The tigecycline evaluation and surveillance trial; assessment of the activity of tigecycline and other selected antibiotics against gram-positive and gram-negative pathogens from France collected between 2004 and 2016.

Background: A high level of antibiotic consumption in France means antimicrobial resistance requires rigorous monitoring. The Tigecycline Evaluation and Surveillance Trial (T.E.S.T.) is a global surveillance study that monitors the in vitro activities of tigecycline and a panel of marketed antimicrobials against clinically important Gram-positive and Gram-negative isolates. Methods: Annually clinically relevant strains were prospectively included in the survey through a national network of hospital-based laboratories. MICs were determined locally by broth microdilution using CLSI guidelines. Antimicrobial susceptibility was assessed using European Committee on Antimicrobial Susceptibility Testing breakpoints. Results: Thirty-three centres in France collected 26,486 isolates between 2004 and 2016. Enterococcus species were highly susceptible (≥94.4%) to linezolid, tigecycline and vancomycin. Staphylococcus aureus, including methicillin-resistant S. aureus (MRSA), were susceptible (≥99.9%) to tigecycline, vancomycin and linezolid. Between 2004 and 2016, 27.7% of S. aureus isolates were MRSA, decreasing from 28.0% in 2013 to 23.5% in 2016. Susceptibility of Streptococcus pneumoniae isolates was 100% to vancomycin, and > 99.0% to levofloxacin, linezolid and meropenem; 3.0% were penicillin-resistant S. pneumoniae (100% susceptibility to vancomycin and linezolid). Escherichia coli isolates were highly susceptible (> 98.0%) to meropenem, tigecycline and amikacin. The rate of extended-spectrum ß-lactamase (ESBL) positive E. coli increased from 2004 (3.0%), but was stable from 2012 (23.1%) to 2016 (19.8%). Susceptibility of Klebsiella pneumoniae isolates was 99.4% to meropenem and 96.5% to amikacin. The proportion of ESBL-positive K. pneumoniae isolates increased from 2004 (7.5%) to 2012 (33.3%) and was highest in 2016 (43.6%). A. baumannii was susceptible to meropenem (81.0%) and amikacin (74.9%); none of the 6.2% of isolates identified as multidrug-resistant (MDR) was susceptible to any agents with breakpoints. P. aeruginosa isolates were most susceptible to amikacin (88.5%), and MDR rates were 13.6% in 2013 to 4.0% in 2016; susceptibility of MDR isolates was no higher than 31.4% to amikacin. Conclusions: Rates of MRSA decreased slowly, while rates of ESBL-positive E. coli and K. pneumoniae increased from 2004 to 2016. Susceptibility of Gram-positive isolates to vancomycin, tigecycline, meropenem and linezolid was well conserved, as was susceptibility of Gram-negative isolates to tigecycline and meropenem. The spread of MDR non-fermentative isolates must be carefully monitored.

Bacterial hypermutation: clinical implications.

Heritable hypermutation in bacteria is mainly due to alterations in the methyl-directed mismatch repair (MMR) system. MMR-deficient strains have been described from several bacterial species, and all of the strains exhibit increased mutation frequency and recombination, which are important mechanisms for acquired drug resistance in bacteria. Antibiotics select for drug-resistant strains and refine resistance determinants on plasmids, thus stimulating DNA recombination via the MMR system. Antibiotics can also act as indirect promoters of antibiotic resistance by inducing the SOS system and certain error-prone DNA polymerases. These alterations have clinical consequences in that efficacious treatment of bacterial infections requires high doses of antibiotics and/or a combination of different classes of antimicrobial agents. There are currently few new drugs with low endogenous resistance potential, and the development of such drugs merits further research.

Curbing methicillin-resistant Staphylococcus aureus in 38 French hospitals through a 15-year institutional control program.

