HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Services Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (7) Citing Articles via Google Scholar Google Scholar Articles by Hajishengallis, G. Articles by Russell, M.W. Search for Related Content PubMed PubMed Citation Articles by Hajishengallis, G. Articles by Russell, M.W.

Immunomodulation with Enterotoxins for the Generation of Secretory Immunity or Tolerance: Applications for Oral Infections

¹ Department of Microbiology, Immunology, and Parasitology, and Center of Excellence in Oral and Craniofacial Biology, Louisiana State University Health Sciences Center, New Orleans, LA, USA;
² Departments of Microbiology and Immunology and
³ Oral Biology,
⁴ The Witebsky Center for Microbial Pathogenesis and Immunology, University at Buffalo, 3435 Main Street/Farber 138, Buffalo, NY 14214, USA;

^* corresponding author, russellm{at}buffalo.edu

	ABSTRACT

TOP ABSTRACT INTRODUCTION TYPE I AND TYPE... MECHANISMS OF ACTION TOLERANCE vs. IMMUNITY APPLICATIONS AND SIGNIFICANCE... REFERENCES

 
The heat-labile enterotoxins, such as cholera toxin (CT), and the labile toxins types I and II (LT-I and LT-II) of Escherichia coli have been extensively studied for their immunomodulatory properties, which result in the enhancement of immune responses. Despite superficial similarity in structure, in which a toxic A subunit is coupled to a pentameric binding B subunit, different toxins have different immunological properties. Administration of appropriate antigens admixed with or coupled to these toxins by oral, intranasal, or other routes in experimental animals induces mucosal IgA and circulating IgG antibodies that have protective potential against a variety of enteric, respiratory, or genital infections. These include the generation of salivary antibodies that may protect against colonization with mutans streptococci and the development of dental caries. However, exploitation of these adjuvants for human use requires an understanding of their mode of action and the separation of their desirable immunomodulatory properties from their toxicity. Recent findings have revealed that adjuvant action is not critically dependent upon the enzymic activity of the A subunits, and that the isolated B subunits may exert different effects on cells of the immune system than do the intact toxins. Interaction of the toxins with immunocompetent cells is not exclusively dependent upon their conventional ganglioside receptors. Immunomodulatory effects have been observed on dendritic cells, macrophages, CD4⁺ and CD8⁺ T-cells, and B-cells. Numerous factors—including the precise form of the toxin adjuvant, properties of the antigen, whether and how they are coupled, route of administration, and species of animal model—affect the outcome, whether this is enhanced humoral and cellular immunity, or specific induced tolerance toward the antigen.

KEY WORDS: cholera toxin • heat-labile enterotoxin • vaccine • adjuvant • immune response

	INTRODUCTION

TOP ABSTRACT INTRODUCTION TYPE I AND TYPE... MECHANISMS OF ACTION TOLERANCE vs. IMMUNITY APPLICATIONS AND SIGNIFICANCE... REFERENCES

 
Among the "witch’s kitchen" of natural products exploited by immunologists, the diarrheagenic heat-labile enterotoxins of certain Gram-negative bacteria have earned a special place in mucosal immunology as potent immunomodulating agents or adjuvants. (Other examples include pokeweed mitogen, keyhole limpet hemocyanin, cobra venom factor, and serum from an unborn calf!) Adjuvants are defined as materials that are co-administered with antigens in vaccines to enhance the desired immune response. The generation of immune responses in mucosal secretions such as saliva, where secretory immunoglobulin A (S-IgA) is the dominant isotype of antibody, requires the stimulation of the mucosal immune system through its inductive sites. The best-known of these sites are the intestinal Peyer’s patches; others include organized mucosa-associated lymphoid tissues of similar structure in the large bowel, and the tonsils and adenoids that form Waldeyer’s ring in the human pharynx. The mechanisms by which antigens applied to these sites are taken up through specialized epithelial cells (M cells) and delivered to underlying antigen-presenting cells (APC)—leading to the stimulation of IgA-committed B-cells that then emigrate and circulate until homing to the effector sites of mucosal immunity, such as the gut lamina propria or salivary glands—have been extensively reviewed elsewhere (Mestecky, 1987; Mestecky et al., 2005). S-IgA is the product of polymeric (p) IgA-secreting plasma cells that come to reside in the subepithelial spaces of these effector sites, after pIgA has been transported through the epithelial cells by means of the polymeric Ig receptor, which becomes the secretory component of S-IgA in the process of transcytosis. However, it has become clear that most foreign antigenic material applied to the mucosal-inductive sites—the great bulk of which consists of harmless food and other environmental antigens, as well as commensal micro-organisms—elicits little or no immune response, or may even induce specific immune tolerance. Effective induction of active immune responses requires the stimulation of cells of the innate immune system, including the major APC, such as dendritic cells and macrophages, by arrays of molecular entities typically found in microbes, whether pathogenic or not, and therefore designated "pathogen- (or, better, microbe-) associated molecular patterns" (PAMP; MAMP) (Janeway and Medzhitov, 2002). These engage pattern-recognition receptors (PRR) on the APC and deliver co-stimulatory signals that result in cellular activation and generation of responses in T- and B-lymphocytes, ultimately leading to the production of antibody or cytotoxic T-cells. The development of vaccines critically depends upon the use of adjuvants that can mimic the stimulus provided by these MAMPs. This is especially true for vaccines that must be administered by a mucosal route to induce S-IgA responses in secretions such as saliva, since non-replicating antigens delivered orally or intranasally readily induce tolerance rather than active immunity. Moreover, the majority of antigenic material consumed orally is degraded by the digestive processes. Several strategies have been developed in recent decades to overcome the inherent unresponsiveness or tolerance of mucosal immune tissues to non-replicating antigens and thereby elicit potent mucosal (and systemic) immune responses (Table 1; Mestecky et al., 1997; Ogra et al., 2001). Some of the most effective of these approaches involve the use of heat-labile enterotoxins, such as cholera toxin (CT), either as adjuvants or as coupled delivery agents; these are the subject of this review.

View larger version (41K): [in this window] [in a new window]   Figure. Molecular structure of heat-labile enterotoxins. A and B polypeptides are translated from a single mRNA, with an excess of B polypeptides. The leader sequences of both allow them to be transported into the periplasm, where B subunits assemble into pentamers. A disulfide bridge spans the junction of the A1 and A2 segments of the A polypeptide, where a proteolytic nick occurs. The A2 subunit non-covalently associates with the central pore in the B pentamer.

	TYPE I AND TYPE II ENTEROTOXINS AS IMMUNOMODULATORY AGENTS

TOP ABSTRACT INTRODUCTION TYPE I AND TYPE... MECHANISMS OF ACTION TOLERANCE vs. IMMUNITY APPLICATIONS AND SIGNIFICANCE... REFERENCES

While the A polypeptides of the type I and type II enterotoxins are highly homologous, significant divergence in amino acid sequences is observed among the B polypeptides (Pickett et al., 1986; Holmes et al., 1995). The B polypeptides of CT and LT-I exhibit over 80% amino acid identity. As a result, the toxicity of CT can be neutralized by anti-LT-I antibodies, and vice versa (Guth et al., 1986; Pickett et al., 1986; Finkelstein et al., 1987; Holmes et al., 1990, 1995). In contrast, the B polypeptides of LT-IIa and LT-IIb have little homology (< 14% sequence identity) to those of CT or LT-I, and are also antigenically distinct from, although cross-reactive with, each other (Holmes et al., 1995).

