ANTIMICROBIAL AGENTS AND CHEMOTHERAPY, Aug. 2003, p. 2606–2614 0066-4804/03/$08.00⫹0 DOI: 10.1128/AAC.47.8.2606–2614.2003 Copyright © 2003, American Society for Microbiology. All Rights Reserved.
Vol. 47, No. 8
Activities of Mutant Prevention Concentration-Targeted Moxifloxacin and Levofloxacin against Streptococcus pneumoniae in an In Vitro Pharmacodynamic Model George P. Allen,1,2† Glenn W. Kaatz,1,3,4 and Michael J. Rybak1,2,3* Anti-Infective Research Laboratory, Department of Pharmacy Practice, Eugene Applebaum College of Pharmacy and Health Sciences,1 and School of Medicine,3 Wayne State University, Detroit Receiving Hospital and University Health Center,2 and the John D. Dingell VA Medical Center,4 Detroit, Michigan 48201 Received 11 June 2002/Returned for modification 8 February 2003/Accepted 10 May 2003
The differential effects of moxifloxacin and levofloxacin on the development of resistance in four Streptococcus pneumoniae isolates were examined by using an in vitro pharmacodynamic model. Therapeutic regimens (moxifloxacin: peak, 4.5 g/ml; half-life [t1/2], 12 h; and levofloxacin: peak, 6 g/ml; t1/2, 6 h) were tested against two fluoroquinolone-susceptible isolates (strains 79 and ATCC 49619) and KD2138 and KD2139 (parC and gyrA mutants, respectively, of ATCC 49619). Mutant prevention concentration (MPC)-targeted regimens with modified pharmacokinetics of each drug were simulated to match the area under the concentration-time curve (AUC) above the MPC for the two fluoroquinolones. Moxifloxacin MICs and MPCs (MIC/MPC) for isolates 79, ATCC 49619, KD2138, and KD2139, respectively, were 0.125 and 0.5, 0.125 and 0.5, 0.25 and 8, and 0.25 and 4 g/ml. Levofloxacin MICs and MPCs for the same isolates were 1 and 4, 0.5 and 2, 1 and 64, and 0.5 and 32 g/ml, respectively. Therapeutic levofloxacin concentrations led to isolation of mutants of ATCC 49619 (S79Y in ParC), KD2138 (S81Y in GyrA), and KD2139 (S79Y in ParC). Therapeutic moxifloxacin concentrations against the gyrA mutant KD2139 resulted in outgrowth of a mutant with a ParC substitution (S79Y) but caused no emergence of mutants of the other three isolates. MPC-targeted moxifloxacin (lower-than-normal peak ⴝ 0.75 to 1.5 g/ml, administered at levofloxacin’s t1/2) caused growth of a GyrA variant (S81Y) of KD2138 and a ParC variant (S79Y) of KD2139, while no mutants of ATCC 49619 were recovered. MPC-targeted levofloxacin (higher-than-normal peak ⴝ 14.5 to 29.5 g/ml, administered at moxifloxacin’s t1/2) against KD2138 and KD2139 did not prevent the development of the mutations observed in therapeutic regimens, but resistance in the fluoroquinolone-susceptible ATCC 49619 was no longer noted. Normalization of the respective AUC/MPC ratios of moxifloxacin and levofloxacin did not eliminate differences in resistance selectivity of the two agents in all cases. We conclude that the reduced recovery of resistant mutants of S. pneumoniae following moxifloxacin exposure compared to levofloxacin may be due to intrinsic differences between the drugs. Increasing the concentration and exposure (t1/2) to exceed the MPC may prevent mutations from occurring in fluoroquinolone-susceptible strains. However, this strategy did not prevent the selection of secondary mutants in strains with preexisting mutations. Further study of the MPC concept to evaluate these relationships is warranted. Fluoroquinolone resistance in Streptococcus pneumoniae is a growing concern, although present global resistance rates remain low. While the apparent high level of fluoroquinolone susceptibility in S. pneumoniae has led some to express optimism regarding the role of fluoroquinolones in antipneumococcal therapy (32), concerns about future reductions in susceptibility continue to be expressed (33). Furthermore, although the overall fluoroquinolone susceptibility of S. pneumoniae remains high, investigators in certain geographic areas have reported alarming increases in fluoroquinolone resistance. For example, Ho et al. reported the results of a multicenter study that found the overall incidence of fluoroquinolone-resistant S. pneumoniae (defined by a levofloxacin MIC of ⱖ4 g/ml) to be 13.3% in Hong Kong in the year 2000 (12). Of particular note are a number of recent reports of fluo-
roquinolone resistance found among clinical isolates (3, 8, 12, 13, 16, 34–38). For example, Weiss et al. described a nosocomial outbreak of S. pneumoniae resistant to ciprofloxacin (36), while Urban et al. reported two cases of fluoroquinoloneresistant S. pneumoniae in levofloxacin-treated patients (34). Cross-resistance to ciprofloxacin, gatifloxacin, grepafloxacin, and trovafloxacin was noted in these isolates. In addition, resistance to gemifloxacin and moxifloxacin was readily obtained by passage on agar containing each of these agents. Given the class-wide fluoroquinolone cross-resistance found in such isolates, it is essential to develop ways to preserve the antimicrobial spectrum of these agents. A multitude of in vitro studies have found that fluoroquinolones possessing a C-8-methoxy (C-8-OMe) substituent are better able to prevent development of resistance than are fluoroquinolones characterized by alternate C-8 moieties (5–7, 11, 14, 15, 18, 20, 21, 39, 41–43). It is thought that the C-8-OMe functional group confers a dual targeting of topoisomerase IV and DNA gyrase, whereas agents lacking this substituent preferentially inhibit one enzyme or the other. The C-8-OMe substituent also has been shown to increase lethality against wild-type and resistant S. pneumoniae (both ParC and GyrA
* Corresponding author. Mailing address: Anti-Infective Research Laboratory, Eugene Applebaum College of Pharmacy and Health Sciences, Wayne State University, Detroit, MI 48201. Phone: (313) 5774376. Fax: (313) 577-8915. E-mail:
[email protected]. † Present address: Oregon State University College of Pharmacy at OHSU, Portland, OR 97239-3098. 2606
MOXIFLOXACIN AND LEVOFLOXACIN AGAINST S. PNEUMONIAE
VOL. 47, 2003
2607
TABLE 1. Derivation of pharmacokinetics of MPC-targeted models MPC
AUC/MPC
Isolate
79 ATCC 49619 KD2138 KD2139
Adjusted t1/2 (h)
Adjusted peaka (g/ml)
Adjusted peakb (g/ml)
Dosec
MXFd
LEVe
MXF
LEV
MXF
LEV
MXF
LEV
MXF
LEV
MXF
LEV
0.5 0.5 8 4
4 2 64 32
96 96 6 12
12 24 0.75 1.5
6 6 6 6
12 12 12 12
0.75 1.5 0.75 0.75
29.5 14.5 29.5 29.5
0.375 0.75 0.375 0.375
20.36 10.01 20.36 20.36
66.7 133.3 66.7 66.7
2,458.3 1,208.3 2,458.3 2,458.3
a