BACKGROUND: The Assistance Publique-Hôpitaux de Paris (AP-HP) institution administers 38 teaching hospitals (23 acute care and 15 rehabilitation and long-term care hospitals; total, 23 000 beds) scattered across Paris and surrounding suburbs in France. In the late 1980s, the proportion of methicillin resistance among clinical strains of Staphylococcus aureus (MRSA) reached approximately 40% at AP-HP. METHODS: A program aimed at curbing the MRSA burden was launched in 1993, based on passive and active surveillance, barrier precautions, training, and feedback. This program, supported by the strong commitment of the institution, was reinforced in 2001 by a campaign promoting the use of alcohol-based hand-rub solutions. An observational study on MRSA rate was prospectively carried out from 1993 onwards. RESULTS: There was a significant progressive decrease in MRSA burden (-35%) from 1993 to 2007, whether recorded as the proportion (expressed as percentage) of MRSA among S aureus strains (41.0% down to 26.6% overall; 45.3% to 24.2% in blood cultures) or incidence of MRSA cases (0.86 down to 0.56 per 1000 hospital days). The MRSA burden decreased more markedly in intensive care units (-59%) than in surgical (-44%) and medical (-32%) wards. The use of ABHR solutions (in liters per 1000 hospital days) increased steadily from 2 L to 21 L (to 26 L in acute care hospitals and to 10 L in rehabilitation and long-term care hospitals) following the campaign. CONCLUSION: A sustained reduction of MRSA burden can be obtained at the scale of a large hospital institution with high endemic MRSA rates, providing that an intensive program is maintained for a long period.

Evaluation of a new test, genotype HelicoDR, for molecular detection of antibiotic resistance in Helicobacter pylori.

The eradication rate of Helicobacter pylori by standard therapy is decreasing due to antibiotic resistance, mainly to clarithromycin. Our aim was to provide a new molecular test to guide the treatment of new and relapsed cases. We first studied 126 H. pylori strains for phenotypic (MIC) and genotypic resistance to clarithromycin (rrl mutation) and levofloxacin (gyrA mutation) and then developed a DNA strip genotyping test on the basis of the correlation results and literature data. Clinical strains (n = 92) and gastric biopsy specimens containing H. pylori (n = 105) were tested blindly with the new molecular test GenoType HelicoDR. The presence of mutations or the absence of hybridization with wild-type sequences was predictive, in rrl for clarithromycin resistance in 91 cases (mostly the A2147G mutation) and in gyrA for levofloxacin resistance in 58 cases (mutations at codon 87 or 91). Genotyping revealed a mix of genotypes in 33% of the cases, reflecting a coinfection or selection for resistant mutants. The sensitivity and specificity of detecting resistance were 94% and 99% for clarithromycin and 87% and 98.5% for levofloxacin, respectively. The concordance scores were 0.96 for clarithromycin and 0.94 for levofloxacin. With global resistance rates of 46% for clarithromycin and 25% for levofloxacin, which were observed for consecutive positive biopsy specimens from 2007 and 2008, the positive and negative predictive values for detecting resistance were 99% and 94% for clarithromycin and 96% and 96% for fluoroquinolone. GenoType HelicoDR is efficient at detecting mutations predictive of antibiotic resistance in H. pylori when applied to strains or directly to gastric biopsy specimens.

The pentapeptide repeat proteins MfpAMt and QnrB4 exhibit opposite effects on DNA gyrase catalytic reactions and on the ternary gyrase-DNA-quinolone complex.