Ganglioside Receptors
The major receptors for the type I and type II heat-labile enterotoxins are gangliosides, a complex family of glycosphingolipids which are normal components of the eukaryotic plasma membrane (Sonnino et al., 1986), and which participate in various cellular functions, including signal transduction (Nagai and Iwamori, 1984; Fishman, 1986; Hannun and Linardic, 1993). Over 100 molecular species of gangliosides have been identified in numerous mammalian tissues, and each cell type appears to express a distinct subset of gangliosides (Nagai and Iwamori, 1984; Sonnino et al., 1986; Hannun and Linardic, 1993). Initially considered to be structural components of the plasma membrane, many gangliosides are now thought to be constituents of signaling pathways that detect external stimuli and transduce those signals to cytoplasmic molecules that modulate gene transcription, cell growth, and differentiation (Nagai and Iwamori, 1984; Sonnino et al., 1986; Hannun and Linardic, 1993). The binding specificities of CT, LT-I, LT-IIa, and LT-IIb for gangliosides, however, are distinctive (Fukuta et al., 1988). CT and LT-I bind to ganglioside G_M1 (GM1), although competitive studies have shown that LT-I also binds to one or more glycoproteins (Critchley et al., 1981; Yamada et al., 1983; Orlandi et al., 1994); LT-IIa binds most avidly to ganglioside G_D1b (GD1b), less to ganglioside G_D1a (GD1a), and has a low but measurable affinity for GM1; and LT-IIb binds with high affinity only to GD1a (Fukuta et al., 1988). Differences in ganglioside-binding activity and specificity are believed to be important in determination of the host specificity of the type I and type II enterotoxins with regard to the animal species, tissues, and cell types that can be intoxicated (Connell et al., 1995; Holmes et al., 1995). Our research has revealed that their divergent ganglioside-binding activities target the enterotoxins to different lymphoid cell subsets (Arce et al., 2005) and determine, at least in part, the molecular pathways by which they augment immune responses to co-administered antigens. Immunomodulation by heat-labile enterotoxins, therefore, is a complex process that likely requires multiple interactions between the enterotoxins and immunocompetent cells via gangliosides or other specific surface receptors.

Holoenterotoxins as Adjuvants
Because intact enterotoxins are too toxic to find application as human adjuvants, it has been important to identify adjuvant activities that are independent of the ability of their A subunits to elevate intracellular cAMP and the resulting enterotoxic effects. In the late 1980s and early 1990s, commercially available preparations of biochemically purified cholera toxin B subunit (CT-B) were effectively used as an adjuvant when mixed with orally administered antigens (Elson and Dertzbaugh, 2005), but initial enthusiasm subsided when it was realized that these preparations were contaminated with bioactive traces of intact CT. As little as 0.1–1 percent contamination of CT-B with intact CT, as typically occurs in these preparations, is sufficient to exert a synergistic adjuvant effect by the intragastric route (Wilson et al., 1990). Moreover, a point substitution mutant (E112K) of LT-I, which lacked ADP-ribosylating activity, was found to be deficient in adjuvant activity upon peroral administration in mice (Lycke et al., 1992), suggesting that adjuvanticity was linked to the catalytic activity of the A subunit. However, coupling of CT-B to protein antigens was found to potentiate induction of mucosal immune responses to the coupled protein, compared with immunization with uncoupled protein (McKenzie and Halsey, 1984; Czerkinsky et al., 1989; Russell and Wu, 1991), but co-administration of intact CT as an adjuvant was necessary, as well as coupling of antigen to CT-B, for effectiveness by the oral route at low doses. These pioneering studies led to the concept that the A subunit is responsible for adjuvant action, which is inseparable from toxicity, whereas the B subunit can provide receptor-binding function that targets coupled antigens for uptake by immunocompetent cells. However, subsequent studies have shown that when antigen is coupled to CT-B in the form of recombinant chimeric proteins (see below), these are immunogenic without a requirement for additional holotoxin adjuvant. Moreover, several studies have since shown that recombinant CT-B or LT-I-B, completely lacking the A subunit, can have adjuvant effects with some antigens, especially when administered by the intranasal route or even orally (Verweij et al., 1998; Wu and Russell, 1998; Plant et al., 2003). An important consideration in the evaluation of these studies is that recombinant proteins may be contaminated with bacterial lipopolysaccharide (LPS) derived from the expression host (usually E. coli), which itself can display adjuvant properties. We have found that recombinant CT-B, having minimal LPS content, has adjuvant activity for AgI/II of Streptococcus mutans administered intranasally, and that 100-fold higher concentrations of LPS are required to reveal adjuvant activity (Wu and Russell, 1998). Similar immunomodulatory effects have been observed by others using recombinant LT-1-B devoid of detectable LPS (Nashar et al., 1996; Fraser et al., 2003), including B subunits obtained from recombinant clones of Bacillus brevis, a Gram-positive organism that does not produce LPS (Maeyama et al., 2001; Yokomizo et al., 2002). In contrast, studies on mucosal immunization conducted in conventional animals cannot readily eliminate the possible synergistic contribution of LPS, or other MAMPs derived from commensal micro-organisms, to the adjuvant effects observed.

B Subunits as Coupled Delivery Agents
The rationale for coupling the B subunits of heat-labile enterotoxins to protein immunogens is to confer on them ganglioside-binding activity, thereby facilitating their uptake at mucosal-inductive sites, such as the intestinal Peyer’s patches. Moreover, the ganglioside-binding ability of CT-B-linked immunogens may also enhance their interactions with APC present in the inductive sites. Antigens may be coupled to enterotoxin B subunits in three ways: (i) by means of chemical cross-linking agents, of which N-succinimidyl-3(2-pyridinedithio) propionate has been widely used; (ii) by genetic fusion of peptides directly to the B subunit at either the N or C terminal, or even internally in an exposed loop; or (iii) by genetic fusion of a protein to the A2 subunit in place of the A1 subunit and co-expression with the B subunit to form chimeric proteins of the form Ag-A2/B₅.

There is experimental evidence in support of the applicability of all three strategies. Intragastric or intranasal immunization with surface protein AgI/II of S. mutans chemically coupled to CT-B induces significantly higher levels of salivary IgA antibodies compared with immunization with unconjugated AgI/II (Russell and Wu, 1991; Wu and Russell, 1993). The structure of chemical conjugates of antigens and B subunits is not accurately known, and it may vary between batches, even under controlled conditions. However, this approach is applicable to a wide variety of antigens, including polysaccharides as well as proteins (Shen et al., 2001).

