VARIABLE EVOLUTIONARY RATES IN THE MOLECULAR EVOLUTION OF MAMMALIAN GROWTH-HORMONES

In mammals pituitary growth hormone (GH) shows a slow basal rate of ev olution (0.22 +/- 0.03 x 10(-9) substitutions/amino acid site/year) wh ich appears to have increased by at least 25-50-fold on two occasions, during the evolution of primates (to at least 10.8 +/- 1.3 x 10(-9) s ubstitutions/amino acid site/year) and artiodactyl ruminants (to at le ast 5.6 +/- 1.3 X 10(-9) substitutions/amino acid site/year). That the se rate increases are real, and not due to inadvertent comparison of n onorthologous genes, was established by showing that features of the G H gene sequences that are not expressed as mature hormone do not show corresponding changes in evolutionary rate. Thus, analysis of nonsynon ymous substitutions in the coding sequence for the mature protein conf irmed the rate increases seen in the primate and ruminant GHs, but ana lysis of nonsynonymous substitutions in the signal peptide sequence, s ynonymous substitutions in the coding sequence for signal peptide or m ature protein, and 5' and 3' untranslated sequences showed no statisti cally significant changes in evolutionary rate. Evidence that the incr eases in evolutionary rate are probably due to positive selection is p rovided by the observation that in the cases of both ruminant and prim ate GHs the periods of rapid evolution were followed by a return to a slow rate similar to the basal late seen in other mammalian GHs. Diffe rences between the biological properties of GHs have been identified w hich may relate to these periods of rapid adaptive molecular evolution . On the basis of sequence data currently available (but excluding rod ent GHs which show an intermediate rate, the basis of which is not cle ar) for most (similar to 90%) of evolutionary time mammalian GHs have been in the slow phase of evolution, with possibly most of the few ami no acid substitutions that have occurred being neutral in nature. But most (similar to 80%) of the amino acid substitutions that have been i ntroduced into GH during the course of mammalian evolution have been a ccepted during the rapid phases and were adaptive in nature.