Cholera and the Cholera Toxin

by Elizabeth Cronenwett

Introduction to Cholera
History of Cholera
History of the Study of Cholera
Action of Cholera
The Cholera Toxin
Assembly and Secretion
Primary Structure
Subunit A
Subunit B
Receptor Binding
Binding Kinetics
Action of Toxin
Entry of the Toxin into the Cytosol
Uses of Cholera Toxin
Treatment for Cholera


Cholera is an intestinal disease common in areas lacking running water and proper sanitation. The causative organism, Vibrio cholerae, colonizes the gut and secretes a toxin which causes massive fluid and electrolyte loss through diarrhea. Cholera is not spread directly from person to person but can be contracted through drinking contaminated water, eating raw or undercooked shellfish from contaminated areas, or eating salads and raw fruits and vegetables washed in contaminated water. The re are simple and effective treatments, but current vaccines have limited protection and a short lifetime.

Introduction to Cholera

Cholera caused as much horror and revulsion in Europe as the plague did before it. It struck without warning, shriveled its victims and could kill relatively quickly after symptoms started. (3) It is still a problem in de veloping nations, and after natural disasters when the water supply is compromised. Initial symptoms occur after a 1 to 3 day incubation period and mainly include abrupt, painless, watery diarrhea. (32) Diarrhea volume often exceeds six liters per hour. (21) Resultant water and electrolyte loss leads to thirst, muscle cramps, weakness, loss of tissue turgor, sunken eyes and wrinkled skin. (32) (33) If untreated, severe metabolic acidosis with potassium depletion, anuria, circulatory collapse and cyanosis can occur, leading to death. (32) (33) Cholera can range from mild, with less than 1% mortality, to severe, with greater than 50% mortality. (3)

History of Cholera

Cholera is of Asian or Indian origin. In India it was endemic, and caused epidemics following each pilgrimage to the Ganges River. During the 18th and 19th centuries it spread slowly north and east by land. Sailors and colonists picked it up and carried it rapidly by ship. (3) It reached Moscow in 1830, causing panic as inhabitants fled the city. It then spread by land through the Russian states, Poland, and into Western Europe. Tens of thousands were killed in the de cade after it reached Europe. Spreading by ship, it reached North America in 1832, appearing first in New York and Philadelphia, then spreading north and south. While cholera can be endemic in the tropics, the epidemics can spread anywhere; 2200 people died in 1832 in Quebec. (34) Several ripples of epidemics then spread outward from India; in 1848 and 1849 over a million people in Russia and 150,000 in France died of cholera. (31) By the end of the nineteenth century however, Europe and North America were free of the disease. The solution was simple; filtration and chlorination of the water supply. (34) (31)

History of the Study of Cholera

A localized London epidemic in 1854 caused John Snow, then the Queen’s physician, to realize that all cases could be traced to a single contaminated well. As the story goes, he removed the pump handle, stopping the epidemic, and completing one of the first epidemiological studies. The causative organism, V. cholerae, was described in 1854 and isolated in pure culture 30 years later. It was not until 1959 in Calcutta, that a cell-free culture filtrate was shown to be capable of produci ng the massive accumulation of water and electrolytes in the intestine. In the same year, 1959, researchers in Bombay described the production of diarrhea in infant rabbits by a crude protein isolate from V. cholerae culture filtrate. The pandemi cs of the 1960’s and early 1970’s in Southeast Asia rekindled an interest in the study of cholera. A protein secreted by the vibrio was isolated and purified. Two factors were identified; the active one, denoted choleragen, is actually the AB5 form of the toxin, while the inactive one is the B5 oligomer, known as choleragenoid. (3)

Action of Cholera

The causative organism, Vibrio cholerae, adheres to the brush border of villous absorptive cells. Adhesion is mediated by bacterial filaments that recognize carbohydrate receptors on microvillous membranes. Variations in specific typ es of filaments and their receptors account for host, site and age (neonate or adult) specificity. V. cholerae causes hypersecretion of electrolytes and water with little or no morphologic damage to the epithelium. The intestinal abnormalities ca using diarrhea are mainly regulatory, not structural. The toxin secreted by V. cholerae acts to alter regulation of secretion and absorption, with the net result being water and electrolyte loss through the intestines. (15)

The Cholera Toxin

The cholera toxin (CT) produced by V. cholerae binds to ganglioside GM1 receptors. This, through an unknown mechanism, causes the enzymatic subunit to be clipped from the rest of the protein, and transferred into the cell. Once inside the cell, the subunit causes an increase in cellular cAMP, leading to overactivity of luminal sodium pumps. The electrolytes in the gut cause water to leave the cells, and produce the characteristic diarrhea.