Better-defined and more consistent coupling is accomplished with recombinant DNA technology to create fusion proteins of peptides with enterotoxin B subunits (Sanchez et al., 1988; Clements, 1990). However, genetic fusion of large peptides to CT-B has been found to disrupt the structure of CT-B, inhibiting pentamer formation and GM1 binding (Dertzbaugh and Elson, 1993). The use of intervening linkers of 6 to 10 amino acids, including two proline residues, was proposed as an approach for overcoming this problem. However, even these fusion proteins, consisting of LT-I-B and the AgI/II-like protein SpaA or dextranase of Streptococcus sobrinus, exhibit reduced GM1-binding activity compared with uncoupled LT-I-B (Jagusztyn-Krynicka et al., 1993).

An alternative strategy for genetically linking large polypeptides to the B subunits exploits the molecular structure of the AB₅ enterotoxins (Fig.). This approach takes advantage of the ability of the A2 subunit to insert into the ring formed by the CT-B pentamer, and thereby to mediate non-covalent association of CT-A with CT-B. Accordingly, other protein Ags genetically fused to the N-terminus of CT-A2 should be able to assemble as chimeric molecules with CT-B, thereby creating novel immunogens. By exploiting the native conformation of AB₅ enterotoxins, these holotoxin-like, non-covalently linked constructs are less disruptive to the B subunit pentameric structure than is direct fusion of antigens to B subunit monomers. Such chimeric proteins were originally made from alkaline phosphatase, mannose-binding protein, and ß-lactamase, but their immunogenicity was not assessed (Jobling and Holmes, 1992). As a first example of the productive use of such chimeric protein immunogens, the CT-A1 subunit was genetically replaced by the saliva-binding region (SBR) of S. mutans AgI/II, which was thus fused to CT-A2, and co-expressed with CT-B to form a chimeric protein that retained GM1-binding activity (Hajishengallis et al., 1995). Another similar chimeric protein was constructed from SBR and the A2/B subunits of LT-IIa (Martin et al., 2001a). Intragastric or intranasal immunization of mice with these Ag-A2/B chimeric proteins induces high levels of salivary IgA antibodies to native AgI/II, even in the absence of intact toxin or other adjuvants. This suggests that ganglioside binding alone is sufficient for enhancing the immune response to the linked immunogen, although the extent to which the A2 subunit might contribute to adjuvant action is currently unknown.

	MECHANISMS OF ACTION

TOP ABSTRACT INTRODUCTION TYPE I AND TYPE... MECHANISMS OF ACTION TOLERANCE vs. IMMUNITY APPLICATIONS AND SIGNIFICANCE... REFERENCES

 
Immunomodulatory Mechanisms Mediated by Holoenterotoxins: Role of the A Subunit
Given the crucial role of DC and other APCs (macrophages and B-cells) in the induction of adaptive immunity (Itano and Jenkins, 2003), the potentiation of APC function is a major aspect of adjuvant action, including that of heat-labile enterotoxins. Most studies have been performed with CT, which up-regulates expression of the B7-2 (CD86) co-stimulatory molecule, and stimulates antigen presentation through enhancement of MHC class II expression and IL-1 production (Bromander et al., 1991; Matousek et al., 1996; Cong et al., 1997; Simmons et al., 2001). CT also interacts with lymphocytes and promotes B-cell isotype-switch differentiation toward IgG1 and IgA in mice (Holmgren et al., 1993). Moreover, CT displays complex stimulatory or inhibitory effects on T-cell proliferation in vitro (Elson and Dertzbaugh, 2005), which are not readily interpreted in terms of its adjuvant effects in vivo. It is generally found that CT induces CD4⁺ T-cell polarization toward the Th2 phenotype (Simmons et al., 2001), although exceptions to this rule have been reported.

Recent findings from our laboratories show that CT, LT-IIa, and LT-IIb bind differentially to immunocompetent cells (Arce et al., 2005), and the precise mechanisms through which these enterotoxins enhance immune responses are being explored. It appears that the different ganglioside-binding specificities of the enterotoxins may be a strong determinant for their distinct immune-regulatory effects on B-cells. Activation of B lymphocytes by CT or LT-I stimulates antigen presentation through induction of enhanced MHC class II expression on B-cells, and by promoting interactions between B-cells and T-cells via induction of CD86, LFA-1, and ICAM-1 on B lymphocytes (Nashar et al., 1997, 2001; Arce et al., 2005). Increased contacts between B- and T-cells may induce B-cell proliferation and eventually differentiation of B-cells into plasma cells or memory cells (Parker, 1993). A mechanism through which CT possibly mediates B-cell differentiation into plasma cells is through inhibition of CD40 ligand (CD40L; CD154) expression on CD4⁺ T-cells (Martin et al., 2001b). Interruption of CD40-CD40L interactions between B-cells and T helper cells is essential to suspend B-cell proliferation and initiate their differentiation into plasma cells (Arpin et al., 1995; Liu and Banchereau, 1997). LT-I can also interfere with CD40-CD40L interactions between APC and T-cells, through its ability to down-regulate CD40 on DC (Petrovska et al., 2003). In contrast, LT-I can provide CD28-mediated co-stimulatory function to T-cells through its ability to stimulate B7-1 (CD80) expression on DC (Petrovska et al., 2003). Since LT-IIa and LT-IIb do not influence CD40L expression on CD4⁺ T-cells (Martin et al., 2001b), these type II enterotoxins may promote mainly the formation of memory B-cells, which depends upon the continuation of CD40-CD40L interactions (Arpin et al., 1995; Liu and Banchereau, 1997). Co-stimulation of T-cells by CD86 expressed on B-cells preferentially elicits IL-4 production by the T-cells, and hence induces polarization of CD4⁺ T-cells toward the Th2 phenotype (Freeman et al., 1995). Th2 cytokines like IL-4, IL-5, IL-6, and IL-10 enhance humoral immune responses by driving the survival, proliferation, and differentiation of B-cells (Parker, 1993; Paul and Seder, 1994; Liu and Banchereau, 1997; Hasbold et al., 1999). In addition, IL-5 and IL-6 promote the survival of long-lived plasma cells and the maintenance of serum antibody titers (Cassese et al., 2003). Another mechanism through which CT induces the Th2 phenotype is by suppressing CD40-CD40L interactions between DC and T-cells, which results in inhibition of the production of IL-12 (p70), a cytokine which favors Th1 polarization (Martin et al., 2001b; Braun et al., 1999). CT also inhibits IL-12 (p70) production directly from dendritic cells and other antigen-presenting cells, and, furthermore, it inhibits expression of the IL-12 receptor on T-cells (Braun et al., 1999). Although Th1 cells can co-operate with B-cells to induce plasma cell differentiation, they do not usually exhibit the full complement of cytokines characteristic of Th2 cells, and therefore Th1 cells are less effective than Th2 cells in providing B-cell help (Smith et al., 2000). Conversely, the lack of these effects in LT-IIa and LT-IIb may account for their ability to bias responses more toward Th1, as observed in vivo especially for LT-IIb (Martin et al., 2000). Furthermore, CT, LT-I (B subunit), or LT-IIa, but not LT-IIb, selectively induces apoptosis in CD8⁺ T-cells (Elson et al., 1995; Nashar et al., 1996; Arce et al., 2005). Although when differentiated, CD8⁺ cells become cytotoxic T-cells, they are also a major source of IFN, a cytokine which is important in driving cell-mediated immunity and promoting the development of Th1 cells (Sad et al., 1995). This effect may contribute to the preferential induction of Th2 responses, especially by CT, and, in combination with other effects, may explain why LT-IIb is biased more toward inducing Th1 responses.

