Effects of Insulin and Glucose on Cultured Rat Hepatocyte Gene Expression

We previously used isolated adult rat hepatocyte cultures to study the ability of glucose to induce several hepatic mRNAs. However, we found that the optimal insulin concentration required to obtain the glucose effect was greater than 10,000 microU/ml. To test the hypothesis that the requirement for high concentrations of insulin in the culture was due to rapid loss of insulin in hepatocyte cultures, serial measurements of insulin were made at different media insulin concentrations (0-500,000 microU/ml) and glucose concentrations (5.5 and 2.75 mM). In addition, a dose-response relationship was established between media insulin concentrations and the pattern of mRNAs present in the hepatocytes determined by two-dimensional gel electrophoresis of in vitro translation products. We found that at low insulin concentrations (less than 1000 microU/ml), greater than 80% of the insulin was lost to the glassware, whereas at high initial insulin concentrations, approximately 23% of the insulin was lost to the glassware. Placement of media into the hepatocyte culture led to further insulin disappearance with a half-life for insulin of 41.5 h at 10,000 microU/ml and 13.8 h at 100 microU/ml. We found 16 mRNAs were altered by insulin at 5.5 mM glucose and 9 mRNAs were changed by insulin at 27.5 mM glucose. After taking into consideration the distributional and metabolic losses of insulin, all but one mRNA responded to insulin within the physiologic range of portal insulin (less than 1-94 microU/ml). Our data indicate that the hepatocyte culture is an excellent model to study the physiologic effects of insulin on hepatic gene expression.