MfpA(Mt) and QnrB4 are two newly characterized pentapeptide repeat proteins (PRPs) that interact with DNA gyrase. The mfpA(Mt) gene is chromosome borne in Mycobacterium tuberculosis, while qnrB4 is plasmid borne in enterobacteria. We expressed and purified the two PRPs and compared their effects on DNA gyrase, taking into account host specificity, i.e., the effect of MfpA(Mt) on M. tuberculosis gyrase and the effect of QnrB4 on Escherichia coli gyrase. Whereas QnrB4 inhibited E. coli gyrase activity only at concentrations higher than 30 microM, MfpA(Mt) inhibited all catalytic reactions of the M. tuberculosis gyrase described for this enzyme (supercoiling, cleavage, relaxation, and decatenation) with a 50% inhibitory concentration of 2 microM. We showed that the D87 residue in GyrA has a major role in the MfpA(Mt)-gyrase interaction, as D87H and D87G substitutions abolished MfpA(Mt) inhibition of M. tuberculosis gyrase catalytic reactions, while A83S modification did not. Since MfpA(Mt) and QnrB4 have been involved in resistance to fluoroquinolones, we measured the inhibition of the quinolone effect in the presence of each PRP. QnrB4 reversed quinolone inhibition of E. coli gyrase at 0.1 microM as described for other Qnr proteins, but MfpA(Mt) did not modify M. tuberculosis gyrase inhibition by fluoroquinolones. Crossover experiments showed that MfpA(Mt) also inhibited E. coli gyrase function, while QnrB4 did not reverse quinolone inhibition of M. tuberculosis gyrase. In conclusion, our in vitro experiments showed that MfpA(Mt) and QnrB4 exhibit opposite effects on DNA gyrase and that these effects are protein and species specific.

A plasmid-borne Shewanella algae Gene, qnrA3, and its possible transfer in vivo between Kluyvera ascorbata and Klebsiella pneumoniae.

The plasmid-borne quinolone resistance gene qnrA1 is prevalent in multidrug-resistant Enterobacteriaceae. A chromosomally encoded homologue in Shewanella algae, qnrA3, has been described. We isolated two qnrA3-positive strains, one of Klebsiella pneumoniae (He96) and one of Kluyvera ascorbata (Kas96), from the feces of an immunocompromised outpatient. The qnrA3 allele was identical to that of S. algae except for 5 nucleotides and differed from qnrA1 by 29 nucleotides affecting three amino acids. The analysis of the qnrA3 genetic environment showed that qnrA3 was inserted downstream from an ISCR1 element at a recombination crossover site described for other resistance genes, including qnrA1, and immediately upstream from IS26, a situation not described before. IS26 preceded an incomplete class 1 integron which contained, among other genes, aac(6')-Ib-cr, another transferable quinolone resistance gene, and the beta-lactamase gene bla(OXA-1/30). The 10-kb fragment encompassing qnrA3 was compared to previously described qnrA1-containing plasmids and multidrug-resistant plasmids; it shares identical sequences with pC15a, pHSH2, pQR1, pQKp311H, and pSAL-1 but with rearrangements, deletions, and mutations. Conjugal transfer of qnrA3 was highly efficient (10(-2)) from K. pneumoniae He96 or K. ascorbata Kas96 to Escherichia coli J53 but less so (10(-5)) from either donor to a clinical strain of Enterobacter cloacae. This first description of a plasmid-borne copy and of the in vitro transfer of qnrA3 is taken to illustrate its likely in vivo transfer from S. algae to the Enterobacteriaceae.

Low selection of topoisomerase mutants from strains of Escherichia coli harbouring plasmid-borne qnr genes.