Adjusted total (non-protein-bound) peak concentration. Adjusted peak (free concentration) after protein binding (moxifloxacin, 50%; levifloxacin, 31%) of each fluoroquinolone considered. c Dose (in milligrams) corresponding to adjusted total (non-protein-bound) peak concentration (usual therapeutic doses for moxifloxacin, 400 mg; and for levofloxacin, 500 mg). d MXF, moxifloxacin. e LEV, levofloxacin. b
variants) (19). The available C-8-OMe fluoroquinolones moxifloxacin and gatifloxacin are potential alternatives to older fluoroquinolones such as levofloxacin, which possess a single topoisomerase target, because of these differences in propensity for mutant selection. The mutant prevention concentration (MPC) has been proposed as a parameter by which the relative potential for selection of resistant mutants by fluoroquinolones may be assessed (for a review, see reference 40). For example, Dong et al. found that agents with the C-8-OMe group possessed lower MPC values than did structural analogs differing only by the functionality at this position (7). Since the MPC is a measure of the MIC using an inoculum size of sufficient magnitude to allow detection of resistant subpopulations, the consequence of such a finding may be that the C-8-OMe fluoroquinolones are more active against isolates possessing preexisting mutations in the genes encoding topoisomerase IV and/or DNA gyrase. This conclusion is supported by the dual-targeting property thought to be fostered by the C-8-OMe functionality. The MPC may represent a way to quantify these attributes. An additional component of the MPC idea is the concept of the mutant selection window, defined as the range of concentrations between the MIC and MPC. In this concentration interval the selective enrichment of resistant subpopulations is proposed to occur (40). It has been found in some cases that the C-8-OMe agents possess a narrower mutant selection window than do alternative fluoroquinolones. Additionally, it has been proposed that a fluoroquinolone (or potentially any other antimicrobial) that achieves concentrations exceeding the MPC (and hence, the mutant selection window) throughout therapy will not selectively enrich the growth of resistant organisms. Thus, the MPC may serve as a predictive pharmacodynamic parameter with respect to selection of resistant mutants. In the present work we compared the in vitro activities of moxifloxacin and levofloxacin against S. pneumoniae isolates possessing mutations in the genes encoding either topoisomerase IV or DNA gyrase, as well as wild-type strains. We also investigated the use of the MPC in pharmacodynamic studies by comparing the activities of MPC-derived concentrations of each antibiotic in an in vitro pharmacodynamic model. (A portion of this work was presented at the 41st Interscience Conference on Antimicrobial Agents and Chemotherapy, Chicago, Ill., December 2001).
MATERIALS AND METHODS Bacterial strains. Isolate 79 (penicillin and macrolide resistant; fluoroquinolone susceptible) was obtained from the Detroit Medical Center Microbiology Laboratory, Detroit, Mich., while ATCC 49619 (penicillin, erythromycin, and fluoroquinolone susceptible), KD2138, and KD2139 were obtained from the Public Health Research Institute, Newark, N.J. KD2138 and KD2139 are singlestep laboratory-derived parC and gyrA mutants, respectively, of ATCC 49619 (KD2138 was created through culturing on medium containing 2 g of levofloxacin/ml; KD2139 on medium containing 0.42 g of moxifloxacin/ml) (19). Medium. All in vitro pharmacodynamic models were performed using ToddHewitt broth (Difco Laboratories, Detroit, Mich.) supplemented with 0.5% yeast extract (THBY). Colony counts for all experiments were determined using tryptic soy agar supplemented with 5% sheep blood (TSA-SRBC; Difco). Antimicrobial agents. Moxifloxacin (lot 502610) was supplied by Bayer, Pharmaceutical Division, West Haven, Conn. Levofloxacin was commercially purchased. Stock solutions of each antibiotic were freshly prepared on the day of use. In vitro susceptibility testing. MICs were determined by microdilution techniques with an inoculum of 5 ⫻ 105 CFU/ml according to NCCLS guidelines (23). Minimum bactericidal concentrations (MBCs), defined as a 99.9% kill of the starting inoculum, were determined by performing colony counts on microtiter wells showing no visible growth. MPC determinations. Each isolate was inoculated on a total of 10 TSA-SRBC plates and was incubated at 37°C in 5% CO2. The resulting confluent lawn of growth was recovered from all plates, transferred to 500 ml of THBY, and incubated for an additional 18 to 24 h. Cultures were then concentrated by centrifugation (5,000 ⫻ g for 30 min), and cells were resuspended in THBY to yield a concentration of 1010 to 1011 CFU/ml. Next, 100 l of this suspension was applied to TSA supplemented with 0.5% lysed horse blood (Rockland, Inc., Gilbertsville, Pa.) containing a known concentration of each fluoroquinolone. A total of 10 plates were utilized per concentration, and a series of at least seven twofold dilutions of each fluoroquinolone in agar were prepared. Inoculated fluoroquinolone-impregnated plates were incubated for 96 h, at which time the MPC was recorded as the lowest fluoroquinolone concentration completely inhibiting bacterial growth. All MPC determinations were performed in duplicate. Inoculum preparation. Colonies recovered after an overnight incubation on TSA-SRBC were added to THBY to obtain a suspension corresponding to a 1010-CFU/ml inoculum. The contents of several plates were used to accomplish this. An aliquot of this suspension was then added to each model in order to achieve the desired initial inoculum. In vitro pharmacodynamic model. An in vitro pharmacodynamic model consisting of a one-compartment 500-ml glass chamber (working model volume, 250 ml) with multiple ports for the removal of THBY, delivery of antibiotics, and collection of bacterial and antimicrobial samples was utilized (1). All model simulations were conducted over 48 h and were performed in duplicate to ensure reproducibility. Each apparatus was placed in a 37°C water bath for the duration of the experiment, with a magnetic stir bar in each model to facilitate continuous mixing of medium. A peristaltic pump (Masterflex; Cole-Parmer Instrument Co., Chicago, Ill.) was used to continually replace antibiotic-containing medium with fresh THBY (at a rate to simulate the half-lives [t1/2] of respective antibiotics). Fluoroquinolone regimen simulations. Regimen simulations of each fluoroquinolone used in the in vitro pharmacodynamic model are described below and
2608
ALLEN ET AL.
