IGF-I and serine protease inhibitor 2·1 nuclear transcript abundance in rat liver during protein restriction

Restriction of dietary protein consumption of young male rats results in decreased growth velocity and a reduction in the abundance of hepatic IGF-I mRNA. It is not known whether the reduction in IGF-I mRNA abundance in the liver of protein-restricted rats results from a decrease in IGF-I gene transcription. In the present study, three experiments were performed with 4-week-old male rats to examine the effect of protein restriction on IGF-I gene transcription in liver. In these experiments, we monitored IGF-I nuclear transcripts (pre-mRNA) within total cellular RNA using a ribonuclease protection assay. In the first experiment, a consistent decrease in IGF-I mRNA from animals fed isocaloric diets containing 20% (control), 12%, 8% and 4% protein (dietary effect, P<0·001) was not paralleled by a decrease (P>0·50) in IGF-I pre-mRNA. Two additional experiments examining the effect of 4% vs 20% protein diets yielded comparable results. Pooled results from these two studies (n=12/treatment) demonstrated that a 64% reduction (P<0·0001) in IGF-I mRNA abundance was not accompanied by a decrease in IGF-I pre-mRNA (1·17 vs 1·31 ±0·21 image density units for 4% and 20% protein treatments). Unlike IGF-I, the abundance of carbamyl phosphate synthetase-I (CPS-I) pre-mRNA and mRNA was comparably reduced (∼70%, P<0·001), indicating that the decrease in mRNA of this urea cycle enzyme during protein restriction occurs predominantly by a transcriptional mechanism. A common feature of all experiments was a pronounced variability in the expression of hepatic IGF-I pre-mRNA among animals, which was not diet specific. To test whether the variability in IGF-I gene transcription was correlated with variability in the transcription of another gene that is regulated by GH, we quantified the abundance of nuclear transcripts for the serine protease inhibitor 2·1 (SPI 2·1 gene. A positive association (r=0·81, P<0·0001) between SPI 2·1 and IGF-I nuclear transcripts was demonstrated. The correlation between IGF-I and SPI 21 transcripts was specific, because the quantity of IGF-I and CPS-I nuclear transcripts was not correlated in this study. Although transcription of the IGF-I and SPI 2·1 genes was similar, the abundance of SPI 21 mRNA was not altered by protein deprivation. In summary, these studies indicated that protein restriction does not substantially alter the mean quantity of IGF-I nuclear transcripts, suggesting that the decrease in IGF mRNA occurs predominantly by a post-transcriptional mechanism. In addition, nuclear transcript abundance of the IGF-I and SPI 21 genes varies in a co-ordinate manner, supporting the hypothesis that transcription of these genes responds rapidly to a common variable factor such as plasma GH. Journal of Endocrinology (1995) 145, 397–407