Resistance to pentamidine is mediated by AdeAB, regulated by AdeRS, and influenced by growth conditions in Acinetobacter baumannii ATCC 17978

Abstract

In recent years, effective treatment of infections caused by Acinetobacter baumannii has become challenging due to the ability of the bacterium to acquire or up-regulate antimicrobial resistance determinants. Two component signal transduction systems are known to regulate expression of virulence factors including multidrug efflux pumps. Here, we investigated the role of the AdeRS two component signal transduction system in regulating the AdeAB efflux system, determined whether AdeA and/or AdeB can individually confer antimicrobial resistance, and explored the interplay between pentamidine resistance and growth conditions in A. baumannii ATCC 17978. Results identified that deletion of adeRS affected resistance towards chlorhexidine and 4',6-diamidino-2-phenylindole dihydrochloride, two previously defined AdeABC substrates, and also identified an 8-fold decrease in resistance to pentamidine. Examination of ΔadeA, ΔadeB and ΔadeAB cells augmented results seen for ΔadeRS and identified a set of dicationic AdeAB substrates. RNA-sequencing of ΔadeRS revealed transcription of 290 genes were ≥2-fold altered compared to the wildtype. Pentamidine shock significantly increased adeA expression in the wildtype, but decreased it in ΔadeRS, implying that AdeRS activates adeAB transcription in ATCC 17978. Investigation under multiple growth conditions, including the use of Biolog phenotypic microarrays, revealed resistance to pentamidine in ATCC 17978 and mutants could be altered by bioavailability of iron or utilization of different carbon sources. In conclusion, the results of this study provide evidence that AdeAB in ATCC 17978 can confer intrinsic resistance to a subset of dicationic compounds and in particular, resistance to pentamidine can be significantly altered depending on the growth conditions.

Fig 1. Global transcriptomic response differences of A. baumannii ATCC 17978 after deletion of adeRS.

Each diamond marker represents a predicted gene within the genome ordered according to locus tag along the X-axis and differential expression generated from normalized reads per kilobase mapped of WT against ΔadeRS are displayed on the Y-axis (Log₂). Positive and negative Log₂-values correlate to up- and down-regulated genes, respectively. Green and red circles highlight genes/gene clusters of interest that have been up- and down-regulated, respectively. See S3 Table for the full list of genes that were differentially expressed ≥ 1 Log₂ fold.

Fig 2. Increased expression of adeA following pentamidine stress is dependent on the presence of AdeRS in ATCC 17978.

Transcriptional levels of adeA (ACX60_09125) from WT and ΔadeRS were determined by qRT-PCR after 30 min shock with 7.8 mg/L of pentamidine (Pent) (0.5 × MIC of ΔadeRS) and corrected to untreated cells (UT) after normalization to 16S. Bars represent the mean fold change (Log₂) of three biological replicates undertaken in triplicate, and error bars represent ± SEM. Statistical analyses were performed by Student’s t-test, two-tailed, unpaired; ** = P < 0.01 and *** = P < 0.001.

Fig 3. Resistance to pentamidine is modulated by carbon sources available in the growth medium.

Ten-fold serial dilutions of A. baumannii ATCC 17978 (WT), ΔadeRS (ΔRS) and ΔadeAB (ΔAB) cells were used to compare the concentration of pentamidine that inhibits growth in different media, Mueller-Hinton agar (MHA) was used as a comparative control. Images display serial 1:10 dilutions after overnight incubation at 37°C, where DF is abbreviated for dilution factor and N represents undiluted cells. Strains were grown in the absence of pentamidine (UT) or presence of 32, 64, 128, 256 and 512 mg/L of pentamidine (P32, P64, P128, P256, and P512, respectively). Carbon sources tested in M9 minimal medium were used at a final concentration of 0.4% (w/v). ND, not done due to precipitation of pentamidine once added into the molten medium. Figures are representative examples of results obtained.

Fig 4. Kinetic response curves paralleling bacterial growth from Biolog PM01 and PM2A plates identify ten carbon sources that increase pentamidine resistance in ΔadeRS.

P8, P16, P32 and P64 represent kinetic response curves at 8, 16, 32, or 64 mg/L of pentamidine compared to the untreated control, respectively. Red curves represent respiration of untreated ΔadeRS, whilst respiratory activity which overlaps between the control and the sample in the different experimental conditions is represented in yellow. Only carbon compounds that promote at least 50% maximal respiration and induce a recovery response by 36 h are shown. See S3 Fig for respiration curves for all tested treatments.