Inquiries into the structure-function relationship of ribonuclease T1 using chemically synthesized coding sequences

Abstract

The genes for ribonuclease T1 and its site-specific mutants were chemically synthesized and introduced to Escherichia coli. All enzymes were fusion products produced by joining the synthetic gene at specific restriction sites to the synthetic gene for human growth hormone in a plasmid containing the E. coli trp promoter. The fusion protein from this plasmid contained 66% of the amino-terminal sequences of the human growth hormone, which were recognizable immunologically. RNase T1 or its mutants were cleaved from the fusion protein with cyanogen bromide. The synthetic RNase T1 endowed with the revised wild-type triad Gly-Ser-Pro, residues 71-73, was fully functional, readily hydrolyzing pGpC bonds, whereas a mutant enzyme having the originally reported, erroneous triad Pro-Gly-Ser was totally inactive. Various amino acid substitutions were also introduced to the guanosine recognition region comprised of residues 42-45, Tyr-Asn-Asn-Tyr. Substitution of either of the tyrosine residues noted above with phenylalanine had no dramatic effect on the enzyme's function. Replacement of asparagine-43 with arginine or alanine also caused only a small change in the hydrolyzing activity--a mutant enzyme maintained greater than 50% of the wild-type activity. In sharp contrast, when aspartic acid or alanine was substituted for asparagine-44, the activity was dramatically reduced to a few percent of the wild-type activity.