In order to study patterns of local antibody responses following mucosal immunization of mice via different routes, a method for collection of secretions directly from mucosal surfaces was developed. Mice were immunized on days 0, 10, 17, and 24 by administration of cholera toxin into the oral cavity, stomach, colon-rectum, or vagina. At sacrifice on day 32, absorbent wicks were placed in the oral cavity and, via an applicator tube, into the vagina and distal colon-rectum and along the entire small intestine after flushing of luminal contents. Protein was quantitatively extracted from wicks, and specific anti-cholera toxin immunoglobulin A (IgA) and IgG were measured by enzyme-linked immunosorbent assay. Concentrations of specific IgA in secretions at various mucosal sites were dramatically influenced by the route of immunization. Oral immunization effectively induced IgA in saliva, and the intragastric route was optimal for induction of IgA in the small intestine. High levels of specific IgA appeared on the colonic-rectal mucosal surface only after rectal delivery of antigen. Oral, gastric, and rectal immunizations also produced distant responses in the vagina. Following vaginal immunization, however, neither local nor distant IgA responses were detected. These results suggest that vaccines intended for protection of colonic-rectal and vaginal mucosal surfaces might best be administered by the rectal route.

Secretory immunoglobulin A (IgA) antibodies (sIgA) directed against cholera toxin (CT) and surface components of Vibrio cholerae are associated with protection against cholera, but the relative importance of specific sIgAs in protection is unknown. A monoclonal IgA directed against the V. cholerae lipopolysaccharide (LPS), secreted into the intestines of neonatal mice bearing hybridoma tumors, was previously shown to provide protection against a lethal oral dose of 107 V. cholerae cells. We show here that a single oral dose of 5 to 50 Lg of the monoclonal anti-LPS IgA, given within 2 h before V. cholerae challenge, protected neonatal mice against challenge. In contrast, an oral dose of 80 tLg of monoclonal IgA directed against CT B subunit (CTB) failed to protect against V. cholerae challenge. A total of 80 pg of monoclonal anti-CTB IgA given orally protected neonatal mice from a lethal (5-.g) oral dose of CT. Secretion of the same anti-CTB IgA antibodies into the intestines of mice bearing IgA hybridoma backpack tumors, however, failed to protect against lethal oral doses of either CT (5 jg) or V. cholerae (107 cells). Furthermore, monoclonal anti-CTB IgA, either delivered orally or secreted onto mucosal surfaces in mice bearing hybridoma tumors, did not significantly enhance protection over that provided by oral anti-LPS IgA alone. These results demonstrate that anti-LPS sIgA is much more effective than anti-CT IgA in prevention of V. chokerae-induced diarrheal disease.

Secretory immunoglobulin A (IgA) antibodies directed against cholera toxin (CT) are thought to be important in resistance to oral challenge with virulent Vibrio cholerae, although alternative mechanisms for protection of intestinal epithelia against CT-induced fluid secretion have been proposed. The ability of anti-CT IgA to block the effects of CT on human enterocytes has not been directly tested because of the lack of a well-defined in vitro intestinal epithelial cell system to directly measure toxin action and the limited availability of purified anti-CT IgA antibodies. We have generated hybridomas that produce monoclonal IgA and IgG antibodies directed against CT by fusion of Peyer's patch cells with mouse myeloma cells after oral-systemic immunization of mice with CT and CT B-subunit protein. All of the anti-CT antibodies recognized the B subunit. Three clones (designated anti-CTB IgA-1, IgA-2, and IgA-3) which produced IgA antibodies in dimeric and polymeric forms were selected. Checkerboard immunoblotting demonstrated that IgA-1 recognized an epitope distinct from that recognized by IgA-2 and IgA-3 and that none of the antibodies were directed against the binding site of GM1, the intestinal cell membrane toxin receptor. The protective capacity of these IgAs was tested in vitro with human T84 colon carcinoma cells grown on permeable supports as confluent monolayers of polarized enterocytes. When each anti-CTB IgA was mixed with 10 nM CT and applied to the apical surfaces of T84 cell monolayers, all three IgAs blocked CT-induced Cl- secretion in a dose-dependent manner and completely inhibited binding of rhodamine-labelled CT to apical cell membranes. Thus, monoclonal anti-CTB IgA antibodies are sufficient to protect human enterocytes in vitro against CT binding and action.