OBJECTIVES: To investigate mutations in the type II topoisomerase genes in quinolone-resistant mutants selected from bacteria harbouring plasmid-borne qnr genes. METHODS: Mutants were selected by nalidixic acid, ciprofloxacin and moxifloxacin from two Escherichia coli reference strains and corresponding transconjugants harbouring qnrA1, qnrA3, qnrB2 or qnrS1 genes. RESULTS: The proportion of resistant mutants selected by the three quinolones was, respectively, in the same range for qnr-positive transconjugants and reference strains. Only 20% (65/329) of the mutants selected from the transconjugants showed a gyrase mutation, whereas 79% (94/119) of those from the reference strains without a qnr gene did (P < 0.0001). At four times the MIC of the selector quinolone, gyrA mutants represented 49% and 95% of the mutants selected with nalidixic acid, 4% and 94% with ciprofloxacin and 0% and 54% with moxifloxacin for qnr-positive transconjugants and reference strains, respectively. Mutations within gyrA were distributed at codon 87 (D87G, H, N or Y) and at codon 83 (S83L) with three novel mutations (gyrA Ser83stop, gyrA Asp82Asn and gyrB insertion of Glu at 465) and three rare mutations (gyrA Gly81Asp, gyrA Asp82Gly and gyrA Ser431Pro), mainly obtained from reference strains after moxifloxacin selection. Strikingly, none of the mutants selected by moxifloxacin from qnr-positive transconjugants harboured a mutation in the topoisomerase genes. CONCLUSIONS: Topoisomerase mutants are rarely selected by ciprofloxacin and moxifloxacin from strains harbouring qnr. This suggests that the quinolone resistance-determining region domains are protected from quinolones by the Qnr protein and consequently other mechanisms are developed to acquire a further step of fluoroquinolone resistance.

Unexpected occurrence of plasmid-mediated quinolone resistance determinants in environmental Aeromonas spp.

We searched for plasmid-mediated quinolone resistance determinants of the Qnr type in several water samples collected at diverse locations from the Seine River (Paris, France). The qnrS2 genes were identified from Aeromonas punctata subsp. punctata and A. media. The qnrS2 gene was located on IncU-type plasmids in both isolates, which resulted in increased MIC values of quinolones and fluoroquinolones, once they were transferred into Escherichia coli. The qnrS2 gene identified in A. punctata was part of novel genetic structure corresponding to a mobile insertion cassette element. This identification of plasmid-mediated qnr genes outside Enterobacteriaceae underlines a possible diffusion of those resistance determinants within gram-negative rods.

Multiplex PCR for detection of plasmid-mediated quinolone resistance qnr genes in ESBL-producing enterobacterial isolates.

OBJECTIVES: To develop a rapid and reliable single-tube-based PCR technique for detecting simultaneously the plasmid-mediated quinolone resistance qnrA, qnrB and qnrS genes. METHODS: After multiple alignments, primers were designed to detect known qnr variants (six for qnrA-, six for qnrB- and two for qnrS-like genes). They were used for screening a collection of 64 expanded-spectrum beta-lactamase (ESBL)-producing enterobacterial isolates from Kuwait, collected from 2002 to 2004, as ESBL genes have been often associated with qnr genes. Sequencing was performed to identify qnr and associated ESBL genes. RESULTS: In optimized conditions, all positive controls (used separately or mixed) confirmed the specificity of the PCR primers. Out of 64 isolates, only 3 isolates were positive for a qnrB-like gene (4.7%), whereas no qnrA-like and qnrS-like gene was detected. A qnrB2 gene was detected in an Enterobacter cloacae K34 (SHV-12+) isolate, whereas qnrB1-like (termed qnrB7) and qnrB6-like (termed qnrB8) genes were identified from E. cloacae K37 (SHV-12+) and Citrobacter freundii K70 (VEB-1b+) isolates, respectively. CONCLUSIONS: We report here a fast and reliable technique for rapid screening of qnr-positive strains to be used for epidemiological surveys. A low prevalence of Qnr determinants among ESBL-producing Enterobacteriaceae was identified in the study with Kuwaiti isolates.

Emergence of a Streptococcus pneumoniae isolate resistant to streptogramins by mutation in ribosomal protein L22 during pristinamycin therapy of pneumococcal pneumonia.

OBJECTIVES: The aim of this study was to characterize the mechanism of resistance to macrolides and streptogramins of a Streptococcus pneumoniae strain isolated from blood cultures in an 80-year-old patient suffering from severe pneumonia unsuccessfully treated with pristinamycin. METHODS: Resistance genes erm(B) and mef(A) were searched for by PCR. Portions of genes for domains V and II of the 23S rRNA (rrl) and genes for ribosomal proteins L4 (rplD) and L22 (rplV) were amplified by PCR from total genomic DNA and sequenced. RESULTS: Resistance genes erm(B) and mef(A) were not detected. Only mutation in the rplV gene encoding ribosomal protein L22 was detected. The strain contained a six amino acid insertion ((107)KRTAHI(108)) in the C-terminus of the ribosomal protein L22. CONCLUSIONS: This is the first report of emergence of a pneumococcus resistant to streptogramins by mutation in ribosomal protein L22 during treatment with pristinamycin.

