the interferences of resistant genotype5,28, explaining the slight delay in the growth rate of the ERT-resistant strain. However, in many instances, it has been shown that compensatory mutations can restore fitness while maintaining antibiotic resistance.29,1, but this was not investigated in this study. To gain insight into the physiological changes conferring resistance to the Ec-ERT strain, we compared protein contents between the resistant and the susceptible wild-type strain using the DiGE technique. The numerous changes in protein abundance observed in the laboratory-derived drug-resistant strain suggest that resistant microorganisms may develop molecular changes in an effort to adapt to adverse environmental conditions4 affecting many aspects of bacterial metabolism, which may be reflected in their virulence parameters. indeed, the OmpA protein identified in Ms analysis plays an important role in bacterial virulence, as it has been associated with the following: invasion of brain microvascular endothelial cells by type 1 fimbrial modulation;30 invasion of the intestinal epithelium;31 the biofilm formation; and the serum resistance by interference with complement activation, inhibition of cytokine induction and the ability to multiply within macrophages.32,33,34 Corroborating the functions of OmpA in the bacterial virulence, the preliminary data of the murine model of intraperitoneal infection showed a higher invasive potential of the resistant strain (Ec-ERT) compared to the parental one. it is known that some antimicrobial drugs can stimulate bacterial adhesion and toxin production that can interfere with the phagocytic process.35,36 We showed recently that the resistance to piperacillin-tazobactam in E. coli ATCC 25922 leads to overall changes in the subproteome of this bacterium, highlighting increased abundance of proteins related to virulence, antibiotic resistance, and DnA protection during stress.5 We recently investigated the relation between resistance and pathogenicity in a B. fragilis strain resistant to piperacillin-tazobactam and er- 138 K. V. sAnTOs - M.A. R. CARVALHO - W. A. MARTins - H. M. AnDRADE - L. C. VELOsO - s.C. COUTinHO - J. L. BAHiA - J.P. L. AnDRADE - A.C. M. APOLôniO - et al. FiGURE 2 - DiGE analysis of differences in protein content between the parent strain (Ec-WT, Cy3) and the laboratory-derived ertapenem-resistant strain (Ec-ERT, Cy5). (A) Gel representative of the internal standard prepared by mixing together equal amounts of each sample (Ec-WT + EcERT, Cy2). (B) and (C) Fluorescent 2D gel images displaying spots of proteins with increased abundance in Ec-WT and Ec-ERT respectively (green marks). These digital images were generated by the scanner Typhoon 9410 Variable Mode imager (GE Healthcare), and were analyzed by the imageMaster 2D Platinum (GE Healthcare). iEF was performed with 480 mg of protein using 18 cm, 3-10nL pH range strips. sDs-PAGE was performed on 12% polyacrylamide gels. This gel is representative of six gel run. The white arrow indicates a protein spot identified in the Ms analysis. FiGURE 3 - MALDi-TOFTOF spectra obtained for OmpA (white arrow in FiGURE 2). Monoisotopic peptide masses were used to search protein databases to match and subsequently identify individual protein spots. The eight masses indicated were matched to OmpA. PHEnOTYPiC CHAnGEs in A LABORATORY-DERiVED ERTAPEnEM-REsisTAnT ESCHERICHIA COLI sTRAin tapenem.17 Diniz and colleagues2,3 also show that the phenotype of a metronidazole-resistant laboratory-derived strain of B. fragilis ATCC 25285 encompasses a broad range of traits, including differences in gene/protein expression3 and pathogenic properties.2 in conclusion, our results suggest that the phenotype of ertapenem-resistance can alter the physiology of E. coli, which may interfere with the management of infectious diseases involving this bacterium, due not only to the selection of resistant strains but also to interference with its pathogenicity. Therefore, further investigations