in Table 1. Therapeutic regimens (simulations of concentrations obtained with usual therapeutic dosing) were as follows: moxifloxacin, 400 mg every 24 h (estimated peak concentration, area under the concentration-time curve [AUC], and t1/2, 4.5 g/ml, 48 h, and 12 h, respectively) (U.S. prescribing information for Avelox, available at http://www.avelox.com); levofloxacin, 500 mg every 24 h (6 g/ml, 48 h, and 6 h) (U.S. prescribing information for Levaquin, available at http://www.levaquin.com). In addition, in order to test the idea that intrinsic differences between the two fluoroquinolones contribute to resistance selection, modified (MPC-targeted) regimens were designed to match the AUC/MPC ratios of moxifloxacin and levofloxacin. For MPC-targeted models of levofloxacin, models were designed so that levofloxacin was given moxifloxacin’s elimination t1/2, and the peak concentration (and resulting AUC) was then adjusted in order to match the AUC/MPC attained by the therapeutic moxifloxacin regimen against each isolate. MPC-targeted moxifloxacin models were designed in the same manner (Table 1). When designing our MPC-targeted simulations, we chose to first alter the elimination t1/2 (administering each fluoroquinolone with its comparator’s elimination t1/2) and then modify only the peak concentration (dose) in each simulation. This was done in order to more accurately match the exposure relative to the MPC attained by moxifloxacin and levofloxacin. Of note, peak concentrations in all MPC-targeted moxifloxacin models were lower than those attained in the corresponding therapeutic simulation, while peak concentrations in all MPCtargeted levofloxacin regimens were higher than those attained in the corresponding therapeutic regimen. Thus, all MPC-targeted models favored levofloxacin. Pharmacokinetic analysis. Fluoroquinolone concentrations were determined from samples drawn in duplicate from each model at 0, 0.5, 1, 2, 4, 6, 8, 24, 28, 32, and 48 h. Samples were stored at ⫺70°C until analysis. Peak and trough concentrations and t1/2 were calculated using concentration-time plots of the model samples. The AUC from 0 to 24 h was calculated using the linear trapezoid method and the PKANALYST program (version 1.10; MicroMath Scientific Software, Salt Lake City, Utah). Pharmacodynamic analysis. Samples from each model were collected at 0, 1, 2, 4, 6, 8, 24, 28, 32, and 48 h and were serially diluted in cold 0.9% sodium chloride. Bacterial quantification was performed by plating triplicate 20-l aliquots of each diluted sample on TSA-SRBC. All samples were diluted 10- to 100-fold before plating in order to minimize antibiotic carryover. In cases where the diluted sample contained a fluoroquinolone concentration at or near the MIC, antibiotic carryover was also prevented by addition of antibiotic-binding resin (Amberlite; Sigma Chemical Co., St. Louis, Mo.). Plated samples were incubated at 37°C for 24 h, and colony counts (log10 CFU per milliliter) were determined manually. The limit of detection for this method of colony count determination is 2 log10 CFU/ml. Time-kill curves were determined by plotting mean colony counts (log10 CFU per milliliter) from each model versus time. Bactericidal activity (99.9% kill) was defined as a reduction of ⱖ3 log10 CFU/ml from the initial inoculum. Reductions in colony counts were determined over a 48-h period and were compared between regimens. The time to achieve 99.9% killing was determined by using linear regression (if r2 ⱖ 0.95) or by visual inspection. Antibiotic assays. Moxifloxacin and levofloxacin concentrations were determined through bioassay using antibiotic assay medium 1 (Difco) and Klebsiella pneumoniae ATCC 33495 as the indicator organism. The limit of detection was 0.31 g/ml. Coefficients of variation for all assays were less than 10%. Detection of resistance. Samples (100 l) from each time point were plated on TSA supplemented with 0.5% lysed horse blood containing an antibiotic concentration of four to eight times the MIC for each organism and were incubated for 48 h at 37°C to monitor the development of resistance. Plates were visually inspected for growth of resistant subpopulations after 24, 32, and 48 h of incubation. The MIC for resistant organisms was determined using Etest methods (AB Biodisk, Solna, Sweden) in order to detect all possible MIC elevations. The MIC for resistant isolates was also determined through microdilution using the common efflux pump substrates acriflavine (ACR), benzalkonium chloride (BAC), ethidium bromide (EtBr), and tetraphenylphosphonium (TPP) (as well as the fluoroquinolone by which resistance was selected) in the presence or absence of 20 g of reserpine (Sigma)/ml to test for the presence of effluxmediated resistance. PCR procedures. Original isolates (before model exposure) and any isolates displaying an MIC elevation after fluoroquinolone exposure were subjected to PCR analysis (as well as quinolone resistance-determining region [QRDR] sequencing; see below). Codons 46 to 172, 371 to 512, 35 to 157, and 398 to 483 of gyrA, gyrB, parC, and parE, respectively, encompassing the QRDR of each gene, were amplified from genomic DNA by using primers and PCR parameters as previously described (9, 10, 22, 24, 25, 30, 31). Primer sequences used for PCR
ANTIMICROB. AGENTS CHEMOTHER. TABLE 2. Susceptibility testing and MPC determinations Moxifloxacin concn Isolate MIC
79 ATCC 49619 KD2138 KD2139 a b
a
0.125 0.125 0.25 0.25
b
MBC
0.125 0.125 0.25 0.25
0.25 0.25 0.5 0.5
MIC
b
Levofloxacin concn MPC
MIC
MIC
MBC
MPC
0.5 0.5 8 4
1 0.75 1.5 0.75
1 0.5 1 0.5
2 1 2 1
4 2 64 32
MIC was determined by Etest. MIC and MBC were determined by microdilution.