The precise role of the enzyme activity of the A1 subunit in the adjuvant action of enterotoxins is uncertain. Up-regulation of cAMP in itself is unlikely to account for adjuvant activity, since other agents that accomplish this (e.g., forskolin) do not serve as adjuvants (Wilson et al., 1993). Furthermore, the toxic effect of this activity in epithelial cells, especially in the intestine (which leads to diarrhea), precludes use of the intact holotoxins as adjuvants in humans. Many studies have attempted to identify adjuvant mechanisms which are independent of the catalytic activity of the enterotoxins by the construction of CT or LT-I holotoxin mutants that lack ADP ribosylation and associated toxicity, while retaining adjuvant action in vivo. Although it is unclear how the enzymatically inactive A subunit would support adjuvant action, one such class of mutants, Ser to Lys substitution at position 63 in the A subunit of LT-I (LTK63) or CT (CTK63), maintained the ability to bind the ADP-ribosylation factor (ARF), which is important in vesicular membrane trafficking (Pizza et al., 2001). This ARF-binding activity is independent of the catalytic activity of the A subunit, although it has not yet been linked to any adjuvant mechanism. Conversely, a role for the ADP-ribosylating activity of the A1 subunit is implied by the adjuvant activity of a construct, designated CTA1-DD, in which the toxic A1 subunit of CT is coupled to the Ig-binding domain of staphylococcal protein A (Ågren et al., 1999). However, this construct is targeted to B-cells in a manner different from that of enterotoxin B subunits.

Comparison with intact LT-I showed that LTK63 is unable to up-regulate surface expression of CD80 or to down-regulate expression of CD40 on DC in vitro, whereas LT-I inhibits the ability of DC to present protein antigen to cognate T-cells (Petrovska et al., 2003). In vivo, however, both wild-type LT-I and LTK63 enhance CD80 expression in peritoneal exudate cells, although only LT-I up-regulates CD86 expression in the same system (Ryan et al., 2000). Moreover, LTK63 appeared to promote Th1 responses, in contrast to wild-type LT-I or a partially detoxified mutant (LTR72), both of which promote Th2 responses. Thus, this study (Ryan et al., 2000) suggested that up-regulation of CD86 and promotion of Th2 responses require enzymatic activity. Both CT and its non-toxic E112K mutant were found to up-regulate CD86 on B-cells and macrophages, in contrast to CT-B (Yamamoto et al., 1999). It may thus be inferred that up-regulation of CD86 is a property of the CT-A subunit, independent of its catalytic activity. This conclusion appears to contradict the study by Ryan and coworkers (Ryan et al., 2000), although the apparent discrepancy could be attributable to the use of different enterotoxins (LT-I vs. CT) in different experimental systems. Moreover, similar to intact CT, the non-toxic mutant was found to inhibit Th1-type CD4⁺ T-cell responses selectively, thus promoting Th2-type immune responses. However, it is notable that the E112K mutant of CT here seems to behave differently from the corresponding mutant of LT-1, which lacks adjuvant activity by the oral route (Lycke et al., 1992). This may be attributable to different routes of immunization (intranasal vs. peroral), since an E112K mutant of LT-I, when tested intranasally by an independent group, also exhibited adjuvant properties (Verweij et al., 1998). Alternatively, minor differences between the respective A1 subunits, the properties of their B subunits, or other extrinsic factors may be involved.

Another point mutation (R192G) in LT-I involves the junction between the A1 and A2 segments of the A subunit. Proteolytic cleavage at this junction is necessary to activate the toxin, and after reduction of the disulfide bridge that spans it, this permits dissociation of the A1 subunit from the A2/B subunit to occur within the Golgi/ER of the targeted cells in the mechanism of intoxication (Fujinaga et al., 2003). R192G is thereby rendered insensitive to proteolytic activation, but it is still an effective mucosal adjuvant and is non-toxic at adjuvant doses (Freytag and Clements, 1999).

Although the A2 subunit in enterotoxins is structurally important in linking the A1 subunit to the B pentamer, it is uncertain whether it has any immunomodulating activities. Interestingly, the A2 subunit contains an ER retention signal, the alteration of which (in CT or LT-I mutants) interferes with toxicity (Lencer et al., 1995), and it may therefore contribute to adjuvant action. It is unknown whether the A2 subunit in the SBR-CTA2/B chimeric immunogen (Hajishengallis et al., 1995) adds to the immuno-enhancing effect mediated by the GM1-binding activity of the B subunit. However, since SBR-CTA2/B induces CD86 expression on B-cells, whereas SBR or CT-B alone is inactive (Yamamoto et al., 1999; Martin et al., 2001a), it can be surmised that CT-A2 or the CT-A2/B complex may be involved in the observed immunostimulatory effect.

Immunomodulatory Mechanisms Mediated by the B Subunit
The use of highly purified recombinant B subunits of type I or type II enterotoxins has generated evidence in support of distinct immunomodulatory properties of the B subunits, suggesting that these molecules may offer more than targeting service in vaccine development. CT-B has been shown to enhance the antigen-presenting function of macrophages (Matousek et al., 1996), to up-regulate expression of MHC class II molecules on B-cells (Francis et al., 1992), and to co-stimulate IL-4 induction in Th2 cells (Li and Fox, 1996). However, CT-B does not up-regulate expression of CD80 or CD86 on B-cells or macrophages (Yamamoto et al., 1999). CT-B indirectly favors Th2 responses by inducing apoptosis of CD8⁺ T-cells, which constitute a major source of IFN- production. Depletion of CD8⁺ T-cells by apoptosis as well as up-regulation of MHC class II on B-cells are properties also shared by LT-I-B (Truitt et al., 1998; Williams et al., 1999).

In contrast to CT-B, however, LT-I-B induces IL-10 production and inhibits IL-12 release in monocytes (Simmons et al., 2001), which further favors Th2-type immune responses. In addition, LT-I-B stimulates TNF- release in monocytes (Turcanu et al., 2002). We similarly found that CT-B lacks detectable cytokine-inducing capacity in human monocytic cells or mouse macrophages, whereas in the same cells, the B subunits of type II enterotoxins (LT-IIa-B and LT-IIb-B) induce release of TNF-, IL-6, and IL-1ß (Hajishengallis et al., 2004). Interestingly, the pro-inflammatory activity of these B subunits is antagonized by their respective holotoxins, probably through a cAMP-dependent mechanism. Strikingly, both LT-IIa-B and LT-IIb-B depend upon Toll-like receptor (TLR) 2 for inducing NF-B activation and cytokine release in human cell lines (Hajishengallis et al., 2005). Moreover, macrophages from TLR2-deficient mice fail to release cytokines in response to LT-IIa-B or LT-IIb-B, in contrast to wild-type or TLR4-deficient cells. Therefore, besides their established binding to gangliosides, the B subunits of type II enterotoxins also engage in TLR2 interactions, which may represent a novel mechanism whereby these molecules exert their immunomodulatory activities. Thus, the LT-II B subunits may promote adaptive immune responses through induction of co-stimulatory molecules and immuno-enhancing cytokines by APC, activities that are readily induced through TLR2 activation (Akira and Takeda, 2004).

