Haemorrhagic toxin and lethal toxin fromClostridium sordelliistrain vpi9048: molecular characterization and comparative analysis of substrate specificity of the large clostridial glucosylating toxins

Abstract

SummaryLarge clostridial glucosylating toxins (LCGTs) are produced by toxigenic strains of Clostridium difficile, Clostridium perfringens, Clostridium novyi and Clostridium sordellii. While most C. sordellii strains solely produce lethal toxin (TcsL), C. sordellii strain VPI9048 co-produces both hemorrhagic toxin (TcsH) and TcsL. Here, the sequences of TcsH-9048 and TcsL-9048 are provided, showing that both toxins retain conserved LCGT features and that TcsL and TcsH are highly related to Toxin A (TcdA) and Toxin B (TcdB) from C. difficile strain VPI10463. The substrate profile of the toxins was investigated with recombinant LCGT transferase domains (rN) and a wide panel of small GTPases. rN-TcsH-9048 and rN-TcdA-10463 glucosylated preferably Rho-GTPases but also Ras-GTPases to some extent. In this respect, rN-TcsH-9048 and rN-TcdA-10463 differ from the respective full-length TcsH-9048 and TcdA-10463, which exclusively glucosylate Rho-GTPases. rNTcsL-9048 and full length TcsL-9048 glucosylate both Rho-and Ras-GTPases, whereas rN-TcdB-10463 and full length TcdB-10463 exclusively glucosylate Rho-GTPases. Vero cells treated with full length TcsH-9048 or TcdA-10463 also showed glucosylation of Ras, albeit to a lower extent than of Rho-GTPases. Thus, in vitro analysis of substrate spectra using recombinant transferase domains corresponding to the auto-proteolytically cleaved domains, predicts more precisely the in vivo substrates than the full length toxins. Except for TcdB-1470, all LCGTs evoked increased expression of the small GTPase RhoB, which exhibited cytoprotective activity in cells treated with TcsL isoforms, but pro-apoptotic activity in cells treated with TcdA, TcdB, and TcsH. All LCGTs induced a rapid dephosphorylation of pY118-paxillin and of pS144/141-PAK1/2 prior to actin filament depolymerization indicating that disassembly of focal adhesions is an early event leading to the disorganization of the actin cytoskeleton.

Large clostridial glycosylating toxins modifying small GTPases

“…Reanalysis of the GTPase specificity using recombinantly expressed N-terminal (rN) glucosyltransferase domains reveals that rN-TcdA (and rN-TcsH) also glucosylate Ras-GTPases to some extent in cell-free systems. This finding is relevant, as the majority of cellular Ras is glucosylated upon long-term treatment of cells with TcdA [74,75]. In fulllength TcdA, the Ras recognition site (ranging from amino acids 408-516) seems to be masked by another domain or kept in a conformation, in which Ras recognition and glucosylation are prevented.…”

Foot Infection by Clostridium sordellii: Case Report and Review of 15 Cases in France

“…To further characterize the C. sordellii isolate, termed CS166.08, whole-genome sequencing was performed as previously described (4), as well as for the reference C. sordellii strains IP82 and VPI 9048 (5,6). While this study was in progress, VPI 9048 sequences were made available in GenBank, and the gene sequence of TcsH was described (7)(8)(9). The DNA sequences of the 16S rRNA genes were 99% identical in CS166.08 and the type strain of the species, further supporting that this isolate belongs to the species C. sordellii (Table 1).…”

Large clostridial cytotoxins modifying small GTpases

“…Toxins from other isoforms are specifically indicated. Main substrates of TcdB appear to be RhoA, B and C, Rac, RhoG, Cdc42, and TC10 [30,44]. TcdA modifies RhoA, B and C, Rac, and Cdc42 [45,46].…”

“…Minor substrates are RhoG and Tc10. In addition, the catalytic domain of TcdA also glucosylates Ras proteins (N-/H-/K-Ras), Rap, and Ral to a minor extent [44,46]. It was observed that the in vitro substrate specificity of TcdA is broader when the isolated domain is employed.…”

“…However, as stated previously, C. difficile toxins are processed by the endogenous CPD to release the glycosyltransferase domain into the cytosol. Therefore, it is likely that Ras subfamily proteins are also modified in intact cells [44]. Although differences in the cytotoxic activities of toxins A and B on specific cells and tissues are dramatic (e.g., toxin B is in general 100-fold to 1000-fold more active than toxin A), the enzyme activities (e.g., with RhoA as protein substrate) are more or less the same [36].…”