Cholera vibrios produce a number of virulence factors in addition to the toxin, such as pili, a polar sheathed flagellum, and several hemagglutinins that could serve as adhesins. (25) These virulence factors are expressed in response to specific environmental conditions. Maximal expression of virulence factors including the toxin is observed when cells are grown in L broth at pH 6.5, whereas such expression is undetectable when cells are grown in L broth at pH 8.5. (28)

The expression of more than 17 virulence genes in V. cholerae is under the coordinate control of the ToxR protein. ToxR is a transmembrane protein that binds to and activates the promoter of the operon encoding cholera toxin. ToxR controls transcription of toxT, whose product in turn is directly responsible for activation of several virulence genes under ToxR control. (28) This includes the ctxAB gene encoding the cholera toxin. (12)

Assembly and Secretion

The cholera toxin (CT) is an oligomeric protein toxin, which is secreted across the bacterial outer membrane into the extracellular environment. V. cholerae normally secretes both the holotoxin and pentameric B subunit alone. The obser vation that A is not secreted in the absence of B implies that although the B oligomer can be secreted independently of A, assembly into holotoxin may be required for the translocation of A. The secretion of enterotoxin subunits from the periplasmic com partment of V. cholerae was found to be specific and was not accompanied by the release or leakage of other periplasmic proteins. (3)

Structural Overview

The cholera toxin (CT) secreted by V. cholerae is composed of five identical B subunits and an A subunit. The five B subunits form a pentagonal structure with a central hole (the "donut"). The A subunit is composed of two par ts. Subunit A2 is a long alpha-helix which is anchored within the central hole of the B5 unit. Subunit A1 is, in the final secreted form, attached to A2 by a single disulfide bond. The B pentamer is the receptor-binding portion of the molecu le, while A1, once cleaved, is the enzymatically active portion. A2 appears merely to mediate the connection between the two.

Primary Structure

The cholera toxin is a substantial protein, with a molecular mass of 85 kDa, and a total of 755 amino acids. The A subunit has a molecular mass of 27,234 Da. The monomeric B subunit has a molecular mass of 11,677 Da, giving the pentamer a mas s of 58,387 Da. The total molecular mass of the holotoxin, AB5, is therefore 85,620 Da. (3) The A subunit is translated as a single 240 amino acid protein that is nicked by a bacterial endoprotease between either residue 193 and 195 to form two chains, A1 and A2. (21) This cleavage occurs in the vibrio, but is not necessary for secretion. (3) The two chains are held together by a disulfide bond between Cys187 and Cys199. (21) The monomeric B subunit has 103 amino acids. (27) Each B subunit has a single disulfide bridge from Cys9 to Cys86. (22)

Subunit A

The A subunit of cholera toxin is translated as a single 240 amino acid protein that is nicked by a bacterial endoprotease to form two chains. These two chains remain held together by extensive non-covalent forces and the single disulfide bond fro m Cys187 to Cys199. (21) The A1 chain is a wedge shaped unit which becomes the active enzyme when separated from the connecting A2 chain. (3)

The A1 subunit has three distinct regions. The first 132 amino acids form a compact globular unit (A11) composed of a mixture of alpha-helices and beta strands. The amino-terminal nitrogen forms a hydrogen bond with the backbone carbon yl oxygen of residue 153, stabilizing a helix-strand transition. The substructure from residue 133 to 161 (A12) forms an extended bridge between the compact A11 and A13 domains. It seems to be simply a molecular tether. The chain is quite flexible, and disordered in the crystal structure. A third globular substructure (A13) is formed from the carboxy-terminal 31 residues that surround the disulfide bridge linking the A1 and A2 fragments. The A13 substructure has a high density of hydrophobic residues including a cluster of four prolines and two tryptophan. The A11 and A12 interface is almost entirely non-polar with only a few polar interactions occurring at the molecular su rface. Thus it is conceivable that at the lipid water interface the A1 subunit undergoes a substantial rearrangement that permits the A11 substructure to make contact with and/or pass through the bilayer. (21)

There are few direct interactions between the A1 chain and the B pentamer. The exceptions include three arginine side-chains (Arg33, Arg143, and Arg148) located along the bottom surface of the A1 subunit that form multiple hydrogen bonds with carb onyl and glutamyl oxygen atoms located along the top of the pentamer’s central alpha-helices. Since these contacts are all located adjacent to the axis of the pore, substantial rotation of the A subunit with regard to the B pentamer is possible without d isruption of the interface. (21)

The A2 chain is an extended alpha helix which begins at residue 196 of the A peptide. (3) The alpha-helical structure is broken only by a central 52 degree kink that is stabilized by a hydrogen bond between the gamma-oxygen of Ser228 and the peptide nitrogen of Asp229. This kink redirects the helix prior to its decent into the pentamer pore. The upper half of the A2 helix from residues 196 to 228 associates extensively with the A1 chain through van der Waals interactions and lies in a shallow groove that extends diagonally across the A1 subunit. (21)

In contrast to the A1 chain, the A2 chain interacts closely with all five B subunits. The A2 subunit passes through the central pore of the B pentamer as a continuous helix. The pore diameter is just wide enough to accommodate the A2 chain as a h elix, therefore it displaces many of the solvent molecules which fill the pore of the isolated pentamer. (21)

View the cholera toxin here.

