To assess the estrogenic potency of benzophenone by in-vivo uterotrophic assay, we gave orally either the compound (100 or 400 mg/kg) or 17beta-estradiol (0.2 mg/kg) as a positive control, once per day for 3 days, to ovariectomized Sprague-Dawley (SD) rats, and all rats were killed 24 h after being given the last dose. The high dose of benzophenone elicited an approximately 1.9-fold increase in absolute and relative uterine weight, and 17beta-estradiol increased uterine weight approximately fivefold relative to the control. The uterine response caused by both compounds was accompanied by an increase in luminal epithelium height and stromal cell numbers in the uterus and an increase in the thickness of vaginal epithelium cell layers with cornification. At 24 h after the last dose, the mean serum concentrations of benzophenone, benzhydrol and p-hydroxybenzophenone in the high-dosed rats were 10.4+/-1.0, 1.5+/-0.3, and 0.7+/-0.2 (mean +/- SE) micro mol/l, respectively, whereas in the serum of low-dosed rats these compounds were not detected. When a single oral administration of benzophenone (100 or 400 mg/kg) was given to intact female rats, serum concentrations of benzophenone, benzhydrol and p-hydroxybenzophenone increased in a dose-dependent manner 6 h later. Previously, Nakagawa et al. (2000) and Nakagawa and Tayama (2001) reported that the subcutaneous injection of p-hydroxybenzophenone into juvenile female rats elicited estrogenic activity in reproductive organs, whereas neither benzophenone nor benzhydrol had such an effect. In addition, p-hydroxybenzophenone itself rather than the parent compound caused a proliferation of estrogen receptor-positive MCF-7 cells in vitro. Based on these findings, it is apparent that the pro-estrogenic compound benzophenone requires biotransformation to p-hydroxybenzophenone, a metabolite with intrinsic hormonal activity.