Signaling Receptors used by Enterotoxins
It has generally been assumed that the immunomodulatory effects of the heat-labile enterotoxins depend upon ganglioside binding, as demonstrated by the finding that a non-binding mutant (G33D) of LT-I-B is deficient in immunogenicity (Nashar et al., 1996; Guidry et al., 1997; Truitt et al., 1998). However, recent evidence suggests that this may not always be the case. For example, the binding-defective point mutant within the B subunit of LT-IIb (T13I) strikingly retains in vivo mucosal adjuvant activity similar to that of the wild-type molecule (Nawar et al., 2005). Certain point mutants of CT-B (H57A) or LT-I-B (H57S) that retain high-affinity binding to GM1 were found defective in inducing apoptosis of CD8⁺ cells (Aman et al., 2001; Fraser et al., 2003). This finding conflicts with the previous demonstration that apoptosis of CD8⁺ cells induced by CT-B or LT-I-B required GM1 binding (Truitt et al., 1998; Williams et al., 1999). It could be speculated that the structural alterations in CT-B (H57A) and LT-I-B (H57S), while not preventing binding to GM1, may preclude interactions with additional receptors required for signaling. Therefore, ganglioside binding alone may not be sufficient to mediate the immunomodulatory effects of type I B subunits.

These observations are consistent with the recently developed concept that cellular activation by microbial molecules involves interactions with several co-operating host receptors within membrane microdomains known as lipid rafts. Gangliosides are known to be present in lipid rafts, which are involved in the formation and function of signaling hetero-oligomeric PRR-TLR complexes (Triantafilou et al., 2004). Multiple receptors may also be involved in cellular activation by LT-IIb-B, which binds to GD1a, leading to NF-B activation and cytokine induction; indeed, these activities are diminished in a GD1a non-binding mutant (T13I) of LT-IIb-B. NF-B activation and cytokine induction by LT-IIb-B are also diminished when TLR2 function is inhibited, either by antibody or genetic deficiency of the receptor (Hajishengallis et al., 2005). It is known that TLR signaling function is facilitated by other PRRs, such as CD14, which mediate direct pathogen recognition (Akira and Takeda, 2004). Thus, it is possible that GD1a provides co-receptor function to TLR2 for cellular activation in response to LT-IIb-B, perhaps by bringing it within range of interaction with TLR2 in lipid rafts, analogous to the way in which CD14 facilitates LPS recognition by TLR4. Alternatively, the propensity of TLR2 to respond to certain lipids attached to proteins or peptides (Takeda et al., 2003) may enable it to recognize GD1a-bound LT-IIb-B as a lipid-protein complex. The concept that gangliosides can cooperate with TLRs is supported by other independent evidence. A glycosphingolipid that binds E. coli P fimbriae through a Gal1-4Galß disaccharide facilitates activation of TLR4 by P fimbriae (Frendéus et al., 2001). Certain gangliosides (GD1a, GD1b, and GT1b among 8 tested) serve as co-receptors with TLR5 for the induction of human ß-defensin-2 in Caco-2 cells by Salmonella enteritidis flagellin (Ogushi et al., 2004). Moreover, asialoGM1 appears to function as a co-receptor for TLR2-dependent NF-B activation and IL-8 induction by Pseudomonas aeruginosa in airway epithelial cells (Soong et al., 2004).

Thus, a full understanding of the mechanisms of heat-labile enterotoxin adjuvant action is not yet available. Effects leading to enhanced immune responses have been observed on DC, macrophages, CD4⁺, and CD8⁺ T-cells, and in B-cells. Gangliosides are not the only cellular receptors involved in host interactions, and TLRs add a new level of complexity; possibly other receptors remain to be identified. An absolute requirement for toxic enzyme activity (ADP ribosylation) by the A1 subunit cannot be sustained. However, it is possible that holotoxins, whether mutant or wild-type, and isolated B subunits exert adjuvant activity in different ways. It is also becoming clear that the different toxins, especially type I and type II, have subtly but importantly different adjuvant effects and modes of action on immunocompetent cells. Some of the main immunological properties of type I and type II enterotoxins are summarized in Table 3.

	TOLERANCE vs. IMMUNITY

TOP ABSTRACT INTRODUCTION TYPE I AND TYPE... MECHANISMS OF ACTION TOLERANCE vs. IMMUNITY APPLICATIONS AND SIGNIFICANCE... REFERENCES

Tolerance induction requires a recombinant B subunit devoid of the toxic enzyme activity of the A1 moiety, and is abrogated by the addition of holotoxin. For example, mice given ovalbumin intragastrically, with or without recombinant LT-B or CT-B or their non-toxic mutants, became tolerant to ovalbumin, whereas native CT or LT blocked tolerance induction (Bagley et al., 2003). Oral immunization with ovalbumin conjugated to recombinant CT-B has also been used to suppress IgE-mediated allergic sensitization to ovalbumin (Rask et al., 2000b). Many studies have used small amounts of intact CT with recombinant CT-B or biochemically purified CT-B containing residual traces of intact toxin to enhance antibody responses or abrogate tolerance to the co-administered antigen (McKenzie and Halsey, 1984; Czerkinsky et al., 1989; Russell and Wu, 1991; Asanuma et al., 1998; Hammond et al., 2001). However, recombinant B subunit alone has been used successfully to induce an immune response to antigen, suggesting that other factors contribute to the tolerogenic outcome (Verweij et al., 1998; Wu and Russell, 1998; Rask et al., 2000a; Wang et al., 2003). It is possible that the properties of the antigen itself contribute to this outcome, since studies demonstrating tolerance have often used auto-antigens, or antigens such as ovalbumin, which readily induces tolerance, whereas microbial antigens result in active immune responses.

Although most studies showing that the recombinant B subunit alone induces tolerance to antigen have used the intragastric route of delivery, it is now clear that "oral tolerance" is a more general phenomenon, since suppression of systemic responses can be induced by application of antigens not only orally, but also intranasally (Waldo et al., 1994), or even intravaginally, at least during estrus (Black et al., 2000). However, the intranasal route of immunization appears to be more efficient than the oral (or intragastric), since smaller doses are required to elicit responses, and the adjuvant effect of detoxified enterotoxin mutants or of their B subunits is more readily demonstrable by this route. Whether this is because of important differences in the uptake and processing of antigens and adjuvants in NALT compared with GALT, or simply because the vaccine materials are less subject to digestion and come into more immediate contact with the inductive site tissues in the nose than in the gastrointestinal tract, is unclear at present.

