Degradation of protease inhibitors, immunoglobulins, and other serum proteins by Serratia protease and its toxicity to fibroblast in culture

We investigated the effect of the extracellular protease of Serratia marcescens on human serum constituents such as immunoglobulins, fibronectin, alpha 1-protease inhibitor, alpha 2-macroglobulin, lysozyme, and transferrin. At a very low concentration of Serratia 56-kilodalton protease (56K protease), purified human plasma fibronectin was degraded rapidly into three structural domains or small fragments. Immunoglobulin G3 (IgG3) and IgA1 were also degraded within 30 min with 1 microgram of this protease per ml, more rapidly than their other subclass of IgG or IgA. alpha 1-Protease inhibitor, which did not inhibit the 56K protease, was degraded similarly by the protease. These events were demonstrated by fluorescence polarization and sodium dodecyl sulfate-polyacrylamide gel electrophoresis. The protease was considerably inhibited by human alpha 2-macroglobulin and chicken ovomacroglobulin. However, when there was a 2 M excess of ovomacroglobulin or a 4 M excess of alpha 2-macroglobulin over the 56K protease, about 25 or 40% proteolytic activity remained, respectively. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis showed that the protease degraded the alpha 2-macroglobulin extensively during prolonged incubation, which paralleled with regeneration of the protease activity. The protease also cleaved human lysozyme, although moderately. Human serum transferrin was degraded slightly, and human serum albumin was almost resistant to the 56K protease. The enzyme seemed to have no effect on reconstituted collagen, but it degraded rat tropocollagen and yielded fragments of beta and gamma chains by cleaving the intramolecular cross-links. Most of the above proteolysis by the 56K protease appears to result in a limited type of substrate specificity. Thus, the present study demonstrates that the protease is capable of degrading defense-oriented humoral proteins and tissue constituents. Furthermore, it is toxic to fibroblasts. These findings also clarified the possible role of Serratia protease as a virulence factor in the pathogenesis of serratial infections. We recently demonstrated this notion in vivo with rabbit cornea (R. Kamata et al., Ophthalmology 92:1452-1459, 1985).