Adaptive strategies and pathogenesis of Clostridium difficile from in vivo transcriptomics

Abstract

Clostridium difficile is currently the major cause of nosocomial intestinal diseases associated with antibiotic therapy in adults. In order to improve our knowledge of C. difficile-host interactions, we analyzed the genome-wide temporal expression of C. difficile 630 genes during the first 38 h of mouse colonization to identify genes whose expression is modulated in vivo, suggesting that they may play a role in facilitating the colonization process. In the ceca of the C. difficile-monoassociated mice, 549 genes of the C. difficile genome were differentially expressed compared to their expression during in vitro growth, and they were distributed in several functional categories. Overall, our results emphasize the roles of genes involved in host adaptation. Colonization results in a metabolic shift, with genes responsible for the fermentation as well as several other metabolic pathways being regulated inversely to those involved in carbon metabolism. In addition, several genes involved in stress responses, such as ferrous iron uptake or the response to oxidative stress, were regulated in vivo. Interestingly, many genes encoding conserved hypothetical proteins (CHP) were highly and specifically upregulated in vivo. Moreover, genes for all stages of sporulation were quickly induced in vivo, highlighting the observation that sporulation is central to the persistence of C. difficile in the gut and to its ability to spread in the environment. Finally, we inactivated two genes that were differentially expressed in vivo and evaluated the relative colonization fitness of the wild-type and mutant strains in coinfection experiments. We identified a CHP as a putative colonization factor, supporting the suggestion that the in vivo transcriptomic approach can unravel new C. difficile virulence genes.

Fig 1

Venn diagrams showing the distributions of all genes that were up- and downregulated (with P values of <0.05) at the two in vivo infection time points (14 and 38 h) compared to their expression at 8 h.

Fig 2

Functional categories of differentially expressed genes (P < 0.05) in the C. difficile in vivo (monoxenic mice) and in vitro (static TY medium) transcriptomes. The genes significantly upregulated or downregulated during growth in vivo (left) and in vitro (right) relative to their 8-h in vivo values are categorized based on the known or predicted functions of the encoded proteins.

Fig 3

Overview of the fermentation pathway involved in the production of butyrate, butanol, and ethanol. Genes upregulated (in black) or constitutively regulated (in gray) in vivo are underlined. Genes that are not expressed are shown in gray and not underlined. Assignments of enzymes to genes which are regulated in vivo are as follows: CD3258 and CD3405 to CD3407, iron hydrogenases; CD2427 to CD2429 and CD2380-CD2381, ferredoxin oxidoreductases; CD2966 and CD0334, aldehyde-alcohol dehydrogenases; CD1059, acetyl-CoA acetyltransferase; CD1058, 3-hydroxybutyryl-CoA-dehydrogenase; CD1057, 3-hydroxybutyryl-CoA dehydratase; CD1054, butyryl-CoA dehydrogenase; CD1055-CD1056, electron transfer flavoproteins; CD0116 to CD0118, ferredoxin oxidoreductases; CD0715 and CD0112, phosphate butyryltransferases; CD2379, butyrate kinase; CD2677 and CD2678, succinyl CoA:3-oxoacid CoA transferase subunits A and B.

Fig 4

C. difficile sporulation gene expression in vivo. The schematic representation shows the main stages of sporulation in C. difficile. (1) Predivisional cell; (2) asymmetric division; (3) intermediate stage in the process of engulfment of the forespore (the future spore) by the larger mother cell; (4) completion of engulfment; (5) synthesis of the spore surface layers (cortex peptidoglycan, coats, and exosporium); (6) free, mature spore, which resumes growth upon germination. The sporulation genes found to be upregulated in our study are indicated in boxes for each stage of the process, with the presumed cellular compartment where their expression takes place. The genes found to be upregulated in vitro are shown in gray, whereas those upregulated both in vivo and in vitro are shown in black. The genes have been placed along the morphological sequence according to their time of expression following infection of mice. The black dots mark genes that belong to the sporulation core machinery (50).

Fig 5

Competitive colonization assays between the parental wild-type strain and the 630 Δerm::CD1581 and 630 Δerm::CD3145 mutant strains. Mice were orally challenged with 5 × 10⁵ CFU of both the WT and mutant strains. The number of bacteria in the feces was determined at 38 h postinfection. Eight mice were used for each competition experiment. Statistically significant differences compared to the wild type are indicated by asterisks (**, P < 0.005).