A tripartite cytolytic toxin formed by Vibrio cholerae proteins with flagellum-facilitated secretion

Abstract

The protein MakA was discovered as a motility-associated secreted toxin from Vibrio cholerae, Here, we show that MakA is part of a gene cluster encoding four additional proteins: MakB, MakC, MakD and MakE. The MakA, MakB and MakE proteins were readily detected in culture supernatants of wild type V. cholerae whereas secretion was very much reduced from a flagellum deficient mutant. Crystal structures of MakA, MakB and MakE revealed structural relationship to a superfamily of bacterial pore-forming proteins. Cloning and expression of MakA/B/E in Escherichia coli resulted in toxicity of the bacteria towards Caenorhabditis elegans used as a predatory organism model. None of these Mak proteins alone or in pairwise combinations were cytolytic but an equimolar mixture of MakA, MakB and MakE acted as a tripartite cytolytic toxin in vitro causing lysis of erythrocytes and cytotoxicity on cultured human colon carcinoma cells. Formation of oligomeric complexes on liposomes was observed by electron microscopy. Oligomer interaction with membranes was initiated by MakA membrane binding followed by MakB and MakE joining in formation of a pore structure. A predicted membrane insertion domain of MakA was shown by site-directed mutagenesis to be essential for toxicity towards C. elegans. Bioinformatic analyses revealed that the makCDBAE gene cluster is present as a novel genomic island in the vast majority of sequenced genomes of V. cholerae and the fish pathogen V. anguillarum. We suggest that the hitherto unrecognized cytolytic MakA/B/E toxin can contribute to Vibrionaceae fitness and virulence potential in different host environments and organisms.

Protein-lipid interaction at low pH induces oligomerisation of the MakA cytotoxin from Vibrio cholerae

“…In the context of pore-forming toxin systems, the initial insertion of one toxin component was described for XaxAB , YaxAB , and suggested for AhlC . Considering that the TMHs of the MakA-tetramers harbor a lipid bilayer, it could be assumed that MakA initiates membrane insertion in a manner similar to its role in pore formation under neutral pH conditions when part of the tripartite MakA/B/E complex 13 . We hypothesize that under low-pH conditions, MakA transitions from inactive to the active stretched conformation and penetrates the membrane as a monomer.…”

“…Upon recognition and binding to the target cell membrane, the toxins undergo a conformational change, interact with the membrane, dimerize, and oligomerize, leading to the formation of membrane pores 5 . alternative route of secretion, in contrast to MakB and MakE, which displayed a more definitive flagellum-dependent secretion 11,13 .…”

“…The V. cholerae gene makA (vca0883) is localized in a gene cluster together with makC (vca0881), makD (vca0880), makB (vca0882) and makE (vca0884). This gene cluster has been found in different V. cholerae strains, including CT-negative isolates 11,12,13 . The crystal structure of MakA revealed that it belongs to the ClyA α-PFT family 11 , named after the potent, one-component, pore-forming toxin ClyA which is expressed from a monocistronic operon in E. coli 14,15 .…”

“…Other family members of α-PFT are found among the bipartite toxins from Yersinia enterocolitica (YaxAB) 16,17 and Xenorhabdus nematophila (XaxAB) 17 , as well as among the tripartite toxins from Bacillus cereus (NheABC and HblL1L2B) 18 , Aeromonas hydrophila (AhlABC) 19 and Serratia marcescens (SmhABC) 20 . The bipartite and tripartite PFTs require the combined action of all protein partners to induce pore formation in the target membranes, and there is evidence that protein interactions occurs in a specific order for maximum cytolytic activity 19,20,13 . However, there are still questions about how many molecules of each subunit protein are required to form a pore, how they interact with each other, how structural conformational changes occur, and how protein moieties are involved in the interaction with the host membrane.…”

“…In addition, it remains possible that some of the subunit proteins can be separately released from the bacteria and thereby exhibit biological effects on their own. Secretion of the MakA/B/E proteins from V. cholerae was shown to be facilitated via the bacterial flagellum 11,13 . However, about 10% of MakA was secreted from V. cholerae lacking the flagellum suggesting an…”