analyses were 5⬘-CCGTCGCATTCTTTACG and 5⬘-AGTTGCTCCATTA ACCA (gyrA), 5⬘-TTCTCCGATTTCCTCATG and 5⬘-AGAAGGGTACGAAT GTGG (gyrB), 5⬘-TGGGTTGAAGCCGGTTCA and 5⬘-TGCTGGCAAGACC GTTGG (parC), and 5⬘-AAGGCGCGTGATGAGAGC and 5⬘-TCTGCTCCA ACACCCGCA (parE). For all QRDR amplifications, PCR parameters were 94°C for 1 min, 55°C for 1 min, and 72°C for 3 min for 30 cycles. DNA sequence determinations. Nucleotide sequences were determined by the dideoxy chain termination method, using the Applied Biosystems 377 capillarybased automated system (Perkin-Elmer Applied Biosystems, Inc., Foster City, Calif.) (29). Sequencing of two independently generated PCR products was performed to control for polymerase-induced errors. Statistical analysis. Differences between regimens with respect to the value for log10 CFU/ml at 48 h, time to 99.9% kill, the emergence of resistance, and all pharmacodynamic variables were determined using analysis of variance with Tukey’s test for multiple comparisons. In addition, all pharmacodynamic parameters (AUC/MIC, AUC/MPC, peak/MIC, peak/MPC, t ⬎ MIC, t ⬎ MPC, and MIC ⬍ t ⬍ MPC [time spent within the mutant selection window]) were correlated with the change in inoculum over 48 h and time to 99.9% kill by using linear regression, while correlation with the emergence of resistance was performed by logistic regression. For all experiments, a P of ⱕ0.05 was considered indicative of statistical significance. All statistical analyses were performed by using Statistical Package for the Social Sciences (version 10; SPSS, Inc., Chicago, Ill.).
RESULTS Susceptibility testing and MPC determinations. The MIC, MBC, and MPC for each isolate are shown in Table 2. Values for all strains were lower for moxifloxacin than for levofloxacin. Of note, MPCs did not correlate with either the MIC or MBC for either fluoroquinolone. Pharmacokinetics. Observed pharmacokinetic parameters (plus or minus standard deviation) for the tested therapeutic regimens were as follows: peak, trough (both in micrograms per milliliter), and t1/2 (in hours): moxifloxacin: 4.90 ⫾ 0.10, 1.38 ⫾ 0.29, and 13.57 ⫾ 0.50; and levofloxacin, 5.77 ⫾ 0.06, 0.36 ⫾ 0.09, and 6.03 ⫾ 0.51. Pharmacokinetic parameters for all other regimens were within 10% of expected values. Pharmacodynamics. Results of 48-h pharmacodynamic models for the tested isolates are shown in Fig. 1 and 2. In general, adjustment of moxifloxacin pharmacokinetics to mimic therapeutic levofloxacin concentrations resulted in diminished killing, while adjustment of levofloxacin to mimic therapeutic moxifloxacin concentrations led to enhanced killing. As shown in Fig. 1A, the therapeutic moxifloxacin regimen achieved bactericidal activity against isolates 79 (at 8 h), ATCC 49619 (at 24 h), and KD2138 (parC mutant) (at 48 h). Bactericidal activity was observed at 28 h against KD2139 (gyrA mutant), but regrowth was noted at 48 h. The therapeutic levofloxacin regimen (Fig. 1B) was bactericidal against isolates 79 (at 24 h), ATCC 49619 (at 8 h), and KD2138 (parC mutant) (at 48 h). Figure 2A and B show the results of MPC-targeted regimens. As shown in Fig. 2A, moxifloxacin (adjusted to mimic
VOL. 47, 2003
MOXIFLOXACIN AND LEVOFLOXACIN AGAINST S. PNEUMONIAE
2609
FIG. 1. Activity of therapeutic regimens of moxifloxacin (MXF) (A) and levofloxacin (LEV) (B) against all isolates (■, 79; ✖, ATCC 49619; }, KD2138 [parC mutant]; 䊐, KD2139 [gyrA mutant]; ✚, growth control 79; *, growth control ATCC 49619; {, growth control KD2138; and j, growth control KD2139). The dotted line indicates the lower limit of detection (2 log10 CFU/ml) used for bacterial quantification.
the AUC/MPC obtained by the therapeutic levofloxacin regimen) achieved bactericidal activity only against ATCC 49619 (at 28 h) and against KD2138 (parC mutant) (at 48 h). MPCtargeted levofloxacin (adjusted to mimic the AUC/MPC obtained by the therapeutic moxifloxacin regimen) (Fig. 2B) was bactericidal for all isolates (isolate 79, at 24 h; ATCC 49619, at 4 h; and KD2138 [parC mutant] and KD2139 [gyrA mutant], at 8 h for each).
Table 3 lists pharmacodynamic parameters for all model regimens of moxifloxacin and levofloxacin, as well as results of postmodel MIC testing, while Tables 4 and 5 list QRDR sequencing and efflux screening results for all isolates. Moxifloxacin (both therapeutic and MPC-targeted models) did not lead to expression of resistance in isolate 79 or ATCC 49619. The therapeutic moxifloxacin regimen also failed to lead to resistance in KD2138 (parC mutant); however, the MPC-tar-
2610
ALLEN ET AL.
ANTIMICROB. AGENTS CHEMOTHER.
FIG. 2. Activity of MPC-targeted regimens of moxifloxacin (⌬MXF) (A) and levofloxacin (⌬LEV) (B) against all isolates (■, 79; ✖, ATCC 49619; }, KD2138 [parC mutant]; 䊐, KD2139 [gyrA mutant]; ✚, growth control 79; *, growth control ATCC 49619; {, growth control KD2138; and j, growth control KD2139). The dotted line indicates the lower limit of detection (2 log10 CFU/ml) used for bacterial quantification.
