Optimal techniques to immunize mucosal surfaces have not yet been devised. Nonreplicating vaccines, such as toxins and pili (fimbriae) offer the advantages of safety and specificity, but are often not sufficiently immunogenic. Earlier studies have shown that immunogenicity of cholera toxin (CT) and its B subunit (CT- B) is based on route of administration, dose, receptor-binding ability, molecular size and ability to stimulate cyclic AMP. Because CT has several unique properties, this project proposes to heneralize the findings on CT to a study of E. coli pili using the rabbit model of RDEC-1 strain disease. Thus, different forms of the RDEC-1 pilus (AF/R1) antigen (intact pili, pilus, subunits, and the live piliated bacillus) will be used to immunize rabbits by various routes and doses, and the IgA- and IgG-specific mucosal and systemic responses will be measured in serum, bile and intestinal lavage specimens. Our recent studies have sought distinguishing features between antigens based on how the host handles antigen absorbed from the intestinal lumen. We found that CT was absorbed from the lumen into thoracic duct lymph and that lymph transport was greater than portal blood transport. There is evidence that in nonimmune rats lymph-borne CT was complexed to its membrane receptor GM1 ganglioside, and probably entered lymph through Peyer's patches. Further studies in the rate model of CT immunity will seek to confirm uptake of CT through Peyer's patches, and confirm the presence of CT-ganglioside complexes in lymph. To test the relevance of lymph-borne antigen, the quantity and form of CT found in lymph after intraduodenal administration will be infused intravenously into nonimmune recipients and subsequent immunity measured. A major effort will relate the different antigen forms of CT (CT- B, toxoid, procholeragenoid) with their uptake into Peyer's patches and their immunogenicity. Pilus antigen will also be sought in Peyer's patches using radiolabelled affinity-purified heterosera and monoclonal antibodies, including a hybridoma recognizing the pilus quaternary structure. The ultimate goal of the project is the correlation between the characteristics of antigen handling, such as antigen absorption into Peyer's patches and lymph, and mucosal immunity to nonreplicating vaccine antigens.
