Glutathione S-transferases of female A/J mouse liver and forestomach and their differential induction by anti-carcinogenic organosulfides from garlic

This study characterizes glutathione (GSH) S-transferase (GST) isoenzymes of the liver and forestomach of the female A/J mouse and compares their specificities in catalyzing the conjugation of GSH with 7 beta, 8-alpha-dihydroxy-9 alpha, 10 alpha-oxy-7,8,9, 10-tetrabydrobenzo[a] pyrene (anti-BPDE), the ultimate carcinogenic metabolite of benzo[a]pyrene (BP). The GST activity in female A/J mouse liver was expressed by a minimum of seven isoenzymes which arose from different homo- or heterodimeric combinations of at least two a class (designated as alpha 1 and alpha 4), four micro class (micro 1 to micro 4), and one pi class GST subunit. The GST isoenzyme composition of A/J mouse forestomach appeared to be different from that of the liver. For example, while GST isoenzymes containing micro 3 and micro 4 type subunits were selectively expressed in the liver, an a class heterodimeric GST isoenzyme (containing alpha 2 and alpha 3 subunits) was expressed in the forestomach but could not be detected in the liver. The (+)-anti-BPDE appeared to be a better substrate than the (-)-enantiomer for all GSTs, except for isoenzymes containing the alpha 4 type GST subunit. The murine pi class GST isoenzyme displayed relatively higher specific activity toward (+)-anti-BPDE compared to other GSTs. The specific activities of mouse GSTs toward (+)-anti-BPDE were in the order of pi > micro > alpha. These results suggest that the pi class GST isoenzyme may play an important role in providing protection against BP-induced cancer. Therefore, it seems logical to postulate that the ability of a chemoprotector to increase the expression of GST pi may be an important determinant of its effectiveness against BP-induced cancer. To test the validity of this contention, we have determined the effects on hepatic and forestomach GST isoenzyme/subunit expression of three naturally occurring organosulfides (OSCs) from garlic, which significantly differ in their effectiveness against BP-induced forestomach cancer. Treatment of mice with diallyl sulfide (DAS) and diallyl trisulfide (DATS), which are potent inhibitors of BP-induced forestomach cancer in mice, resulted in a significant increase in hepatic and forestomach GST activity to ward anti-BPDE. On the contrary, this activity was no, increased in either organ by dipropyl sulfide (DPS) which is ineffective against BP-induced forestomach cancer. The chemopreventive efficacy of these OSCs correlated with their ability to increase the expression of GST pi. For example, DAS treatment resulted in approximate increases of 1.7- and 2.2-fold in hepatic and forestomach GST pi expression, respectively, over the control. Treatment of mice with DATS, which is a relatively more potent inhibitor of BP-induced forestomach cancer than DAS, resulted in about 3.8- and 3.2-fold increases, respectively, in hepatic and forestomach GST pi expression over the control. On the contrary, the expression of hepatic and forestomach GST pi was increased only marginally (10-20%) upon DPS administration. In conclusion, the results of the present study suggest that induction of GST pi can be used as a bioassay for screening potential inhibitors of BP-induced cancer.