geted moxifloxacin regimen led to growth of a GyrA variant (S81Y) of KD2138 that exhibited a 12-fold elevation of MIC from that found for the original strain. Both moxifloxacin models (therapeutic and MPC targeted) selected a ParC derivative (S79Y) of the gyrA mutant KD1239 (resulting in a 12-fold MIC elevation) (Table 4). Levofloxacin (both therapeutic and MPC-targeted models) also did not lead to resistance in isolate 79 (Table 5). However,
the therapeutic levofloxacin regimen led to a twofold MIC increase in the control strain ATCC 49619 (corresponding amino acid substitution, S79Y in ParC). The MPC-targeted levofloxacin model (adjusted to mimic the AUC/MPC attained by the therapeutic moxifloxacin regimen; resultant peak, 14.5 g/ml) did not lead to outgrowth of a resistant variant of ATCC 49619. For strain KD2138 (parC mutant), both levofloxacin models (therapeutic and MPC targeted) resulted in
MOXIFLOXACIN AND LEVOFLOXACIN AGAINST S. PNEUMONIAE
VOL. 47, 2003
2611
TABLE 3. IVPM pharmacokinetics and pharmacodynamics Regimena
t ⬎ MPC (h)
AUC/MIC
9 1.5
24 4
384 48
48 384
1.5 29.5
4 24
1.125 0.094
48 12
9 3
0.375 3.625
48 192
3 7.25
1.125 0.047
48 6
0.375 7.375
Pkb (g/ml)
Tr (g/ml)c
AUC
4.5 0.75
1.125 0.047
48 6
6 29.5
0.375 7.375
4.5 1.5 6 14.5
S. pneumoniae 79 MXF Therapeutic MPC targeted LEV Therapeutic MPC targeted S. pneumoniae ATCC 49619 MXF Therapeutic MPC targeted LEV Therapeutic MPC targeted S. pneumoniae KD2138 MXF Therapeutic MPC targeted
4.5 0.75
LEV Therapeutic MPC targeted
6 29.5
S. pneumoniae KD2139 MXF Therapeutic MPC targeted
4.5 0.75
LEV Therapeutic MPC targeted
6 29.5
Pre-IVPMd
Post-IVPMe
96 12
0.125
NC f NC
48 384
12 96
1
NC NC
24 10
384 96
96 24
0.125
NC NC
10 24
64 256
24 96
0.75
1.5 NC
0.5625 0.094
0 0
192 24
6 0.75
0.25
NC 3
48 384
0.094 0.461
0 0
32 256
0.75 6
1.5
32 32
1.125 0.047
48 6
1.125 0.187
2 0
192 24
12 1.5
0.25
3 3
0.375 7.375
48 384
0.188 0.922
0 0
64 512
1.5 12
0.75
32 32
Pk/MPC
AUC/MPC
a
FQ, fluoroquinolone; IVPM, in vitro pharmacodynamic model; MXF, moxifloxacin; and LEV, levifloxacin. Pk, peak. c Tr, trough. d Pre-IVPM, MIC for original isolate. e Post-IVPM, MIC for resistant isolate recovered from IVPM. f NC indicates post-IVPM MIC equivalent to preexposure MIC. b
outgrowth of an isolate with an approximate 21-fold MIC increase and expression of a GyrA substitution (S81Y). For isolate KD2139 (gyrA mutant), both levofloxacin models (therapeutic and MPC targeted) selected a mutant expressing a
substitution in ParC (S79Y) associated with an approximate 42-fold elevation in MIC. As shown by the presence of fourfold-or-greater decreases in MIC with the addition of reserpine to the efflux pump
TABLE 4. Resistance selection by moxifloxacin MIC reductionb by reserpine with:
MIC (g/ml) Isolate
49619 MXF/49619f ⌬MXF/49619 KD2138 MXF/KD2138 ⌬MXF/KD2138 KD2139 MXF/KD2139 ⌬MXF/KD2139 a
Pre-IVPMa g
NA 0.125 NA 0.25 NA 0.75
Post-IVPMa
NA 0.125 0.125 NA 0.25 3 NA 3 3
FQc
0 NDe ND 0 ND 0 0 0 0
ACR
8 ND ND 8 ND 4 16 8 16
BAC
2 ND ND 2 ND 2 2 0 4
MIC determined by Etest (IVPM, in vitro pharmacodynamic model). Values listed are reduction (n-fold) in MIC with addition of reserpine. FQ, fluoroquinolone (selecting FQ from model). d WT, wild-type sequence. e ND, not determined. f MXF, therapeutic moxifloxacin regimen; ⌬MXF, MPC-targeted moxifloxacin regimen. g NA, not applicable. b c
EtBr
8 ND ND 8 ND 4 16 8 16
QRDR substitution TPP
ParC
GyrA
ParE
GyrB
2 ND ND 0 ND 0 0 0 0
d
WT ND ND WT ND S81Y S81F S81F S81F
WT ND ND WT ND WT WT WT WT
WT ND ND WT ND WT WT WT WT
WT ND ND S79Y ND S79Y WT S79Y S79Y
2612
ALLEN ET AL.
ANTIMICROB. AGENTS CHEMOTHER. TABLE 5. Resistance selection by levofloxacin MIC reduction by reserpine withb:
MIC (g/ml) Isolate Pre-IVPM
49619 LEV/49619f ⌬LEV/49619 KD2138 LEV/KD2138 ⌬LEV/KD2138 KD2139 LEV/KD2139 ⌬LEV/KD2139
g
NA 0.75 NA 1.5
NA 0.75
a
Post-IVPM
NA 1.5 0.75 NA 32 32 NA 32 32
FQ
c
0 0 NDe 0 0 0 0 0 0
ACR
8 8 ND 8 8 8 16 16 8
BAC
2 0 ND 2 4 0 2 0 2
EtBr
8 8 ND 8 8 16 16 8 8
QRDR substitution TPP
ParC
GyrA
ParE
GyrB
2 2 ND 0 0 0 0 0 0
d
WT WT ND WT S81Y S81Y S81F S81F S81F
WT WT ND WT WT WT WT WT WT
WT WT ND WT WT WT WT WT WT
WT S79Y ND S79Y S79Y S79Y WT S79Y S79Y
a
MIC was determined by Etest. IVPM, in vitro pharmacodynamic model. Values listed are reduction (n-fold) in MIC with addition of reserpine. FQ, fluoroquinolone (selecting FQ from model). d WT, wild-type sequence. e ND, not determined. f LEV, therapeutic levofloxacin regimen; ⌬LEV, MPC-targeted levofloxacin regimen. g NA, not applicable. b c
substrates ACR and EtBr, efflux was noted in all isolates with MIC elevations (Tables 4 and 5). Efflux of moxifloxacin and levofloxacin, however (by either parent isolates or their resistant derivatives), was not detected in our assays. The parameters AUC/MIC (r2 ⫽ 0.502), AUC/MPC (r2 ⫽ 0.571), peak/MIC (r2 ⫽ 0.496), peak/MPC (r2 ⫽ 0.391), t ⬎ MPC (r2 ⫽ 0.529), and MIC ⬍ t ⬍ MPC (r2 ⫽ 0.261) correlated with the inoculum change from 0 to 48 h (P ⱕ 0.05). None of these parameters in these experiments were associated with time to 99.9% kill or with the emergence of resistance. DISCUSSION Fluoroquinolones are widely utilized in the treatment of streptococcal infections due to their excellent activity and ease of use; however, recent reports of diminished susceptibility have raised concerns regarding the future utility of these agents (3, 8, 12, 13, 16, 34–38). Thus, attention is being directed towards optimization of fluoroquinolone activity and resistance prevention through use of pharmacodynamically derived dosing strategies. The MPC is presently receiving notice as a potential pharmacodynamic parameter whose use may play a role in slowing the selective enrichment of resistant mutants by the fluoroquinolones. Although the MPC has been proposed as a means to construct a hierarchy of available fluoroquinolones with respect to their potential for selection of resistant organisms (2), little work evaluating the pharmacodynamic applicability of the MPC has been performed. It has been suggested that maintenance of antimicrobial concentrations within the mutant selection window will lead to enrichment of resistant subpopulations (40). Ideally, then, optimization of antimicrobial dosing from the standpoint of resistance prevention would necessitate maintenance of supra-MPC concentrations for the maximum time possible. In order to investigate the possibility that moxifloxacin and levofloxacin inherently differ in their ability to prevent enrichment of resistant subpopulations, we chose to match the pharmacodynamics of the two agents with respect to the MPC. This was done by devising MPC-targeted regimens of each
