Purification and Characterization of Protease So, a Cytoplasmic Serine Protease in Escherichia coli

Abstract

A new cytoplasmic endoprotease, named protease So, was purified to homogeneity from Escherichia coli by conventional procedures with casein as the substrate. Its molecular weight was 140,000 when determined by gel filtration on Sephadex G-200 and 77,000 when estimated by polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate. Thus, it appears to be composed of two identical subunits. Protease So had an isoelectric point of 6.4 and a Km of 1.4 ,uM for casein. In addition to casein, it hydrolyzed globin, glucagon, and denatured bovine serum albumin to acid-soluble peptides but did not degrade insulin, native bovine serum albumin, or the "auto a" fragment of ,B-galactosidase. A variety of commonly used peptide substrates for endoproteases were not hydrolyzed by protease So. It had a broad pH optimum of 6.5 to 8.0. This enzyme is a serine protease, since it was inhibited by diisopropyl fluorophosphate and phenylmethylsulfonyl fluoride. Although it was not inhibited by chelating agents, divalent cations (e.g., Mg2+) stabilized its activity. Protease So was sensitive to inhibition by N-tosyl-L-phenylalanine chloromethyl ketone but not by N-tosyl-L-lysine chloromethyl ketone. Neither ATP nor 5'-diphosphate-guanosine-3'-diphosphate affected the mte of casein hydrolysis. Protease So was distinct from the other soluble endoproteases in E. coli (including proteases Do, Re, Mi, Fa, La, Ci, and Pi) in its physical and chemical properties and also differed from the membrane-associated proteases, protease IV and V, and from two amino acid esterases, originally named protease I and II. The physiological function of protease So is presently unknown.

A periplasmic insulin‐cleaving proteinase (ICP) from Acinetobacter calcoaceticus sharing properties with protease III from Escherichia coli and IDE from eucaryotes

“…The pure enzyme is a neutral metalloprotease with a molecular mass of 108 kDa (gel filtration, native PAGE) (FRICKE and AURICH 1992), which corresponds well to the results for the protease I11 (CHENG and ZIPSER 1979, DYKSTRA and KUSHNER 1985, FINCH et al 1986). The molecular mass of the IDE is described as being 110 kDa for the monomeric protein (KIRSCHNER andGOLDBERG 1981, SHII et al 1986).…”

“…Three different proteinases are located in the periplasmic space of E. coli, the protease Pi , the protease Mi (COOK 1988) and a nonpolar C-terminini-specific protease, called Tsp (SILBER et al 1992). The protease 111, which is identical with the protease Pi , was isolated and characterized by CHENG and ZIPSER (1979).…”

“…Three different proteinases are located in the periplasmic space of E. coli, the protease Pi , the protease Mi (COOK 1988) and a nonpolar C-terminini-specific protease, called Tsp (SILBER et al 1992). The protease 111, which is identical with the protease Pi , was isolated and characterized by CHENG and ZIPSER (1979). It prefers the cleavage of polypeptides like glucagon and insulin to proteins, and shows closed similarities in its cleavage specificity and inhibition behaviour with the insulin-destroying enzyme (IDE EC 3.4.22.1 1 from eucaryotes (BARRETT and RAWLINGS 1991, DING et al 1992, at first purified by DUCKWORTH et al (1972) with means of insulin-specific affinity chromatography.…”

Peptidases and proteases of Escherichia coli and Salmonella typhimurium

“…Two enzymes from the cytoplasmic fraction responsible for the degradation of the signal peptide have been purified and identified. The one is similar to protease So [15] and represents less than 10% of the activity. The other is similar to the oligopeptidase described by Vimr et al [110].…”

“…Proteases So and Re have been extensively studied and shown to be responsible for the rapid degradation of oxidatively damaged glutamine synthetase in vitro and perhaps in vivo [15,58,82,91]. Thus there may exist, in E. coli, an ATPindependent proteolytic process that specifically eliminates oxidatively damaged proteins.…”