BacA: a possible regulator that contributes to the biofilm formation of Pseudomonas aeruginosa

Abstract

Previously, we pointed out in P. aeruginosa PAO1 biofilm cells the accumulation of a hypothetical protein named PA3731 and showed that the deletion of the corresponding gene impacted its biofilm formation capacity. PA3731 belongs to a cluster of 4 genes (pa3732 to pa3729) that we named bac for "Biofilm Associated Cluster." The present study focuses on the PA14_16140 protein, i.e., the PA3732 (BacA) homolog in the PA14 strain. The role of BacA in rhamnolipid secretion, biofilm formation and virulence, was confirmed by phenotypic experiments with a bacA mutant. Additional investigations allow to advance that the bac system involves in fact 6 genes organized in operon, i.e., bacA to bacF. At a molecular level, quantitative proteomic studies revealed an accumulation of the BAC cognate partners by the bacA sessile mutant, suggesting a negative control of BacA toward the bac operon. Finally, a first crystallographic structure of BacA was obtained revealing a structure homologous to chaperones or/and regulatory proteins.

Keywords: Pseudomonas aeruginosa; Psp system; biofilm; crystallography; proteome; rhamnolipids.

Figure 1

Decrease of rhamnolipid secretion. (A) Extracted Ion Chromatograms (EIC) peaks detected for the most abundant rhamnolipids collected from bacterial cultures. (B) Molecular representation of the expected fragmentation pattern of the Rha-Rha-C₁₀-C₁₀ (m/z: 649.38). (C) Calibration curve obtained from commercial rhamnolipids. The EIC areas were measured from MS data for all detected rhamnolipids injected at concentrations covering nearly 2 orders of magnitude (2.5, 5, 10, 15, 20, 25, 30, 40, 50, 75, 100 ng). (D) Rhamnolipids quantification in supernatants of 24 h or 48 h culture of the WT and the ΔbacA strains. Data are expressed as mean values (±SEM) from three independent biological experiments (****p < 0.0001).

Figure 2

Differential phenotypic traits. (A) Swarming motility assay in agar plates after 16 h of incubation at 37°C. (B) Resulting covered distances for the WT and ∆bacA strains (**p < 0.01). (C) Percentage of adhesion based on the crystal violet test of the WT and ΔbacA strains. The 100% value represents the OD value obtained with the WT. Data represent mean values (±SEM) from at least three independent biological experiments (***p < 0.001). (D) Representative images of the crystal violet coloration obtained for the WT and ΔbacA strains after 4 h and 6 h of incubation. (E) Quantification of the biofilm formation by the WT and ∆bacA strains using the Biofilm Ring Test^®. Data represent mean values (±SEM) from at least four independent biological experiments (ns: non-significant; *p < 0.05; **p < 0.01). (F) Representative images of the migration of the magnetic beads in contact with the WT and ΔbacA strains after 1 h and 6 h of incubation. (G) Quantitative comparison of biofilm biovolume (μm³/μm²) based on the integration of microscopy images. Data represent mean values (±SEM) from at least three independent biological experiments (ns: non-significant; *p < 0.05; ****p < 0.0001). (H) Representative confocal microscopy images of the WT and ∆bacA biofilms labeled with Syto9.

Figure 3

Differential quantitative proteomic analyses of WT strain and ∆bacA mutant grown as biofilms. (A) Volcano plot representation of differential proteomics analysis. Circles representing proteins are placed according to their statistical p-value and their relative abundance ratio (log 2-fold change). Gray circles correspond to proteins that do not present significant differential abundance. (B) Classification of the differentially represented proteins (under-represented on the left and over-represented in the right) according to their biological processes using the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway database. (C) Protein–protein interaction network for proteins displaying the most differential abundances between the biofilm formed by the WT or ∆bacA strains. Nodes are colored according to protein abundance: red color corresponds to accumulated proteins in the mutant whereas blue color is used for proteins of decreased abundance. Edges indicate protein–protein interactions.

Figure 4

BAC system is coding by an operon made up by 6 genes and regulated by a σ⁵⁴ (RpoN) sigma factor. (A) Electrophoretic gel from RT-PCR experiment. Lane 1: transcript bacA-bacB, lane 2: transcript bacB-bacC, lane 3: transcript bacC-bacD, lane 4: transcript bacD-PA14_16190, lane 5: transcript PA14_16190–16,200, lane M: 100 bp ladder molecular size marker (bp). (B) σ⁵⁴ sigma factor -binding sites of P. aeruginosa PA14_16,140 (bacA). The highly conserved −24 and −12 dinucleotides are shown in bold.

Figure 5

Crystallographic structure of BacA. (A) Schematic representation of the DUF2170 domain identified using Interpro database. (B) Structure of one of the two dimers of BacA, reconstructed after application of the crystallographic symmetries. The 3D structure is oriented in order to highlight the regions expected to interact with the polypeptide it is supposed to chaperon as solved in the structure of two other complexes involving a chaperon CesT/Tir (5wez) and ExsC/ExsE (3kxy) (see C where the interacting molecules are colored in yellow). The two monomers are colored in blue and green, respectively. One molecule of glycerol has been identified in the structure, localized in the pocket formed by four residues (Tyr-72, Phe-84, Glu-95, Leu-97) in the hollow of the concave beta sheet. The two residues solved in two alternative positions (Glu-90, His-155, see main text) are also indicated. The missing loop 130–136 is schematized as a dotted line. (C) Crystal structure of the Tir-CesT effector-chaperone complex (5wez, left part) and of the ExsC-ExsE complex (3kxy, right part). The PDB numbers and structures presented were obtained from the PDB website (https://www.rcsb.org/).