REGULATION MECHANISM OF SPECIFIC EXPRESSION OF TUMOR-MARKER GENE DURING CARCINOGENESIS

Rat glutathione transferase P (GST-P) is expressed at low levels in th e normal liver but becomes highly expressed in hyperplastic nodules an d in hepatocellular carcinomas during chemical hepatocarcinogenesis. T o understand the regulation mechanisms of this gene, we have character ized the 5'-flanking region and have found that GST-P gene is regulate d by at least two elements: one is a strong enhancer and the other is a silencer. GST-P enhancer I (GPEI) located at -2.5 Kb, consists of tw o TPA-responsive element (TRE)-like sequences that are palindromically oriented with 3 bp in between. It is well known that TRE is activated by two nuclear oncogenes, c-Jun and c-Fos. Although GPEI is trans-act ivated by these oncogenes, it is also active in F9 embryonal carcinoma cells that lack c-Jun protein, suggesting that it can function with s ome trans-activator other than AP-1 (c-Jun/c-Fos heterodimer). Indeed, another protein is identified from the F9 nuclear extract. We have al so identified a silencer element at 300 bp upstream from the cap site. There are several cis-elements in this region and at least three tran s-acting factors bind to these elements. We purified SF-A (silencer fa ctor A) which binds to several regions in this silencer, and determine d the partial amino and sequence. Interestingly, SF-A seemed to be a r elated protein to NF1 (nuclear factor 1) which is an activator for the transcription and DNA replication. Another factor SF-B (silencer fact or B) has been cloned and found to be the same as LIP (liver inhibitor y protein) which is a competitor for LAP (liver activator protein), bo th ar from the same gene designated as C/EBP beta. By transfection ana lysis using GAL4 DNA binding domain we found LIP is not only a competi tor but a direct repressor. In the normal liver, another C/EBP family member, C/EBP alpha also acts as a negative regulator, and this expres sion decreases during hepatocarcinogenesis, resulting in the loss of s ilencer function. We carried out the carcinogenesis experiments using transgenic rats harboring a chloramphenicol acetyltransferase (CAT) re porter gene which -2900 to +59 of the GST-P gene. Liver foci and nodul es produced by chemical carcinogens were found to express high levels CAT activity by both CAT assay and immunohistochemical study, while no rmal liver cells did not express any CAT activity. These results demon strate that the GST-P gene is trans-activated locus-independently duri ng rat hepatocarcinogenesis. Moreover, the similar results were obtain ed using transgenic rats carrying GPEI-CAT, indicating that GPEI is an important cis-element for activation of GST-P gene during hepatocarci nogenesis.