Table 1.
Bacterial strains/ plasmids used in this study.
Fig 1.
GDH is required for C. difficile virulence.
Kaplan-Meier survival curve of clindamycin-treated Syrian hamsters inoculated with 2,000 vegetative cells of C. difficile JIR8094 (Parent) or C. difficile JIR8094::gluD (mutant). Animals (n = 7 per group) were monitored every four hours for the symptoms of wet tail, poor fur coat, lethargy, or hunched posture. Moribund animals were euthanized. Log rank statistical analysis was performed; p <0.0001.
Fig 2.
C. difficile JIR8094::gluD mutant does not colonize or induce inflammation in hamsters.
A. Representative colonic histologic images (hematoxylin and eosin (H&E) staining). Cecal tissues from parental strain-infected hamsters were harvested at the time of sacrifice. Cecal tissues from surviving gluD mutant-infected (and uninfected) hamsters were harvested 15 days post-infection. B. Histology scores were evaluated as described in the Materials and Methods. C. C. difficile colonization levels for each of the two groups in CFU per gram of cecal content at the time of necropsy. In three of seven gluD mutant-infected animals, C. difficile was not detected and are not represented in the figure (n = 7 per group). Error bars represent SD.
Fig 3.
Complementation of gluD mutant with various gluD constructs.
The gluD homologues from closely related bacterial species were expressed in the C. difficile gluD mutant strain, and their secretion from C. difficile was analyzed. Cytosolic (cyt) and concentrated supernatants (sup) from the bacterial cultures expressing various gluD constructs were separated by SDS-PAGE, and were analyzed by Coomassie staining (A) and by zymogram (B). C. difficile gluD constructs with deletions of their N-terminus (panels C, D, and E) or of C-terminus (panels F, G and H) were expressed in C. difficile gluD mutant, and their secretion from C. difficile was analyzed by zymogram (D&G) and ELISA (E&H).
Fig 4.
Extracellular GDH enables rapid progression of C. difficile infection in hamsters.
C. difficile gluD mutant complemented with secretable C. difficile GDH or with nonsecretable C. sordellii GDH were used to infect the clindamycin-treated hamsters. Survival rate was plotted using Kaplan-Meier survival curve. Comparisons of C. difficile GDH-WT vs. C. sordellii GDH survival curves were made using long rank test; p = 0.035.
Fig 5.
C. difficile gluD mutant complemented with secretable forms of GDH colonized better and induced more inflammation than the mutant expressing nonsecretable GDH.
A. Representative image of H&E stained colonic specimens. B. Histology scores evaluated as described in the Materials and Methods section. C. C. difficile colonization levels for each of the groups (CFU per gram of cecal content at the time of necropsy). Unpaired t test was performed for statistical analysis (n = 7 per group). Error bars represent SD.
Fig 6.
Colonization of hamster gut by C. difficile gluD mutant in the presence of parental C. difficile strain.
Hamsters were gavaged with a bacterial mixture containing 1000 parent and 1000 gluD mutant cells. Bacterial load of each strain at the time of necropsy was measured and presented as CFU per gram of cecal content. Unpaired t test was performed for statistical analysis (n = 7 per group). Error bars represent SD.
Table 2.
Amino acid analyses of hamster cecal contents.