Surveillance of Fluoroquinolone-Resistant Clinical Isolates of Pseudomonas aeruginosa

Ciprofloxacin (CPFX) and pazufloxacin (PZFX) have strong antibacterial activity against Pseudomonas aeruginosa. We investigated the sensitivity of P. aeruginosa to CPFX and PZFX in 373 strains isolated from inpatients (321 strains) and outpatients (52 strains) during September 2010 to September 2011 at Toho University Ohashi Medical Center. The percentage of CPFX-non-susceptible (≥3.91 μg/mL) among inpatients was 22.4%, but that among outpatients was 1.9%. As the major resistance mechanism to fluoroquinolones in P. aeruginosa involves modification of type II topoisomerases (DNA gyrase and topoisomerase IV), we examined mutations in the quinolone-resistance-determining regions (QRDRs) of gyrA and parC of P. aeruginosa isolates. Among the 373 isolates, 73 isolates had reduced CPFX-susceptibility and 88 had reduced PZFX-susceptibility. Sequencing of gyrA and parC revealed base substitutions that resulted in amino acid replacements in QRDR of GyrA in 70 P. aeruginosa isolates, while Thr83Ile (in GyrA) and Ser87Leu (in ParC) substitutions were found in 12 strains. These replacements were clearly associated with reduced susceptibility to CPFX and PZFX. However, we also found strains with high MICs to quinolones without mutations in either gyrA or parC. We then investigated the effect of efflux pumps in CPFX-resistance in these isolates. In the presence of an efflux pump inhibitor, MIC values in 12 of 66 strains decreased to 1/2. We also sequenced genes related to overexpression of efflux pumps, viz., mexZ, mexR, and nfxB. Eight of the strains without mutations in QRDRs had a mutation in mexZ, 7 strains had a mutation in mexR, but no mutation was identified in nfxB.


Introduction
Pseudomonas aeruginosa, a ubiquitous Gram-negative soil organism, is widely distributed in nature.P. aeruginosa is noted for its metabolic versatility and its exceptional ability to colonize a wide variety of environments, and also for its intrinsic resistance to a wide variety of antimicrobial agents.Such characteristics result in a higher prevalence of this organism in hospital environments, where it is a clinically significant pathogen causing opportunistic infections and nosocomial outbreaks [1,2].
One of the challenges in managing P. aeruginosa infections is an inherent resistance mechanism, referred to as intrinsic resistance.Because of its virulence and the limited choices of effective antimicrobial agents, treatment of P. aeruginosa infection is often complicated.The present study aimed to study the antimicrobial susceptibility patterns of P. aeruginosa isolated from outpatients and inpatients, using the agar dilution method, and by defining mutations in gyrA, parC, mexZ, mexR and nfxB, which contribute to fluoroquinolon resistance, in these isolates.

Antibacterial Agents and Bacterial Strains
Ciprofloxacin, meropenem, and amikacin were purchased from Wako Pure Chemical Industries, Ltd., Japan, and pazufloxacin was purchased from Toyama Chemical Co., Ltd., Japan.
In this study, 373 clinical isolates of P. aeruginosa were collected during September 2010 to September 2011 at Toho University Ohashi Medical Center.Of these, 321 isolates of P. aeruginosa were obtained from inpatients with respiratory tract infections, urinary tract infections, surgical site infection, gastrointestinal tract infections, or bacteremia, whereas the remaining 52 isolates were obtained from outpatients with tympanitis, bacterial skin infections, respiratory tract infections, or urinary tract infections.All the P. aeruginosa strains isolated were maintained in Microbank TM (Pro-Lab, Toronto, Canada) and kept at -80˚C.

Statistical Analysis
Mann-Whitney's U-test was used to evaluate the rela-tionship between the distribution of MIC of CPFX, MEPM, PZFX and AMK in the strains of P. aeruginosa isolated from outpatients and inpatients.

PCR Amplification and DNA Sequence Determination and Analysis
The strains were incubated in MH broth at 37˚C overnight, and 1 mL of the culture was centrifuged (15,000 rpm, 10 min, at 4˚C  determined after overnight incubation of isolates at 37˚C.

Antibiogram Pattern of Pseudomonas aeruginosa
The results of MH agar-based antibiogram pattern study of the strains of P. aeruginosa isolated from inpatients (n = 321) and outpatients (n = 52) is shown in Figure 1.
The MIC range of CPFX in the strains isolated from outpatients, except for 1 strain, was between <0.06 and 1.95 μg/mL, but that of strains isolated from inpatients was between <0.06 and >62.5 μg/mL (Figure 1(a)).Thus, strains from outpatients were sensitive to CPFX, except for 1 strain that was isolated from an outpatient who had received ofloxacin treatment for chronic otitis media.The percentage of CPFX-non-susceptible strains (≥3.91 μg/mL) among inpatients was 22.4%, but that among outpatients was 1.9%.The MIC range of the strains from outpatients for PZFX was between 0.06 and 7.81 μg/mL, except for 1 strain, and that from inpatients was between <0.06 and 62.5 μg/mL (Figure 1
The MIC ranges of the strains from outpatients for AMK was between <0.49 and 15.62 μg/mL, and that from inpatients was between <0.49 and 62.5 μg/mL.The percentage of AMK-non-susceptible strains (≥31.25 μg/ mL) from inpatients was 5.6%, but no AMK-non-susceptible strains were found among outpatients (Figure 1(d)).
Our results indicated that most of the strains isolated from outpatients were susceptible to CPFX and PZFX, but 22.4% of the strains isolated from inpatients were resistant to CPFX; strains isolated from outpatient were susceptible to MEPM, but 30.5% of the strains isolated from inpatients were resistant to MEPM (Figure 1(b)).Similarly, strains isolated from outpatients were susceptible to AMK, but 5.6 % of the strains isolated from inpatients were resistant to this antimicrobial agent (Figure 1(d)).
Statistical analysis of the distribution of MIC of CPFX, MEPM, PZFX and AMK in the strains of P. aeruginosa isolated from outpatients and inpatients showed significant differences (P < 0.05) in each antibiotics.Surveys of antimicrobial susceptibilities in clinical isolates of P. aeruginosa in 2004 and 2008 in Japan showed that 22% of the 90 isolates collected from 13 medical facilities We found no CPFX-or PZFX-resistant strain with a mutation only in parC.In Gram-positive cocci, mutations in parC are seen commonly in fluoroquinolone-resistant strains, and it is well-known that these mutations precede

