Host associations of Campylobacter jejuni and Campylobacter coli isolates carrying the L-fucose or d-glucose utilization cluster

Campylobacter was considered asaccharolytic, but is now known to carry saccharide metabolization pathways for L-fucose and d-glucose. We hypothesized that these clusters are beneficial for Campylobacter niche adaptation and may help establish human infection. We investigated the distribution of d-glucose and L-fucose clusters among ∼9600 C. jejuni and C. coli genomes of different isolation sources in the Netherlands, the United Kingdom, the United States of America and Finland. The L-fucose utilization cluster was integrated at the same location in all C. jejuni and C. coli genomes, and was flanked by the genes rpoB, rpoC, rspL, repsG and fusA, which are associated with functions in transcription as well as translation and in acquired drug resistance. In contrast, the flanking regions of the d-glucose utilization cluster were variable among the isolates, and integration sites were located within one of the three different 16S23S ribosomal RNA areas of the C. jejuni and C. coli genomes. In addition, we investigated whether acquisition of the L-fucose utilization cluster could be due to horizontal gene transfer between the two species and found three isolates for which this was the case: one C. jejuni isolate carrying a C. coli L-fucose cluster, and two C. coli isolates which carried a C. jejuni L-fucose cluster. Furthermore, L-fucose utilization cluster alignments revealed multiple frameshift mutations, most of which were commonly found in the non-essential genes for L-fucose metabolism, namely, Cj0484 and Cj0489. These findings support our hypothesis that the L-fucose cluster was integrated multiple times across the C. coli/C. jejuni phylogeny. Notably, association analysis using the C. jejuni isolates from the Netherlands showed a significant correlation between human C. jejuni isolates and C. jejuni isolates carrying the L-fucose utilization cluster. This correlation was even stronger when the Dutch isolates were combined with the isolates from the UK, the USA and Finland. No such correlations were observed for C. coli or for the d-glucose cluster for both species. This research provides insight into the spread and host associations of the L-fucose and d-glucose utilization clusters in C. jejuni and C. coli, and the potential benefits in human infection and/or proliferation in humans, conceivably after transmission from any reservoir.