Researchers discovered a new enzyme called mirolase from the gum disease bacterium Tannerella forsythia. This enzyme needs calcium to become active and stays partially blocked by a piece of itself until that piece is broken down. Importantly, mirolase can chew up proteins like fibrinogen, hemoglobin, and the human antimicrobial peptide LL‑37, which helps protect us from infections.

The genome of Tannerella forsythia, an etiological factor of chronic periodontitis, contains several genes encoding putative proteases. Here, we characterized a subtilisin-like serine protease of T. forsythia referred to as mirolase. Recombinant full-length latent promirolase [85 kDa, without its signal peptide (SP)] processed itself through sequential autoproteolytic cleavages into a mature enzyme of 40 kDa. Mirolase latency was driven by the N-terminal prodomain (NTP). In stark contrast to almost all known subtilases, the cleaved NTP remained non-covalently associated with mirolase, inhibiting its proteolytic, but not amidolytic, activity. Full activity was observed only after the NTP was gradually, and fully, degraded. Both activity and processing was absolutely dependent on calcium ions, which were also essential for enzyme stability. As a consequence, both serine protease inhibitors and calcium ions chelators inhibited mirolase activity. Activity assays using an array of chromogenic substrates revealed that mirolase specificity is driven not only by the substrate-binding subsite S1, but also by other subsites. Taken together, mirolase is a calcium-dependent serine protease of the S8 family with the unique mechanism of activation that may contribute to T. forsythia pathogenicity by degradation of fibrinogen, hemoglobin, and the antimicrobial peptide LL-37.