Immunization via intragastric, intranasal, intravaginal, or transcutaneous routes with admixed or coupled B subunits as adjuvant induces mucosal and systemic immune responses to antigen across many species. Previously, studies have shown that intragastric immunization of mice and intranasal immunization of rats or rhesus monkeys with S. mutans AgI/II, chemically conjugated to or admixed with CT-B, induces antigen-specific antibodies in plasma as well as mucosal secretions like saliva (Russell and Wu, 1991; Katz et al., 1993; Russell et al., 1996). In humans, intranasal delivery of recombinant CT-B admixed with whole killed vibrios elicits CT-B-specific titers in serum, as well as genital and rectal secretions (Kozlowski et al., 2002). Intravaginal or intranasal immunization of mice with CT-B or antigen conjugated to CT-B induces antibodies and antibody-secreting cells in the female genital tract (Johansson et al., 1998). However, by comparison with intranasal immunization, intravaginal immunization with AgI/II-CT-B conjugate plus CT adjuvant is ineffective at disseminating responses among remote mucosal sites (Wu et al., 2000). Novel immunization strategies like transcutaneous immunization have also engaged the B subunits of holotoxins to induce antigen-specific immunity (Hammond et al., 2001).

Furthermore, the method of co-administering antigen and adjuvant may affect the immune response to the antigen. For example, the tolerance observed with the use of antigen admixed with or chemically coupled to recombinant B subunits might be overcome by the use of genetic coupling of B subunits to antigen. Chimeric proteins prepared by genetic replacement of the toxic A1 moiety of CT with an antigen segment of choice have been found to be immunogenic by either intragastric or intranasal routes, without a requirement for additional toxin adjuvant (Hajishengallis et al., 1995; Sultan et al., 1998; Martin et al., 2001a; Sheoran et al., 2002; Li et al., 2004; Gockel and Russell, 2005). Using different techniques, others have genetically coupled peptides to recombinant CT-B and have shown that intranasal immunization induces antigen-specific antibody production in the liver, lung, and serum, as well as reduces T-cell-mediated pathology in schistosome-parasitized mice (Lebens et al., 2003).

	APPLICATIONS AND SIGNIFICANCE WITH RESPECT TO ORAL DISEASE

TOP ABSTRACT INTRODUCTION TYPE I AND TYPE... MECHANISMS OF ACTION TOLERANCE vs. IMMUNITY APPLICATIONS AND SIGNIFICANCE... REFERENCES

 
An extensive body of literature exists on the use of enterotoxins, especially CT and LT-I or their non-toxic derivatives, as mucosally applied adjuvants for numerous antigens and experimental vaccines against a wide range of infections. Among these, and the focus of much of our work, is the application of these approaches for the development of a vaccine against dental caries (Russell, 1992; Russell et al., 1999, 2004; Smith, 2002). Thus, we have demonstrated ‘proof of concept’, in rodent and monkey models, that intragastric or intranasal immunization with AgI/II (or its SBR component), coupled to CT-B, either chemically or genetically induces salivary IgA antibodies against AgI/II that have protective potential, and protects against oral challenge with S. mutans and inhibits the development of caries (Hajishengallis et al., 1992, 1998; Katz et al., 1993; Russell et al., 1996). Further studies have shown that responses induced by these approaches to immunization can persist and be recalled by booster immunization during the lifetime of a mouse (Hajishengallis et al., 1996b; Harokopakis et al., 1997; Harrod et al., 2001; Gockel and Russell, 2005). Attempts to apply the same approach to glucosyltransferase (GTF), which is another prime candidate antigen for a vaccine against caries, have been limited by the inability of E. coli to express the cloned components in soluble form (Jespersgaard et al., 1999). However, expression of GTF and SBR components, possibly together with CT-B, in a live attenuated vector organism might permit these constructs to be used effectively in a vaccine (Hajishengallis et al., 1996a; Jespersgaard et al., 2001).

Smith and colleagues have shown, in rats, that serum IgG and salivary IgA antibody responses to synthetic peptides derived from GTF sequences can be induced by the co-administration of CT, or the R192G mutant of LT-I, as an intranasal adjuvant (Smith et al., 2001). The same adjuvants, given together with microencapsulated GTF by the rectal route, enhanced salivary IgA antibody responses that were associated with diminished development of caries lesions after oral challenge with S. sobrinus (Smith et al., 2003).

Others have used CT or LT-I, or their B subunits, either as adjuvants or as coupling agents or genetic fusion partners for peptides derived from antigens of mutans streptococci. The initial demonstration of intranasal immunization by S. mutans AgI/II (PAc) was augmented by CT-B as an adjuvant (Takahashi et al., 1990). Subsequently, the same group showed that intranasal immunization of mice with a synthetic peptide representing residues 301–319 from S. mutans PAc, coupled to CT-B, induced serum IgG antibodies to the parent antigen and suppressed oral colonization by S. mutans (Takahashi et al., 1991). A fusion protein representing a peptide from GTF coupled to the N-terminus of CT-B retained the GM1-binding capacity of CT-B and was orally immunogenic in mice (Dertzbaugh et al., 1990). Even large protein antigens, such as SpaA and dextranase from S. sobrinus, have been successfully fused to the C-terminus of LT-I-B by the interposition of a 6-residue linker sequence containing 2 prolines, and the fusion proteins retained the antigenicity of the S. sobrinus antigens as well as some GM1-binding activity of the LT-I-B (Jagusztyn-Krynicka et al., 1993).

Similarly, several studies have demonstrated that immune responses to antigens of periodontal pathogens can be induced by oral or intranasal immunization with CT, LT-I, or LT-II as adjuvant. Examples include Porphyromonas gingivalis fimbriae (Connell et al., 1998; Nagasawa et al., 1999; Yanagita et al., 1999), outer membrane protein of P. gingivalis (Namikoshi et al., 2003), and Actinobacillus actinomycetemcomitans polysaccharide-protein conjugate or fimbrial peptide (Takamatsu-Matsushita et al., 1996; Honma et al., 1999), but these studies assessed the outcomes only in terms of antibody or cellular immune responses. A recent study, however, has reported that intranasal immunization of mice with a recombinant hemagglutinin/adhesin domain of P. gingivalis gingipain (Kgp), either mixed with or conjugated to CT-B, resulted in enhanced serum and mucosal antibody responses to Kgp, and that the antibodies could inhibit invasion of epithelial cells by P. gingivalis (Zhang et al., 2005).

Most applications of enterotoxin-enhanced mucosal immunization have focused on the generation of IgA antibody responses in secretions, and an enormous body of literature has accumulated in which enhanced immune responses to a wide variety of microbial antigens have been demonstrated in this way. It should be noted that circulating IgG antibodies are also effectively stimulated by these approaches, thereby expanding the range of infections that can potentially be addressed by mucosal immunization (Lycke et al., 1983; Jackson et al., 1993). Cell-mediated immunity, including the generation of cytotoxic T lymphocytes (CTL), has been less frequently studied in this context; however, it is noteworthy that CTL with activity against HIV-infected cells can be induced by mucosal immunization with HIV peptides administered together with CT or LT(R192G) (Porgador et al., 1997; Belyakov et al., 2000).