Subunit B

The five B subunits of the cholera toxin form a highly stable pentamer with nearly perfect 5-fold symmetry. (21) X-ray crystallography reveals that there is very little deviation, less than 0.5 angstroms r.m.s., from exact rotational symmetry. (22) This ring has a central pore lined by five amphipathic alpha-helices, one from each subunit, which are involved in pentamer stabilization. (21)

The 103 amino acids of each B subunit make up a small N-terminal helix from residues 4 to 12, a large central helix containing residues 58 to 79 and ten beta-strands. (22) Six of the beta strands form two triple stranded antip arallel sheets. The beta sheets on the outside of the pentamer give the outer surface of the ring a smooth appearance. (3) The overall fold consists of six antiparallel beta strands forming a closed beta barrel, capped by the large a lpha helix between the fourth and fifth strands. The beta sheets on the outside of the pentamer give the outer surface of the ring a smooth appearance. This long helical cap is slightly curved with the hydrophilic face forming the boundary wall of the c entral "pore". The "pore" is approximately 30 angstroms long with an effective diameter of 16 angstroms at the amino end, and narrowing to 11 angstroms at the carboxyl end. The sequential alternation of positive and negatively char ged residues along each alpha helix promotes the formation of salt bridges between neighboring helices and results in a pore with only a modest net positive charge. (22)

The B subunit’s sole disulfide bridge from Cys9 to Cys86 anchors the short solvent exposed amino terminal helix to an interior beta strand. (22) The pentamer has an overall diameter of approximately 64 angstroms (6.4 nm) and a height of 40 angstroms (4 nm). (3)

View the B pentamer here.


The A subunit of CT is loosely folded with almost complete loss of secondary structure occurring above 46 degrees C. The B pentamer on the other hand is very stable. The presence or absence of the A subunit has little effect on the stability of the B pentamer. (21) This is probably due to the number of well ordered solvent molecules seen in the pore of the B pentamer in the X-ray structure. In the holotoxin these are replaced by the carboxy-terminal tail of the A2 chai n. The holotoxin cannot be reconstituted by mixing A subunits and preformed B5 in vitro however. This reflects either a size and/or electrostatic barrier to passing a charged peptide through the preformed pore. (22)

Receptor Binding

The cholera toxin recognizes and binds to the pentasaccharide chain of ganglioside GM1 present on the outer surface of the plasma membrane of cells lining the lumen of the intestine. (17) Five molecules of GM1 on the membrane surface are bound by the five identical binding site on the B pentamer. (4) Cells lacking GM1 are immune to the effects of cholera toxin, conversely the addition of GM1 can induce sens itivity in cells normally lacking the receptor. (30) The normal physiological function of GM1 in these epithelial cells is not well understood, but GM1 has been implicated in various signal transduction pathways . (30) The structure of GM1 is Gal(beta1-3)GalNAc(beta1-4)(NeuAc(alpha2-3))Gal(beta1-4)Glc(beta1-1)-ceramide. (4) Cholera toxin binds specifically to GM1. (4) CT also binds weakly (about a thousand fold weaker) to ganglioside GD1b, which differs from GM1 by the addition of a second sialic acid residue to O8 of the GM1 sialic acid. (30)

The association of GM1 to cholera toxin is considered a "2-fingered grip" (30). The primary binding interactions are due to the terminal sugar of each of the two branches, galactose and sialic acid (N-acety l neuraminic acid). (30) The Gal(beta1-3)GalNAc "forefinger" is inserted into a deep pocket in the binding site, while the sialic acid "thumb" occupies a shallower depression on the surface of the toxin. (30) If the terminal galactose is not present, then CT does not bind the resulting shorter ganglioside. (4) Although sialic acid is necessary, the presence of the free carboxyl group on sialic acid is not essential for b inding but does enhance toxin binding. (17) The binding site is sufficiently open to accommodate additional substituents on the galactose, e.g. the addition of a fucose linked to O2. (4) (6) T he lipid portion of GM1 is not necessary for toxin binding, but serves to present the oligosaccharides in the clustered orientation apparently preferred by the toxins. (17)

There are no large scale conformational changes in CT upon binding to GM1. (30) Infrared spectroscopy and circular dichroism have shown that no greater than a 3-4% change in beta-sheet or alpha-helix content could o ccur in the presence of the receptor. (29)

Each of the five identical GM1 binding sites lies primarily within a single monomer of the B pentamer. The binding pocket is formed on one side by the loop (residues 51-60) connecting the beta4 strand to the central helix, and on the ot her side by the 2 loops connecting alpha1 to beta1 (residues 10-14) and beta5 to beta6 (resides 89-93). One end of this pocket is closed by the beta2-beta3 loop (residues 31-36) from an adjacent monomer, but these residues do not interact directly with t he bound oligosaccharide. (30)

