Role of the regulatory domain of calpastatin in the autoproteolytic activation of calpain

Calpastatin (CST), the natural inhibitor of calpain (CALP), is a protein composed by four repetitive inhibitory domains and a regulatory domain without inhibitory properties. Depending on the CST form, the regulatory region can be constituted by L- or XL-L domains which exist in different splicing forms. We have previously observed that all the CSTs, containing the XL-L and L-domains, lose inhibitory efficiency at concentrations expected to promote complete CALP inhibition. It is known that the CST inhibitory domain binds to CALP catalytic site in the presence of Ca2+, and that the L-domain binds at the N-terminus of CALP in a region close to the same site. Hence, we explored the possible role of the L-domain during the activation of human erythrocyte calpain. Activation of CALP proceeds following conformational changes reorganizing the catalytic triad, through the conversion of the 80kD to 75kD which requires lower Ca2+ for activity. Our experiments indicate that CALP conversion is reduced in the presence of cast600 (containing the L-domain and one inhibitory unit), and is delayed in the presence of cast300 (single inhibitory unit, without L-domain). Moreover, while addition of free L-domain to cast300 reduces CALP autoproteolysis, its addition to cast600 has no effect. We can postulate that both CSTs prevent the access of the substrate to CALP active site, but cast600 acts also on the autolytic process, inhibiting the formation of the 75kD form. Hence, the CALP/CST interaction is controlled by the L-domain whose presence can address the formation of different enzyme-inhibitor complexes. Since the L-domain binds to CALP also when the protease is in the inactive conformation, we explored if it affects CALP activation at low [Ca2+]. We analyzed whether the 75kD CALP can activate native 80kD CALP in 5 µM Ca2+, a concentration too low for activating erythrocyte calpain. We observed that, the active 75kD CALP does not catalyze the intermolecular conversion of the native CALP to the 75kD form, but it degrades the protease to inactive fragments. When we performed the same experiments with increasing amounts of free L-domain, we observed that CALP degradation mediated by the 75kD CALP is nearly abolished. This was evidenced both from the limited appearance of calpain fragments and, most importantly, by the activity of the 80kD CALP assayed in the presence of substrate. Thus, at low [Ca2+] the L-domain bound to native CALP protects the protease from the degradation mediated by activated CALP and leaves the active site available for substrate digestion. Since the recombinant L-domain (RNCAST110) used in our experiments is a truncated mRNA for CST, detected in rat brain, we are now looking for a similar mRNA in human nervous tumors where we have already observed different ratios of native/activated calpain. These data can be the starting point for exploring new possibilities, not based on CALP inhibition, to correct aberrant CALP activity.