fluoroquinolone that attained an AUC/MPC equivalent to that attained by the therapeutic regimen of the comparator fluoroquinolone. It should be noted that our normalized MPC-targeted regimens for each agent were vastly different from those used therapeutically (Table 1). For the fluoroquinolone-susceptible control strain ATCC 49619, therapeutic levofloxacin concentrations led to outgrowth of a parC mutant. Adjustment of levofloxacin concentrations to match the AUC/MPC of moxifloxacin for this isolate overcame this result (although the peak concentration used in this MPC-targeted levofloxacin regimen is not attainable clinically). Neither moxifloxacin regimen (therapeutic or MPC-targeted concentrations) resulted in isolation of resistant mutants of ATCC 49619. We were unable to obtain mutant derivatives of clinical isolate 79 in any simulation performed with either moxifloxacin or levofloxacin; it was previously found (using an inoculum approximately 10,000-fold lower than that used in the present models) that this isolate did not readily express resistance upon fluoroquinolone exposure (4). All levofloxacin regimens led to outgrowth of resistant subpopulations when tested against the mutant isolates KD2138 (parC) and KD2139 (gyrA). Against KD2138, the MPC-targeted moxifloxacin regimen (with a peak concentration sixfold less than that obtained with usual therapeutic dosing) led to outgrowth of a GyrA variant, while therapeutic moxifloxacin concentrations did not. In contrast, both moxifloxacin regimens (therapeutic and MPC targeted) led to selection of a ParC variant of the gyrA mutant KD2139. Regarding the utility of the MPC as a pharmacodynamic parameter, we failed to determine a consistent breakpoint value for either t ⬎ MPC, peak/MPC, or AUC/MPC that was predictive of the emergence of resistance. We found that the parameters AUC/MIC, AUC/MPC, peak/MIC, peak/MPC, t ⬎ MPC, and MIC ⬍ t ⬍ MPC correlated with the population reduction over 48 h, but these parameters were not associated with either time to 99.9% kill or the emergence of resistance. Elucidation of a statistical relationship supporting the use of the MPC as a predictive parameter may be obtained with analysis of a greater number of isolates and/or dosing regimens; this work was not the primary goal of the present re-
VOL. 47, 2003
MOXIFLOXACIN AND LEVOFLOXACIN AGAINST S. PNEUMONIAE
search and would require separate analysis of a significant range of fluoroquinolone concentrations relative to the MPC. However, we did observe a consistency in the hierarchy of MPC values, since for all isolates moxifloxacin exhibited a lower MPC than did levofloxacin. This is consistent with the work of Blondeau et al. (2), as well as with previous results that we observed with Staphylococcus aureus (G. P. Allen, G. W. Kaatz, J. M. Blondeau, and M. J. Rybak, Abstr. 40th Intersci. Conf. Antimicrob. Agents Chemother., abstr. 299, 2000). A potential limitation to our results is the failure in some instances to reach the targeted inoculum of 109 to 1010 CFU (both in MPC determinations and model experiments). This is a common problem when studying the MPC in S. pneumoniae, since autolysis by this organism often limits the inoculum that is obtained (2). However, our finding of MIC elevations at the conclusion of our model simulations leads us to believe that we achieved an inoculum size sufficient to contain resistant subpopulations. It is thought that the C-8-OMe functionality of moxifloxacin confers a dual targeting of topoisomerase IV and DNA gyrase, thus helping to explain the effect of this functional group in restricting selection of resistant organisms (11, 26, 27). Levofloxacin, in contrast, appears to preferentially target topoisomerase IV in S. pneumoniae. This, along with the enhanced lethality of moxifloxacin against mutant variants of S. pneumoniae (19), may explain our finding that moxifloxacin and levofloxacin differed in their ability to prevent further resistance in the parC mutant. In contrast, the presence of a preexisting gyrA mutation (KD2139) negated this difference (although the magnitude of MIC elevation observed in moxifloxacin-resistant mutants of KD2139 was lower than that seen in levofloxacinresistant mutants of this strain). The recessive nature of gyrase mutations (as opposed to topoisomerase IV mutations, which are codominant) may also contribute to the differences in resistance selectivity between the two agents (17). Of note, we cannot definitively conclude that the presence or absence of the C-8-OMe moiety is the factor explaining our findings of differential resistance selection by moxifloxacin and levofloxacin. However, this idea is consistent with research performed by other investigators (5–7, 11, 14, 15, 18, 20, 21, 39, 41–43). It is noteworthy that emergence of resistant subpopulations occurred more readily when isolates containing preexisting mutations (either in parC or gyrA) were exposed to the fluoroquinolones. This