Both terminal sugars of the branched GM1 pentasaccharide, galactose and sialic acid, exhibit substantial specific binding interactions with the toxin. A smaller contribution to the binding interactions comes from the N-acetyl galactose residue. The remaining 2 sugars apparently interact only indirectly with the toxin. (30)

The terminal galactose is held in place due to sugar atoms O2, O3 and O4 accepting hydrogen bonds directly from nitrogen donors in the side chains of Asn90 and Lys91. There are additional direct and solvent mediated hydrogen bonds from galactose t o Asn14 and Glu51. (30) The galactose O6 is hydrogen bonded to two solvent molecules in addition to accepting a hydrogen bond from the amine N of Gln61. (4)

The galactose ring lies parallel to the ring system of Trp88; the stacking arrangement of the hydrophobic surface of the sugar against the tryptophan ring stabilizes the binding conformation. (4) (30) Repla cing the Trp88 with another amino acid destroys the ability of CT to bind to GM1. (27) Early studies had shown that the single Trp residue in the B subunit must be associated with the GM1 binding site due to a b lue shift of approximately 12 nm in the fluorescence emission maximum of the single tryptophan residue. (29) (27)

The sialic acid of GM1 has hydrophobic interactions with Tyr12 and directly hydrogen bonds with the backbone of residues Glu11 and His13. Within the GM1 molecule sialic acid also makes at least one direct hydrogen bond with t he N-acetyl galactosamine and 2 water-mediated interactions with the terminal galactose. (30)

The N-acetyl galactosamine, which is a third specificity determinant, does not exhibit any direct or water mediated hydrogen bonds with the protein although its methyl group is in contact with the beta carbon of His13. (30)

GM1 is bound by both the holotoxin and the B pentamer, but not by monomeric B subunits. Each receptor binding site on the toxin lies primarily within a single B-subunit, with a single solvent-mediated hydrogen bond from the backbone N o f residue Gly33 on an adjacent subunit to O6 of the terminal galactose. (30) This Gly33 was examined through site-directed mutagenesis, and it was found that Ala33 was wild-type in interactions with GM1. It appears that t he residue at position 33 can be replaced by several other amino acids as long as there are no charge or steric hindrances to forming the hydrogen bond to the backbone. (27)

The main difference between the structure of the free toxin and the structure of the toxin bound to GM1 is the degree to which the loop consisting of residues 51-60 is well-ordered. This loop protrudes from the bottom surface of the ass embled AB5 holotoxin. Residues Glu51, Gln56, His57, and Gln61 are involved in sugar binding, and cause the loop to be well ordered. Presumably this loop exhibits flexibility in solution, allowing easy entry of the receptor into the binding si te. It has been shown that a 15-residue synthetic peptide (CTP3) spanning this loop and corresponding in sequence to residues 50-64 can elicit antibodies that are cross-reactive with native CT. Antibodies to this peptide were partially effective in neu tralizing the biological activity of the toxin, and may act by blocking the binding site. (4)

The binding of CT to GM1 can be effectively blocked using low concentrations of the appropriate nonlipid compound, which in one study was an o-GM1-poly-L-lysine derivative. This suggests that it may be possible to prevent ch olera toxin from binding GM1 by providing a compound containing polyvalent clusters of the GM1 pentasaccharide. Exogenous GM1 is a poor choice since it has been found to become a functional part of the cell’s membrane. < a href="#Bib3">(17)

Binding Kinetics

There are five binding sites in the holotoxin, one per B monomer, and the binding of GM1 to the 5 sites is known to be cooperative. (3) (30) Results of several experiments show that CT must bind to more than one GM1 molecule on the cell surface to subsequently activate adenylate cyclase. (3) One set of experiments determined that the dissociation constant for the five binding sites were equal, and ranged from 10-9 to 10-10 M depending on the technique. (21) A quantitative molecular interpretation based on an analysis involving equilibrium dialysis and the Hill equation indicated positive cooperativity and provided affini ty constants for binding of each of the five subunits to GM1 in the range of 2.0 x 106 to 3.0 x 106 M-1. (3)

Action of Toxin

There is a delay of approximately 15 minutes between the binding of the holotoxin to the GM1 receptors on the cell surface and the rise in cAMP. It has been postulated that during this delay the A subunit is translocated across the memb rane and reduced to form A1. (26) The disulfide bond between A1 and A2 keeps the protein inactive; this bond must be cleaved for the A1 subunit to become enzymatically active.

