Genomic and phenotypic characterization of a Clostridioides difficile strain of the epidemic ST37 type from China

Abstract

Clostridioides difficile strains of sequence type (ST) 37, primarily including PCR ribotype (RT) 017, are prevalent in mainland China. Our study aimed to compare the major virulence factors of an epidemic C. difficile isolate of ST37 type (Xy06) from China with the well-characterized C. difficile reference strains R20291 (RT027) and CD630E (ST54), as well as a Chinese ST54 strain (Xy07) isolated from the same hospital. The Xy06 genome was predicted to harbor two complete prophages and several transposon-like elements. Comparative analysis of PaLoc revealed a truncated tcdA gene, a functional tcdB gene, a functional tcdC gene, and well-conserved tcdR and tcdE genes. Phenotypic comparisons showed that Xy06 was a robust producer of TcdB, readily sporulated and germinated, and strongly bound to human gut epithelial cells. In a mouse model of C. difficile infection, Xy06 was more virulent than strains CD630E and Xy07 and was comparable to strain R20291 in virulence. Our data suggest the potential threat of the epidemic ST37 strains in China.

Keywords: Clostridioides difficile; RT017; ST37; pathogenicity locus (PaLoc); toxin; virulence.

Figure 1

Phylogenetic tree of A-B+ C. difficile strains. A minimum evolution phylogenetic tree was generated using the Type (Strain) Genome Server. Xy06 is indicated in red. The scale bar indicates 0.01 nucleotide substitutions per site. Dotted lines separate select branches for easier viewing and do not indicate evolutionary distance.

Figure 2

Comparative sequence analysis of PaLoc region. C. difficile typically encodes the genes cdu2 (pink) and cdu1 (light blue) upstream of PaLoc and cdd2 (pink), cdd3 (maroon), and cdd4 (orange) downstream of PaLoc, so these were used as boundaries for visualizing the genome below. PaLoc-adjacent genes were visualized for strains lacking the above surrounding genes (e.g., 173070). A-B+ C. difficile strains display an atypical organization of their PaLoc region and surrounding genes, such as deletions in tcdA (e.g., Xy06) or multiple genes (e.g., 173070, ES130, J9965). Blue BLAST alignments indicate regions with relatively lower (64%) shared identity, while red regions indicate high similarity.

Figure 3

Phylogenetic tree of A-B+ C. difficile TcdB subtypes. A neighbor-joining phylogenetic tree of TcdB amino acid sequences from the A-B+ strains examined in this study and other strains representing the various subtypes of TcdB. Xy06 TcdB is clustered with several A-B+ strains that encode TcdB3 (light blue). The scale bar indicates 0.020 amino acid substitutions per site. Bootstrap values of 50 or greater are displayed.

Figure 4

Toxin production in strains Xy06 and Xy07. TcdA (A) and TcdB (B) concentrations of C. difficile culture supernatants at different time points. (C) TcdB concentrations of C. difficile culture supernatants were compared at the same time point. *P < 0.05 (t-test). (D) Cytotoxicity of C. difficile culture supernatants. C. difficile culture supernatants (16 μl) collected at 72 hours post-inoculation in TY medium were added to CT26 cells seeded in 96-well plates. CT26 cell rounding was visualized by phase-contrast microscopy (40X) at 20 hours post-treatment. (E) Percentage of round cells at different time points (****P<0.0001, one-way ANOVA). Data are presented as mean ± SEM (n = triplicate). ns- not significant.

Figure 5

Adhesion of C. difficile spores to HCT-8 cells. HCT-8 cells were incubated with spores for 100 minutes under anaerobic conditions, and the relative number of adherent spores was calculated. Data are from three biological replicates and reported as mean ± SEM. **p=0.0097 (Xy06 VS R20291); p=0.8011 (Xy06 VS CD630E); *** p<0.0001 (R20291 VS Xy07).

Figure 6

Sporulation and germination rates of C. difficile strains. (A) Sporulation. *P<0.05; **P<0.01. (B) Germination: Purified C. difficile spores were suspended in buffer supplemented with taurocholic acid and glycine to induce germination. Germination was monitored by plotting the ratio of the OD₆₀₀ at a given time to the OD₆₀₀ at time zero. Data are from three biological replicates and reported as mean ± SEM. ****p=0.0001(CD630E vs Xy06); ***p=0.0021 (Xy07 vs Xy06); ns: p=0.0714 (R20291 vs Xy06). (C) Spore colony-forming efficiency. Spore suspensions were serially diluted and plated on BHIS agar supplemented with 0.1%(w/v) taurocholic acid (TA). The spore colony-forming efficiency was defined as the percentage of total spores that gave rise to colonies on BHIS agar with 0.1% (w/v) TA, calculated by c.f.u. per ml colony count/c.f.u. per ml direct count by microscopy × 100%. Data are from three biological replicates and reported as mean ± SEM. * p=0.0276 (CD630E vs Xy06); ns: no significance (R20291, Xy07 vs Xy06). ns- not significant.

Figure 7

Virulence of Xy06 in the mouse model of CDI. Four groups of mice (n = 20) were challenged with 10⁶ spores of Xy06, Xy07, CD630E and R20291, respectively. Mice were monitored for one week for weight changes (A), survival (B), and diarrhea (C). Animal survival was analyzed by Kaplan–Meier survival analysis with a log-rank test of significance. The mean relative weight of mice was analyzed for significance using a Student’s t-test. Xy06 challenged group lost significantly more weight at day 2 post-infection compared to CD630E challenged group (p=0.0145). R20291 challenged group lost significantly more weight at days 2 and 3 post-infection than Xy06 challenged group (p=0.0015 and p=0.0076, respectively) (A). (*p < 0.05, **p < 0.01 and **p < 0.01). ns- not significant.