Update on fluoroquinolone resistance in Helicobacter pylori: new mutations leading to resistance and first description of a gyrA polymorphism associated with hypersusceptibility.

Helicobacter pylori eradication by standard therapy is decreasing due to clarithromycin and metronidazole resistance. Fluoroquinolones are valuable drugs for alternative therapy, but their activity needs to be updated. We determined minimum inhibitory concentrations (MICs) of the newly marketed fluoroquinolones (levofloxacin, moxifloxacin and gatifloxacin) and assessed the prevalence of resistance in 128 H. pylori strains isolated in 2004-2005. The quinolone resistance-determining region (QRDR) of gyrA was sequenced for all strains. Gatifloxacin MICs (MIC(50) = 0.25 mg/L) were two- to four-fold lower than those of the other fluoroquinolones. The prevalence of resistance (ciprofloxacin MIC > 1 mg/L) was 17.2% (22 strains). All resistant strains harboured one gyrA mutation at codons 86, 87 or 91, including three new mutations (Asp86Asn, Thr87Ile and Asn87Tyr). Ciprofloxacin-susceptible strains were devoid of such gyrA mutations, but harboured a polymorphism at codon 87 that distinguished 18 isolates (17%) with a Thr87 like the reference strain J99 from 88 strains with Asn87 like the reference strain 26695. Strains with Thr87 were four-fold more susceptible to nalidixic acid, pefloxacin, ciprofloxacin and levofloxacin and were equally susceptible to moxifloxacin and gatifloxacin. The high rate of quinolone resistance in H. pylori requires the use/implication of a 'test and treat' strategy that can confidently rely on QRDR gyrA sequencing.

Prevalence of qnr genes in Salmonella in France.

OBJECTIVES: To detect the qnrA, qnrB and qnrS genes among Salmonella isolates received at the French National Reference Centre for Salmonella in Paris, France. METHODS: Antibiotic susceptibility was determined by disc diffusion for 499 Salmonella isolates including 320 Salmonella Typhimurium, 100 Salmonella Enteritidis and 79 Salmonella Hadar collected in 2002. Amplification with specific primers of qnrA, qnrB and qnrS genes was performed for all Salmonella Typhimurium, Salmonella Enteritidis and Salmonella Hadar isolates resistant to quinolones and for 17 additional isolates that produced expanded-spectrum beta-lactamases (ESBLs). RESULTS: Prevalence of quinolone resistance was 3.75%, 11% and 79.7% for Salmonella Typhimurium, Salmonella Enteritidis and Salmonella Hadar serovars, respectively. A single isolate (0.2%) was qnrA-positive (QnrA1 determinant) being a Salmonella serovar Concord carrying also the ESBL gene bla(CTX-M-15). This strain was probably from East Africa. No qnrB or qnrS genes were identified. CONCLUSIONS: Whereas plasmid-mediated quinolone resistance of the Qnr type is emerging in Enterobacteriaceae worldwide, it remains rare in Salmonella in France.

Type II topoisomerase mutations in clinical isolates of Enterobacter cloacae and other enterobacterial species harbouring the qnrA gene.

