A new protein phosphatase from skeletal muscle: myosin light-chain phosphatase.

All the vertebrate muscle myosins so far investigated have been shown to possess a lightchain component of 18000-20000 molecular weight, the P light chain (Frearson & Perry, 1975), which is phosphorylated by a highly specific enzyme, myosin light-chain kinase (Perrie et al., 1973; Pires et af . , 1974; Frearson & Perry, 1975; Frearson et al., 1976). Studies (Perrie & Perry, 1970; Perry et al., 1975) have also indicated that an enzymic system is present in rabbit skeletal muscle for the dephosphorylation of the phosphorylated form of this light chain. A soluble protein extract made by extraction of minced rabbit white skeletal muscle with 50m~-Tris/40m~-HCl (pH7.6)/10m~-magnesium acetate/15m~-dithiothreitol was fractionated by a procedure involving (NH&S04 precipitation, chromatography on DEAE-cellulose and gel filtration on Sephadex G-200. This resulted in an 800-fold purification of the enzyme. The final purification step was carried out by affinity chromatography on Sepharose, to which a phosphorylated wholelight-chain fraction of myosin from white skeletal muscle of the rabbit had been linked. By addition of 10mM-EDTA to the buffer the enzyme was eluted as a single component of molecular weight about 72000. The complete procedure resulted in an approx. 10000-fold purification of the myosin light-chain phosphatase with a 2% yield. The enzyme probably exists in the monomeric form, for on gel filtration on Sephadex G-200 in 50m~-Tris/40mM-HC1 (pH7.6)/10m~-magnesium acetate/l5m~-dithiothreitol, myosin light-chain phosphatase was eluted with a retardation corresponding to a molecular weight of 78000. With phosphorylated isolated whole light chains of myosin as substrate, the enzyme was most active in the presence of Mgz+. This ion appeared to act as a stabilizer rather than an activator, for the activity wasnot inhibited by 5Orn~-NaF. The enzyme possessed pH optima at 6.5 and 8.0 and had a K,,, of 4 8 p ~ at both pH values. The specific activity of the enzyme in 50m~-Tris/3omM-HCI/lOm~magnesium acetate/l5m~-dithiothreitol, pH6.5, was 1 pmol of 32P released/min per mg. The rates of phosphate release from AMP, ADP, ATP, aor /3-glycerophosphate, p-nitrophenyl phosphate, phosphorylase a and phosphorylated forms of histone, casein, troponin, glycogen synthetase and phosphorylase kinase were less than 2% of that obtained with the phosphorylated whole light-chain fraction of myosin as substrate. On incubation with radioactively labelled 2m~~potass ium phosphate, 32P was incorporated into the enzyme up to a maximum of 0.8mo1/70000g of protein. The incorporated phosphate was stable to washing with 15 % (w/v) trichloroacetic acid at 4°C. By analogy with the similar exchange obtained with Escherichia coli alkaline phosphatase (Engstrom, 1959), the 32P was assumed to be linked to a serine residue at the active site of the enzyme. The existence of highly specific enzymes for phosphorylation and dephosphorylation of the myosin raises the question of the biological role of phosphorylation of the P light chain. After phosphorylation with myosin light-chain kinase no significant differences (+lo%) in the CaZ+-, Mgz+and K+-stimulated ATPase* activities of myosin and heavy meromyosin (prepared by chymotryptic digestion; Leadbetter & Perry, 1963) when tested at pH7.6, either alone or in the presence of actin were detected. Likewise the binding of actin to myosin and heavy meromyosin, determined by reciprocal plots of the actomyosin ATPase (Eisenberg & Moos, 1968), by viscosity measurements and by coprecipitation experiments, also appeared not to be significantly affected by phosphorylation.