is consistent with the stepwise development of fluoroquinolone resistance that is known to occur. This finding has potential clinical implications, since the emergence of S. pneumoniae possessing first-step fluoroquinolone mutations may be the initial step in the proliferation of fluoroquinolone-resistant S. pneumoniae and since such organisms are generally not detected by standard susceptibility testing (28). Our research shows that prevention of the development and/or proliferation of first-step fluoroquinolone mutants is an important goal that may limit the emergence of further resistance. In conclusion, we found that moxifloxacin is superior to levofloxacin in preventing enrichment of resistant subpopulations contained within a large inoculum of S. pneumoniae. This disparity was noted in both a fluoroquinolone-susceptible isolate containing no preexisting target site mutations, as well as a parC mutant, while the difference in prevention of resistance was less evident in the case of a gyrA mutant. Increasing levo-
2613
floxacin concentrations in order to exceed the MPC was effective in preventing the selection of mutants of the fluoroquinolone-susceptible strain. For the parC and gyrA mutants, however, our MPC-targeted levofloxacin regimens did not lead to concentrations exceeding the MPC, so further study of the influence of supra-MPC concentrations is required. Our results may have implications for the future role of fluoroquinolones in the therapy of infections caused by S. pneumoniae. ACKNOWLEDGMENTS We thank Susan Seo for performing PCRs; Raymond Cha, Chrissy Cheung and Patrick Prajzner for technical assistance; and Karl Drlica for critical comments on the manuscript. This work was supported by an award from the Society of Infectious Diseases Pharmacists. REFERENCES 1. Akins, R. L., and M. J. Rybak. 2000. In vitro activities of daptomycin, arbekacin, vancomycin, and gentamicin alone and/or in combination against glycopeptide intermediate-resistant Staphylococcus aureus in an infection model. Antimicrob. Agents Chemother. 44:1925–1929. 2. Blondeau, J. M., X. Zhao, G. Hansen, and K. Drlica. 2001. Mutation prevention concentrations of fluoroquinolones for clinical isolates of Streptococcus pneumoniae. Antimicrob. Agents Chemother. 45:433–438. 3. Chen, D. K., A. McGeer, J. C. de Azavedo, and D. E. Low. 1999. Decreased susceptibility of Streptococcus pneumoniae to fluoroquinolones in Canada. N. Engl. J. Med. 341:233–239. 4. Coyle, E. A., G. W. Kaatz, and M. J. Rybak. 2001. Activities of newer fluoroquinolones against ciprofloxacin-resistant Streptococcus pneumoniae. Antimicrob. Agents Chemother. 45:1654–1659. 5. Dalhoff, A. 2001. Comparative in vitro and in vivo activity of the C-8 methoxy quinolone moxifloxacin and the C-8 chlorine quinolone BAY y 3118. Clin. Infect. Dis. 32(Suppl. 1):S16–S22. 6. Dong, Y., C. Xu, X. Zhao, J. Domagala, and K. Drlica. 1998. Fluoroquinolone action against mycobacteria: effects of C-8 substituents on growth, survival, and resistance. Antimicrob. Agents Chemother. 42:2978–2984. 7. Dong, Y., X. Zhao, J. Domagala, and K. Drlica. 1999. Effect of fluoroquinolone concentration on selection of resistant mutants of Mycobacterium bovis BCG and Staphylococcus aureus. Antimicrob. Agents Chemother. 43:1756– 1758. 8. Empey, P. E., H. R. Jennings, A. C. Thornton, R. P. Rapp, and M. E. Evans. 2001. Levofloxacin failure in a patient with pneumococcal pneumonia. Ann. Pharmacother. 35:687–690. 9. Ferrero, L., B. Cameron, and J. Crouzet. 1995. Analysis of gyrA and grlA mutations in stepwise-selected ciprofloxacin-resistant mutants of Staphylococcus aureus. Antimicrob. Agents Chemother. 39:1554–1558. 10. Fukuda, H., and K. Hiramatsu. 1999. Primary targets of fluoroquinolones in Streptococcus pneumoniae. Antimicrob. Agents Chemother. 43:410–412. 11. Fukuda, H., R. Kishii, M. Takei, and M. Hosaka. 2001. Contributions of the 8-methoxy group of gatifloxacin to resistance selectivity, target preference, and antibacterial activity against Streptococcus pneumoniae. Antimicrob. Agents Chemother. 45:1649–1653. 12. Ho, P.-L., T.-L. Que, D. N.-C. Tsang, T.-K. Ng, K.-H. Chow, and W.-H. Seto. 1999. Emergence of fluoroquinolone resistance among multiply resistant strains of Streptococcus pneumoniae in Hong Kong. Antimicrob. Agents Chemother. 43:1310–1313. 13. Ho, P. L., W. S. Tse, K. W. T. Tsang, T. K. Kwok, T. K. Ng, V. C. C. Cheng, and R. M. T. Chan. 2001. Risk factors for acquisition of levofloxacin-resistant Streptococcus pneumoniae: a case-control study. Clin. Infect. Dis. 32:701–707. 14. Ince, D., and D. C. Hooper. 2001. Mechanisms of resistance to gatifloxacin in comparison to AM-1121 and ciprofloxacin in Staphylococcus aureus. Antimicrob. Agents Chemother. 45:2755–2764. 15. Ito, T., M. Matsumoto, and T. Nishino. 1995. Improved bactericidal activity of Q-35 against quinolone-resistant staphylococci. Antimicrob. Agents Chemother. 39:1522–1525. 16. Jorgensen, J. H., L. M. Weigel, J. M. Swenson, C. G. Whitney, M. J. Ferraro, and F. C. Tenover. 2000. Activities of clinafloxacin, gatifloxacin, gemifloxacin, and trovafloxacin against recent clinical isolates of levofloxacin-resistant Streptococcus pneumoniae. Antimicrob. Agents Chemother. 44:2962–2968. 17. Khodursky, A. B., and N. R. Cozzarelli. 1998. The mechanism of inhibition of topoisomerase IV by quinolone antibacterials. J. Biol. Chem. 273:27668– 27677. 18. Kitamura, A., K. Hoshino, Y. Kimura, I. Hayakawa, and K. Sato. 1995. Contribution of the C-8 substituent of DU-6859a, a new potent fluoroquinolone, to its activity against DNA gyrase mutants of Pseudomonas aeruginosa. Antimicrob. Agents Chemother. 39:1467–1471. 19. Li, X., X. Zhao, and K. Drlica. 2002. Selection of Streptococcus pneumoniae
2614
20.
21.
22.
23.
24. 25. 26.
27. 28. 29. 30. 31.
