The Relationship between Quinolone Exposures and Resistance ...

3 downloads 0 Views 65KB Size Report
Mar 17, 2006 - Vincent H. Tam,* Arnold Louie, Mark R. Deziel,† Weiguo Liu, and George L. Drusano. Emerging Infections .... ery 24 h (terminal half-life. 12.8 h ...
ANTIMICROBIAL AGENTS AND CHEMOTHERAPY, Feb. 2007, p. 744–747 0066-4804/07/$08.00⫹0 doi:10.1128/AAC.00334-06 Copyright © 2007, American Society for Microbiology. All Rights Reserved.

Vol. 51, No. 2

The Relationship between Quinolone Exposures and Resistance Amplification Is Characterized by an Inverted U: a New Paradigm for Optimizing Pharmacodynamics To Counterselect Resistance䌤 Vincent H. Tam,* Arnold Louie, Mark R. Deziel,† Weiguo Liu, and George L. Drusano Emerging Infections and Host Defense Laboratory, Ordway Research Institute, Albany, New York Received 17 March 2006/Returned for modification 9 August 2006/Accepted 31 October 2006

We determined the relationship between garenoxacin exposure and quinolone-resistant subpopulations for three bacterial isolates in an in vitro hollow-fiber infection model. An “inverted-U” relationship was identified wherein resistant subpopulations rose initially and then declined with increasing exposure, until reaching a threshold that prevented resistance amplifications. Different targets for the area under the concentration-time curve over 24 h/MIC ratio were required for different bacteria. time. This may be perceived clinically as the emergence of resistance. We have previously demonstrated in an animal system the possibility of dosing to suppress the amplification of this resistant subpopulation (6). In the present study, we used a hollowfiber infection model (HFIM) to examine the relationship between garenoxacin (a des-fluoroquinolone) exposures and the likelihood to counterselect resistance. Antimicrobial agent. Garenoxacin powder [lot 807T-3(A)] was supplied by Bristol-Myers Squibb Pharmaceutical, Inc. (Princeton, NJ). Ciprofloxacin powder (lot 110) was supplied by Bayer Corp. (Kankakee, IL). Stock solutions of garenoxacin and ciprofloxacin were prepared at a concentration of 1 mg/ml in sterile water and stored at ⫺20°C. Microorganism. Klebsiella pneumoniae ATCC 13883 (American Type Culture Collection, Manassas, VA) and a clinical strain of methicillin-resistant, ciprofloxacin-susceptible Staphylococcus aureus (MRSA-CS) were used. An isogenic ciprofloxacin-resistant MRSA strain (MRSA-CR) was generated by culturing approximately 109 CFU of MRSA-CS onto Trypticase soy agar plates (Difco, Detroit, MI) containing 3⫻ MIC of ciprofloxacin. All of the bacterial isolates were stored at ⫺70°C in skim milk. Fresh isolates were grown on blood agar plates (BBL Microbiology Systems, Cockeysville, MD) for 24 h at 35°C before each investigation. Susceptibility studies. The MICs and MBCs of the bacterial isolates to garenoxacin were determined in cation-adjusted Mueller-Hinton II broth (Difco) by using a serial twofold broth macrodilution method described by the Clinical and Laboratory Standards Institute (CLSI; formerly NCCLS) (8). The MICs of ciprofloxacin for MRSA-CS and MRSA-CR were also determined. The studies were conducted in duplicate and repeated once on a separate day. HFIM. The schematic diagram of the system (1) and details of the experimental setup (5) have been published previously. A few colonies of bacteria were inoculated in cation-adjusted Mueller-Hinton II broth and incubated at 35°C until reaching the late log phase growth. The bacteria were inoculated into the HFIM at a concentration of approximately 3 ⫻ 108 CFU/ ml. The high inoculum was used to simulate the bacterial load in severe infections (e.g., nosocomial pneumonia) and to allow

The emergence of resistance has become a worldwide problem that challenges the ability of physicians to provide appropriate therapy to seriously infected patients. As resistance to our current antimicrobial agents continues to rise, the toll in mortality and morbidity, as well as the cost of care, will rise in parallel. It is imperative, therefore, to identify ways to suppress the emergence of resistance. There are multiple mechanisms by which organisms may become resistant to antimicrobials (12). The acquisition of plasmids bearing resistance determinants and the acquisition of new genes through the production of mosaic chromosomes are two common ways for resistance to emerge. However, a third important way is the spontaneous production of a clone bearing a point DNA mutation that provides a competitive advantage when the organism is under selective pressure. Some examples include mutations in the control sequence of the AmpC ␤-lactamase, point mutations resulting in the overexpression of efflux pumps, and point mutations in topoisomerase II and IV genes. Each of the examples of this mechanism cited above occurs with a specific frequency that differs among antibiotics and species. When the population burden seen at an infection site (e.g., nosocomial pneumonia) exceeds the inverse of the mutational frequency substantially, there will be a high probability that multiple subpopulations of organisms will exist. The largest subpopulation will be the susceptible population of organisms. The more frequent the occurrence of the point mutation, the larger the baseline population containing that specific mutation will be. Antimicrobial pressure may have differential effects on these subpopulations. A specific drug exposure may be adequate to mediate a substantial (multilog) decline in total bacterial burden. However, that exposure may have only a minor effect on the resistant subpopulation or may even amplify that group of organisms, so that the resistant clones represent a higher fraction of the total population over

