Identification of Ata, a multifunctional trimeric autotransporter of Acinetobacter baumannii

Abstract

Acinetobacter baumannii has recently emerged as a highly troublesome nosocomial pathogen, especially in patients in intensive care units and in those undergoing mechanical ventilation. We have identified a surface protein adhesin of A. baumannii, designated the Acinetobacter trimeric autotransporter (Ata), that contains all of the typical features of trimeric autotransporters (TA), including a long signal peptide followed by an N-terminal, surface-exposed passenger domain and a C-terminal domain encoding 4 β-strands. To demonstrate that Ata encoded a TA, we created a fusion protein in which we replaced the entire passenger domain of Ata with the epitope tag V5, which can be tracked with specific monoclonal antibodies, and demonstrated that the C-terminal 101 amino acids of Ata were capable of exporting the heterologous V5 tag to the surface of A. baumannii in a trimeric form. We found that Ata played a role in biofilm formation and bound to various extracellular matrix/basal membrane (ECM/BM) components, including collagen types I, III, IV, and V and laminin. Moreover, Ata mediated the adhesion of whole A. baumannii cells to immobilized collagen type IV and played a role in the survival of A. baumannii in a lethal model of systemic infection in immunocompetent mice. Taken together, these results reveal that Ata is a TA of A. baumannii involved in virulence, including biofilm formation, binding to ECM/BM proteins, mediating the adhesion of A. baumannii cells to collagen type IV, and contributing to the survival of A. baumannii in a mouse model of lethal infection.

Fig 1

(A) Immunoblot of Ata-V5-His fusion protein expressed in E. coli. Outer membrane proteins (OMPs) of E. coli LMG194 harboring pAta-V5-6×His were prepared after induction with either 0.2% glucose (lane 1) or 0.2% arabinose (lane 2) and then tested by Western blotting with a monoclonal antibody (MAb) directed against the V5 peptide. Samples from arabinose-induced OMPs were treated with 70% formic acid and detected as before (lane 3). The predicted monomeric mass of Ata-V5-His fusion proteins is 13.8 kDa. Migration distances of molecular mass markers (in kDa) are indicated on the right. (B) Detection of Ata-V5-His fusion protein in E. coli pAta-V5-6×His by CLSM. E. coli pAta-V5-6×His cultures were grown in M9 plus 0.25% Casamino Acids and induced with 0.2% arabinose or 0.2% glucose. The V5 epitope was labeled with a mouse anti-V5 MAb and a secondary goat anti-mouse Alexa 488 fluorescent antibody (green channel), and cell nucleic acids were stained with DAPI (blue channel). (a) E. coli pAta-V5-6×His induced with arabinose (green channel) or (b) glucose (green channel). Also shown are DAPI-stained nuclei of E. coli cells (blue channel) after induction with arabinose (c) or glucose (d). (C) CLSM detection of Ata in A. baumannii ATCC 17978, Δata, and Δata-c strains. A. baumannii ATCC 17978, Δata, and Δata-c induced with arabinose or glucose were labeled with rabbit anti-Ata antisera and a secondary goat anti-rabbit Alexa 488 fluorescent antibody. Nucleic acids were stained with DAPI. Merged images of Alexa 488- and DAPI-stained A. baumannii ATCC 17978 (a) and Δata (b), Δata-c induced with glucose (c), and Δata-c induced with arabinose (d) are shown.

Fig 2

Quantification of Ata expression in A. baumannii ATCC 17978 by flow cytometry during various growth phases. A. baumannii was grown in LB to early exponential (OD₆₅₀ of 0.025 or 0.1), mid-exponential (OD₆₅₀ of 0.4), late exponential (OD₆₅₀ of 0.8), or stationary phase (OD₆₅₀ of 1.2), labeled with rabbit antibody to Ata and secondary goat antibody to rabbit IgG conjugated to Alexa 488 fluorescent dye, and analyzed by flow cytometry. Results represent the mean fluorescent intensity (MFI) of 500,000 cells, and bars indicate the averages from three independent experiments ± standard errors of the means (SEM).

Fig 3

Quantitative biofilm formation by A. baumannii ATCC 17978, Δata, and Δata-c on polystyrene surfaces. A. baumannii 17978 and Δata were grown in LB and ATCC 17978 Δata-c in LB plus 2% arabinose or 0.2% glucose under static conditions for 24 h at 37°C. Total biofilm formation (OD₅₉₅) was normalized by bacterial growth (OD₆₅₀). The bars indicate the means of 15 tubes from five independent experiments ± SEM. P values were determined by one-way ANOVA with Tukey's post hoc analysis. ns, not significant.

Fig 4

Adhesive properties of Ata. (A) Dose-dependent binding of Ata to selected ECM/BM components. ECM/BM proteins, including collagen types I, II, III, IV, and V, heparan, fibronectin, vitronectin, and laminin, and BSA (as a control) were used at 5 μg/well, and binding of Ata was quantified by ELISA. Data points represent the means from four independent experiments ± SEM. (B) Binding of A. baumannii ATCC 17978, Δata, and Δata-c induced with 2% arabinose or 0.2% glucose to immobilized collagen type IV. The bars indicate the means from three independent experiments ± SEM. P values were determined by one-way ANOVA with Tukey's post hoc analysis. ns, not significant.

Fig 5

Survival curves of mice (n = 8; C57BL/6) following intraperitoneal infection with A. baumannii ATCC 17978 (1.6 × 10⁷ CFU/mouse), Δata (1.6 × 10⁷ CFU/mouse), Δata-pLVB-Ata (ata gene with its native promoter cloned into pBAD18Kan-Ori; 1.4 × 10⁷ CFU/mouse), and Δata-pBAD18Kan-Ori (empty vector pBAD18Kan-Ori; 1.2 × 10⁷ CFU/mouse). P = 0.0005 by log-rank test in a Kaplan-Meier analysis of wild-type ATCC 17978 versus Δata and ATCC 17978 versus Δata-pBAD18kan-Ori.

Fig 6

Analysis of Ata expression among A. baumannii clinical isolates. (A) Flow-cytometric analysis of Ata production among A. baumannii clinical isolates positive for the ata gene by PCR. A. baumannii strains were grown in LB and probed with rabbit anti-Ata antibodies and a secondary goat anti-rabbit Alexa 488 fluorescent antibody. Approximately 500,000 cells were then analyzed by flow cytometry, and antibody binding was expressed as mean fluorescent intensity (MFI). Bars represent the fold ratio (in MFI) of each A. baumannii clinical isolate to the MFI of the ATCC 17978 Δata strain. (B) Western blot (WB) analysis of Ata levels in 4 strains producing high levels of Ata (ATCC 17978, S19, I30, and N10) and in 4 strains producing low levels of Ata (I42, I25, I31, and I28) as determined by FACS and the Ata-negative strain ATCC 17978 Δata, which was used as a control. Migration distances of molecular mass markers (in kDa) are indicated on the left. (Strains used for WB analysis are highlighted with arrows in panel A). Outer membrane proteins were extracted, resolved by SDS-PAGE, and transferred to a PVDF membrane, and Ata was detected by WB with anti-Ata rabbit antibodies and a secondary goat anti-rabbit IgG.