ALLEN ET AL.
mutants having reduced susceptibility to moxifloxacin and levofloxacin. Antimicrob. Agents Chemother. 46:522–524. Lu, T., X. Zhao, and K. Drlica. 1999. Gatifloxacin activity against quinoloneresistant gyrase: allele-specific enhancement of bacteriostatic and bactericidal activities by the C-8-methoxy group. Antimicrob. Agents Chemother. 43:2969–2974. Lu, T., X. Zhao, X. Li, A. Drlica-Wagner, J.-Y. Wang, J. Domagala, and K. Drlica. 2001. Enhancement of fluoroquinolone activity by C-8 halogen and methoxy moieties: action against a gyrase resistance mutant of Mycobacterium smegmatis and a gyrase-topoisomerase IV double mutant of Staphylococcus aureus. Antimicrob. Agents Chemother. 45:2703–2709. Mun ˜ oz, R., and A. G. De La Campa. 1996. ParC subunit of DNA topoisomerase IV of Streptococcus pneumoniae is a primary target of fluoroquinolones and cooperates with DNA gyrase A subunit in forming resistance phenotype. Antimicrob. Agents Chemother. 40:2252–2257. National Committee for Clinical Laboratory Standards. 1997. Approved standard. Methods for dilution antimicrobial susceptibility tests for bacteria that grow aerobically, 4th ed. NCCLS document M7-A4. National Committee for Clinical Laboratory Standards, Wayne, Pa. Pan, X.-S., J. Ambler, S. Mehtar, and L. M. Fisher. 1996. Involvement of topoisomerase IV and DNA gyrase as ciprofloxacin targets in Streptococcus pneumoniae. Antimicrob. Agents Chemother. 40:2321–2326. Perichon, B., J. Tankovic, and P. Courvalin. 1997. Characterization of a mutation in the parE gene that confers fluoroquinolone resistance in Streptococcus pneumoniae. Antimicrob. Agents Chemother. 41:1166–1167. Pestova, E., R. Beyer, N. P. Cianciotto, G. A. Noskin, and L. R. Peterson. 1999. Contribution of topoisomerase IV and DNA gyrase mutations in Streptococcus pneumoniae to resistance to novel fluoroquinolones. Antimicrob. Agents Chemother. 43:2000–2004. Pestova, E., J. Millichap, G. Noskin, and L. Peterson. 2000. Intracellular targets of moxifloxacin: a comparison with other fluoroquinolones. J. Antimicrob. Chemother. 45:583–590. Richardson, D. C., D. Bast, A. McGeer, and D. E. Low. 2001. Evaluation of susceptibility testing to detect fluoroquinolone resistance mechanisms in Streptococcus pneumoniae. Antimicrob. Agents Chemother. 45:1911–1914. Sanger, F., S. Nicklen, and A. R. Coulson. 1977. DNA sequencing with chain-terminating inhibitors. Proc. Natl. Acad. Sci. USA 74:5463–5467. Taba, H., and N. Kusano. 1998. Sparfloxacin resistance in clinical isolates of Streptococcus pneumoniae: involvement of multiple mutations in gyrA and parC genes. Antimicrob. Agents Chemother. 42:2193–2196. Tankovic, J., B. Perichon, J. Duval, and P. Courvalin. 1996. Contribution of mutations in gyrA and parC genes to fluoroquinolone resistance of mutants
ANTIMICROB. AGENTS CHEMOTHER.
32. 33. 34. 35.
36.
37.
38. 39.
40. 41. 42. 43.
of Streptococcus pneumoniae obtained in vivo and in vitro. Antimicrob. Agents Chemother. 40:2505–2510. Thornsberry, C., J. A. Karlowsky, and D. F. Sahm. 2001. Levofloxacinresistant Streptococcus pneumoniae: second look. Antimicrob. Agents Chemother. 45:2183–2184. Tillotson, G., X. Zhao, and K. Drlica. 2001. Fluoroquinolones as pneumococcal therapy: closing the barn door before the horse escapes. Lancet Infect. Dis. 1:145–146. Urban, C., N. Rahman, X. Zhao, N. Mariano, S. Segal-Maurer, K. Drlica, and J. J. Rahal. 2001. Fluoroquinolone-resistant Streptococcus pneumoniae associated with levofloxacin therapy. J. Infect. Dis. 184:794–798. Weigel, L. M., G. J. Anderson, R. R. Facklam, and F. C. Tenover. 2001. Genetic analyses of mutations contributing to fluoroquinolone resistance in clinical isolates of Streptococcus pneumoniae. Antimicrob. Agents Chemother. 45:3517–3523. Weiss, K., C. Restieri, R. Gauthier, M. Laverdie`re, A. McGeer, R. J. Davidson, L. Kilburn, D. J. Bast, J. de Azavedo, and D. E. Low. 2001. A nosocomial outbreak of fluoroquinolone-resistant Streptococcus pneumoniae. Clin. Infect. Dis. 33:517–522. Whitney, C. G., M. M. Farlewy, J. Hadler, L. H. Harrison, C. Lexau, A. Reingold, L. Lefkowitz, P. R. Cieslak, M. Cetron, E. R. Zell, J. H. Jorgensen, and A. Schuchat. 2000. Increasing prevalence of multidrug-resistant Streptococcus pneumoniae in the United States. N. Engl. J. Med. 343:1917–1924. Wortmann, G. W., and S. P. Bennett. 1999. Fatal meningitis due to levofloxacin-resistant Streptococcus pneumoniae. Clin. Infect. Dis. 29:1599–1600. Zhao, B. Y., R. Pine, J. Domagala, and K. Drlica. 1999. Fluoroquinolone action against clinical isolates of Mycobacterium tuberculosis: effects of a C-8 methoxyl group on survival in liquid media and in human macrophages. Antimicrob. Agents Chemother. 43:661–666. Zhao, X., and K. Drlica. 2001. Restricting the selection of antibiotic-resistant mutants: a general strategy derived from fluoroquinolone studies. Clin. Infect. Dis. 33(Suppl. 3):S146–S157. Zhao, X., J. Wang, C. Xu, Y. Dong, J. Zhou, J. Domagala, and K. Drlica. 1998. Killing of Staphylococcus aureus by C-8-methoxy fluoroquinolones. Antimicrob. Agents Chemother. 42:956–958. Zhao, X., C. Xu, J. Domagala, and K. Drlica. 1997. DNA topoisomerase targets of the fluoroquinolones: a strategy for avoiding bacterial resistance. Proc. Natl. Acad. Sci. USA 94:13991–13996. Zhou, J., Y. Dong, X. Zhao, S. Lee, A. Amin, S. Ramaswamy, J. Domagala, J. M. Musser, and K. Drlica. 2000. Selection of antibiotic-resistant mutants: allelic diversity among fluoroquinolone-resistant mutations. J. Infect. Dis. 182:517–525.
