Background: Kazal-like serine protease inhibitors are defined by a conserved sequence motif. A typical Kazal domain contains six cysteine residues leading to three disulfide bonds with a 1-5/2-4/3-6 pattern. Most Kazal domains described so far belong to this class. However, a novel class of Kazal domains with two disulfide bridges resulting from the absence of the third and sixth cysteines have been found in biologically important molecules, such as human LEKTI, a 15-domain inhibitor associated with the severe congenital disease Netherton syndrome. These domains are referred to as atypical Kazal domains. Previously, EPI1, a Kazal-like protease inhibitor from the oomycete plant pathogen Phytophthora infestans, was shown to be a tight-binding inhibitor of subtilisin A. EPI1 also inhibits and interacts with the pathogenesis-related P69B subtilase of the host plant tomato, suggesting a role in virulence. EPI1 is composed of two Kazal domains, the four-cysteine atypical domain EPI1a and the typical domain EPI1b.
Results: In this study, we predicted the inhibition constants of EPI1a and EPI1b to subtilisin A using the additivity-based sequence to reactivity algorithm (Laskowski algorithm). The atypical domain EPI1a, but not the typical domain EPI1b, was predicted to have strong inhibitory activity against subtilisin A. Inhibition assays and coimmunoprecipitation experiments showed that recombinant domain EPI1a exhibited stable inhibitory activity against subilisin A and was solely responsible for inhibition and interaction with tomato P69B subtilase.
Conclusion: The finding that the two disulfide bridge atypical Kazal domain EPI1a is a stable inhibitor indicates that the missing two cysteines and their corresponding disulfide bond are not essential for inhibitor reactivity and stability. This report also suggests that the Laskowski algorithm originally developed and validated with typical Kazal domains might operate accurately for atypical Kazal domains.