* Corresponding author. Mailing address: University of Houston College of Pharmacy, 1441 Moursund St., Houston, TX 77030. Phone: (713) 795-8316. Fax: (713) 795-8383. E-mail: [email protected]. † Deceased. 䌤 Published ahead of print on 20 November 2006. 744

VOL. 51, 2007

resistant subpopulation(s) to be present at baseline. Each HFIM was exposed to garenoxacin with different and escalating free area under the concentration-time curve over 24 h (AUC24)/MIC exposures for 48 h. The AUC24/MIC exposures simulated the steady-state human pharmacokinetic profile of unbound garenoxacin when the drug is administered once every 24 h (terminal half-life ⫽ 12.8 h, clearance ⫽ 6 liters/h) (2). Seven regimens were used for each bacterial isolate; simulated free AUC24/MIC values ranged from 0 to 800 for K. pneumoniae and MRSA-CS. The free AUC24/MIC exposures of 0 to 2,000 were used for MRSA-CR. Serial samples were obtained from the HFIM for garenoxacin concentrations analysis using a validated high-performance liquid chromatography method as described previously (10). A one-compartment open model with zero-order, time-limited infusion was fit to the concentration-time profiles to determine the clearance by using the ADAPT II program (4). The AUC24 was derived by using the relationship: AUC24 ⫽ dose/clearance. Microbiologic responses. Samples were also obtained from each HFIM for quantitative culture at 0 and 48 h to define the effect of the various drug exposures on the selection of resistant bacterial subpopulations. Prior to plating the bacteria quantitatively, the bacterial samples were washed once with sterile normal saline in order to minimize drug carryover effect. Resistant subpopulations were quantified by culturing onto heart infusion agar plates (Difco) containing garenoxacin at a concentration 3⫻ MIC to garenoxacin. The medium plates were incubated at 35°C for up to 72 h, and then colonies on each plate were counted. Breakpoints were defined as the minimal AUC24/MIC needed to prevent resistant mutant population from amplifying above the absolute number present at baseline. At 48 h, one mutant clone growing on each antibiotic containing plate was picked, and its susceptibility to garenoxacin was redetermined in parallel to the baseline isolate to confirm the presence of resistance. For selected S. aureus isolates, the quinolone resistance determining regions (QRDRs) of the gyrA, gyrB, grlA, and grlB genes were amplified by PCR and sequenced as previously described (9). Susceptibilities of bacterial isolates to garenoxacin. The susceptibilities of various bacteria and their respective isolates obtained from garenoxacin-supplemented plates are shown in Table 1. Minimal change in garenoxacin MIC was observed with the MRSA-CS (parent) isolate in the presence of reserpine (10 ␮g/ml). However, a fourfold reduction in MIC was seen in the MRSA-CR (daughter) strain with the addition of reserpine. The MICs of ciprofloxacin for MRSA-CS and MRSA-CR were found to be 0.5 and 4 ␮g/ml, respectively. DNA sequencing studies of both isolates did not reveal any point mutation in the QRDRs of the gyrA, gyrB, grlA, and grlB genes (data not shown). Microbiologic responses. The system setups were reasonable for simulating the target drug exposures, as indexed to AUC24 (data not shown). The fraction of drug-resistant organisms in the total population observed at baseline and at 48 h did not differ with the placebo (data not shown). As seen in Fig. 1, low AUC24/MIC ratios amplified the resistant mutant population. Apart from K. pneumoniae, for which the peak amplification was at a very low AUC24/MIC ratio (AUC24/MIC ratio ⬵ 10), the maximal amplification of the resistant subpopulation of S. aureus isolates occurred at AUC24/MICs ratios between 20 and

NOTES

745

TABLE 1. Susceptibility of various bacterial isolates to garenoxacin Bacterium

K. pneumoniae ATCC 13883 MRSA-CS MRSA-CR

Baseline concn (␮g/ml)

