Unexpectedly, we found much stronger signals for MMP-2 and TIMP-2 mRNAs within the mesenchymal cells in the desmoplastic stroma, of endothelial and/or (myo)fibroblastic nature, rather than in tumor epithelial cells in which localization of MMP-2 was anticipated.
However, MMP-2 transcripts appeared to be more abundant in stromal cells, TIMP-1 and TIMP-2 transcripts were evenly distributed over tumor and stromal cells and relatively more MMP-9 transcripts were found in tumor cells.
Elevated mRNA levels of MMP-2 and Col IV were demonstrated in all the cases examined and were associated with in situ disruption of basement membranes around the tumor nests.
MMP-2 mRNA was expressed in the stromal cells in all cases and was more marked in the less-differentiated gastric and colonic carcinomas; it was also detected in the neoplastic cells of poorly differentiated tumors, particularly in those of the signet-ring cell type, both in the colon and stomach.
We did not detect a mutation in the activation locus of the enzyme in any of the tumors studied, which suggests that activation may be due to up-regulation of a tumor-associated gelatinase A activating species.
We conclude that the stromal elements of ovarian tumors express MMP-2 and 9 and their specific inhibitors, but these do not seem to be controlled by endogenous TNF in the tumor microenvironment.
An inactive precursor of gelatinase A (pro-gelatinase A) is secreted and activated in invasive tumour tissue as a result of proteolysis which is mediated by a fraction of tumour cell membrane that is sensitive to metalloproteinase inhibitors.
Besides tumour cells, the MMP-2 mRNA was expressed by microvascular endothelial cells of intratumour and peri-tumour vessels, and by fibroblasts which were especially abundant in the stroma adjacent to the tumour nodules.
In situ hybridization using 35S-labeled riboprobes was used to analyze 33 human ovarian tumors and mouse xenografts of human ovarian (DOV 13, SKOV3) and breast (MCF 7) tumor cell lines known to express MT1-MMP and MMP-2.
When B16-BL6 melanoma cells or Lewis lung carcinoma cells were implanted intradermally, the tumor volumes at 3 weeks after implantation in the gelatinase A-deficient mice decreased by 39% for B16-BL6 melanoma and by 24% for Lewis lung carcinoma (P < 0.03 for each tumor).The number of lung colonies of i.v. injections fell by 54% for B16-BL6 melanoma and 77% for Lewis lung carcinoma (P < 0.014 and P < 0.0015, respectively).
We used northern blot analysis of RNA from fresh-frozen tumour specimens to study the patterns of expression of genes for proteolytic enzymes (cysteine proteinase cathepsin L and aspartyl proteinase cathepsin D; matrix metalloproteinases MMP-9 and MMP-2), tissue inhibitors of metalloproteinases TIMP-1 and TIMP-2, and TP53.
The ratio of type IV collagenase expression (mean of the expression of MMP-2 and MMP-9) to E-cadherin expression (MMP:E-cadherin ratio) at the periphery of the tumors was significantly higher in patients with recurrent disease (4.7 +/- 2.1) than in patients who were disease free (2.3 +/- 1.7; P = 0.0008).
Gelatin zymography indicates that the pro-MMP-2 activation ratio is significantly higher in the glioblastomas than in other astrocytic tumors (P<0.01), metastatic brain tumors (P<0.01), and normal brains (P<0.01).
More importantly, mouse survival was zero after 4-6 months in mice bearing TGF-beta1- and MMP-2-expressing tumors and increased significantly in mice implanted with IL-10- and TIMP-1-expressing tumors (i.e., to >80% survival).
Among the MMP's associated with the tumor cell surface, gelatinase A is believed to be particularly important, since it degrades type IV collagen, and is activated in a tumor specific manner, correlating with tumor spread and poor prognosis.
Tumor growth factor beta (TGFbeta) increased keratinocyte migration as well as both cell-associated and secreted MMP-2 production in wounded cell cultures.