CHARACTERIZATION AND REGULATION OF THE MOUSE INSULIN-RECEPTOR SUBSTRATE GENE PROMOTER

To evaluate the potential for regulation of the insulin receptor subst rate IRS-1, we have cloned the mouse IRS-1 gene, identified its promot er, and analyzed promoter activity in the basal state and in response to stimulation. The 5'-region of the mouse IRS-1 gene lacks typical CA AT and TATA boxes but contains nine potential Spl binding sites consis tent with a housekeeping gene. The 5'-region of the IRS-1 gene also ha s significant regions of homology with the promoters of the progestero ne receptor gene, the insulin-like growth factor I receptor gene, and the androgen receptor gene. Multiple transcription start sites were id entified 0.4-1.2 kilobases (kb) upstream from the start codon. Using a chloramphenicol acetyl transferase assay in Chinese hamster ovary (CH O) cells, basal promoter activity was present in the 3.2 kb 5'-flankin g region of IRS-1 gene. Within this region, there were 184-base pair a nd 60-base pair negative regulatory elements at -3.2 kb and -1.6 kb su rrounded by positive elements. By gel shift assay, a nuclear factor wa s identified in CHO cells which binds to -1606 and -1586 sequence in t he negative regulatory element and appears to be distinct from C/EBP, CREB, and AP-1. In 3T3-F442A adipocytes dexamethasone treatment signif icantly decreased IRS-1 mRNA and IRS-1 protein. This was due to a decr ease in the half-life of IRS-1 mRNA, with no change in IRS-1 promoter- chloramphenicol acetyl transferase activity. Insulin also decreased IR S-1 protein by similar to 60% within 9 h but did so without altering I RS-1 mRNA levels or chloramphenicol acetyl transferase activity. Thus, both insulin and dexamethasone down-regulate IRS-1 expression at the posttranscriptional level; with insulin this is probably due to an eff ect on protein half-life, whereas with dexamethasone the effect is due to a change in the half-life of IRS-1 mRNA.