are needed to elucidate the level of interference of the phenotype of ertapenem-resistance on the virulence of this resistant E. coli strain, selected in vitro after drug exposure. ACKnOWLEDGEMEnTs. The study was supported by grants from Conselho nacional de Desenvolvimento Científico e Tecnológico (CnPq) and Fundação de Amparo à Pesquisa do Estado de Minas Gerais (FAPEMiG). The authors are grateful to Luzia Rosa Resende and José sérgio Barros de souza for technical support. DisCLOsURE sTATEMEnT: The authors declare that they have no competing financial interests. REFEREnCEs 1 Wright GD. The antibiotic resistome: the nexus of chemical and genetic diversity. nat Rev Microbiol 2007; 5: 175-186. 2 Diniz CG, Cara DC, nicoli JR, Farias LD, De Carvalho MA. Effect of metronidazole on the pathogenicity of resistant Bacteroides strains in gnotobiotic mice. Antimicrob Agents Chemother 2000; 44: 2419-2423. 3 Diniz CG, Farias LM, Carvalho MA, Rocha ER, smith CJ. Differential gene expression in a Bacteroides fragilis metronidazole-resistant mutant. J Antimicrob Chemother 2004; 54: 100-108. 4 Linares-Rodriguez JF, Martinez-Menendez JL. [Antimicrobial resistance and bacterial virulence]. Enferm infecc Microbiol Clin 2005; 23: 86-93. 5 dos santos KV, Diniz CG, Veloso LC, de Andrade HM, Giusta Ms, Pires sF, et al. Proteomic analysis of Escherichia coli with experimentally induced resistance to piperacillin/tazobactam. Res Microbiol 2010; 161: 268-275. 6 Martinez JL, Baquero F. Mutation frequencies and antibiotic resistance. Antimicrob Agents Chemother 2000; 44: 1771-1777. 7 van der Waaij D, nord CE. Development and persistence of multi-resistance to antibiotics in bacteria; an analysis and a new approach to this urgent problem. int J Antimicrob Agents 2000; 16: 191-197. 8 Olofsson sK, Geli P, Andersson Di, Cars O. Pharmacodynamic model to describe the concentration-dependent selection of cefotaxime-resistant Escherichia coli. Antimicrob Agents Chemother 2005; 49: 5081-5091. 9 Gustafsson i, sjolund M, Torell E, Johannesson M, Engstrand L, Cars O, Andersson Di. Bacteria with increased mutation frequency and antibiotic resistance are enriched in the commensal flora of patients with high antibiotic usage. J Antimicrob Chemother 2003; 52: 645-650. 10 nys s, Tjhie JH, Bartelds Ai, Heijnen ML, Peeters MF, stobberingh EE. Erythromycin resistance in the commensal throat flora of patients visiting the general practitioner: a reservoir for resistance genes for potential pathogenic bacteria. int J Antimicrob Agents 2005; 26: 133-137. 11 salyers AA, Gupta A, Wang Y. Human intestinal bacteria as reservoirs for antibiotic resistance genes. Trends Microbiol 2004; 12: 412-416. 12 sjolund M, Tano E, Blaser MJ, Andersson Di, Engstrand L. Persistence of resistant Staphylococcus epidermidis after single course of clarithromycin. Emerg infect Dis 2005; 11: 1389-1393. 13 Wexler HM. in vitro activity of ertapenem: review of recent studies. J Antimicrob Chemother 2004; 53 suppl 2: ii11-ii21. 14 nord CE, Brismar B, Kasholm-Tengve B, Tunevall G. Effect of piperacillin/tazobactam treatment on human bowel microflora. J Antimicrob Chemother 1993; 31 suppl A: 61-65. 15 Pletz MW, Rau M, Bulitta J, De RA, Burkhardt O, Kruse G, et al. Ertapenem pharmacokinetics and impact on intestinal microflora, in comparison to those of ceftriaxone, after multiple dosing in male and female volunteers. Antimicrob Agents Chemother 2004; 48: 3765-3772. 139 16 Elliott E, Brink AJ, van GJ, Els Z, Woodford n, Turton J, et al. In vivo development of ertapenem resistance in a patient with pneumonia caused by Klebsiella pneumoniae with an extended-spectrum beta-lactamase. Clin infect Dis 2006; 42: e95-e98. 17 Valeria Dos santos K, Roque DE Carvalho MA, Amancio MW, Coutinho sC, Bahia JL, et al. in vitro selection of ertapenem and piperacillin/tazobactam-resistant strains of Bacteroides fragilis and analysis of their virulence in gnotobiotic mice. J Chemother 2010; 22: 259-263. 