In Vitro And In Vivo Study On The Synergistic Effect Of Colistin Combined With Glycopeptide Against Extensively Drug-resistant Acinetobacter Baumannii

Objectives To investigate the in vitro activities of colistin combined with glycopeptide vancomycin or daptomycin against extensively drug-resistant Acinetobacter baumannii.To evaluate the potential of Galleria mellonella larvae as an in vivo infection model for Acinetobacter baumannii, and employ the model to assess the in vivo antibacterial activity of glycopeptide combined with colistin, in an attempt to gain further insights into whether these therapies should be explored further for the treatment of extensively drug-resistant Acinetobacter baumannii infection.To compare the in vitro and in vivo growth rate, as well as differences in virulence and growth competitiveness between colistin-resistant and colistin-susceptible Acinetobacter baumannii strains, in order to discuss whether it have biological cost in the process to become resistant to colistin.Meterials and Methods Bacterial strainsIn this study, we used six strains of Acinetobacter baumannii, including a standard Acinetobacter baumannii ATCC 19606, 2 colistin-susceptible strains isolated from clinical specimens from 2012 to 2013 in Anhui Province. And another 3 colistinresistant strains were mutant of above 3 isolates obtained in vitro. Quality control strains Escherichia coli ATCC 25922 and Pseudomonas aeruginosa ATCC 27853 were stored by Anhui Center for Surveillance of Bacterial Resistance.Methods 1. ATCC19606 and two clinical isolates(GN2231 and GN0653) are used as parent strains. Three colistin-resistant strains(19606R, GN2231 R and GN0653R) were obtained by multiple passages in increasing concentrations of colistin. Minimum inhibitory concentrations(MICs) of antibiotics were obtained by the standard agar dilution method according to CLSI recommendations. Carbapenems and colistin resistant genes were identified with polymerase chain reaction(PCR). Population analysis profiles(PAPs) were used to determine heteroresistance to colistin-susceptible strains. Synergy between colistin and glycopeptides(vancomycin and daptomycin) were tested using the microtiter plate chequerboard assay and time-kill methodology. Inhibitory activity of antibiotics against biofilms and the mutant prevention concentrations were also studied in vitro.2. To investigate the virulence of two strains and the differences in mortality, we injected different concentrations of live bacteria(ATCC19606 and GN2231) into the hemolymph of each Galleria mellonella larva via the last left proleg. A heat-killed inoculum experiment was carried out to investigate the impact of dead strains on G. mellonella survival. To investigate the effect of post-inoculation incubation temperature on the survival of infected larvae, larvae were inoculated with same CFU of live bacteria and incubated at different temperatures. To test the protective effect of five antibiotics(cefepime, cefoperazone/sulbactam, imipenem, gentamicin and levofloxacin), we injected those drugs into G. mellonella and finally drew the survival curves.3. We determined the virulence and lethal dose of the 6 strains using the G. mellonella larvae infection model. G. mellonella larvae were infected with a lethal inoculum of A. baumannii, and colistin alone or combined with vancomycin or daptomycin were then administered by injection into a different proleg within 2h. G. mellonella larvae that were injected twice with PBS were used as a further control group for these experiments. The doses were based on those used for humans. The larvae were observed for survival every 24 hours for 4 days. Survival analysis was determined using the log-rank test, with a p value ≤ 0.05 considered statistically significant. Larvae were inoculated with a lethal dose of GN2231. Vancomycin at 10 mg/kg were administered within 2 h of inoculation. This dose was chosen to mimic those used to treat human infections. To study the dose-dependent effect, doses of 1 mg/kg and 20 mg/kg also were administered. To assess the ability of vancomycin to protect against subsequent A. baumannii infection, single treatment dose of vancomycin(1 mg/kg, 10 mg/kg, or 20 mg/kg) were given 2 h prior to inoculation with lethal dose of bacteria, as described above.4. In order to measure the relative fitness of each colistin-resistant mutant compared with the parental strain, an in vitro competitive index(CI) was determined. Exponentially growing cells of each mutant and the wild-type strain were mixed in a 1:1 proportion and grown in LB broth at 37°C for 48 h. The number of cells corresponding to each strain of the pair was determined at 0, 4, 8, 12, 24, and 48 h by spreading serial 10-fold dilutions onto LB agar plates with or without 8 mg/L of colistin. For in vivo bacterial growth and CI experiments in G. mellonella larvae mode, 2 groups of 30 larvae were inoculated with 10 μL of each strain, 19606 and 19606 R, separately and with a mixed inoculum(50% of each strain). Subgroups of 5 larvae were homogenized at 0, 4, 8, 12 and 24 h, and CFU were determined after plating in MH agar with or without colistin, and the CI was calculated as described above.Results 1. ATCC 19606 was susceptible to all the tested antimicrobials except glycopeptide. GN2231 and GN0653 were resistant to all the tested antimicrobials