Bacterial toxins can induce characteristic changes in the function of G proteins by catalyzing covalent modification of the alpha chains. (16) The G proteins are a family of membrane-bound nucleotide-binding proteins which pla y key roles in transducing hormonal and sensory signals. (8) These proteins are heterotrimers of alpha, beta and gamma chains and are distinguished principally by their structurally distinct alpha chains, which bind and hydrolyze GTP . The more highly conserved beta and gamma chains serve to attach the G proteins to the cytoplasmic face of the plasma membrane and to present the alpha chain to the receptor. The hormone activated receptor promotes binding of GTP by the G protein alph a chain, which in turn stimulates the appropriate effector enzyme; stimulation of the effector is terminated when the alpha chain hydrolyzes its bound GTP. (16)

The A1 subunit of cholera toxin is an ADP-ribosyltransferase that catalyzes the transfer of an ADP-ribose from NAD+ to residue Arg187 in the alpha chain of Gs. (30) (26) ADP-ribosylation of alpha chain of Gs (a s) by cholera toxin stabilizes the GTP-bound conformation of a s and decreases its intrinsic GTPase activity, thereby producing increased stimulation of adenylyl cyclase and elevated intracellular cAMP. (16) (30) The ribosylated G protein activates luminal sodium pumps via a cAMP-dependent protein kinase. (22) There is also electrogenic Cl- secretion leading to decreased electrolytes in the cell. (14) This causes water to move with the electrolytes, and more electrolytes and water move into the cell from the bloodstream. Eventually this depletes the body’s electrolytes and causes dehydration.

The reaction is (NAD+ + acceptor --> ADP-ribose-acceptor + nicotinamide + H+) (21)

Several factors are necessary for the activation of adenylate cyclase by the A1 subunit. It has been shown that GTP or an analog is needed for the completion of the reaction. Removal of magnesium from cells inhibits the toxin-catalyzed reaction, because magnesium is needed for resident nucleoside diphosphate kinases to generate GTP from GMP or GDP. NAD+ is also needed to supply the ADP-ribose. (24)

Degradation of G protein

Surprisingly, cholera toxin also appears to induce a 74 to 95% decrease in the amount of a s. This decrease proceeds through an unknown mechanism, although it has been postulated that the cell recognizes that the G protein has been modified and begins to break it down. The fact that levels of cAMP remain high and GTP-dependent aden ylyl cyclase activity is still elevated despite the loss of a substantial fraction of a s suggests that the amount of a s in membranes is greater than the amount necessary for maximal activation of cAMP synthesis by cholera toxin. The toxin-induced increase in GTP-dependent adenylyl cyclase activity is maximal at 1 hour and adenylyl cyclase remains elevated for at least 32 hours. The decrease in immunoreactive a s begins after 1 hour of toxin treatment and is complete by 8 hours. (16)

Entry of the Toxin into the Cytosol

While the structure of the cholera toxin has been well defined, and its actions and receptor binding characterized, the method by which it moves across the cell membrane is still unknown. Early studies assumed that the B5 ring bound with the A subunit facing the cell membrane. This would force the A1 subunit through the membrane during binding. This theory was almost immediately challenged because it would force a number of hydrophilic regions into the membrane as well. More rece nt data has shown that the binding sites on the B subunits can only bind with the A facing away. (3)

A second possibility was that the B subunit ring somehow formed a channel into the cell that the A1 unit passed through. This would be difficult because the outside surface of the ring is strongly hydrophilic. Several studies now show that the B pentamer does not enter the membrane. (3)

There are now two models for the possible entry of cholera toxin into the cell. Both involve endocytosis of the toxin-receptor complex, which is apparently stimulated by the binding of B5 to the GM1 receptors.

In the first model, endosomal processing appears to involve CT migration to the Golgi and possibly other membrane-lined regions of the target cell cytoplasm. (1) A study has shown that the toxin-receptor complex remains assoc iated with the membrane until the lysosome associates with the Golgi region. (1) (2) In the same study processing in the Golgi region has been shown to be necessary for the production of the A1 subunit, and is blo cked by chemical which prevents the functioning of the Golgi region. (1) This model does not adequately explain how the A1 fragment crosses out of the lysosome.

Another model proposes that the toxin-receptor complex moves by transcytosis from the apical membrane of intestinal cells to the basolateral membrane where the G protein target is located. Recent studies demonstrate that endosomal pH induces confo rmational changes in the B subunit, which then influence the properties of associated GM1-containing membranes. There is a localized disturbance of membrane phospholipid packing that has been shown to increase the membrane’s permeability to a number of substances. This disturbance has been shown to be specifically induced by acidic pH. It is not yet known if this leads directly to the translocation of the A1 peptide. The transcytosis model would indicate that cholera toxin spends a consider able amount of time within an acidified endosomal compartment. (18) Lowered pH has been shown to be essential in the cholera toxin intoxication of rat hepatocytes, among other cells. (2)

There is a pH-dependent increase in the generation of the A1 peptide, which implies that in a lysosome the A1 subunit can be cleaved from the rest of the peptide. Additional photolabeling, light scattering, and electron-microscope data support a model in which the B pentamer remains intact, on or within the surface of the membrane, and subunit A1 is inserted into the bilayer. (3) The generation of A1 is a fairly quick process, and probably occurs in either model.