Given that enterotoxins and their B subunits profoundly modulate immune responses, even responses that have already been established by previous immunization, it is tempting to speculate that these adjuvants, or antigens coupled to enterotoxin B subunits, might be used to ameliorate chronic inflammatory disease (Russell, 2003). Periodontal disease (Russell, 1998; Russell and Sibley, 1999), inflammatory bowel disease, and Helicobacter pylori infection (Raghavan et al., 2002) are examples of infection-driven inflammatory conditions that might be amenable to such an approach. In these diseases, pathology arises, at least in part, from an inappropriate, or dysregulated, immune response to the provoking bacteria, that not only fails to eliminate the pathogen but also inflicts and perpetuates inflammatory damage to host tissues. The validity of this approach is strengthened by the findings, discussed above, that CT and the type II enterotoxins, or their B subunits, have differential effects on CD4⁺, CD8⁺, and B lymphocytes that may be amenable to manipulation aimed at reprogramming the immune response to a mode that destroys the pathogen and alleviates damage to host tissues. Whether this can be successfully achieved remains to be determined. Findings that adverse immune responses in animal models of auto-immune disease or allergy can be corrected by the administration of the relevant antigen coupled to CT-B suggest that it might be possible.

Human application of enterotoxins or their derivatives in vaccines raises questions of safety, particularly with respect to the holotoxins themselves. It is noteworthy that oral administration of pure or recombinant CT-B has a record of safety and acceptability, having been used as a component of a licensed oral vaccine against cholera (Van Loon et al., 1996). However, concerns have arisen recently over intranasal administration of enterotoxins, because of their potential for retrograde trafficking through the olfactory nerve, which offers a direct pathway into the brain without an intervening synapse (Van Ginkel et al., 2000). An intranasal vaccine against influenza "adjuvanted" with a small dose of LT-I was withdrawn because of its association with increased incidence of Bell’s palsy, although the mechanism underlying this adverse effect was unclear (Mutsch et al., 2004). It remains to be determined whether these potential problems can be circumvented by the use of B subunits or non-toxic derivatives of enterotoxins as adjuvants or carriers for intranasal administration.

	ACKNOWLEDGMENTS

 
Studies in the authors’ laboratories are supported by US-PHS grants from the National Institute of Dental and Craniofacial Research: DE015254 (GH), DE013833 (TDC), and DE006746 (MWR); and in part (SA) by the Ralph Hochstetter Medical Research Fund in honor of Dr. Henry C. and Bertha H. Buswell.

	FOOTNOTES

 
⁵ Present address, University of Louisville, Center for Oral Health and Systemic Disease, Louisville, KY, USA;

Received March 31, 2005; Accepted August 3, 2005

	REFERENCES

TOP ABSTRACT INTRODUCTION TYPE I AND TYPE... MECHANISMS OF ACTION TOLERANCE vs. IMMUNITY APPLICATIONS AND SIGNIFICANCE... REFERENCES

 
Ågren LC, Ekman L, Löwenadler B, Nedrud JG, Lycke NY (1999). Adjuvanticity of the cholera toxin A1-based gene fusion protein, CTA1-DD, is critically dependent on the ADP-ribosyltransferase and Ig-binding activity. J Immunol 162:2432–2440.[Abstract/Free Full Text]

Åhrén C, Jertborn M, Svennerholm A-M (1998). Intestinal immune responses to an inactivated oral enterotoxigenic Escherichia coli vaccine and associated immunoglobulin A responses in blood. Infect Immun 66:3311–3316.[Abstract/Free Full Text]

Akira S, Takeda K (2004). Toll-like receptor signalling. Nat Rev Immunol 4:499–511.[ISI][Medline]

Aman AT, Fraser S, Merritt EA, Rodigherio C, Kenny M, Ahn M, et al. (2001). A mutant cholera toxin B subunit that binds GM1-ganglioside but lacks immunomodulatory or toxic activity. Proc Natl Acad Sci USA 98:8536–8541.[Abstract/Free Full Text]

Arce S, Nawar H, Russell MW, Connell TD (2005). Differential binding of Escherichia coli enterotoxins LT-IIa, LT-IIb and of cholera toxin elicits differences in apoptosis, proliferation, and activation in lymphoid cells. Infect Immun 73:2718–2727.[Abstract/Free Full Text]

Arpin C, Déchanet J, Van Kooten C, Merville P, Grouard G, Brière F, et al. (1995). Generation of memory B cells and plasma cells in vitro. Science 268:720–722.[Abstract/Free Full Text]

Asanuma H, Aizawa C, Kurata T, Tamura S (1998). IgA antibody-forming cell responses in the nasal-associated lymphoid tissue of mice vaccinated by intranasal, intravenous and/or subcutaneous administration. Vaccine 16:1257–1262.[ISI][Medline]

Bagley KC, Abdelwahab SF, Tuskan RG, Lewis GK (2003). An enzymatically active A domain is required for cholera-like enterotoxins to induce a long-lived blockade on the induction of oral tolerance: new method for screening mucosal adjuvants. Infect Immun 71:6850–6856.[Abstract/Free Full Text]

Belyakov IM, Ahlers JD, Clements JD, Strober W, Berzofsky JA (2000). Interplay of cytokines and adjuvants in the regulation of mucosal and systemic HIV-specific CTL. J Immunol 165:6454–6462.[Abstract/Free Full Text]

Bergerot I, Ploix C, Petersen J, Moulin V, Rask C, Fabien N, et al. (1997). A cholera toxoid-insulin conjugate as an oral vaccine against spontaneous autoimmune diabetes. Proc Natl Acad Sci USA 94:4610–4614.[Abstract/Free Full Text]

Black CA, Rohan LC, Cost M, Watkins SC, Draviam R, Alber S, et al. (2000). Vaginal mucosa serves as an inductive site for tolerance. J Immunol 165:5077–5083.[Abstract/Free Full Text]

Braun MC, He JP, Wu CY, Kelsall BL (1999). Cholera toxin suppresses interleukin (IL)-12 production and IL-12 receptor b1 and b2 chain expression. J Exp Med 189:541–552.[Abstract/Free Full Text]

Bromander A, Holmgren J, Lycke N (1991). Cholera toxin stimulates IL-1 production and enhances antigen presentation by macrophages in vitro. J Immunol 146:2908–2914.[Abstract]

Cassese G, Arce S, Hauser AE, Lehnert K, Moewes B, Mostarac M, et al. (2003). Plasma cell survival is mediated by synergistic effects of cytokines and adhesion-dependent signals. J Immunol 171:1684–1690.[Abstract/Free Full Text]

Challacombe SJ, Tomasi TB (1980). Systemic tolerance and secretory immunity after oral immunization. J Exp Med 152:1459–1472.[Abstract/Free Full Text]

