Direct exposure of mouse ovaries and oocytes to high doses of an adenovirus gene therapy vector fails to lead to germ cell transduction.

The risk of insertion of adenovirus gene therapy DNA into female germ cells during the course of somatic gene therapy was stringently tested in the mouse by injecting up to 10(10) infectious particles directly into the ovary and by incubating naked oocytes in a solution of 2 x 10(8) particles/ml for 1 h prior to in vitro fertilization (IVF). The vector used was a recombinant adenovirus carrying the bacterial lacZ gene driven by the cytomegalovirus promoter (Adbeta-gal). Ovaries were stained for LacZ activity, or immunochemically for LacZ, 5-7 days after injection. Although very large amounts of LacZ activity and protein were detected, all positive staining was in the thecal portion of the ovary, with no staining seen in oocytes. In another series of experiments, mice with injected ovaries were mated, and preimplantation embryos or fetuses were analyzed either for LacZ expression or by PCR for lacZ DNA. None of 202 preimplantation embryos stained positively for LacZ and none of 58 fetuses were positive for DNA by PCR analysis. Finally, more than 1400 eggs were fertilized after exposure to the vector prior to IVF and stained as morulae for LacZ activity. Fewer than 2% of the embryos stained positively for LacZ, and experiments indicated that the staining was due to incomplete washing of the eggs prior to IVF. These data provide strong evidence that adenoviruses cannot infect oocytes and that the risk of female germ-line transduction with such vectors is very low.