In Vitro And In Vivo Studies Of Mutant Selection Window For Methicillin-resistant Staphylococcus Aureus

Objective:To test the susceptibility of the resistant mutants of S. aureus ATCC43300by threedifferent MIC testing methods, which direct to explain the relationship between theMIC creep and the mutant selection window (MSW) for vancomycin.To study the comparison of vancomycin (VAN) separately combined withlevofloxacin (LVX), rifampincin (RFP) and fosfomycin (FOM) for decreasing mutantprevention concentration (MPC) of S. aureus ATCC43300in vitro, which direct rationalapplications of available antibiotics in clinical practice and provide evidence for newstrategies for restricting the development of resistance.To test the MSW hypothesis with Methicillin-resistant Staphylococcus aureus(MRSA) exposed to vancomycin in the tissue-cage infection model and to compare invivo and in vitro exposures that restrict the enrichment of resistant mutants.Meterials and Methods:S. aureus ATCC43300was stored at the Anhui Center for Surveillance of BacterialResistance. Female New Zealand White rabbits weighing2.5–3kg each were suppliedby the Anhui Medical University Experimental Animal Center. M-H agar dilution method was used to determine minimum inhibitoryconcentration (MIC) of VAN against S. aureus ATCC43300.The MPC of vancomycin was tested by the M-H agar dilution method. The S.aureus ATCC43300was enriched the concentration to1010colony forming units permilliliter in broth. High-density cultures were prepared from overnight cultures grownin MH agar medium followed of incubation with shaking at37℃. Inoculated drugs-impregnated plates were incubated for72h, and the provisional mutant preventionconcentration (MPCpr) was recorded as the lowest concentration completely inhibitingbacterial growth at this time.To estimate the MPC and MIC99, logarithms of bacterial numbers were plotted againstantibiotic concentrations.The resistant mutants were selected from different antibiotic concentrations in theMSW (lower, middle and upper portions of the window). The Mueller–Hinton Agar(MHA) and Brain–Heart Infusion Agar (BHIA) plates were inoculated with10μl of asuspension of the testing isolate equivalent to104CFU/ml. Test plates were incubated in ambientair at35℃and were observed after24h. MIC testing for the resistant mutants and S.aureus ATCC43300was performed using the Etest method, following manufacturer’sguidelines. The Spa typing was performed on each of S. aureus ATCC43300and theresistant mutants.M-H agar dilution method was used to determine the MICs of VAN、LVX、RFPand FOM alone against S.aureus ATCC43300.The simplified chessboard method is determinated the MIC value of VAN combined with LVX, RFP, FOM. The FICI (fractional inhibitory concentration index)was calculated for the three combined treatment.The MPC of each agent and combination of them were tested by the M-H agardilution method. The S. aureus ATCC43300was enriched the concentration to1010colony forming units per milliliter in broth. High-density cultures were prepared fromovernight cultures grown in MH agar medium followed of incubation with shaking at37℃. Inoculated drugs-impregnated plates were incubated for72h, and the MPC wasrecorded as the lowest concentration completely inhibiting bacterial growth at this time.Local infection with MRSA was established in rabbits, and the infected animalswere treated with various doses of twice-daily vancomycin (half-life6h) for threeconsecutive days to provide antibiotic concentrations below the MIC, between the MICand the MPC, and above the MPC. The total drug concentration of vancomycin wasdetermined by high-performance liquid chromatography (HPLC). Thepharmacokinetic/pharmacodynamic (PK/PD) indices were calculated according to anoncompartmental model by use of DAS software (version2.0; Wuhu, China). Allindices were determined after the sixth dose, at which time steady-state kinetics werereached.In each experiment, multiple sampling of bacteria-containing medium from thecentral compartment was performed throughout the observation period. One hundredmicrolitre samples were plated onto MHA plates. In order to account for antibioticcarryover, all samples were diluted sufficiently prior to plating, therefore reducing theantibiotic concentration below the MIC of the drug. The lower limit of accuratedetection was2×102CFU/mL. To detect changes in susceptibility during treatment,multiple samples from the model were determined every12hours before vancomycin intravenous drip during vancomycin treatment and24,48,72h after the termination ofvancomycin treatment. The each sample was plated onto agar plates containing2×and4×MIC of vancomycin (detection limit10CFU/mL). In addition, the MICs ofvancomycin for each rabbit were determined prior to and after treatment.Fisher’s exact test was used for statistical analysis of the PK/PD data, with aninfected but untreated set of rabbits as a control. P<0.05was considered to bestatistically significant.Results:The MIC of VAN was estimated at2μg/ml on the MHA plate only for the resistantmutant that was selected from the plate of vancomycin concentration12μg/ml. Theobvious changes of susceptibility testing were found between the resistant mutants andS. aureus ATCC43300on the BHIA plates. There were subtle changes in the MIC trendwithin the susceptible range with the result of Etest for the resistant mutants. Thesusceptibility for the subcultures of resistant mutants would fall back when the externaldrugs environment disappeared. In comparison with the S. aureus ATCC43300,sequence analysis revealed that there were no mutations in the Spa sequencing of theresistant mutants. The Spa tape is t421for all isolates.MICs of VAN, LVX, RFP and FOM for S. aureus ATCC43300were1μg/ml,0.25μg/ml,0.008μg/ml,2μg/ml; MPCs were16μg/ml,2μg/ml,>32μg/ml,64μg/ml; andMSWs (MPC/MIC) were16,8,>4000,32; The MIC values of VAN combined withLVX, RFP, FOM were decreased for different degrees. The FICIs for the threecombined treatment were1,1,0.75. VAN combined with LVX can decrease toMPCVANused alone1/2times. VAN combined with RFP can decrease to MPCVANusedalone1/4times. VAN combined with FOM can decrease to MPCVANused alone1/16 times. FOM combined with the concentration of6.4μg/ml (8MIC×80%) of VAN candecrease MPCFOMto2μg/ml and narrow MSW of FOM used alone1/32times. VANcombined with the concentration of28.8μg/ml (16MIC×90%) of FOM can decreaseMPCVANto1μg/ml and narrow MSW of VAN used alone1/16times.S. aureus ATCC43300lost vancomycin susceptibility when drug concentrations atthe site of infection fluctuated between the lower and upper boundaries of the MSW,defined in vitro as the MIC99and the MPC, respectively. Both boundaries aredetermined in vitro, before starting animal studies. The value at which resistant mutantsare not enriched in vivo was estimated as an AUC24/MPC value of～15h, whereAUC24is the area under the drug concentration time curve in a24-h interval. Theestimated anti-mutant AUC/MIC ratio in vivo was≥200h.Conclusions:A broad plateau in mutant recovery is observed for vancomycin with S. aureusATCC43300. Perhaps it is the decisive reason for the treatment failures. Due to thedifference of drug distribution, the concentration of vancomycin may fall in MSW sothat resistant mutants were enriched, with a concomitant loss in susceptibility. Sovancomycin MIC creeps were found over time for long-term and/or extensive use ofvancomycin.Compared with each one alone, the combination of VAN separately with otherthree antibiotics can obviously decrease MPCVAN. The combination of VAN and FOMwas more evidently in dropping MPC and narrowing MSW of VAN.These findings of in vivo studies support the MSW hypothesis and the anti-mutantAUC/MIC ratio estimated in vivo is consistent with that reported in in vitro studies. Keeping vancomycin concentrations above the MPC or AUC24/MPC>15h is astraightforward way to restrict the acquisition of resistance.