18 CLsi. Methods for Dilution Antimicrobial susceptibility Tests for Bacteria that Grow Aerobically; Approved standard. 2009; 8: 19 Valeria dos santos K, nicoli JR, Martins WA, Coutinho sC, Apolonio AC, Diniz CG, et al. Comparative activity of ertapenem and piperacillin tazobactam in a murine systemic infection model with Bacteroides fragilis and Escherichia coli. J Med Microbiol 2007; 56: 1576-1579. 20 Hoban DJ, Bouchillon sK, Hawser sP, Badal RE. Trends in the frequency of multiple drug-resistant Enterobacteriaceae and their susceptibility to ertapenem, imipenem, and other antimicrobial agents: data from the study for Monitoring Antimicrobial Resistance Trends 2002 to 2007. Diagn Microbiol infect Dis 2010; 66: 78-86. 21 Yan JJ, Wu JJ, Lee CC, Ko WC, Yang FC. Prevalence and characteristics of ertapenem-nonsusceptible Escherichia coli in a Taiwanese university hospital, 1999 to 2007. Eur J Clin Microbiol infect Dis 2010; 22 Kiremitci A, Dinleyici EC, Erben n, Durmaz G, Yargic ZA, Aybey AD, et al. in vitro activity of ertapenem and other carbapenems against extendedspectrum beta-lactamase producing Escherichia coli and Klebsiella pneumoniae clinical isolates in a tertiary care center in Turkey. Expert Opin Pharmacother 2008; 9: 1441-1449. 23 Girlich D, Poirel L, nordmann P. CTx-M expression and selection of ertapenem resistance in Klebsiella pneumoniae and Escherichia coli. Antimicrob Agents Chemother 2009; 53: 832-834. 24 Jacoby GA, Mills DM, Chow n. Role of beta-lactamases and porins in resistance to ertapenem and other beta-lactams in Klebsiella pneumoniae. Antimicrob Agents Chemother 2004; 48: 3203-3206. 25 Quale J, Bratu s, Gupta J, Landman D. interplay of efflux system, ampC, and oprD expression in carbapenem resistance of Pseudomonas aeruginosa clinical isolates. Antimicrob Agents Chemother 2006; 50: 16331641. 26 szabo D, silveira F, Hujer AM, Bonomo RA, Hujer KM, Marsh JW, et al. Outer membrane protein changes and efflux pump expression together may confer resistance to ertapenem in Enterobacter cloacae. Antimicrob Agents Chemother 2006; 50: 2833-2835. 27 skurnik D, Lasocki s, Bremont s, Muller-serieys C, Kitzis MD, Courvalin P, eet al. Development of ertapenem resistance in a patient with mediastinitis caused by Klebsiella pneumoniae producing an extended-spectrum beta-lactamase. J Med Microbiol 2010; 59: 115-119. 28 Andersson Di. Persistence of antibiotic resistant bacteria. Curr Opin Microbiol 2003; 6: 452-456. 29 Andersson Di. The biological cost of mutational antibiotic resistance: any practical conclusions? Curr Opin Microbiol 2006; 9: 461-465. 30 Teng CH, xie Y, shin s, Di CF, Paul-satyaseela M, Cai M, et al. Effects of ompA deletion on expression of type 1 fimbriae in Escherichia coli K1 strain Rs218 and on the association of E. coli with human brain microvascular endothelial cells. infect immun 2006; 74: 5609-5616. 31 Mohan nair MK, Venkitanarayanan K. Role of bacterial OmpA and host cytoskeleton in the invasion of human intestinal epithelial cells by Enterobacter sakazakii. Pediatr Res 2007; 62: 664-669. 32 Orme R, Douglas CW, Rimmer s, Webb M. Proteomic analysis of Escherichia coli biofilms reveals the overexpression of the outer membrane protein OmpA. Proteomics 2006; 6: 4269-4277. 33 Wooster DG, Maruvada R, Blom AM, Prasadarao nV. Logarithmic phase Escherichia coli K1 efficiently avoids serum killing by promoting C4bp-mediated C3b and C4b degradation. immunology 2006; 117: 482493. 34 smith sG, Mahon V, Lambert MA, Fagan RP. A molecular swiss army knife: OmpA structure, function and expression. FEMs Microbiol Lett 2007; 273: 1-11. 35 Lorian V, Gemmel GC. Effect of low antibiotic concentrations on bacteria: effects on ultrastructure, virulence, and susceptibility to immunodefenses. 1991; 493-555. 36 Ohlsen K, Ziebuhr W, Koller KP, Hell W, Wichelhaus TA, Hacker J. Effects of subinhibitory concentrations of antibiotics on alpha-toxin (hla) gene expression of methicillin-sensitive and methicillin-resistant Staphylococcus aureus isolates. Antimicrob Agents Chemother 1998; 42: 2817-2823.