ORAL TOLERIZATION TO ADENOVIRAL PROTEINS PERMITS REPEATED ADENOVIRUS-MEDIATED GENE-THERAPY IN RATS WITH PREEXISTING IMMUNITY TO ADENOVIRUSES

Exposure to wild-type adenoviruses is common in humans and results in immune response against adenoviruses. The pre-existing antibodies and a strong secondary humoral and cellular immune response would interfer e with gene transfer using recombinant adenoviral vectors. To test whe ther the secondary immune response can be abrogated by oral tolerizati on to adenoviral antigens, we immunized bilirubin -UDP- glucuronosyltr ansferase (BUGT)- deficient jaundiced Gunn rats with a recombinant ade novirus (5 x 10(9) pfu/rat) expressing the human UDP- glucouronosyltra nsferase (BUGT(1)) gene (Ad-hBUGT). Transgene expression was shown by reduction of mean serum bilirubin levels from 7.0 mg/dL to 2.3 mg/dL i n 14 days, which then increased gradually to pretreatment. levels in 6 weeks. All recipients developed antibodies (1:2(10)) and cytotoxic ly mphocytes against the adenovirus. For oral tolerization, we administer ed to the immunized rats protein extracts of a recombinant adenovirus type 5 (1-1.5 mg/day) via duodenostomy tubes 10 to 40 days after the i nitial virus injection; control rats received bovine serum albumin. In rats fed adenoviral proteins and the BSA-fed controls, the antibody t iters decreased to 1:2(7) and 1:2(9), respectively, in 70 days. Lympho cytes from the tolerized rats expressed TGF-beta(1) upon exposure to a ntigen-presenting cells primed with adenoviral antigens, whereas IFN-g amma expression was undetectable. In contrast, lymphocytes from the BS A-treated control rats expressed IFN-gamma but not transforming growth factor beta(1) (TGF-beta(1)). Seventy days after the first injection in the orally tolerized rats, but not in the controls, a second Ad-hBU GT injection caused human BUGT(1) expression again, reducing serum bil irubin levels to those observed after the first injection. In the tole rized rats, serum antibody titers and anti-adenoviral cytotoxic lympho cyte activities continued to decline despite the second injection, whe reas the antibody levels were boosted in the non-tolerized group. This results show that by preventing the secondary booster response, oral tolerization permits repeated adenovirus-directed gene transfer despit e the presence of a residual antibody titer from a previous adenoviral exposure.