The qnr genes are transferable genes that confer low-level quinolone resistance by protection of topoisomerase. The occurrence of mutations in DNA gyrase (gyrA, gyrB) and topoisomerase IV (parC, parE) genes in strains harbouring qnr was investigated in 28 qnrA-positive clinical isolates, among which 7 strains also harboured qnrS. Topoisomerase mutations were found in 25 (89%) of the 28 strains, with at least two mutations (gyrA and parC) in 13 strains and one mutation in 12 strains. Isolates of the Enterobacter cloacae complex were compared with reference strains of the new Enterobacter species. gyrA mutations were found at position 83 (Ser or Thr for Ile, Tyr, Leu or Phe depending on the species), and new gyrB mutations were described (S463A, S464F). qnrA had an additive effect of a 10-fold increase in the minimum inhibitory concentration (MIC) whatever the number of topoisomerase mutations, and qnrS was additive to qnrA with a further 2- to 10-fold increase in the MIC. Comparison of MICs with susceptibility breakpoints showed that strains combining qnrA and topoisomerase mutations were resistant to fluoroquinolones, but the three strains lacking a topoisomerase mutation were susceptible using ciprofloxacin and levofloxacin but not using nalidixic acid or moxifloxacin testing.

Routine use of real-time PCR for detection of Helicobacter pylori and of clarithromycin resistance mutations.

OBJECTIVES: We previously showed that real-time PCR was a reliable technique for coupled detection of Helicobacter pylori and clarithromycin resistance mutations directly from biopsies. After one year of use, we compared its performances to those of histology, which remains the most employed method for H. pylori detection from gastric biopsies. MATERIALS AND METHODS: 518 subjects underwent endoscopy during the year 2003 with biopsies taken for H. pylori detection by histology, PCR, and in case of discrepancy between the two techniques, by culture. RESULTS: The prevalence of infection, defined as positive PCR and histology, and in case of discrepancy as a positive culture, was 30% (163/518). The percentage of concordance between the two tests was 87.8% (455/518). The sensitivity, specificity, positive and negative predictive values of PCR were 98.2%, 97.5%, 94.7%, and 99.1%, respectively. The corresponding performances of histology were 87.7%, 91.3%, 82.2%, and 94.2%, respectively (p<0.001). The prevalence of clarithromycin resistance was 30%. CONCLUSIONS: PCR is more accurate in routine than histology and permits easy determination of clarithromycin resistance, which is useful in countries like France where the prevalence of resistance is high.

Novel Ambler class A beta-lactamase LAP-1 and its association with the plasmid-mediated quinolone resistance determinant QnrS1.

The plasmid-mediated quinolone resistance determinant QnrS1 was identified in non-clonally related Enterobacter cloacae isolates in association with a transferable narrow-spectrum beta-lactam resistance marker. Cloning experiments allowed the identification of a novel Ambler class A beta-lactamase, named LAP-1. It shares 62 and 61% amino acid identity with the most closely related beta-lactamases, TEM-1 and SHV-1, respectively. It has a narrow-spectrum hydrolysis of beta-lactams and is strongly inhibited by clavulanic acid and sulbactam and, to a lesser extent, by tazobactam. Association of the blaLAP-1 gene with the qnrS1 gene was identified in E. cloacae isolates from France and Vietnam. These genes were plasmid located and associated with similar insertion sequences but were not associated with sul1-type class 1 integrons, as opposed to the qnrA genes.

Bacterial identification, clinical significance, and antimicrobial susceptibilities of Acinetobacter ursingii and Acinetobacter schindleri, two frequently misidentified opportunistic pathogens.

