Funding provided by the Department for Environment, Food and Rural Affairs (DEFRA) and the Food Standards Agency (FSA) has enabled sentinel surveillance of human campylobacteriosis isolates obtained from individuals in Oxfordshire, UK, between 27th June 2011 and 30th June, 2014 using whole genome sequencing (WGS) technology. This continues seven locus multi-locus sequence type (MLST) sequencing that had been in place since 2003.
Methods
Molecular characterisation of isolates obtained from the John Radcliffe Hospital, Oxfordshire, UK, was achieved by WGS, which was interpreted with a gene-by-gene approach. Sequences for 1,802 defined loci were automatically identified by the Bacterial Isolate Genome Sequence database (BIGSdb) software (Jolley and Maiden, 2010. BMC Bioinformatics 11:595), providing summary reports by established (MLST, Dingle et al., 2001. J Clin Microbiol 39(1):14-23; Dingle et al., 2005. J Clin Microbiol 43(1):340-7) and novel methods (ribosomal sequence typing, rMLST, Jolley et al., 2012. Microbiology 158(4):1005-15; core genome MLST, cgMLST, Maiden et al., 2013. Nat Rev Microbiol 11(10):728-36). Ciprofloxacin resistance was determined from the automatically assigned gyrA gene sequence for each isolate (Wang et al., 1993. Antimicrob Agents Chemother 37(3):457-63).
Findings
Complete MLST data were obtained for 2,549 unique isolates, 2,267 (88.9%) of which were C. jejuni and 282 (11.1%) C. coli. The distribution of the 33 clonal complexes identified did not change significantly over the three-year study period although comparison of longer term trends from 2003 and 2013 indicated a decline in isolates belonging to three clonal complexes and highlighted a sustained increase of isolates belonging to a fourth. All four of these clonal complexes are chicken-associated and may reflect changes in the dynamics of these genotypes in poultry flocks.
The number of human campylobacteriosis cases in Oxfordshire did not reduce significantly over the study period, nor since DEFRA-funded Campylobacter surveillance in Oxfordshire began in 2003.
Attribution of human isolates to their potential source, utilising MLST data, demonstrated that farm animals remain the predominant source of human campylobacteriosis. The number of isolates attributable to chicken, cattle, sheep or pigs did not change significantly over the three-year study, and this finding was confirmed by MLST data from C. jejuni isolates collected since 2003. Although chicken-associated C. jejuni isolates remain the most important single source of disease, the contribution of this source may be under estimated, as disease isolates attributed to sheep are partly represented by a genotype found in several hosts and therefore could be of poultry origin.
There has been an increase of 36.8% in the number of ciprofloxacin resistant isolates causing human disease since 2003. This most recent study noted an increase from 38.1% in 2011-2012 to 43.8% in 2013-2014 and identified four chicken-associated clonal complexes that were significantly associated with ciprofloxacin resistance.
Conclusions
This project has contributed to the development of a world-leading, cost-effective, WGS analysis methodology that has produced data for over 2,500 human campylobacteriosis isolates which are available on-line with open access. No variation was detected in either the incidence or source of human campylobacteriosis, indicating the need for further evidence based intervention strategies aimed at combating this disease. As WGS Campylobacter reference populations from sources of human infection become available it will be possible to develop and apply WGS based attribution to refine the estimates available from MLST data. Levels of ciprofloxacin-resistant isolates have continued to increase with particularly high levels among several chicken associated clonal complexes. These data are publicly available and can be accessed by members of the wider Campylobacter research community (http://pubmlst.org/campylobacter/).