Three practical diets were formulated to contain 0, 500, or 1000 ppm genistein. The three diets were distributed for 1 year to groups of rainbow trout undergoing their first gametogenesis and until spawning. Growth performance of rainbow trout was not affected by dietary treatments. Plasma cholesterol levels were equivalent between groups. In males, a slight but constant induction of vitellogenin (VTG) synthesis and a decrease in testosterone levels were observed. A slight decrease in plasma levels of betaFSH and betaLH was noticed at the end of spermatogenesis in the male fish fed a diet with 500 ppm (genistein) (from 2.16 +/- 0.39 to 1.47 +/- 0.23 for betaFSH and from 0.44 +/- 0.09 to 0.31 +/- 0.09 for betaLH). There was a significantly reduced 17alpha,20beta(OH)(2)-progesterone (from 10.93 +/- 0.88 in control to 5.46 +/- 0.92 in males and from 251.22 +/- 21.40 to 183.22 +/- 13.48 in females). Testicular development was accelerated in genistein-fed fish, and sperm motility and concentration were decreased in a dose-dependent manner at spawning. In females, a significant increase in plasma VTG occurred only at the beginning and at the end of oogenesis. Testosterone levels were decreased at the beginning of oogenesis. Both betaFSH and betaLH were decreased by genistein (from 6.38 +/- 1.55 to 3.44 +/- 0.82 for betaFSH and from 15.18 +/- 3.00 to 6.93 +/- 0.99 for betaLH in females), whereas spawning was delayed only in females fed the diet with 500 ppm of genistein. Gamete quality was impaired only in this group, as underlined by a lower percentage of ovulating females (from 100 to 79% at the end of the trial), a lower fertilization rate, and a lower viability of fry. These results may be explained by the agonistic/antagonistic effect of genistein on estrogen function related to the tissue ratio between endogenous estrogens/genistein.