Clemens JD, van Loon F, Sack DA, Chakraborty J, Rao MR, Ahmed F, et al. (1991). Field trial of oral cholera vaccines in Bangladesh: serum vibriocidal and antitoxic antibodies as markers of the risk of cholera. J Infect Dis 163:1235–1242.[ISI][Medline]

Clements JD (1990). Construction of a nontoxic fusion peptide for immunization against Escherichia coli strains that produce heat-labile and heat-stable enterotoxins. Infect Immun 58:1159–1166.[Abstract/Free Full Text]

Cong Y, Weaver CT, Elson CO (1997). The mucosal adjuvanticity of cholera toxin involves enhancement of costimulatory activity by selective up-regulation of B7.2 expression. J Immunol 159:5301–5308.[Abstract]

Connell TD, Metzger DJ, Wang M, Jobling MG, Holmes RK (1995). Initial studies of the structural signal for extracellular transport of cholera toxin and other proteins recognized by Vibrio cholerae. Infect Immun 63:4091–4098.[Abstract]

Connell TD, Metzger D, Sfintescu C, Evans RT (1998). Immunostimulatory activity of LT-IIa, a type II heat-labile enterotoxin of Escherichia coli. Immunol Lett 62:117–120.[ISI][Medline]

Critchley DR, Magnani JL, Fishman PH (1981). Interaction of cholera toxin with rat intestinal brush border membranes. Relative roles of gangliosides and galactoproteins as toxin receptors. J Biol Chem 256:8724–8731.[Free Full Text]

Czerkinsky C, Russell MW, Lycke N, Lindblad M, Holmgren J (1989). Oral administration of a streptococcal antigen coupled to cholera toxin B subunit evokes strong antibody responses in salivary glands and extramucosal tissues. Infect Immun 57:1072–1077.[Abstract/Free Full Text]

Dertzbaugh MT, Elson CO (1993). Reduction in oral immunogenicity of cholera toxin B subunit by N-terminal peptide addition. Infect Immun 61:384–390.[Abstract/Free Full Text]

Dertzbaugh MT, Peterson DL, Macrina FL (1990). Cholera toxin B-subunit gene fusion: structural and functional analysis of the chimeric protein. Infect Immun 58:70–79.[Abstract/Free Full Text]

Elson CO, Dertzbaugh MT (2005). Mucosal adjuvants. In: Mucosal immunology. 3rd ed. Mestecky J, Bienenstock J, Lamm ME, Mayer L, Strober W, McGhee JR, editors. San Diego: Elsevier/Academic Press, pp. 967–986.

Elson CO, Holland SP, Dertzbaugh MT, Cuff CF, Anderson AO (1995). Morphologic and functional alterations of mucosal T cells by cholera toxin and its B subunit. J Immunol 154:1032–1040.[Abstract]

Finkelstein RA, Burks MF, Zupan A, Dallas WS, Jacob CO, Ludwig DS (1987). Epitopes of the cholera family of enterotoxins. Rev Infect Dis 9:544–561.[ISI][Medline]

Fishman PH (1986). Recent advances in identifying the functions of gangliosides. Chem Phys Lipids 42:137–151.[ISI][Medline]

Francis ML, Ryan J, Jobling MG, Holmes RK, Moss J, Mond JJ (1992). Cyclic AMP-independent effects of cholera toxin on B cell activation. II. Binding of ganglioside GM1 induces B cell activation. J Immunol 148:1999–2005.[Abstract]

Fraser SA, de Haan L, Hearn AR, Bone HK, Salmond RJ, Rivett AJ, et al. (2003). Mutant Escherichia coli heat-labile toxin B subunit that separates toxoid-mediated signaling and immunomodulatory action from trafficking and delivery functions. Infect Immun 71:1527–1537.[Abstract/Free Full Text]

Freeman GJ, Boussiotis VA, Anumanthan A, Bernstein GM, Ke XY, Rennert PD, et al. (1995). B7-1 and B7-2 do not deliver identical costimulatory signals, since B7-2 but not B7-1 preferentially costimulates the initial production of IL-4. Immunity 2:523–532.[ISI][Medline]

Frendéus B, Wachtler C, Hedlund M, Fischer H, Samuelsson P, Svensson M, et al. (2001). Escherichia coli P fimbriae utilize the Toll-like receptor 4 pathway for cell activation. Mol Microbiol 40:37–51.[ISI][Medline]

Freytag LC, Clements JD (1999). Bacterial toxins as mucosal adjuvants. Curr Top Microbiol Immunol 236:215–236.[ISI][Medline]

Fujinaga Y, Wolf AA, Rodighiero C, Wheeler H, Tsai B, Allen L, et al. (2003). Gangliosides that associate with lipid rafts mediate transport of cholera and related toxins from the plasma membrane to endoplasmic reticulum. Mol Biol Cell 14:4783–4793.[Abstract/Free Full Text]

Fukuta S, Magnani JL, Twiddy EM, Holmes RK, Ginsburg V (1988). Comparison of the carbohydrate-binding specificities of cholera toxin and Escherichia coli heat-labile enterotoxins LTh-I, LTIIa, and LTIIb. Infect Immun 56:1748–1753.[Abstract/Free Full Text]

Gill DM, Clements JD, Robertson DC, Finkelstein RA (1981). Subunit number and arrangement in Escherichia coli heat-labile enterotoxin. Infect Immun 33:677–682.[Abstract/Free Full Text]

Gockel CM, Russell MW (2005). Induction and recall of immune memory by mucosal immunization with a non-toxic recombinant enterotoxin-based chimeric protein. Immunology (in press).

Gockel CM, Bao SS, Beagley KW (2000). Transcutaneous immunization induces mucosal and systemic immunity: a potent method for targeting immunity to the female reproductive tract. Mol Immunol 37:537–544.[ISI][Medline]

Guidry JJ, Cárdenas L, Cheng E, Clements JD (1997). Role of receptor binding in toxicity, immunogenicity, and adjuvanticity of Escherichia coli heat-labile enterotoxin. Infect Immun 65:4943–4950.[Abstract]

Guth BE, Pickett CL, Twiddy EM, Holmes RK, Gomes TA, Lima AA, et al. (1986). Production of type II heat-labile enterotoxin by Escherichia coli isolated from food and human feces. Infect Immun 54:587–589.[Abstract/Free Full Text]

Hajishengallis G, Nikolova E, Russell MW (1992). Inhibition of Streptococcus mutans adherence to saliva-coated hydroxyapatite by human secretory immunoglobulin A (S-IgA) antibodies to cell surface protein antigen I/II: reversal by IgA1 protease cleavage. Infect Immun 60:5057–5064.[Abstract/Free Full Text]

Hajishengallis G, Hollingshead SK, Koga T, Russell MW (1995). Mucosal immunization with a bacterial protein antigen genetically coupled to cholera toxin A2/B subunits. J Immunol 154:4322–4332.[Abstract]

Hajishengallis G, Harokopakis E, Hollingshead SK, Russell MW, Michalek SM (1996a). Construction and oral immunogenicity of a Salmonella typhimurium strain expressing a streptococcal adhesin linked to the A2/B subunits of cholera toxin. Vaccine 14:154