Induction of nuclear catechol-O-methyltransferase by estrogens in hamster kidney: implications for estrogen-induced renal cancer.

Catecholestrogens are postulated to contribute to carcinogenesis by causing DNA damage mediated by reactive oxygen species generated during redox cycling between catechol and quinone estrogens, and by quinone estrogens that can form depurinating adducts. The above hypothesis is based principally on studies of the cancers that develop in renal cortex of hamsters treated with primary estrogens: Hamster kidney can catalyze 2- and 4-hydroxylation of estrogens and support their redox cycling, and the kidneys of estradiol-treated hamsters show evidence of oxidative cellular and DNA damage. Here we used immunocytochemisty to test the postulate that catechol-O-methyltransferase (COMT), the enzyme that can prevent oxidation of catecholestrogens to their quinone derivatives, would be induced in renal cortex of hamsters treated with estradiol or ethinyl estradiol. In kidneys of control hamsters, COMT was localized in cytoplasm of epithelial cells of proximal convoluted tubules, predominantly in the juxtamedullary region where the estrogen-induced cancers arise. After 2- or 4-weeks of treatment with either estrogen, COMT was seen in epithelial cells of proximal convoluted tubules throughout the cortex, and many cells also showed intense nuclear COMT immunoreactivity. Estradiol-induced renal cancers were negative for COMT, but were surrounded by tubules with intense cytoplasmic and nuclear immunostaining. The nucleus-associated COMT was shown by immunoblot analysis to be the soluble form of the enzyme. Using reverse transcription-polymerase chain reaction amplification, hamster kidney COMT was shown to lack the putative nuclear localization signal sequence present in human COMT. A second phase II enzyme, CuZn-superoxide dismutase (CuZnSOD), was shown by immunocytochemistry to remain extranuclear in proximal convoluted tubules of estrogen-treated hamsters, which indicates entry of COMT into the nucleus to be selective. The findings are consistent with the catechol/quinone estrogen hypothesis of estrogen-induced cancer, while the translocation of the enzyme to the nucleus following estrogen treatment suggests a response to a threat to the genome by electrophilic products of catechols.