An investigation into the targeting of vaccines to the colon

The colon has a unique immunological environment which makes it an interesting target for vaccine delivery. In addition, it has a near neutral pH and low proteolytic activity (favourable for antigen delivery). Before colonic immunisation was attempted in the laboratory using rodent models, these latter were characterised in terms of lymphoid tissue, gastrointestinal pH and gastrointestinal water content. There were more lymphoid follicles per centimetre in the colon of mice and rats than in the small intestine. The mean intestinal pH was lower in rodents than that in man, being less than pH 5.2 in the mouse and less than pH 6.6 in the rat. The mouse and rat had less than 1ml and 8ml of water respectively in the gastrointestinal tract. Immunisation studies were carried out in mice using ovalbumin as a model antigen (encapsulated in sustained release PLGA nanoparticles). The hypothesis was that delivering a vaccine to the colon would give different responses to orally administered antigen (antigen expected to be taken up by the small intestine) due to different immunological environments. Intramuscular and rectal deliveries were investigated as controls. Generally colonic administration gave levels of mucosal antibody (measured by IgA in faecal material) which were higher than those produced by oral or intramuscular delivery, and produced high levels of antibody on the vaginal mucosa. Following booster doses, colonic delivery also produced high levels of IgG in the serum which were similar to intramuscular levels. Subsequently, ways of delivering vaccines to the colon by oral administration were investigated. Amylose, which is degraded by colonic microbial enzymes, did not form microparticles except after exposure to high temperatures and was therefore unsuitable. Chitosan did form microparticles, and its digestion by human colonic bacteria was tested to determine its applicability for colon-specific targeting. Addition of the crosslinking agent tripolyphosphate to chitosan inhibited pancreatic (porcine) digestion but not human faecal digestion, and also reduced swelling which suggests its potential for colon-specific dosage forms in man. However, microparticles prepared from chitosan and tripolyphosphate were unable to control the release of a model drug (prednisolone) in low pH (stomach) conditions. Consequently, the exploitation of the changes in pH along the gastrointestinal tract was investigated for colonic targeting. Polymers were chosen which could be tested in rodents but the principle extrapolated to man. Hydroxypropylmethylcellulose phthalate polymers (dissolution threshold of pH 5 or 5.5) were identified as having potential for testing colonic-specific targeting of vaccines in the mouse. Vaccine nanoparticles, as used for immunisation, were successfully entrapped within the pH-responsive microparticles, and these nanoparticles can be released when pH 5 or 5.5 is reached. Further work is required to establish the in vivo release site of these microparticles, before orally administered colon-specific vaccination can be successfully tested