Intoxication of cultured cells by cholera toxin: evidence for different pathways when bound to ganglioside GM1 or neoganglioproteins

Plasma membrane reorganization: A glycolipid gateway for microbes

Ligand-receptor interactions, which represent the core for cell signaling and internalization processes are largely affected by the spatial configuration of host cell receptors. There is a growing piece of evidence that receptors are not homogeneously distributed within the plasma membrane, but are rather pre-clustered in nanodomains, or clusters are formed upon ligand binding. Pathogens have evolved many strategies to evade the host immune system and to ensure their survival by hijacking plasma membrane receptors that are most often associated with lipid rafts. In this review, we discuss the early stage molecular and physiological events that occur following ligand binding to host cell glycolipids. The ability of various biological ligands (e.g. toxins, lectins, viruses or bacteria) that bind to glycolipids to induce their own uptake into mammalian cells by creating negative membrane curvature and membrane invaginations is explored. We highlight recent trends in understanding nanoscale plasma membrane (re-)organization and present the benefits of using synthetic membrane systems. This article is part of a Special Issue entitled: Nanoscale membrane organisation and signalling.

Glycosphingolipid functions

The combination of carbohydrate and lipid generates unusual molecules in which the two distinctive halves of the glycoconjugate influence the function of each other. Membrane glycolipids can act as primary receptors for carbohydrate binding proteins to mediate transmembrane signaling despite restriction to the outer bilayer leaflet. The extensive heterogeneity of the lipid moiety plays a significant, but still largely unknown, role in glycosphingolipid function. Potential interplay between glycolipids and their fatty acid isoforms, together with their preferential interaction with cholesterol, generates a complex mechanism for the regulation of their function in cellular physiology.

Cholera toxin: A paradigm for multi-functional engagement of cellular mechanisms (Review)

Cholera toxin (Ctx) from Vibrio cholerae and its closely related homologue, heat-labile enterotoxin (Etx) from Escherichia coli have become superb tools for illuminating pathways of cellular trafficking and immune cell function. These bacterial protein toxins should be viewed as conglomerates of highly evolved, multi-functional elements equipped to engage the trafficking and signalling machineries of cells. Ctx and Etx are members of a larger family of A-B toxins of bacterial (and plant) origin that are comprised of structurally and functionally distinct enzymatically active A and receptor-binding B sub-units or domains. Intoxication of mammalian cells by Ctx and Etx involves B pentamer-mediated receptor binding and entry into a vesicular pathway, followed by translocation of the enzymatic A1 domain of the A sub-unit into the target cell cytosol, where covalent modification of intracellular targets leads to activation of adenylate cyclase and a sequence of events culminating in life-threatening diarrhoeal disease. Importantly, Ctx and Etx also have the capacity to induce a wide spectrum of remarkable immunological processes. With respect to the latter, it has been found that these toxins activate signalling pathways that modulate the immune system. This review explores the complexities of the cellular interactions that are engaged by these bacterial protein toxins, and highlights some of the new insights to have recently emerged.

Floating cholera toxin into epithelial cells: functional association with caveolae-like detergent-insoluble membrane microdomains

In polarized cells, signal transduction by cholera toxin (CT) requires apical endocytosis and retrograde transport into Golgi cisternae and likely endoplasmic reticulum (ER) (Lencer et al., J. Cell Biol. 131, 951-962 (1995)). We have recently found that the toxin's apical membrane receptor ganglioside GM1 acts specifically in this signal transduction pathway, likely by coupling CT with caveolae or caveolae-related membrane domains (lipid rafts) (Wolf et al., J. Cell Biol. 141, 917-927 (1998)). Work in progress shows that 1) cholesterol depletion uncouples the CT-GM1 receptor complex from signal transduction, a characteristic of lipid rafts; 2) the GM1 acyl chains rather than the carbohydrate head groups appear to account for the structural basis of ganglioside specificity in toxin trafficking; and 3) intestinal epithelial cells obtained from normal adult humans exhibit lipid rafts which differentiate between CT-GM1 and LTIIb-GD1a complexes and which contain caveolin 1.

Transferrin trojan horses as a rational approach for the biological delivery of therapeutic peptide domains

One novel approach for the biological delivery of peptide drugs is to incorporate the sequence of the peptide into the structure of a natural transport protein, such as human serum transferrin. To examine whether this is feasible, a peptide sequence cleavable by the human immunodeficiency virus type 1 protease (VSQNYPIVL) was inserted into various regions of human serum transferrin, and the resultant proteins were tested for function. Experimentally, molecular modeling was used to identify five candidate insertion sites in surface exposed loops of human serum transferrin that were distant from biologically active domains. These insertions were cloned using polymerase chain reaction mutagenesis, and the proteins were expressed using a baculovirus expression vector system. Analysis of the mutant proteins provided a number of important findings: (a) they retained native human serum transferrin function, (b) the inserted peptide sequence was surface exposed, and most importantly, (c) two of these mutants could be cleaved by human immunodeficiency virus-1 protease. In conclusion, this investigation has validated the use of human serum transferrin as a carrier protein for functional peptide domains introduced into its structure using protein engineering. These findings will be useful for developing a novel class of therapeutic agents for a broad spectrum of diseases.

Membrane traffic and the cellular uptake of cholera toxin

In nature, cholera toxin (CT) and the structurally related E. coli heat labile toxin type I (LTI) must breech the epithelial barrier of the intestine to cause the massive diarrhea seen in cholera. This requires endocytosis of toxin-receptor complexes into the apical endosome, retrograde transport into Golgi cisternae or endoplasmic reticulum (ER), and finally transport of toxin across the cell to its site of action on the basolateral membrane. Targeting into this pathway depends on toxin binding ganglioside GM1 and association with caveolae-like membrane domains. Thus to cause disease, both CT and LTI co-opt the molecular machinery used by the host cell to sort, move, and organize their cellular membranes and substituent components.