except tigecycline and colistin. The MICs of 19606 R, GN2231 R, and GN0653 R to colistin were 8 mg/L, 32 mg/L, and 16 mg/L, respectively. The bla OXA-51 gene was detected in all the six strains, and the bla OXA-23 gene was also detected in those strains except 19606 and 19606 R. The mutation of pmr B gene caused amino acid substitution of Ala227 Val. Heterogeneously resistant subpopulation was detected in all the 3 colistin-susceptible strains. Strong synergic effect was showed for colistin when combined with vancomycin or daptomycin using in vitro checkerboard test, mutant prevention concentration test and biofilms inhibitory test. In the study of time-killing test, vancomycin or daptomycin monotherapy did not show any bactericidal activity against all the 6 strains. But colistin combined with vancomycin or daptomycin showed strong synergic effect against colistin-susceptible strains, and vancomycin-colistin combination showed synergy against colistin-resistant strains 19606 R and GN2231 R.2. The effect of infection with 19606 and GN2231 on survival of G. mellonella larvae was shown. With live bacterial inocula, killing was significantly dependent on the number of A. baumannii cells injected in a dose-dependent manner during 96 h of incubation. And the lethal doses of different strains were different, reflecting the different virulence between strains. With both strains, the heat-killed A. baumannii inoculum had no significant effect on larval survival. Further, we observed that post-inoculation temperature affected larval survival after injection of live A. baumannii. It was shown that survival is reduced with increasing post-inoculation temperature(incubated at 25, 30 and 37°C). Larvae infected with the susceptible strain 19606 showed that all five antibiotics resulted in significantly greater percentage survival, compared with those larvae treated with PBS only(P<0.05). In contrast, when G. mellonella larvae were infected with extensively drug-resistant A. baumannii GN2231, these treatment regimens conferred no therapeutic benefit. In conclusion, the efficacy of the antibiotics in vivo closely correlated with the measured in vitro sensitivities. And we also showed that successful antibiotic therapy results in a parallel reduction in larval burden of A. baumannii.3. Larvae infected with colstin-susceptible strains showed that colistin resulted in significantly greater percentage survival, compared with those larvae treated with PBS only(P<0.05), especially for ATCC 19606. But the colistin monotherapy did not show protective effect to larvae infected with colstin-resistant strains. Treatment of G. mellonella larvae infected with lethal doses of A. baumannii resulted in significantly increased survival rates when vancomycin was given with colistin compared to colistin treatment alone(P<0.05). Daptomycin monotherapy did not confer therapeutic benefit for infected larvae. But daptomycin-colistin combination therapy significantly increased survival rates compared to monotherapy. Interestingly, it is contradictory to the in vitro susceptibility data of vancomycin, that the apparent effectiveness of vancomycin monotherapy in improving larval survival. A statistically significant increase in survival at all time points was detected for 10 mg/kg, or 20 mg/kg vancomycin treatment compared to the control for GN2231. Daptomycin monotherapy did not show the same protective effect as vancomycin.4. Colistin-resistant mutants showed more susceptible to other antimicrobials except colistin compared with wild type. Colistin-resistant strains showed significantly decreased biofilm formation. To test for potential loss of virulence following colistin adaptation, Galleria larvae were infected with each of the colistin-adapted mutants and compared with wild type infection. GN2231 R and GN0653 R showed no change in virulence, but 19606 R appeared to have lost their virulence after adaptation to colistin. The growth of strain 19606R(colistin resistant) was less than that of strain 19606(colistin susceptible) when the strains were grown in competition(48-h competition index, 0.84). In Galleria larvae model, 19606 R reached hemolymph concentrations of 8 log10 CFU/m L when coinfecting with 19606, which was 2.2 log10 CFU/m L lower(24-h competition index, 0.78).Conclusions 1. The combination of colistin with vancomycin or daptomycin have in vitro synergistic antibacterial activity against extensively drug-resistant strains of A. baumannii, and also can inhibit the biofilm formation and reduce mutant prevention concentration. Heteroresistance to colistin is one of the reasons for regrowth of colistin-susceptible strains when colistin used alone. 2. G. mellonella was developed to investigate the pathogenicity of A. baumannii, and the model was also able to assess the effects of antimicrobial treatment on sensitive and resistant strains of A. baumannii.3. This work suggests that glycopeptide-colistin combinations are highly active against A. baumannii both in vitro and in G. mellonella model of infection. It is contradictory to the in vitro susceptibility data of vancomycin, that the apparent effectiveness of vancomycin monotherapy in improving larval survival. 4. We have shown decreased fitness, growth capacity, and lower virulence associated with the acquisition of colistin resistance in A. baumannii.