There is evidence that the transduction mechanism occurs at different rates in different cell lines. Since the model for the transduction of A1 is not fully understood, the reasons for these differences are not readily apparent. Studies with leuk emia cell lines show a difference of several orders of magnitude in the amount of cholera toxin needed to affect the two different cells even after correcting for the amount of GM1. (9)

Uses of Cholera Toxin

Determining the mechanisms responsible for cholera toxin A1 transport is important because, in addition to its well-known toxic effect, CT possesses mucosal adjuvant properties that would be useful in oral immunization strategies. (19) (20) Many microorganisms invade their hosts through mucosal membrane surfaces and the development of a protective mucosal immune response can provide a preemptory line of defense against a variety of infectious diseases. Immunization against these pathogens by a parenteral route is useful in eliciting a circulatory immune response but doesn’t generate effective protection against disease onset at mucosal surfaces due to the absence of an antibody response in mucosal sec retions. An enhanced mucosal immune response to orally administered antigens can be achieved when the antigens are mixed with CT or conjugated to CTB. The ability of CT and CTB to act as mucosal adjuvants may stem from the adhesive and epithelial cell m embrane-permeabilizing effects of these protein complexes. (18)

Cholera toxin is used in many areas of research. It is particularly useful because of its receptor specificity. (3) It has been used to study myelin basic protein (MBP) which is one of the major protein constituents of myeli n. Each MBP has multiple sites which can be ADP-ribosylated by cholera toxin. In rat brains, MBP was ADP-ribosylated 4 times faster than Gs. Other than Gs and MBP, virtually nothing in the rat brain can be ADP-ribosylated by CT. This makes CT a useful tool for studying myelin and myelin defects. (13)

Using CT as well as other bacterial toxins, including the pertussis toxin and the toxin produced by E. coli, a team of researchers was able to determine which forms of G protein interact with a specific set of cell surface receptors. (5)

CT is used extensively for cytochemical and immunohistochemical studies on the basis of its ability to bind preferentially to ganglioside GM1, which is found in high concentration in the membranes of neurons. (3) Ea rly work made use of the receptor specificity of CT to demonstrate that glycolipids inserted into lipid vesicles retained their biological function and orientation. (3)

Treatment for cholera

Early research into treatment of cholera revolved around the fact that in vitro high doses of choleragenoid (the inactive B pentamer) blocked the binding of cholera toxin by competitive inhibition. Early experiments dealt with the difficult y of introducing a sufficient amount of B5 into the intestine to block all the receptors; this therapy was only partially successful. (3)

The current treatment for cholera involves oral rehydration therapy and antibiotics. The toxin itself, once inside a cell, does not spread, and therefore is passed out of the body when the cell dies. The antibiotics, usually a course of tetracycl ine, kill the V. cholerae colonizing the gut. The toxin subunits cannot recombine once separated, so the active A1 subunit inside an affected cell does not enter other cells, but is simply passed out of the body once the host cell dies. Oral rehy dration therapy is effective because while the Na+ mucosal carrier is pumping Na+ out of the cell, the Na+/glucose pump is not affected. The oral solution recommended by the World Health Organization (WHO) contains 20 gm glucose; 3.5 gm sodium chloride; 2.9 gm trisodium citrate, dihydrate (or 2.5 gm sodium bicarbonate); and 1.5 gm potassium chloride per liter of drinking water. For severe dehydration, venous infusion is recommended until blood pressure has risen to normal levels. Because the underlying problem is fluid and electrolyte loss, not blood pressure, plasma and vasopressors are not recommended. (33)


The most effective treatment would obviously be to prevent the disease. This requires both good sanitation and an effective vaccine. The disease itself provides lifelong immunity to cholera, but attenuated whole-cell vaccines administered parente rally provide protection in only about 50% of cases, and that protection last only about 3 months. (32) (33) A substantial serum IgG response occurs and persists for several years following cholera infection, and there is evidence that the antitoxin can neutralize the effect of CT in vivo. (3) This provides hope that an effective vaccine can be found. There are two candidate vaccines currently being tested on volunteers and in field trials in Bangladesh and Pakistan. One is a mixture of killed whole bacterial cells and B subunit, while the other is based on a live attenuated strain of V. cholerae. Both have been effective in tests, but both have problems that will require further w ork to resolve. (3)

In a trial of nearly 90,000 individuals, a single oral dose of the combined B pentamer and killed whole cell vaccine produced an initial 64% protective rate, much better than any previous vaccine. This large-scale field trial showed that responses remained protective for 3 years in more than 40% of those vaccinated. (3)

Much of the current difficulty in providing effective immunization is due to the ineffective generation of the intestinal immunological response. The toxin delivered to the gut in most vaccination schemes has been denatured either in the preparati on and purification, or on its trip through the stomach. A vaccine currently being tested involves the use of live V. cholerae cells which are capable of colonization in the gut but which are genetically unable to produce the A subunit. The B5 subunit is capable of being assembled and secreted without the A subunit, and is therefore delivered in near-native form directly into the gut. In recent small-scale trials, ingestion of a single oral dose provided initial protection of 95%. Te sts in children are in progress, and field trials are planned. (3)

