Relationship between cytocidal activity and glutathione-S-transferase inhibition using doxorubicin coupled to stereoisomers of glutathione with different substrate specificity

Conformational change in the active center region of GST P1-1, due to binding of a synthetic conjugate of DXR with GSH, enhanced JNK-mediated apoptosis

Treatment of cells with a synthetic conjugate of DXR with GSH via glutaraldehyde (GSH-DXR) caused cytochrome c to be released from the mitochondria to the cytosol following potent activation of caspase-3 and -9 by typical DNA fragmentation. This apoptosis was regulated by the JNK-signaling pathway. In the present experiment, binding of GSH-DXR to GST P1-1 allosterically led to the disappearance of its enzyme activity and activated the kinase activity of JNK without dissociation of the JNK-GST P1-1 complex. The recombinant GST P1-1 molecule with a mutation in the active center region (W38H and C47S) lost its GST activity when bound to JNK to the same degree as the wild-type, with the mutated GST P1-1 molecule failing to inhibit the activity of JNK. It has been reported that JNK-signaling is regulated by GST P1-1 via interaction with the C-terminus. We confirmed that GST P1-1 deletion mutant (Delta194-209) and a site-directed mutant (R201A) in the C-terminal region failed to bind and inhibit JNK. These results indicated that not only binding of the C-terminal region of GST P1-1 to the JNK molecule, but also the active center region of GST P1-1 play important roles in the regulation of JNK enzyme activity. The findings suggested that allosteric inhibition of GST P1-1 activity by the binding of GSH-DXR following conformational change may activate JNK and induce apoptosis via the mitochondrial pathway in the cells.

Glutathione-S-transferase-pi expression regulates sensitivity to glutathione-doxorubicin conjugate

We have reported that glutathione-doxorubicin conjugate (GSH-DXR) exhibited potent cytotoxicity against tumor cells and inhibited glutathione-S-transferase (GST) enzyme activity. In order to determine whether or not the expression of GST-pi lowered the cytotoxicity of GSH-DXR, cytocidal activity of the conjugate was examined using tumor cells in which the level of GST-pi expression was regulated by transfecting GST-pi cDNA in the correct or reverse direction and comparing with that of DXR. Enhancement of GST-pi expression by transfecting GST-pi sense cDNA into human hepatoblastoma HepG2 cells in which GST-pi expression was extremely low caused an increase in GST activity from 0.26 to 55.0 nmol/mg/min and a marked reduction in transfectant sensitivity to GSH-DXR to 1/120 (0.15-18 nM IC50) although the sensitivity to DXR was slightly decreased to 1/2.6 (380-990 nM IC50). By contrast, a high GST-pi-expressing human colon cancer cell line, HT29, showed a decrease in GST enzyme activity from 72.0 to 45.9 nmol/mg/min after transfecting GST-pi antisense cDNA and a marked improvement in transfectant sensitivity to GSH-DXR was observed (28-2.9 nM IC50) compared with the transfectant sensitivity to DXR (1020-320 nM IC50). Additionally, the expression of GST-pi in HepG2 cells caused a decrease in GSH-DXR-induced activation of caspase-3, which was an apoptotic marker, whereas the suppression of GST-pi in HT29 cells showed an increase in caspase-3 activation. These results suggested that the cytocidal efficacy of GSH-DXR, but not that of DXR, was controlled by the level of GST-pi expression in the cells.