Back

Determination of the QNR–DNA gyrase interaction site by a genetic approach

Abstract number: o53

Cesaro  A., Cattoir  V., Lascols  C., Soussy  C.J., Cambau  E.

Objectives: 

QNR is a new mechanism of quinolone resistance due to the protection of gyrase by the QNR protein. Since quinolone inhibition of gyrase is hampered by Qnr, we hypothesized that the Qnr-gyrase interaction occurs at a site closed to that of interaction of quinolones, i.e. the quinolone resistance-determining region (QRDR) in the GyrA and GyrB subunits.

Methods: 

First, in vitro selection of quinolone resistant mutants was performed comparatively with a wild-type (WT) reference Escherichia coli strain and with an E. coli transconjugant harbouring the qnrA gene. Selection was carried out using nalidixic acid (NAL), ciprofloxacin (CIP) and moxifloxacin (MOX). Mutants were chosen on the basis of quinolone MICs and mutations in the gyrA and gyrB QRDRs were determined. Second, qnrA was introduced by conjugation into E. coli strains referenced for gyrA mutations (S83L, D87Y, D87G, D87N), gyrB mutations (D426N, K447E), gyrA + parC (S83L + S80R), and gyrA + parE (S83L + L445H).

Results: 

For the WT E. coli, gyrA mutation was observed in 10/10 mutants selected by NAL, 7/7 mutants selected by CIP and in 5/12 mutants selected by MOX. For the E. coliqnr+, gyrA mutation was observed in 18/22 mutants selected by NAL, 1/11 mutants selected by CIP and 0/18 mutants selected by MOX. Analysis of the gyrA mutations showed that whereas 68% (15/22) of the mutations occurred at the codon 87 (D87G, D87H, D87N or D87Y) in mutants selected from the WT E. coli, only 1/19 harboured a D87Y mutation in mutants selected from E. coliqnr+. Conversely, 95 % (18/19) of the mutations observed in mutants selected from the E. coli qnr+ occurred at the codon 83 (all S83L) but 7/22 for WT E. coli mutants. In all 18 one step-mutants and in all 8-second-step mutants selected by MOX from E. coli qnr+, no mutation was observed either in the QRDRs in gyrA, gyrB, parC and parE, or in the entire gyrA and gyrB genes, whereas selection by MOX from the E. coli wild-type resulted in 5 gyrA mutants (D87N [n = 3] or D87G [n = 2]). After introduction of qnr in the strains referenced for topoisomerase mutations, MICs of CIP and MOX increased 2- to 16-fold. NAL MICs did not increase whatever the topoisomerase mutation harboured by the recipient strain.

Conclusion: 

From the results of the two approaches, D87 seems to be one of the amino acid involved in QNR-gyrase interaction, and S83 seems to preferentially interact with NAL and CIP in presence of QNR but not with MOX.

Session Details

Date: 01/08/2007
Time: 00:00-00:00
Session name: XXIst ISTH Congress
Subject:
Location: Oxford, UK
Presentation type:
Back to top