A functional phenylacetic acid catabolic pathway is required for full pathogenicity of Burkholderia cenocepacia in the Caenorhabditis elegans host model

Abstract

Burkholderia cenocepacia is a member of the Burkholderia cepacia complex, a group of metabolically versatile bacteria that have emerged as opportunistic pathogens in cystic fibrosis and immunocompromised patients. Previously a screen of transposon mutants in a rat pulmonary infection model identified an attenuated mutant with an insertion in paaE, a gene related to the phenylacetic acid (PA) catabolic pathway. In this study, we characterized gene clusters involved in the PA degradation pathway of B. cenocepacia K56-2 in relation to its pathogenicity in the Caenorhabditis elegans model of infection. We demonstrated that targeted-insertion mutagenesis of paaA and paaE, which encode part of the putative PA-coenzyme A (CoA) ring hydroxylation system, paaZ, coding for a putative ring opening enzyme, and paaF, encoding part of the putative beta-oxidation system, severely reduces growth on PA as a sole carbon source. paaA and paaE insertional mutants were attenuated for virulence, and expression of paaE in trans restored pathogenicity of the paaE mutant to wild-type levels. Interruption of paaZ and paaF slightly increased virulence. Using gene interference by ingested double-stranded RNA, we showed that the attenuated phenotype of the paaA and paaE mutants is dependent on a functional p38 mitogen-activated protein kinase pathway in C. elegans. Taken together, our results demonstrate that B. cenocepacia possesses a functional PA degradation pathway and that the putative PA-CoA ring hydroxylation system is required for full pathogenicity in C. elegans.

FIG. 1.

Proposed PA catabolic pathway of B. cenocepacia strain J2315. (A) Genetic organization of the PA catabolic gene clusters in B. cenocepacia strain J2315. (B) PA catabolic enzymes and putative intermediates of the PA catabolic pathway. Genes disrupted by insertional mutagenesis are shown in bold. Disrupted steps are marked with an “X” and the observed pathogenic phenotype summarized as follows: solid lines, attenuated pathogenicity; dashed lines, increased pathogenicity. The gene names are in accordance with those listed in reference .

FIG. 2.

The virulence of paaA and paaE mutants is diminished in the C. elegans infection model. (A) Kaplan-Meier survival plots for DH26 worms fed with mutant strains STC179-paaA and STC155-paaE. The killing ability of the wild-type B. cenocepacia strain K56-2 (n = 66) was compared with that of STC179-paaA (n = 113; P < 0.0001) or STC155-paaE (n = 67; P < 0.0001) in slow-killing assays using C. elegans strain DH26. solid lines, K56-2; dashed lines, STC155-paaE and STC179-paaA. (B) Appearance of worms after 2 days of bacterial exposure. Worms exposed to the nonpathogenic E. coli OP50 or B. cenocepacia strains were randomly chosen and photographed (magnification, ×80). (C) Wild-type B. cenocepacia K56-2 and the mutant STC155-paaE accumulate to similar levels in the C. elegans intestine. Data represent the mean numbers of CFU per worm from five independent experiments, with error bars signifying standard errors of measurement. gray bars, K56-2; lined bars, STC155-paaE. P values for 8, 24, and 48 h were 0.8766, 0.1666, and 0.5745, respectively.

FIG. 3.

The virulence of the paaZ and paaF mutants is enhanced in comparison with that of B. cenocepacia K56-2 in the C. elegans infection model. Kaplan-Meier survival plots for DH26 worms fed with the STC199-paaZ (A) or STC183-paaF (B) mutant strain are shown. The killing ability of the wild-type B. cenocepacia strain K56-2 (n = 118) was compared with that of STC183-paaZ (n = 99; P < 0.006) or STC199-paaF (n = 114; P < 0.0001) in slow-killing assays using C. elegans strain DH26. solid lines, K56-2; dashed lines, STC183-paaZ or STC199-paaF.

FIG. 4.

Complementation of STC155-paaE with the paaE gene in trans restores full pathogenicity in C. elegans DH26. Kaplan-Meier survival plots for DH26 worms fed with K56-2 (n = 67), STC155-paaE/pAP20 (n = 91; P < 0.0001), or STC155-paaE/pAS1 (n = 78; P = 0.05740) are shown. Squares and solid lines, K56-2; crosses and solid lines, STC155-paaE/pAP20; triangles and dashed lines, STC155-paaE/pAS1.

FIG. 5.

pmk-1 (RNAi) worms are hypersusceptible to B. cenocepacia K56-2 and paaA mutant strains. (A) Kaplan-Meier survival plots for DH26 and pmk-1 (RNAi) worms, fed with B. cenocepacia K56-2 or STC179-paaA. The killing abilities of B. cenocepacia K56-2 (solid lines; P < 0.0001) and STC179-paaA (dashed lines; P < 0.0001) were assayed in killing assays using C. elegans DH26 (triangles) or pmk-1 (RNAi) worms (circles). (B) Appearance of worms exposed to B. cenocepacia K56-2 or STC179-paaA for 2 days. Five to ten worms were chosen randomly, and pictures were taken. Representative pictures are shown at magnification ×80.

FIG. 6.

Kaplan-Meier survival plots for DH26 and pmk-1 (RNAi) worms, fed with B. cenocepacia STC155-paaE or STC155-paaE/pAS1. The killing ability of STC155-paaE (dashed lines) was compared with that of STC155-paaE/pAS1 (solid lines) in slow-killing assays using C. elegans DH26 (triangles) (P < 0.0016) or pmk-1 (RNAi) (circles) (P = 0.4354) worms.