Peptide bond formation on the ribosome. Structural requirements for inhibition of protein synthesis and of release of peptides from peptidyl-tRNA on bacterial and mammalian ribosomes by aminoacyl and nucleotidyl analogues of puromycin

Abstract

3′- N-Aminoacyl analogues of puromycin and 5′- O-phosphoryl and 5′- O-nucleotidyl derivatives of 3′- N-glycyl-puromycin aminonucleoside (PANS-Gly) were used to investigate the acceptor substrate specificity of peptidyl transferase in animal (rabbit reticulocyte) and bacterial ( Escherichia coli) ribosomes. A survey of the activity of these analogues at 1 · 10 −5–3 · 10 −4 M, relative to that of puromycin, was carried out in the following assay systems; (i) the poly U-directed synthesis of poly-phenylalanine, (ii) the release of nascent peptides (coded for by natural mRNA) from peptidyl-tRNA on the ribosome, and (iii) the natural mRNA-directed protein synthesis in vivo and in vitro. The results are summarized as follows. 3′-N- l-Aminoacyl-PANS derivatives possessing a benzyl group in the aminoacyl moiety ( S- benzyl- l-cysteinyl , O- benzyl- l-seryl , l-tyrosyl and l-phenylalanyl derivatives) gave moderate (25–50 % of that of puromycin) to high (> 50 % of that of puromycin) inhibition of the reticulocyte poly U system. The same derivatives also gave moderate to high activity in releasing peptides from reticulocyte and E. coli ribosomes. PANS- l-phenylalanine was also highly effective on both reticulocyte and E. coli in vivo and in vitro protein synthesis. PANS- l-tryptophan had negligible or low (5–25 % of that of puromycin) activity in the reticulocyte poly U system and in releasing peptides from E. coli ribosomes but had moderate activity in releasing peptides from reticulocyte ribosomes. PANS- d-phenylalanine had negligible or low activity in the reticulocyte poly U system. It had negligible activity in releasing peptides from reticulocyte ribosomes but high activity at 3 · 10 −4 M in releasing peptides from E. coli ribosomes. The analogue was also inactive (at 1 · 10 −4 M) on E. coli in vivo and reticulocyte in vivo and in vitro protein synthesis but had moderate activity on E. coli in vitro protein synthesis. The glycyl, l-alanyl, l-leucyl, and l-prolyl derivatives of PANS gave negligible or low inhibition in the reticulocyte poly U system and in releasing peptides from E coli and reticulocyte ribosomes except for the l-leucyl and l-prolyl derivatives which gave low to moderate activity (at 3 · 10 −4 M) in releasing peptides from reticulocyte ribosomes. However, PANS- l-leucine gave low inhibition of E. coli and reticulocyte protein synthesis in vivo but moderate activity in vitro while PANS-glycine was inactive. 5′- O-Phosphoryl-PANS-glycine and the 5′- O-nucleotidyl (Ap, Gp, and Up) derivatives of PANS-glycine gave negligible or low inhibition in the reticulocyte poly U system while CpPANS-glycine gave moderate to high activity. CpPANS-glycine had low activity in releasing peptides from reticulocyte ribosomes while the 5′- O-phosphoryl, Ap, Gp and Up derivatives of PANS-glycine had lower or negligible activity. The activity of these derivatives of PANS-glycine for releasing peptides from E. coli ribosomes was similar except that CpPANS-glycine was moderately active. From these results, it was concluded that a hydrophobic binding site exists on both E. coli and rabbit reticulocyte peptidyl transferase acceptor sites which is relatively specific for the aminoacyl R groups of phenylalanyl- and tyrosyl-tRNA and a second site relatively specific for binding the penultimate 3′-CMP residue of aminoacyl-tRNA. Interpretation of the data was made difficult by the variation in activity of some of the analogues with the type of assay system used.