Variability of glutathione S-transferase isoenzyme patterns in matched normal and cancer human breast tissue

The determination of GST levels in blood has been proposed to a marker of tumour burden in general, whereas level of the P1 isoenzyme has been identified as a prognostic factor for breast-cancer patients receiving no adjuvant chemotherapy. Particular glutathione S-transferase (GST) isoenzymes differ in their substrate specificity, however, and their presence or absence might therefore account for the resistance of tumours to particular chemotherapeutic drugs, as already established for cultured cell lines. Determination of the GST isoenzyme profile of a cancer tissue could have prognostic value in the selection of treatment if the levels of expression/activity show a degree of variation comparable with that exhibited by actual patient responses. Using reversed-phase h.p.l.c. to quantify affinity-isolated GSTs, we have analysed full isoenzyme profiles in the first large sample of matched normal and cancer human tissues (18 breast-cancer patients). In no patients did the tumour tissues express any isoenzymes that were not found in normal breast tissue. In addition to the GSTs, another enzyme, identified as enoyl-CoA isomerase, was regularly found in breast tissue cytosol following elution from a hexyl-glutathione affinity column. In most cases, the average level of GST was substantially elevated in the cancer tissues above the levels in normal breast tissue from the same patient. Furthermore, the relative levels of the isoenzymes were substantially more variable in the cancer samples than in the normal breast tissue, providing a plausible mechanism for the well established variable response to treatment.