MIC (␮g/ml) at 48 h

MIC

MBC

0.16

0.32

0.64–10.24

0.04 0.16

0.32 0.32

0.08–0.32 2.56

35. It was clear by inspection that the resistant population declined as a function of increasing drug exposure. The minimal exposures estimated to hold the number of resistant clones at or below the number present at baseline were AUC24/MIC ratios of 67 for K. pneumoniae, 143 for MRSA-CS, and 431 for MRSA-CR. The MRSA-CR strain was derived from the parent strain of MRSA-CS by passage on a plate containing ciprofloxacin. The resistance counterselective exposure for this parent-daughter pair of organisms is vastly different (more than threefold), even though the exposures were altered to address the baseline change in MIC. Figure 1D shows a stylized response of the number of resistant clones as a function of the AUC24/MIC ratio. It was on the basis of this response that we used the term “inverted-U response.” Susceptibilities to garenoxacin were rechecked at the end of the experiment. MIC increases were, in the main ⬎4-fold, with a few isolates demonstrating a 2-fold change. The greatest MIC changes were 8to 16-fold for S. aureus and 64-fold for K. pneumoniae. Because of the crisis of drug resistance among bacteria, particularly those found in the nosocomial setting, any simple methodology that could guide the appropriate dosing of antimicrobial agents to help suppress amplification of the resistant subpopulation would be important. Such a methodology would allow drugs currently in use to remain in the physician’s armamentarium for a longer period, since the pressure to switch to newer agents would be reduced because of decreased resistance. Furthermore, new agents could be developed from the beginning so that the dosing schedule used would suppress the emergence of resistance. Previous uses of the HFIM system have focused on the overall relationship between drug exposure and killing of the total microbial population. In the present investigation, we examined the effect of differing drug exposures on the size of the resistant subpopulation over 48 h in three different organisms often seen in nosocomial (and community-acquired) infections. Low AUC24/MIC ratios (10 to 35) were optimal for maximally amplifying the preexisting resistant subpopulation. Exposures in excess of this amplified the resistant subpopulation suboptimally, until an exposure was achieved that kept the number of resistant clones at or below the number present at the initiation of the therapeutic pressure. This sequence gives the plot an “inverted-U” shape. This relationship has been postulated for drug therapy in human immunodeficiency virus infections (3). However, we are unaware of experimental data demonstrating this relationship for any other class of pathogens. This direct demonstration of an “inverted-U” shape for the function linking drug exposure to suppression of the resistant subpopulation is, to our knowledge, the first. This concept

746

NOTES

ANTIMICROB. AGENTS CHEMOTHER.

FIG. 1. Resistant populations at 48 h under various drug exposures. The data at AUC24/MIC 0 represent the densities of resistant subpopulation as baseline (0 h). All other datum points denote the densities of resistant subpopulations present at 48 h.

has inspired several subsequent investigations, repeatedly demonstrating a similar trend (5, 10, 11). K. pneumoniae required a relatively low AUC24/MIC ratio to counterselect resistance (i.e., an AUC24/MIC ratio of 67). The MRSA-CS strain, however, required an AUC24/MIC ratio of 143. This finding is concordant with the clinical observations that the emergence of fluoroquinolone resistance in general and of ciprofloxacin in particular occurs much more readily for methicillin-resistant strains. Somewhat to our surprise, the MRSA-CR strain required a much larger AUC24/MIC ratio (about 431) to counterselect resistance. An obvious observation of Fig. 1C was that the estimate of an AUC24/MIC ratio of 431 was probably imprecise. There were no other observations between an AUC24/MIC ratio of 200 and 450. Therefore, the slope of the line may be flatter than would have been found had there been other observations in the range. However, the overall finding that a greater AUC24/MIC ratio needed for the prevention of amplification of the resistant subpopulation in the ciprofloxacin-resistant strain was certainly valid. An AUC24/MIC ratio of 200 still allowed near-maximal amplification of the resistant subpopulation. This value was already considerably in excess of that necessary for the MRSA-CS parent isolate (AUC24/MIC ratio of 143). The addition of reserpine showed minimal effect in the MRSA-CS isolate on its susceptibility to garenoxacin. However, the MRSA-CR strain demonstrated a fourfold reduction in MIC with the addition of the reserpine. It is highly likely that the daughter isolate has overexpressed the norA pump, which is analogous to the quinolone-effluxing pmrA pump in Streptococcus pneumoniae. Sequencing through the QRDRs of the MRSA-CR isolate demonstrated no mutation likely to result in

