Purification and Properties of Ribonuclease Ii from Escherichia Coli.

Three enzymes in Escherichia coli which break down ribonucleic acid are now known. One is polynucleotide phosphorylase (I), which in the presence of phosphate yields ribonucleoside 5’-diphosphates. A ribonuclease, which will be called ribonuclease I, is bound to 30 S ribosomes in latent form. When released from the ribosomes it attacks RNA to yield ribonucleoside 3’-phosphates via a 2’) 3’-cyclic phosphate intermediate (2, 3). The third enzyme, which here will be called ribonuclease II, liberates ribonucleoside 5’-phosphates. Its presence was first reported by Wade (4) and by Wade and Lovett (5), and it was implicated by Sekiguchi and Cohen (6) in the breakdown of messenger RNA in vitro. Our more recent experiments (7) show that this enzyme acts only in the presence of K+ (or NH*+) and Mg ++ (or Mn++) ions, a feature which clearly distinguishes it from both ribonuclease I and polynucleotide phosphorylase. RNase I has no activity as long as it remains attached to a ribosome, and apparently it normally never functions inside an intact, growing cell. Instead it may function only outside the cell membrane, a hypothesis favored by the recent experiments of Neu and Heppel (8) showing the preferential release of RNase I from spheroplasts. In contrast, however, polynucleotide phosphorylase and RNase II are highly active in cell-free extracts. Their activity may in large measure explain why cell-free E. coli extracts quickly lose the ability to synthesize protein. Depending upon their exact environment, both enzymes can exist either free or attached to ribosomes. Unlike RNase I, however, they possess enzymatic activity even when attached to ribosomes. One or both of these enzymes is likely to be involved in the normal turnover of E. coli messenger RNA in viva, but as yet no conclusive data exist as to which of these enzymes performs this function, or whether both are involved. The search will clearly be aided by the purification of RNase II and the study of its mode of action. In this paper is reported a method for obtaining both approximately 30-fold purification of RNase II and its complete separation from polynucleotide phosphorylase, from RNase I, and from the several E. coli deoxyribonucleases (see the review of E. coli nucleases by Lehman (9)) and the acid and alkaline phosphatases. This has enabled us to show that RNase II can act as both an exoand an endonuclease, depending on the specific polynucleotide substrate. This clearly differentiates RNase II from polynucleotide phosphorylase, which acts only as an exo-