Latent and Active Human Polymorphonuclear Leukocyte Collagenases

Latent and active human polymorphonuclear leukocyte collagenase could be purified to apparent homogenei by a rapid and reproducible method involving affinity chromatography on concanavalin‐A–Sepharose, collage Sepharose and activated thiol‐Sepharose 4‐B, gel filtration on Sephacryl S‐300 and ion‐exchange chromatograpl on DEAE‐Sephacel. The latent and active collagenase were shown to be single entities with apparent M r 91000 and 6400 respectively, as determined by sodium dodecyl sulfate/polyacrylamide gel electrophoresis without mercapto ethanol. Dodecyl sulfate gel electrophoresis of the latent enzyme in the presence of mercaptoethanol showed two bands with apparent M r of 64000 and 24000 indicating the cleavage of a disulfide bond. Activation of the native latent collagenase could be achieved by_ disulfide‐containing compounds. A concomitant decrease in molecular weight of approximately 25000 was observed on activation of the latent enzyme. The activators proved ineffective after alkylation of the latent enzyme with iodoacetamide. This lends support to the proposed thiol/disulfide interchange mechanism in the activation reaction [Macartney, H. W. and Tschesche, H (1980) FEBS Lett. 119 , 327–332]. The active collagenase could be completely inhibited by an inhibitor dissociated from the latent enzyme, i.e. a complex of enzyme and inhibitor. Regulation of the collagenase activity could be achieved by the redox potential of the glutathione cycle [Tschesche, H. and Macartney, H. W. (1981) Eur. J. Biochem. 120 , 183–190]. The latent and active enzymes were shown to be acidic proteins with isoelectric points at pH 5.5 and 6.4 respectively. Amino acid analyses of both enzymes have been obtained and indicate a difference of about 222–223 amino acid residues between the latent and active collagenases. This corresponds to a difference of approximately M r 25000 between the active and the latent enzyme, the latter being a mixed disulfide of the enzyme and its inhibitor. The optimum pH for collagenolytic activity was observed near neutrality. The enzyme specificity was revealed to be that of a true mammalian collagenase splitting the native type I tropocollagen molecule into three‐quarter and one‐quarter fragments according to dodecyl sulfate/polyacrylamide gel electrophoresis. No other contaminating proteolytic activity was observed in the purified latent and active collagenase preparations. The active enzyme was found to be stable for over one year when stored at ‐ 70°C. The collagenase was capable of reducing the specific viscosity of a native type I collagen solution by approximately 40%. The amount of latent collagenase present in the leukocytes was estimated to 0.65 mg/10 10 cells. The overall yield obtained after the purification to homogeneity was 75–80% of the theoretical value.