Bacteriophages infecting the food-borne human pathogen Campylobacter jejuni could potentially be exploited to reduce bacterial counts in poultry prior to slaughter. This bacterium colonizes the intestinal tract of poultry in high numbers, and contaminated poultry meat is regarded as the major source of human campylobacteriosis. In this study, we used phage F336 belonging to the Myoviridae family to select a C. jejuni NCTC11168 phage-resistant strain, called 11168R, with the aim of investigating the mechanisms of phage resistance. We found that phage F336 has reduced adsorption to 11168R, thus indicating that the receptor is altered. While proteinase K-treated C. jejuni cells did not affect adsorption, periodate treatment resulted in reduced adsorption, suggesting that the phage binds to a carbohydrate moiety. Using high-resolution magic angle spinning nuclear magnetic resonance (NMR) spectroscopy, we found that 11168R lacks an O-methyl phosphoramidate (MeOPN) moiety attached to the GalfNAc on the capsular polysaccharide (CPS), which was further confirmed by mass spectroscopy. Sequence analysis of 11168R showed that the potentially hypervariable gene cj1421, which encodes the GalfNAc MeOPN transferase, contains a tract of 10 Gs, resulting in a nonfunctional gene product. However, when 11168R reverted back to phage sensitive, cj1421 contained 9 Gs, and the GalfNAc MeOPN was regained in this strain. In summary, we have identified the phase-variable MeOPN moiety, a common component of the diverse capsular polysaccharides of C. jejuni, as a novel receptor of phages infecting this bacterium.

The pH of the human stomach is dynamic and changes over time, depending on the composition of the food ingested and a number of host-related factors such as age. To evaluate the number of bacteria surviving the gastric acid barrier, we have developed a simple gastric acid model, in which we mimicked the dynamic pH changes in the human stomach. In the present study, model gastric fluid was set up to imitate pH dynamics in the stomachs of young and elderly people after ingestion of a standard meal. To model a serious foodborne pathogen, we followed the survival of Salmonella enterica serotype Dublin, and found that the addition of proteins such as pepsin, ovalbumin, and blended turkey meat to the simple gastric acid model significantly delayed pathogen inactivation compared with the control, for which no proteins were added. In contrast, no delay in inactivation was observed in the presence of bovine serum albumin, indicating that protection could be protein specific. The simple gastric acid model was validated against a more laborious and complex fermenter model, and similar survival of Salmonella Dublin was observed in both models. Our gastric acid model allowed us to evaluate the influence of food components on survival of pathogens under gastric conditions, and the model could contribute to a broader understanding of the impact of specific food components on the inactivation of pathogens during gastric passage.