HqiA, a novel quorum-quenching enzyme which expands the AHL lactonase family

Abstract

The screening of a metagenomic library of 250,000 clones generated from a hypersaline soil (Spain) allowed us to identify a single positive clone which confers the ability to degrade N-acyl homoserine lactones (AHLs). The sequencing of the fosmid revealed a 42,318 bp environmental insert characterized by 46 ORFs. The subcloning of these ORFs demonstrated that a single gene (hqiA) allowed AHL degradation. Enzymatic analysis using purified HqiA and HPLC/MS revealed that this protein has lactonase activity on a broad range of AHLs. The introduction of hqiA in the plant pathogen Pectobacterium carotovorum efficiently interfered with both the synthesis of AHLs and quorum-sensing regulated functions, such as swarming motility and the production of maceration enzymes. Bioinformatic analyses highlighted that HqiA showed no sequence homology with the known prototypic AHL lactonases or acylases, thus expanding the AHL-degrading enzymes with a new family related to the cysteine hydrolase (CHase) group. The complete sequence analysis of the fosmid showed that 31 ORFs out of the 46 identified were related to Deltaproteobacteria, whilst many intercalated ORFs presented high homology with other taxa. In this sense, hqiA appeared to be assigned to the Hyphomonas genus (Alphaproteobacteria), suggesting that horizontal gene transfer had occurred.

Figure 1

Diffusion agar-plate assay to detect AHL degradation using the biosensors Chromobacterium violaceum CV026, C. violaceum VIR07 and Agrobacterium tumefaciens NTL4 (pZLR4). Cell-free LB medium (1), E. coli S17 λ pir (2) and E. coli S17 λ pir harbouring the fosmid f10/17.1H (3) were supplemented with the same quantities of C₄-HSL, C₆-HSL, 3-O-C₆-HSL, C₈-HSL, 3-O-C₈-HSL, C₁₀-HSL, 3-OH-C₁₀-HSL, C₁₂-HSL, 3-O-C₁₂-HSL and C₁₄-HSL.

Figure 2

Confirmation of AHL-degradation activity. (a) TLC analysis using the biosensor Agrobacterium tumefaciens NTL4 (pZLR4). (b) HPLC/MS measurements of remaining C₆-HSL and C₁₂-HSL after 24 h incubation with E. coli S17 λ pir and E. coli S17 λ pir::f10/17.1H. LB medium was used as control. Initial AHL concentration was 25 µM. Error bars represent standard deviations. Different letters above the bars indicate that the values are significantly different according to a one-factor (treatment) ANOVA (P < 0.05) and the Tukey test.

Figure 3

Genetic map of the DNA insert of the fosmid f10/17-1H. The 46 ORFs are numbered and coloured according to their phylogeny, ORF29 being (in black) hqiA. The chemotaxis related genes are also presented as square dots.

Figure 4

Phylogenetic analysis of the ORFs found in fosmid f10/17.1H based on protein sequences. (a) ORF6 as an example of affiliation of ORFs by phylogenetic analysis of representative proteins of each genus identified based on maximum likelihood (phyML). (b) Phylogenetic analysis of HqiA showed that it is not related to any known AHL lactonases, otherwise clustering in the cysteine hydrolase group with other hypothetical (HYPO), isochorimatase-like (ISO) and N-carbamoylsarcosine amidase (NCAR)-like enzymes.

Figure 5

Functional characterization of purified HqiA. (a) Detection of the quorum-quenching activity against C₁₂-HSL of the purified HqiA protein and the negative control RE (reaction buffer) using the biosensor Agrobacterium tumefaciens NTL4 (pZLR4). (b) HPLC/MS analysis of the C₁₂-HSL degradation by HqiA. RE and HqiA sample chromatograms are shown. In the RE and HqiA samples, the main peaks masses are 282 and 300 which corresponds to the closed and open-ring form of the C₁₂-HSL respectively, therefore indicating a lactonase activity.

Figure 6

Quorum quenching of AHLs and QS-regulated functions in Pectobacterium carotovorum. (a) AHL detection using the biosensors Chromobacterium violaceum CV026 and Agrobacterium tumefaciens NTL4 (pZRL4). (b) Phenotypes tested in pectin, casein, DNase, PVK (alkaline phosphatase), swarming and swimming media. Error bars represent standard deviations. For each assay, the P. carotovorum strain without plasmid (1), with the empty plasmid (pME6010) (2) and the plasmid carrying hqiA (3) were compared. For the bioassays presented in panel b, different letters above the bars indicate that the values are significantly different according to a one-factor (treatment) ANOVA (P < 0.05) and the Tukey test.

Figure 7

Virulence assay on potato tubers. Potato tuber slices inoculated with (a) Pectobacterium carotovorum without plasmid, (b) P. carotovorum (pME6010), (c) P. carotovorum expressing the hqiA gene and (d) cell-free LB medium.