Open Collections

Abstract

The activation of gelatinase A (matrix metalloproteinase-2) occurs on the cell surface via a unique mechanism that requires a ternary complex of membrane-type 1 matrix metalloproteinase (MT1-MMP) and tissue inhibitor of metalloproteinases (TIMP)-2, which acts as a cell surface receptor for progelatinase A. TIMP-2 binds to both the MT1-MMP and the hemopexin C-domain of progelatinase A thus linking progelatinase A to the cell surface where a second MT1-MMP can begin the activation process by cleaving the propeptide of progelatinase A. Recently, our lab has shown that TIMP-4, which shares a high degree of homology with TIMP-2, can bind and inhibit MT1-MMP and independently bind the hemopexin C-domain of progelatinase A in a manner similar to TIMP-2 but is unable to support the activation of progelatinase A. TIMPs consist of two domains: the N-terminal inhibitory domain and the C-terminal domain. Additionally, TIMPs-2, -3, and -4 contain nine residues at the C-terminus known as the C-terminal tail. TIMPs-2 and -4 are hypothesized to bind to progelatinase A through their non-inhibitory C-domains. To investigate the differences in the interactions of TIMP-2 and TIMP-4 with progelatinase A, the C-terminal domains of TIMPs-2 and -4 were each cloned as a C-terminal fusion partner to horse heart myoglobin. The anionic C-terminal tail of TIMP-2 is hypothesized to bind a cationic site on the hemopexin C-domain of progelatinase A during activation. Therefore, to investigate the aspects of TIMP-4 that renders it deficient in the activation of progelatinase A, a set of mutations were introduced into the C-terminal tail of TIMP-2 and -4 and the activities of 8 fusion proteins were investigated. Myoglobin-C-TIMP-2 (MbcT2) bound progelatinase A with reduced affinity relative to full length TIMP-2, whereas deletion of the tail decreased the affinity further. In contrast, deletion of the C-terminal tail of C-TIMP- 4 did not alter the binding of Mb-C-TIMP-4 (MbcT4) to progelatinase A and both proteins bound with only slightly less affinity than full length TIMP-4. The similarity of the binding affinities of TIMP-4 and MbcT4 to progelatinase A indicates that the N-terminal domain and C-terminal tail of TIMP-4 is unlikely to contribute to its interaction with progelatinase A. Mutations that removed the negative charges in the C-terminal tail of MbcT2 (E192V/D193Q) also resulted in a reduction in binding, as did a substitution of the C-terminal tail of MbcT2 with that of TIMP-4. Conversely, addition of negative charges to the C-terminal tail of MbcT4 (V193E/Q194D) resulted in an increased affinity for progelatinase A, as did a substitution of its tail with that of TIMP-2. The overall analysis of these fusion protein constructs show that the C-terminal tail of TIMP-2 is integral to its ability to support the activation of progelatinase A, specifically the negatively charged residues Glu192 and Asp193. The data also indicate that the reason TIMP-4 cannot support the activation of progelatinase A is because of the lack of these negatively charged residues in the TIMP-4 C-terminal tail.