an MIC change to a quinolone. However, the isolates recovered at the end of the experiment were found to have a garenoxacin MIC 16-fold higher than the baseline MIC. QRDR sequencing of these mutant isolates (derived from MRSA-CR) demonstrated mutations at S84L of gyrA and E84K of grlA, both known quinolone resistance mutations. This indicated that the likely explanation for the difference between MRSA-CS and MRSA-CR was the frequency with which these organisms developed a mutation to overexpress the norA pump. Further, this finding suggested that the much higher exposures required to counterselect resistance in the ciprofloxacin-resistant daughter isolate were because the presence of pump overexpression enhanced the mutational frequency of altered target site, as we have shown previously in S. pneumoniae (7). The presence of these secondary mutation(s) resulted in a greater drug exposure necessary to counterselect the amplification. The counterselective AUC24/MIC ratios differed considerably among the strains tested. It is unlikely to be straightforward in identifying a single AUC24/MIC ratio for a drug that will counterselect all isolates, unless a very large value is used reflecting the genus and species that is the most difficult to suppress. In summary, there is a considerable between genus and species difference in the intensity of the exposures required for resistance suppression. The presence of efflux pump overexpression may influence the likelihood in which S. aureus acquires a (secondary) target mutation. For this type of point mutation-based resistance, the ability to control resistance can be influenced by the dose of drug selected. Such findings will be crucial to fighting our current crisis of resistance and warrants further investigations in the clinical setting.

VOL. 51, 2007 This study was supported by Bristol-Myers Squibb Pharmaceuticals, Inc., Princeton, NJ. REFERENCES 1. Bilello, J. A., G. Bauer, M. N. Dudley, G. A. Cole, and G. L. Drusano. 1994. Effect of 2⬘,3⬘-didehydro-3⬘-deoxythymidine in an in vitro hollow-fiber pharmacodynamic model system correlates with results of dose-ranging clinical studies. Antimicrob. Agents Chemother. 38:1386–1391. 2. Bristol-Myers Squibb. 1999. Investigator brochure (BMS-284756). BristolMyers Squibb Pharmaceutical Research Institute, Wallingford, CT. 3. Condra, J. H., and E. A. Emini. 1997. Preventing HIV-1 drug resistance. Sci. Med. 4:14–23. 4. D’Argenio, D. Z., and A. Schumitzky. 1997. ADAPT II user’s guide: pharmacokinetic/pharmacodynamic systems analysis software. Biomedical simulations resource, University of Southern California, Los Angeles. 5. Gumbo, T., A. Louie, M. R. Deziel, L. M. Parsons, M. Salfinger, and G. L. Drusano. 2004. Selection of a moxifloxacin dose that suppresses drug resistance in Mycobacterium tuberculosis, by use of an in vitro pharmacodynamic infection model and mathematical modeling. J. Infect. Dis. 190:1642–1651. 6. Jumbe, N., A. Louie, R. Leary, W. Liu, M. R. Deziel, V. H. Tam, R. Bachhawat, C. Freeman, J. B. Kahn, K. Bush, M. N. Dudley, M. H. Miller, and G. L. Drusano. 2003. Application of a mathematical model to prevent in vivo amplification of antibiotic-resistant bacterial populations during therapy. J. Clin. Investig. 112:275–285.

NOTES

747

7. Jumbe, N. L., A. Louie, M. H. Miller, W. Liu, M. R. Deziel, V. H. Tam, R. Bachhawat, and G. L. Drusano. 2006. Quinolone efflux pumps play a central role in emergence of fluoroquinolone resistance in Streptococcus pneumoniae. Antimicrob. Agents Chemother. 50:310–317. 8. NCCLS. 2003. Methods for dilution antimicrobial susceptibility tests for bacteria that grow aerobically; Approved standard, 6th ed. NCCLS, Wayne, PA. 9. Schmitz, F. J., B. Hofmann, B. Hansen, S. Scheuring, M. Luckefahr, M. Klootwijk, J. Verhoef, A. Fluit, H. P. Heinz, K. Kohrer, and M. E. Jones. 1998. Relationship between ciprofloxacin, ofloxacin, levofloxacin, sparfloxacin, and moxifloxacin (BAY 12-8039) MICs and mutations in grlA, grlB, gyrA, and gyrB in 116 unrelated clinical isolates of Staphylococcus aureus. J. Antimicrob. Chemother. 41:481–484. 10. Tam, V. H., A. Louie, M. R. Deziel, W. Liu, R. Leary, and G. L. Drusano. 2005. Bacterial-population responses to drug-selective pressure: examination of garenoxacin’s effect on Pseudomonas aeruginosa. J. Infect. Dis. 192:420– 428. 11. Tam, V. H., A. N. Schilling, S. Neshat, K. Poole, D. A. Melnick, and E. A. Coyle. 2005. Optimization of meropenem minimum concentration/MIC ratio to suppress in vitro resistance of Pseudomonas aeruginosa. Antimicrob. Agents Chemother. 49:4920–4927. 12. Walsh, C. 2000. Molecular mechanisms that confer antibacterial drug resistance. Nature 406:775–781.