Effect of feeding solanidine, solasodine and tomatidine to non-pregnant and pregnant mice

The aglycone forms of three steroidal glycoalkaloids—solanidine (derived by hydrolytic removal of the carbohydrate side chain from the potato glycoalkaloids α-chaconine and α-solanine), solasodine (derived from solasonine in eggplants) and tomatidine (derived from α-tomatine in tomatoes)—were evaluated for their effects on liver weight increase (hepatomegaly) in non-pregnant and pregnant mice and on fecundity in pregnant mice fed for 14 days on a diet containing 2.4 mmol/kg of aglycone. In non-pregnant mice, observed ratios of % liver weights to body weights (%LW/BWs) were significantly greater than those of the control values as follows (all values in % vs matched controls±S.D.): solanidine, 25.5±13.2; solasodine 16.8±12.0; and tomatidine, 6.0±7.1. The corresponding increases in pregnant mice were: solanidine, 5.3±10.7; solasodine, 33.1±15.1; tomatidine, 8.4±9.1. For pregnant mice (a) body weight gains were less with the algycones than with controls: solanidine, −36.1±14.5; solasodine, −17.9±14.3; tomatidine, −11.9±18.1; (b) litter weights were less than controls: solanidine, −27.0±17.1; solasodine, −15.5±16.8; tomatidine, no difference; (c) the %LTW/BW ratio was less than that of the controls and was significant only for solasodine, −8.7±13.7; and (d) the average weight of the fetuses was less than the controls: solanidine, −11.2±15.2; solasodine, −11.4±9.4; tomatidine, no difference. Abortion of fetuses occurred in five of 24 pregnant mice on the solanidine and none on the other diets. To obtain evidence for possible mechanisms of the observed in vivo effects, the four glycoalkaloids (α-chaconine, α-solanine, solasonine and α-tomatine) mentioned above and the aglycones solanidine and tomatidine were also evaluated in in vitro assays for estrogenic activity. Only solanidine at 10 μm concentration exhibited an increase in the MCF-7 human breast cancer cell proliferation assay. Generally, the biological effects of solanidine differ from those of the parent potato glycoalkaloids. Possible mechanisms of these effects and the implication of the results for food safety and plant physiology are discussed.