Recent discoveries indicate that the genes coding for the cholera toxin are borne on, and can be infectiously transmitted by, a filamentous bacteriophage. This raises a number of questions about the evolution of bacterial pathogenesis. The early pandemics of cholera were caused by the classical biotype of V. cholerae serogroup O1, while the current pandemic, ongoing since 1961, is caused by the ‘El Tor’ biotype. The two biotypes are very different in phenotypic assays and in enzymes, but the sequences of the toxin genes are the same in the classical and many El Tor strains, whereas other El Tor strains have different sequences. This indicates that the evolutionary history of the toxin is to some degree independent of the evolution of th e host bacterium. Among the large number of different non-O1 serotypes of V. cholerae, the presence of cholera toxin genes is very rare. Even within the serogroup O1 a broad variety of strains do not have the toxin genes. (10) This could present a problem for vaccination schemes that target the whole cell. The genes for the toxin could easily move to another serotype, which would render previous vaccines useless.


While oral rehydration and antibiotics form a safe and effective treatment for cholera, the disease will not be eradicated until an effective vaccine is found, and sanitation in many areas is improved. Today in developing countries, waterborne and sanitation-related diseases kill over three million people annually, most of them younger than five years of age. Around the world a billion people lack access to safe water, and 1.8 billion do not have adequate sanitary facilities. According to one es timate, providing safe water and decent sanitation facilities for all people would cost $68 billion over the next 10 years--a sum equivalent to only 1% of the world’s military expenditures for the same period. (31)


1. Nambiar MP, Oda T, Chen C, Kuwazuru Y, Wu HC (1993) Involvement of the Golgi Region in the Intracellular Trafficking of Cholera Toxin J Cell Physiol 154: 222-228

2. Janicot M, Fouque F, Desbuquois B (1991) Activation of Rat Liver Adenylate Cyclase by Cholera Toxin Requires Toxin Internalization and Processing in Endosomes J Biol Chem 266:12858-12865

3. Spangler BD (1992) Structure and Function of Cholera Toxin and the Related Escherichia coli Heat-Labile Enterotoxin Microbiol Rev 56: 622-647

4. Merritt EA, Sixma TK, Kalk KH, van Zanten BAM, Hol, WGJ (1994) Galactose-Binding Site in Escherichia coli Heat-Labile Enterotoxin (LT) and Cholera Toxin (CT) Mol Microbiol 13: 745-753

5. Rajagopalan-Gupta RM, Rasenick MM, Hunzicker-Dunn M (1997) Luteinizing Hormone/Choriogonadotropin-Dependent, Cholera Toxin-Catalyzed Adenosine 5’-Diphosphate (ADP)-Ribosylation of the Long and Short Forms of Gsa< /FONT> and Pertussis Toxin-Catalyzed ADP-Ribosylation of Gia Mol Endocrinol 11: 538-549

6. van den Akker F, Steensma E, Hol WGJ (1996) Tumor Marker Disaccharide d-Gal-beta-1,3-GalNAc complexed to Heat-Labile Enterotoxin from Escherichia coli. Protien Sci 5:1184-1188

7. Stentebjerg-Olesen B, Pallesen L, Jensen LB, Christiansen G, Klemm P (1997) Authentic Display of a Cholera Toxin Epitope by Chimeric Type 1 Fimbriae: Effects of Insert Position and Host Background Microbiology 143: 2027-2038

8. Sugden D, Klein DC (1987) A Cholera Toxin Substrate Regulates Cyclic GMP Content of Rat Pinealocytes J Biol Chem 262: 7447-7450

9. Giuliani A, Calappi E, Mineo E, Neri MG, Gallina A, Pessina A (1995) The Different Inhibiting Effect of Cholera Toxin on Two Leukemia Cell Lines Does Not Correlate With Their Toxin Binding Capacity Mol and Cell Biochem 152: 103-112

10. Levin BR, Tauxe RV (1996) Cholera: Nice Bacteria and Bad Viruses Curr Biol 6: 1389-1391

11. Spiegel S (1989) Inhibition of Protein Kinase C-Dependent Cellular Proliferation by Interaction of Endogenous Ganglioside GM1 with the B Subunit of Cholera Toxin J Biol Chem 264: 16512-16517

12. Pfau JD, Taylor RK (1996) Genetic Footprint of the ToxR-Binding Site in the Promoter for Cholera Toxin Mol Microbiol 20:213-222

13. Enomoto K, Asakawa T (1990) ADP-Riboxylation of Myelin Basic Protein by Cholera Toxin Biochim Biophys Acta 1036: 188-192