Filipin-dependent inhibition of cholera toxin: evidence for toxin internalization and activation through caveolae-like domains

The mechanism by which cholera toxin (CT) is internalized from the plasma membrane before its intracellular reduction and subsequent activation of adenylyl cyclase is not well understood. Ganglioside GM1, the receptor for CT, is predominantly clustered in detergent-insoluble glycolipid rafts and in caveolae, noncoated, cholesterol-rich invaginations on the plasma membrane. In this study, we used filipin, a sterol-binding agent that disrupts caveolae and caveolae-like structures, to explore their role in the internalization and activation of CT in CaCo-2 human intestinal epithelial cells. When toxin internalization was quantified, only 33% of surface-bound toxin was internalized by filipin-treated cells within 1 h compared with 79% in untreated cells. However, CT activation as determined by its reduction to form the A1 peptide and CT activity as measured by cyclic AMP accumulation were inhibited in filipin-treated cells. Another sterol-binding agent, 2-hydroxy-beta-cyclodextrin, gave comparable results. The cationic amphiphilic drug chlorpromazine, an inhibitor of clathrin-dependent, receptor-mediated endocytosis, however, affected neither CT internalization, activation, nor activity in contrast to its inhibitory effects on diphtheria toxin cytotoxicity. As filipin did not inhibit the latter, the two drugs appeared to distinguish between caveolae- and coated pit-mediated processes. In addition to its effects in CaCo-2 cells that express low levels of caveolin, filipin also inhibited CT activity in human epidermoid carcinoma A431 and Jurkat T lymphoma cells that are, respectively, rich in or lack caveolin. Thus, filipin inhibition correlated more closely with alterations in the biochemical characteristics of CT-bound membranes due to the interactions of filipin with cholesterol rather than with the expressed levels of caveolin and caveolar structure. Our results indicated that the internalization and activation of CT was dependent on and mediated through cholesterol- and glycolipid-rich microdomains at the plasma membrane rather than through a specific morphological structure and that these glycolipid microdomains have the necessary components required to mediate endocytosis.

Role of a potential endoplasmic reticulum retention sequence (RDEL) and the Golgi complex in the cytotonic activity of Escherichia coli heat-labile enterotoxin

Recent experimental evidence indicates that Escherichia coli heat-labile enterotoxin and the closely related cholera toxin gain access to intracellular target substrates through a brefeldin A-sensitive pathway that may involve retrograde transport through the Golgi-endoplasmic reticulum network. The A subunits of both toxins possess a carboxy-terminal tetrapeptide sequence (KDEL in cholera toxin and RDEL in the heat-labile enterotoxins) that is known to mediate the retention of eukaryotic proteins in the endoplasmic reticulum. To investigate the potential role of the RDEL sequence in the toxic activity of the heat-labile enterotoxin we constructed mutant analogues of the toxin containing single substitutions (RDGL and RDEV) or a reversed sequence (LEDR). The single substitutions had little effect on Chinese hamster ovary cell elongation or the ability to stimulate cAMP accumulation in Caco-2 cells. Reversal of the sequence reduced the ability of the toxin to increase cAMP levels in Caco-2 cells by approximately 60% and decreased the ability to elicit elongation of Chinese hamster ovary cells. The effects of the heat-labile enterotoxin were not diminished in a mutant Chinese hamster ovary cell line (V.24.1) that belongs to the End4 complementation group and possesses a temperature-sensitive block in secretion that correlates directly with the disappearance of the Golgi stacks. Collectively, these findings suggest that the brefeldin A-sensitive process involved in intoxication by the heat-labile enterotoxin does not involve RDEL-dependent retrograde transport of the A subunit through the Golgi-endoplasmic reticulum complex. The results are more consistent with a model of internalization involving translocation of the A subunit from an endosomal or a trans-Golgi network compartment.

The heat-labile enterotoxin of Escherichia coli binds to polylactosaminoglycan-containing receptors in CaCo-2 human intestinal epithelial cells

The E. coli type I heat-labile enterotoxin (LT-I) shares considerable functional, structural, and immunological homology with cholera toxin (CT). Although the ganglioside GM1 is the sole receptor for CT, LT-I also appears to utilize additional, unique receptors on intestinal cells not recognized by CT. We characterized this second class of LT-I receptors using the human intestinal epithelial cell line, CaCo-2. CaCo-2 cells bound 8-fold more LT-I than CT, and some of these additional LT-I receptors appeared to be functional, as CT-B only partially inhibited LT-I activity at concentrations that completely inhibited CT activity. Membranes from unlabeled or [3H]galactose-labeled cells were incubated with toxin B subunits and extracted with Triton X-100, and the solubilized toxin B-receptor complexes were immunoabsorbed with anti-B bound to protein A-Sepharose. When organic extracts of the complexes were separated by thin-layer chromatography and overlayed with [125I]toxin, both toxins were found to bind only GM1. Separation of the complexes from [3H]galactose-labeled membranes by sodium dodecyl sulfate-polyacrylamide gel electrophoresis revealed a series of galactoproteins specifically recognized by LT-I but not by CT. Similar proteins were detected on Western blots probed with [125I]toxin. LT-I activity on intact cells and binding to membranes and the above galactoproteins were enhanced by neuraminidase treatment even in the presence of CT-B. beta-1,4-Galactosidase and endo-beta-1,4-galactosidase, but not beta-1,3-galactosidase, significantly reduced LT-I binding. LT-I binding to fetuin and transferrin exhibited a similar glycosidase sensitivity.(ABSTRACT TRUNCATED AT 250 WORDS)