Extracellular Lytic Enzymes of Myxococcus virescens II. Purification of Three Bacteriolytic Enzymes and Determination of their Molecular Weights and Isoelectric Points

The bacteriolytic enzymes produced by Myxococcus virescens and previously concentrated and separated from most of the non-bacteriolytic proteins have been further separated and purified. The bacteriolytic enzyme solution was concentrated by lyo-philization. When applied to a Sephadex G-100 column, three peaks of bacteriolytic activity were eluted. Polyacrylamide gel electrophoresis showed that all the three enzyme fractions were contaminated with at least four non-bacteriolytic proteins. In the first enzyme fraction the bacteriolytic enzymes could be freed from the contaminating proteolytic activity by adsorption on a hydroxylapatite column. The bacteriolytic enzymes could then be adsorbed on a CM-cellulose column. The remaining contaminating proteins passed the column un-adsorbed while the bacteriolytic enzymes could be eluted with a gradient of 0.02–0.10 M ammonium hydrogen carbonate solution. The second enzyme fraction was adsorbed on a CM-cellulose column and then eluted with 0.03–0.15 M NH4 HCO₃. After rechromatography on a new column under the same conditions, all of the contaminating proteins had disappeared. For purification of the third enzyme fraction chro-matography on one single CM-cellulose column was sufficient. The elution of the adsorbed enzymes was performed with a gradient of 0.15–0.30 M NH₄HCO₃. The recovery of activity for each of the ion-exchange chromatography separations was at least 90%. The purity of the enzymes was tested by polyacrylamid gel electrophoresis. Each of the purified enzymes gave only one coloured band which coincided with the enzyme activity assayed in sliced gels. The molecular weights of the enzymes were determined by electrophoresis on acryl-amide gels containing sodiumdodecylsulphate. The molecular weights determined in this way (about 40,000, 30,000 and 20,000, respectively) were about 10,000 daltons higher than those obtained by gel chromatography on Sephadex G-100. This discrepancy seems to depend on interactions between the enzymes and the dextran molecules probably caused by the strongly basic nature of the enzymes or by formation of enzyme-substrate complexes.