The species belonging to the Acinetobacter genus are currently reported as opportunistic pathogens in hospitalized patients with underlying predispositions. However, except for the Acinetobacter calcoaceticus-Acinetobacter baumannii complex, the identification of other species is frequently unreliable, especially for Acinetobacter ursingii and Acinetobacter schindleri, newly described in 2001. Thus, the clinical significance, phenotypic features, and antimicrobial susceptibilities of these two misidentified species remain unclear. Of 456 Acinetobacter sp. clinical strains isolated from 2002 to 2005 in Henri Mondor Hospital, 15 isolates (10 A. ursingii and 5 A. schindleri isolates) were studied. They were characterized using a phenotypic approach (API 20 NE and VITEK 2 systems), 16S rRNA gene sequencing, and susceptibility to antimicrobial agents with evaluation of impact in clinical relevance. The two corresponding type strains were also included for comparison. All isolates were identified to the species level using molecular tools, whereas the phenotypic methods remained unreliable due to the absence of these two species in the manufacturers' databases. However, the API 20 NE system appeared to be a reasonably reliable phenotypic alternative for the identification of A. ursingii when the numerical code 0000071 was found. Conversely, no discriminative phenotypic alternative existed for A. schindleri isolates. Concerning antimicrobial susceptibility, A. ursingii strains appeared to be more resistant to antibiotics than A. schindleri strains, which could imply therapeutic consequences. Finally, the prevalence of infections caused by A. ursingii and A. schindleri (representing 9.7% and 4.8% of non-A. calcoaceticus-A. baumannii complex strains, respectively) seems to be underestimated.

[Management of urinary tract infections in women. Epidemiologic survey of 7916 women in general practice]. / Les infections urinaires de la femme en médecine générale. Résultats d'un observatoire réalisé auprès de 7916 patientes.

OBJECTIVE: To evaluate the management of urinary tract infections in women by general practitioners and compare it with official French guidelines. METHODS: This survey enrolled 1587 general practitioners in France and 7916 adult women. Exclusion criteria for patients included: pregnancy, diabetes, neurogenic bladder, or urinary catheters. During the visit at which the diagnosis was made, physicians completed a questionnaire that included diagnostic and management details, in particular, prescription of further examinations. RESULTS: According to the French guidelines, 37% of women had an upper or complicated urinary tract infection, although one third of the complicated infections were so defined only by the patient's age (>65 years). Additional testing was prescribed for 36% of the women with acute uncomplicated cystitis. CONCLUSION: This study shows that the management of urinary tract infections in women does not comply with current guidelines, especially in cases of acute uncomplicated cystitis. The use of age alone as a complicating factor should be reconsidered.

In vivo selection during ofloxacin therapy of Escherichia coli with combined topoisomerase mutations that confer high resistance to ofloxacin but susceptibility to nalidixic acid.

OBJECTIVES: To investigate quinolone resistance mechanisms in an Escherichia coli clinical isolate (Ar2) resistant to ofloxacin but susceptible to nalidixic acid selected after 10 days of ofloxacin therapy in a patient with prostatitis. METHODS: Molecular typing (ERIC-PCR and RAPD), antibiotic susceptibility and gyrA, gyrB, parC and parE QRDR sequences were compared for E. coli Ar2 and a wild-type E. coli (Ar1) isolated 2 months earlier in the same patient. Ofloxacin-resistant mutants were selected in vitro in order to reproduce the mutations observed and the original phenotype. RESULTS: The two strains were similar with regard to antibiotic susceptibility except quinolones and for ERIC-PCR and RAPD patterns, suggesting a clonal relationship and acquisition of quinolone resistance by chromosomal mutation. Quinolone MICs were 3, 0.12, 0.05 and 0.02 mg/L of nalidixic acid, ofloxacin, levofloxacin and ciprofloxacin, respectively, for E. coli Ar1 and 6, 32, 8 and 1 mg/L, respectively, for E. coli Ar2. The strain Ar2 harboured two substitutions, Gly-81-->Asp in GyrA and Ser-80-->Arg in ParC. Introduction into E. coli Ar2 of the wild-type gyrA fully complemented fluoroquinolone resistance. Although the strain was not a hypermutator, ofloxacin first-step resistant mutants with gyrA mutations were easily obtained from E. coli Ar1 and 25% of them were at codon 81. In vitro stepwise combination of Gly-81-->Asp in GyrA and Ser-80-->Arg in ParC reproduced the original phenotype in E. coli KL16. CONCLUSIONS: A double topoisomerase mutant was selected in vivo by 10 days ofloxacin. The mutations were originally combined for a result of ofloxacin resistance but nalidixic acid susceptibility.