Characterization of genetically matched isolates of Campylobacter jejuni reveals that mutations in genes involved in flagellar biosynthesis alter the organism's virulence potential

Abstract

Phenotypic and genotypic evidence suggests that not all Campylobacter jejuni isolates are pathogenic for humans. We hypothesized that differences in gene content or gene expression alter the degree of pathogenicity of C. jejuni isolates. A C. jejuni isolate (Turkey) recovered from a turkey and a second C. jejuni isolate (CS) recovered from a chicken differed in their degrees of in vitro and in vivo virulence. The C. jejuni Turkey isolate invaded INT 407 human epithelial cells and secreted the Cia (Campylobacter invasion antigen) proteins, while the C. jejuni CS isolate was noninvasive for human epithelial cells and did not secrete the Cia proteins. Newborn piglets inoculated with the C. jejuni Turkey isolate developed more severe clinical signs of campylobacteriosis than piglets inoculated with the C. jejuni CS isolate. Additional work revealed that flagellin was not expressed in the C. jejuni CS isolate. Microarray and real-time reverse transcription-PCR analyses revealed that all flagellar class II genes were significantly downregulated in the C. jejuni CS isolate compared to the C. jejuni Turkey isolate. Finally, nucleotide sequencing of the flgR gene revealed the presence of a single residue that was different in the FlgR proteins of the C. jejuni Turkey and CS isolates. Complementation of the C. jejuni CS isolate with a wild-type copy of the flgR gene restored the isolate's motility. Collectively, these findings support the hypothesis that critical differences in gene content or gene expression can alter the pathogenic potential of C. jejuni isolates.

FIG. 1.

Some C. jejuni isolates recovered from poultry do not secrete the Cia proteins. C. jejuni isolates 81-176 (lane 1), CS (lane 2), S1 (lane 3), S2B (lane 4), S3 (lane 5), SC (lane 6), and Turkey (lane 7) were incubated in MEM lacking methionine and containing 1% FBS and labeled with [³⁵S]methionine for 3 h. The supernatant fluids were analyzed by SDS-PAGE and autoradiography as outlined in Materials and Methods. Molecular mass standards, in kDa, are indicated on the left.

FIG. 2.

MRP-PFGE profiles of C. jejuni isolates. Lanes 1 (A and B), PFGE lambda ladder (New England Biolabs, Beverly, MA). The sizes of the molecular mass markers are shown on the left of each panel in kilobase pairs. (A) MRP-PFGE of SmaI-digested DNA from 81-176 (lane 2), CS (lane 3), S1 (lane 4), S2B (lane 5), S3 (lane 6), SC (lane 7), and Turkey (lane 8). (B) MRP-PFGE of SalI-digested DNA from 81-176 (lane 2), Turkey (lane 3), and CS (lane 4).

FIG. 3.

Electrophoretic and immunoblot analysis of C. jejuni Turkey (T) and CS whole-cell lysates (WCL), OMP, and flagellin preparations. Proteins were separated in SDS-12.5% polyacrylamide gels and either stained with CBB-R250 (A) or transferred to PVDF membranes and reacted a 1:250 dilution of a C. jejuni antiflagellin serum that contains antibodies reactive with both the FlaA and FlaB proteins (B). The arrows indicate the FlaA flagellar protein. Molecular mass standards, in kDa, are indicated on the left.

FIG. 4.

(A) Transcript levels of σ⁵⁴-regulated genes in the C. jejuni Turkey isolate are increased relative to those of the C. jejuni CS isolate. The flaB, flaD, flgB, flgD, flgE, flgG, flgH, and flgK transcripts were measured by real-time RT-PCR using total RNA extracted from the C. jejuni CS and Turkey isolates grown to mid-log phase in MH broth. The change in transcript levels was measured using the comparative C_T method, where glnA was used as the internal control and ΔC_T of the C. jejuni CS isolate was used as the calibrator. (B) The flaB gene, which is σ⁵⁴ regulated, is not expressed in the C. jejuni CS isolate. β-Galactosidase activity from the C. jejuni CS and F38011 isolates harboring the PmetK-pMW10 and PflaB-pMW10 constructs was measured. The bacteria were grown in MH broth for 16 h. The values are averages of three independent experiments, with the error bars representing the standard deviations.

FIG. 5.

(A) Multiple sequence alignment of the flgR gene from C. jejuni 11168, Turkey, and CS isolates. Shown is a segment of the nucleotide sequence of the flgR gene and its predicted translation from C. jejuni strains NCTC 11168 (11168), Turkey (Turkey), and CS (CS). The FlgR receiver and σ⁵⁴-interacting domains are indicated. Also highlighted (underlined) is the residue that differed in the C. jejuni CS isolate compared to the Turkey isolate. (B) Complementation of the C. jejuni CS isolate with a wild-type copy of flgR gene restores the isolate's motility. Both the C. jejuni CS wild-type and CS flgR pRY107-transformed isolates were spotted onto MH plates supplemented with 0.4% agar, and motility was assessed after 48 h of incubation.

FIG. 6.

Growth of the C. jejuni Turkey and CS isolates in MH broth. MH broth was inoculated with the C. jejuni Turkey and CS isolates to an OD₅₄₀ of 0.01, and the cultures were incubated at 37°C and 150 rpm, under microaerobic conditions. The points represent the time points at which the optical density of each bacterial culture was determined at a wavelength of 540 nm. Circles, C. jejuni CS isolate; squares, C. jejuni Turkey isolate.

FIG. 7.

The C. jejuni CS isolate and the C. jejuni Turkey flgR mutant display greater resistance to sodium deoxycholate (DOC) than the C. jejuni Turkey isolate. The C. jejuni CS isolate (open bars), C. jejuni Turkey isolate (gray bars), and C. jejuni Turkey flgR mutant (black bars) were incubated in MH broth with 0.1%, 0.05%, 0.025%, and 0% sodium deoxycholate under microaerobic conditions for 48 h at 37°C with shaking. The y axis represents the optical density of bacterial culture relative to the growth achieved without deoxycholate (0% sodium deoxycholate). The error bars represent the standard deviations.

FIG. 8.

(A) The rpoN operon of the C. jejuni S2B isolate was sequenced, and a single thymine base insertion was identified between bp 1004 and 1005. This base insertion results in a termination codon at bp 1011 as opposed to the normal termination codon at bp 1248. Shown is a segment of the nucleotide sequence of the rpoN gene and its predicted translation from the C. jejuni NCTC 11168 (11168) and S2B (S2B) isolates. Also shown (bottom) is the number of bp from the start codon of rpoN. *, termination codon. (B) Complementation of the C. jejuni S2B strain with a wild-type copy of rpoN restored the motility phenotype. The C. jejuni S2B wild-type and S2B rpoN pRY111-transformed isolates were spotted onto MH plates supplemented with 0.4% agar, and motility was assessed after 48 h of incubation.