Standing genetic variation in contingency loci drives the rapid adaptation of Campylobacter jejuni to a novel host

Abstract

The genome of the food-borne pathogen Campylobacter jejuni contains multiple highly mutable sites, or contingency loci. It has been suggested that standing variation at these loci is a mechanism for rapid adaptation to a novel environment, but this phenomenon has not been shown experimentally. In previous work we showed that the virulence of C. jejuni NCTC11168 increased after serial passage through a C57BL/6 IL-10(-/-) mouse model of campylobacteriosis. Here we sought to determine the genetic basis of this adaptation during passage. Re-sequencing of the 1.64 Mb genome to 200-500 X coverage allowed us to define variation in 23 contingency loci to an unprecedented depth both before and after in vivo adaptation. Mutations in the mouse-adapted C. jejuni were largely restricted to the homopolymeric tracts of thirteen contingency loci. These changes cause significant alterations in open reading frames of genes in surface structure biosynthesis loci and in genes with only putative functions. Several loci with open reading frame changes also had altered transcript abundance. The increase in specific phases of contingency loci during in vivo passage of C. jejuni, coupled with the observed virulence increase and the lack of other types of genetic changes, is the first experimental evidence that these variable regions play a significant role in C. jejuni adaptation and virulence in a novel host.

Figure 1. Microevolution of contingency loci.

For each contingency gene, the percentage of sequence reads with a particular base count in the homopolymeric tract is graphed. The distribution of homopolymeric variations before (wild-type C. jejuni NCTC11168), and after (mouse-adapted C. jejuni NCTC11168) passage, for each variable gene in the genome is shown. Cj0628 contains 2 variable tracts, but is not pictured. The distribution of variation at this tract was stable through passage. * p≤0.01 and ** p≤0.001 by chi-square test of distribution for open reading frame variations.

Figure 2. Absence of large genomic changes during passage.

(A) Pulse field gel electrophoresis image to compare wild-type and mouse-adapted bacteria. (B) The same gel with an image filter as an attempt to detect rare bands.

Figure 3. In vitro phenotypes of mouse-adapted C. jejuni.

(A) Soft agar plates allow motile C. jejuni to spread from the center of inoculation. The ability to spread is based on flagellar motility. The top right spot is the wild-type variant and shown going clockwise are mouse-adapted variants passaged one, two, or three times through mice. All have spread an equal amount after 48 hours. (B) A higher final OD₆₀₀ for the mouse-adapted variant indicates a decreased ability to autoagglutinate. Standard error bars are shown.