Impact of Efflux Pump Inhibition on MICs of Ciprofloxacin
Interestingly, we found strains with high MICs for fluoroquinolones without mutations in either gyrA or parC.These higher CPFX-resistant strains were assumed to have other fluoroquinolone-resistance mechanisms, such as an overproduction of efflux pumps.To investigate this possibility, we used phenylalanine arginyl β-naphtylamide (PAβN), which inhibits efflux pumps in P. aeruginosa.[20].Table 2 shows the effect of PAβN on the MICs of CPFX-resistant strains.Twelve of 66 CPFX- effect on CPFX-resistance.Two strains had an 11-bp deletion in nfxB leading to a frame-shift following Val35; these strains also had other alterations affecting GyrA, ParC, MexZ, and MexR.Amino acid substitutions in quinolone resistance-determining region (QRDR) of GyrA (Thr83Ile or Asp-87Asn) were found in 60 of 120 isolates; moreover, the Thr83Ile substitution was the mutation most frequently reported previously in CPFX-resistant clinical isolates of P. aeruginosa [21,22].Fifteen strains in Group V with an amino acid substitution in QRDR of GyrA (Thr83Ile) were resistant to 3.91 μg/mL CPFX, but 2 of 6 strains in Group IV with another amino acid substitution in QRDR of GyrA (Asp87Asn) were resistant to 3.91 μg/mL CPFX.
The only amino acid substitution found in QRDR of the ParC subunit of topoisomerase IV, another target of CPFX, was Ser87Leu; this alteration was always found along with a GyrA alteration (Groups IX-XII).These results suggested that CPFX-resistance in P. aeruginosa is mainly mediated through alterations in GyrA, and that alterations in ParC are secondary, as reported previously [23].Alterations in ParC, MexZ, MexR, or NfxB in addition to mutation in GyrA (Thr83Ile) increased the MICs for CPFX, although the number of strains with multiple alterations was limited.
Our results revealed that multiple alterations in GyrA, ParC, MexZ, MexR, and NfxB were associated with a higher-level resistance to CPFX than that caused by a single alteration in GyrA.Although 2 strains in Group I (Table 3), in which we could not identify mutatition in QRDRs of gyrA, parC, or the efflux pump regulator genes mexZ, mexR, and nfxB, were resistant to >7.81 µg/mL CPFX, the MICs of CPFX in strains 286 and 297 decreased from 15.6 µg/mL to 1.95 or 0.98 µg/mL in the presence of PAN, respectively (Table 2).A similar effect of PAN on the MICs of Group III strains, which had a GyrA (Thr83Ile) alteration, but no alterations in efflux pump regulators, was observed in 9 of 14 strains.These results suggested that over expression of other active efflux systems or decreased permeability could contribute to CPFX-resistance in these strains [24].
were determined by the standard agar dilution assay, according to guidelines of the National Committee for Clinical Laboratory Standards (NCCLS), using Mueller-Hinton (MH) agar (Becton, Dickinson and Company, USA).Strains preserved on Microbank TM were inoculated into 2 mL of MH broth and incubated at 37˚C for 18 -20 h.After dilution with MH broth, 2 μL of the bacterial suspension containing 10 4 CFU was spotted onto drug-containing agar plates.The plates were incubated at 37˚C for 24 h.The MIC was defined as the lowest drug concentration that prevented visible growth of bacteria.

Figure 2 .
Figure 2. The relation with MICs of CPFX (a) or PZFX (b) and mutations in gyrA and parC genes.mutations in gyrA[16,17].The inverse, i.e., that mutations in gyrA precede mutations in parC, is typically seen in Gram-negative bacteria.In this study, CPFX-resistant strains had mutations in gyrA, while some strains had mutations in both gyrA and parC, similar to the finding in fluoroquinolone-resistant Gram-negative bacteria.Moreover, the strains with mutations in both gyrA and parC had higher MICs for CPFX[18,19].

). A sterilized toothpick was used to lightly touch the precipitate, and cells on the toothpick were suspended in 25 µL of PCR mixture: KOD FX 0.5 U (TOYOBO Co., Ltd., Japan), 0.3 µM of each primer, 0.4 mM of each dNTP. Primers used for PCR are listed in Table 1. The amplification procedure comprised de- naturation at 95˚C for 2 min, followed by 40 cycles, each consisting of denaturation for 10 s at 98˚C, annealing for 30 s at 55˚C, and polymerization for 30 s at 74˚C. Am- plified DNA was directly sequenced by the dideoxy chain termination method, employing a Thermo Seque- nase II dye terminator cycle sequencing kit (Amersham Pharmacia Biotech Inc., NJ, USA), BigDye ® Terminator v1.1 Cycle Sequencing Kit (Applied Biosystems, Inc., CA, USA) according to the manufacturer's protocol. The products were automatically analyzed in a model ABI 3500 DNA autosequencer (Applied Biosystems, Inc.). DNA sequences have been deposited in DDBJ (Acces- sion
numbers: AB753299-AB753440).