baumannii has been demonstrated with mutants created by gene inactivation/deletion or by creating spontaneous efflux pump overexpressing mutants via selection on antibiotic gradients, but with some inconsistencies in antimicrobial susceptibilities

depending on how selleck the genes were inactivated [5]. For example, inactivation of adeABC in a clinical MDR isolate by insertion of a ticarcillin-resistance gene conferred increased susceptibility to aminoglycosides, β-lactams, fluoroquinolones, chloramphenicol, tetracycline, macrolides and trimethoprim [7]. However when adeABC was deleted and an apramycin resistance cassette was inserted in the same MDR isolate, the ΔadeABC mutant showed increased susceptibility to fluoroquinolones, chloramphenicol, tetracycline, tigecycline and macrolides but no change in susceptibility to aminoglycosides, trimethoprim and β-lactams [4, 6]. We hypothesized that the antibiotic resistance gene used in the creation of pump gene mutants complicated the interpretation of antimicrobial susceptibility data and hence which agents were putative substrates of each A.

baumannii efflux pump. When adeIJK was inactivated using the marker-less method, the MDR isolates became more susceptible to nalidixic acid, chloramphenicol, clindamycin, tetracycline, minocycline, tigecycline and trimethoprim. It is interesting to note that the DBΔadeIJK and R2ΔadeIJK mutants showed increased susceptibility to nalidixic acid without affecting susceptibility to ciprofloxacin, suggesting AdeIJK may be specific for quinolones LXH254 but not fluoroquinolones. We also noted that, buy Nintedanib although the AdeIJK pump confers increased resistance to exactly the same antibiotics in both DB and R2, the host genotype had an influence on the magnitude of resistance to each antibiotic. The successful creation of adeFGH and adeIJK gene deletions, separately and together, in two MDR A. baumannii isolates demonstrates the robustness of the method and its application across different MDR A. baumannii isolates. The antibiotic substrates revealed with our mutants are in general agreement

with those described by Damier-Piolle et al (2008) in which adeIJK was inactivated in an MDR isolate by gene deletion together with insertion of a kanamycin-resistance cassette [6]. However, in our study the DBΔadeIJK and R2ΔadeIJK mutants were also more susceptible to trimethoprim, but not to β-lactams. It should be noted that differences between these studies may be due to the presence of different antibiotic resistance genes on the host genome, e.g. R2 had bla OXA-23 like, bla OXA-51 like genes, bla TEM , bla OXA and bla ADC that confer β-lactam resistance. The MICs of antibiotics for double mutants R2ΔadeFGHΔadeIJK and DBΔadeFGHΔadeIJK were the same as for the corresponding single mutants R2ΔadeIJK and DBΔadeIJK. This was expected, as a single deletion of adeFGH had minimal effect on MICs of antibiotics in either strain.