AbaM Regulates Quorum Sensing, Biofilm Formation, and Virulence in Acinetobacter baumannii

Abstract

Acinetobacter baumannii possesses a single divergent luxR/luxRI-type quorum-sensing (QS) locus named abaR/abaI This locus also contains a third gene located between abaR and abaI, which we term abaM, that codes for an uncharacterized member of the RsaM protein family known to regulate N-acylhomoserine lactone (AHL)-dependent QS in other beta- and gammaproteobacteria. Here, we show that disruption of abaM via a T26 insertion in A. baumannii strain AB5075 resulted in increased production of N-(3-hydroxydodecanoyl)-l-homoserine lactone and enhanced surface motility and biofilm formation. In contrast to the wild type and the abaI::T26 mutant, the virulence of the abaM::T26 mutant was completely attenuated in a Galleria mellonella infection model. Transcriptomic analysis of the abaM::T26 mutant revealed that AbaM differentially regulates at least 76 genes, including the csu pilus operon and the acinetin 505 lipopeptide biosynthetic operon, that are involved in surface adherence, biofilm formation and virulence. A comparison of the wild type, abaM::T26 and abaI::T26 transcriptomes, indicates that AbaM regulates ∼21% of the QS regulon including the csu operon. Moreover, the QS genes (abaI and abaR) were among the most upregulated in the abaM::T26 mutant. A. baumanniilux-based abaM reporter gene fusions revealed that abaM expression is positively regulated by QS but negatively autoregulated. Overall, the data presented in this work demonstrates that AbaM plays a central role in regulating A. baumannii QS, virulence, surface motility, and biofilm formation.IMPORTANCEAcinetobacter baumannii is a multiantibiotic-resistant pathogen of global health care importance. Understanding Acinetobacter virulence gene regulation could aid the development of novel anti-infective strategies. In A. baumannii, the abaR and abaI genes that code for the receptor and synthase components of an N-acylhomoserine (AHL) lactone-dependent quorum sensing system (QS) are separated by abaM Here, we show that although mutation of abaM increased AHL production, surface motility, and biofilm development, it resulted in the attenuation of virulence. AbaM was found to control both QS-dependent and QS-independent genes. The significance of this work lies in the identification of AbaM, an RsaM ortholog known to control virulence in plant pathogens, as a modulator of virulence in a human pathogen.

Keywords: Acinetobacter; N-acylhomoserine lactones; RsaM; biofilm; quorum sensing; virulence.

FIG 1

(A) Schematic of the abaRMI QS locus in A. baumannii AB5075 showing the organization of the three QS genes and their orientations. Green boxes represent predicted lux boxes. Curved arrows represent the predicted abaM and abaI promoters. (B) Multiple sequence alignment of A. baumannii AB5075 AbaM (Abau) with previously characterized orthologs in other bacterial species: Bcen, Burkholderia cenocepacia J2315 BcRsaM; Pfus, Pseudomonas fuscovaginae UPB0736 RsaM; Btha-2, Burkholderia thailandensis E264 RsaM-2; Bglu, Burkholderia glumae BGR1 TofM; Btha-1, Burkholderia thailandensis E264 RsaM-1. The MUSCLE algorithm (48) was used to create the alignment, and ESPript (49) was used to render residue similarities and generate the final figure. The red background indicates conserved residues. Red residues indicate conservative substitutions. Blue frames indicate highly conserved regions. The secondary structures in B. cenocepacia BcRsaM (PDB entry 4O2H) are displayed above the alignment. η, 3₁₀-helix; α, α-helices; β, β-strands; TT, strict β-turns; TTT, strict α-turns.

FIG 2

AHL production in wild-type, abaI::T26, and abaM::T26 strains. (A) Chemical structures of the AHLs produced by A. baumannii AB5075. (B and C) Quantification of OHC12 (B) and OHC10 (C) production throughout growth. Error bars represent the standard deviations between three biological replicates. Asterisks indicate statistically significant differences: **, P ≤ 0.01; ***, P ≤ 0.001.

FIG 3

Phenotypic characterization of the Acinetobacter abaM and abaI mutants. (A) Surface motility on agar. (B) Biofilm formation on polypropylene. For both biofilm and surface motility assays, error bars indicate the standard deviations. Asterisks indicate statistically significant differences compared to the wild-type AB5075 strain. ***, P ≤ 0.001; ****, P ≤ 0.0001. (C) Galleria mellonella larva killing after inoculation of approximately 2 × 10⁴ (left) or 2 × 10⁵ (right) CFU/larva. Each graph represents data from three independent biological replicates together. At least 30 larvae were used for each strain and assay. None of the control larvae died after 5 days. Asterisks indicate statistically significant differences compared to the wild-type AB5075 strain. ****, P ≤ 0.0001.

FIG 4

Comparison of the transcriptomes of the abaI::T26 and abaM::T26 mutants. (A and B) Genes differentially expressed in abaM::T26 (A) and abaI::T26 (B) strains compared to the wild type. Blue circles indicate upregulated genes, red circles indicate downregulated genes, and gray circles represent genes where changes in expression are unlikely to be biologically significant. (C) Venn diagram showing that AbaM regulates genes that are both QS dependent and QS independent.

FIG 5

Validation ABUW_3773 and csu expression by quantitative real-time PCR. The relative expression of csuA/B and ABUW_3773 in mutants was compared to the wild-type AB5075 strain. The expression was normalized in relation to an endogenous control gene (rpoB). Error bars indicate standard deviations between three independent biological replicates. Hashtags indicate a biologically significant difference [|log₂(fold change)| ≥ 1] compared to the wild type.

FIG 6

Expression of abaM and abaI. (A) abaM promoter activity in A. baumannii A5075 wild-type, abaM::T26, and abaI::T26 strains relative to the wild-type strain. (B) abaM promoter activity in response to exogenous OHC12 as a function of growth (RLU/OD₆₀₀). (C) abaI promoter activity in the wild type and an abaI mutant in response to exogenous OHC12 as a function of growth (RLU/OD₆₀₀). Error bars indicate standard deviations between three independent biological replicates. Asterisks indicate statistically significant differences compared to the wild-type AB5075 strain. ***, P ≤ 0.001; ****, P ≤ 0.0001.

FIG 7

Proposed model for the QS/AbaM IFFL in A. baumannii 5075. AbaR activated by OHC12 positively activates expression of abaM and abaI and hence OHC12 production. AbaM is negatively autoregulated and also represses expression of both abaR and abaI. Under the growth conditions used here, AbaM negatively regulates surface motility, biofilm, and Csu pili but is required for virulence as abaM mutants are avirulent in Galleria mellonella larvae. Green arrows and red lines represent positive and negative regulation, respectively.