Abstract
This study aimed to investigate the genetic basis of chicken colonisation potential variations in virulence between two Campylobacter jejuni strains. Genomic subtractive hybridisation was developed for C. jejuni and to compare the genetic complement of a poor coloniser, strain NCTC11168, and a good coloniser, strain 81116. Twenty-three DNA fragments were identified that were specific for strain 81116 and absent in strain NCTC11168. Five of these DNA fragments, inserts 31, 50, 183, 186 and 188, were identified as a putative arsenite export gene (arsB), di- and tri-peptide transporter gene (dtpT), transfer gene (traG), arsenate reductase gene (arsC) and a gene encoding a putative integral membrane protein with an unknown function (Cj0033B), respectively. Obtention of this information significantly benefited from the known genome sequence of NCTC11168 (Parkhill et al. , 2000). The recent, albeit preliminary, sequence from C. jejuni strain RM1221 (http://www. tigr. org/) was used to compare the genomic locations of these inserts in strain 81116, as well as hypothesise mechanisms for their absence in strain NCTC11168. The results suggest that the tRNA loci and the occurrence of direct repeats in C. jejuni play key roles in the generation of genetic diversity, and therefore, the evolution of this species. Two of the 81116-specific DNA fragments, inserts 50 and 236, identified as part of a di- and tri-peptide transporter (dtpT) gene and y-glutamyl transpeptidase (ggt) gene, respectively, were further investigated for potential involvement in chicken colonisation. Mutants were generated in strain 81116 for each of these genes and characterised by a number of in vitro and in vivo methods. Using a chicken colonisation model, results suggested that DtpT was required for optimal bacterial colonisation of chicken caeca. This DtpT protein appears to be important in the acquisition of intestinal di- and tripeptides as energy sources, and thus contributing to the survival of the organism in the host gut. GGT was also required for persistent chicken colonisation. Inactivating GGT in strain 81116 appeared to cause a reduction in bacterial intracellular glutathione over time. Once glutathione levels had been significantly reduced, the mutant became sensitive to the toxic metabolites including those seen during host inflammatory responses. Thus, the reduction in colonisation may have been due to depleted intracellular glutathione levels. Overall, this study has confirmed the value of subtractive hybridisation in identifying genetic diversity in C. jejuni. The generation of such genetic variants of C. jejuni may enable adaptation to different environments or as illustrated by phenotypic variation, including in colonisation potential.