The role of matrix metalloproteinases in cell-matrix interactions.

Cellular interaction with the extracellular matrix influences basic processes such as proliferation, differentiation, migration and invasion. Matrix metalloproteinase (MMP) activity has been implicated in the remodelling of tissue associated with these events and in the migration of cells through tissue barriers. Evidence suggests that MMP expression and activation is influenced by the matrix components to which the cell is exposed. The recognition sites of the integrins α1β1 and α2β1 within collagen IV are located in the cyanogen bromide fragment CB3[IV] region, close to the reported gelatinase A (GLA) cleavage site. The susceptibility of solubilised forms of collagen IV to GLA cleavage and the concomitant effects on cell adhesion were assessed. Preparations of collagen IV monomers with disrupted intramolecular disulphide bonds in the CB3[IV] region were cleaved by GLA into the two characteristic 100nm-300nm fragments at 30°C and were totally degraded at 37°C. This resulted in the partial or total inhibition of cell adhesion to the collagen via integrin receptors. Preparations of monomers with intact disulphide bonds in the CB3[IV] region were not susceptible to GLA cleavage. Furthermore, no effect on binding of cells to the GLA-treated collagen could be detected after treatment at 37°C. Dimeric collagen IV with intact disulphide bonds in the CB3[IV] region was not degraded by GLA at 37°C. Tetrameric collagen was highly susceptible to GLA degradation at 37°C, but this did not affect cell adhesion, indicating that the CB3[IV] region of the tetramers remained intact. The effects of fibronectin and laminin-1 matrices on the activation of GLA was assessed. HT1080 fibrosarcoma cells cultured on fibronectin activated endogenous GLA to a level comparable with that elicited by phorbol ester treatment. In contrast, cells cultured on laminin-1 secreted mainly proGLA. Cells cultured on Augments of fibronectin derived from the central cell binding domain secreted levels of active GLA similar to those observed for full length fibronectin. A substrate of immobilised anti-α5 or anti-β1 integrin antibodies promoted GLA activation, whereas an anti-α6 antibody did not. The data demonstrate that that signals via the α5β1 integrin receptor may be involved in the fibronectin-induced up-regulatiun of GLA activation. The effect of fibronectin on the components of the putative MT1-MMP/TIMP-2 'receptor' complex for GLA was assessed. Levels of TIMP-2 protein expressed by HT1080 cells did not vary detectably between cells cultured on fibronectin or laminin-1. However, a fibronectin substrate increased the processing of MT1-MMP to a truncated 45 kDa form, which was concomitant with GLA activation. Inhibitor studies showed that the truncation of MT1-MMP to 45 kDa is MMP mediated, although not inhibited by TIMP-1. This study raises the possibility that truncation of MT1-MMP to 45 kDa form represents a regulatory end-point in the activation pathway of GLA