14. Sandvig K, Garred O, van Deurs B (1997) Intracellular Transport and Processing of Protein Toxins Produced by Enteric Bacteria p. 225-232 In Paul et al (ed.) Mechanisms in the Pathogenesis of Enteric Diseases. New York: Plenum Pre ss

15. Moon HW (1997) Comparative Histopathology of Intestinal Infections p. 1-19 In Paul et al (ed.) Mechanisms in the Pathogenesis of Enteric Diseases. New York: Plenum Press

16. Chang F, Bourne HR (1989) Cholera Toxin Induces cAMP-Independent Degredation of Gs J Biol Chem 264: 5352-5357

17. Schengrund C, Ringler NJ (1989) Binding of Vibrio cholera Toxin and the Heat-Labile Enterotoxin of Escherichia coli to GM1, and Nonlipid Oligosaccharide Polyvalent Ligands J Biol Chem 264: 13233-13237

18. McCann JA, Mertz JA, Czworkowski J, Picking WD (1997) Conformational Changes in Cholera Toxin B Subunit--Ganglioside GM1 Complexes Are Elicited by Environmental pH and Evoke Changes in Membrane Structure Biochemistry 36: 9169-9178

19. Francis ML, Moss J, Fitz TA, Mond JJ (1990) cAMP-Independent Effects of Cholera Toxin on B Cell Activation: A Possible Role for Cell Surface Ganglioside GM1 in B Cell Activation J Immunol 145: 3162-3169

20. Anastassiou ED, Yamada H, Francis ML, Mond JJ, Tsokos GC (1990) Effects of Cholera Toxin on Human B Cells: Cholera Toxin Induces B Cell Surface DR Expression While It Inhibits Anti-mu Antibody-Induced Cell Proliferation J Immunol 1 45: 2375-2380

21. Zhang RG, Scott DL, Westbrook ML, Nance S, Spangler BD, Shipley GG, Westbrook EM (1995) The Three-Dimensional Crystal Structure of Cholera Toxin J Mol Biol 251:563-573

22. Zhang RG, Westbrook ML, Westbrook EM, Scott DL, Otwinowski Z, Maulik PR, Reed RA, Shipley GG (1995) The 2.4 Angstrom Crystal Structure of Cholera Toxin B Subunit Pentamer: Choleragenoid J Mol Biol 251:550-562

23. Hu SH, Peek JA, Rattigan E, Taylor RK, Martin JL (1997) Structure of TcpG, the DsbA Protein Folding Catalyst from Vibrio cholerae. J Mol Biol 268:137-146

24. Enomoto K, Gill DM (1980) Cholera Toxin Activation of Adenylate Cyclase J Biol Chem 255: 1252-1258

25. Sun D, Lafferty MJ, Peek JA, Taylor RK (1997) Domains Within the Vibrio cholerae Toxin Coregulated Pilin Subunit that Mediate Bacterial Colonization Gene 192: 79-85

26. Kassis S, Hagmann J, Fishman PH, Chang PP, Moss J (1982) Mechanism of Action of Cholera Toxin on Intact Cells J Biol Chem 257: 12148-12152

27. Jobling MG, Holmes RK (1991) Analysis of Structure and Function of the B Subunit of Cholera Toxin by the Use of Site-Directed Mutagenesis Mol Microbiol 5: 1755-1767

28. DiRita VJ, Parsot C, Jander G, Mekalanos JJ (1991) Regulatory Cascade Controls Virulence in Vibrio cholerae. Proc Natl Acad Sci USA 88: 5403-5407

29. Surewicz WK, Leddy JJ, Mantsch HH (1990) Structure, Stability, and Receptor Interaction of Cholera Toxin as Studied by Fourier-Transform Infrared Spectroscopy Biochemistry 29: 8106-8111

30. Merritt EA, Sarfaty S, van den Akker F, L’Hoir C, Martial JA, Hol WGJ (1994) Crystal Structure of Cholera Toxin B-pentamer Bound to Receptor GM1 Pentasaccharide Protein Sci 3:166-175

31. Doyle R (1997) Access to Safe Drinking Water Scientific American 277:38

32. Travel Health Online (11/24/97)

33. The Merck Manual Last updated 1992, visited 11/04/97 - 11/30/97

34. Moore TG (11/24/97)

Key to Journal Abbreviations:

J Biol Chem

Journal of Biological Chemistry

J Cell Physiol

Journal of Cellular Physiology

Microbiol Rev

Microbiological Reviews

Mol Microbiol

Molecular Microbiology

Mol Endocrinol

Molecular Endocrinology

Protien Sci

Protein Science



Mol and Cell Biochem

Molecular and Cellular Biochemistry

Curr Biol

Current Biology

Biochim Biophys Acta

Biochimica et Biophysica Acta



J Immunol

Journal of Immunology

J Mol Biol

Journal of Molecular Biology



Proc Natl Acad Sci USA

Proceedings of the National Academy of Sciences, USA

Copyright 1997 by Elizabeth Cronenwett