Search

Although it is generally accepted that proteolytic degradation is an important mechanism used by malignant cells in the process of metastasis, comparatively little is known about the regulation of molecules responsible for proteolysis and how they become de-regulated during human tumour progression. Using a genetically related pair of human melanoma cell lines, derived from the same patient at different stages of disease, we analysed differences in the cytokine-mediated regulation of gelatinase B (MMP-9), an enzyme thought to play an important role in cellular invasiveness, and TIMP-1, a physiological inhibitor of this enzyme. Whereas the advanced stage (i.e. metastatic) partner of this pair (WM 239) could produce gelatinase B upon induction with interleukin (IL)-1beta or tumour necrosis factor alpha (TNF-alpha), the early stage (i.e. primary) partner (WM 115) could not. In sharp contrast, we found that TIMP-1 displayed an opposite pattern of induction in these cell lines. Specifically, the early stage cell line, WM 115, demonstrated a marked increase in the production of TIMP-1 when treated with IL-1beta or TNF-alpha whereas the advanced cell line, WM 239, showed no such increase. Treatment with the DNA demethylating agent, 2-deoxy-5-azacytidine, resulted in a marked up-regulation of both gelatinase B and TIMP-1 in both cell lines. It was further found that constitutive overexpression of gelatinase B in WM 239 cells and an additional melanoma cell line (MeWo), derived from a metastatic lesion, was able to greatly enhance lung colonization in an experimental metastasis assay while we did not observe differences in tumorigenicity. From these results we conclude that an altered responsiveness of gelatinase B and TIMP-1 to induction by similar agents is associated with disease progression in human melanoma and that this altered responsiveness can have consequences to the aggressive nature of the disease.

Although a number of effective therapies are available for localized prostate cancer, metastatic prostate cancer is difficult to treat and impossible to cure. Identification of the gene products that enable a prostatic carcinoma cell to metastasize should facilitate an understanding of the processes leading to metastasis. To characterize the contribution of matrix metalloproteinase-9 (MMP-9, gelatinase B or the 92-kd type IV gelatinase/collagenase) to the development of metastasis in prostate cancer, we reduced MMP-9 expression in metastatic murine prostatic carcinoma cells using a ribozyme. The ribozyme transfected cells had lower basal levels of MMP-9 as well as decreased levels after stimulation by transforming growth factor-beta or phorbol 12-myristate 13-acetate when compared with the parental cells or with control transfectants. The cells with down-regulated MMP-9 were unable to form lung colonies in the experimental metastasis assay, whereas the controls and parental cells readily formed metastases. All cell types readily formed tumors after injection and down-regulation of MMP-9 did not adversely affect the rate of tumor growth. Thus, MMP-9 expression is required for hematogenous metastasis in a murine prostate model system raising the possibility that it may play an equivalent role in human prostate cancer.

The 92-kDa type IV collagenase (MMP-9) is a metalloproteinase frequently localized in both tumor stroma and in tumor cells, particularly at the tumor invasion front. To explore the factors regulating transcriptional activation of MMP-9 in stromal cells, we used a model system in which fibroblast MMP-9 expression can be upregulated by cell-cell contact with metastatic transformed rat embryo cells. Using transient transfection of reporter gene constructs containing 5'-deleted or mutated MMP-9 promoter fragments, as well as electrophoretic mobility shift assays, the upstream NFkappaB, SP-1, and Ets sites and the downstream AP-1 site and retinoblastoma binding element were shown to be necessary for basal transcriptional activity of fibroblast MMP-9. In contrast only Ets or SP-1 appeared to be involved in contact-mediated induction of MMP-9. Mutation of the upstream AP-1 site increased both basal and contact-stimulated promoter activation. Deletion of the alternating purine-pyrimidine repeat in the downstream promoter decreased transcriptional activity. Together these findings suggest that Ets and SP-1 are the central transcriptional activators of MMP-9 gene expression in fibroblasts specifically responding to tumor cell contact, and that promoter conformation may regulate MMP-9 expression.

The 92 kd type IV collagenase/gelatinase (MMP-9) is important in mediating basement membrane and extracellular matrix degradation in metastasis. Because MMP-9 is made in tumor cells, but not in quiescent normal cells, we wished to identify the transcriptional elements responsible for its synthesis in tumor cells. We chose to characterize transcriptional regulation of the MMP-9 gene in a highly metastatic H-ras and v-myc transformed rat embryo cell line which overexpresses MMP-9. Using transient transfection of reporter gene constructs containing either 5'-deleted or mutated MMP-9 promoter fragments, as well as electrophoretic mobility shift assays, we have demonstrated that multiple transcription factor consensus binding motifs in the promoter, including those for NFkappaB, SP-1, Ets, AP-1, and a retinoblastoma binding element, participate in transcriptional regulation of MMP-9 expression in this cell line. Also, deletion of an alternating purine-pyrimidine tract in the downstream promoter was found to decrease transcriptional activity, suggesting that promoter conformation may be important in MMP-9 regulation. Thus multiple pathways leading to activation of NFkappaB, SP-1, Ets, AP-1, and retinoblastoma binding factors in tumor cells all may contribute to MMP-9 transcription and hence to metastasis.

Tumor-stromal interactions appear to play an important role in the induction of metalloproteinase expression in malignant tumors. We describe a tissue culture system in which expression of MMP-9 (gelatinase B or the 92 kDa type IV collagenase/gelatinase) was induced by co-cultivation of fibroblasts with breast cancer cell lines. While neither the breast cancer cells nor the normal rat embryo fibroblasts made MMP-9 alone in culture, human MMP-9 was made in the co-cultures. The MMP-9 was secreted in a latent form. The induction occurred at least in part through increases in the MMP-9 mRNA levels in the breast cancer cells. These increases did not appear to require protein synthesis. Conditioned medium from the fibroblasts could duplicate the induction of MMP-9 in the breast cancer cell lines. The active factor in the medium was inactivated by heat or by trypsin suggesting that it was a protein. This protein was in the size range of 30-100 kDa. Thus, fibroblasts could secrete a factor which was able to regulate the expression of MMP-9 in breast cancer cells.

Considerable remodeling of the extracellular matrix as well as cellular migration takes place during embryogenesis. Since the metalloproteinase MMP-9 is implicated in these functions in cancer cells, we studied the patterns of expression of MMP-9 mRNA during the development of post-implantation mouse embryos. MMP-9 mRNA was detected using the ribonuclease protection assay in poly A+RNA from 13 to 17 day embryos, but not at 11 days. In order to localize these transcripts, in situ hybridization was performed on sections of murine embryos from 7.5 to 15 days of gestation. At the time of implantation, MMP-9 mRNA was localized to the invading trophoblast cells. Strong signals were also seen in the yolk sac. No signal for MMP-9 mRNA was seen by in situ hybridization in the embryo until day 11 when detectable reaction was seen in the central nervous system. By day 15 strong signals were seen in the liver, in the developing bronchial epithelium of the lungs and in the primordial alveoli, in the epithelium of the thyroid gland, in the thymus, in the endochondrial plates of the bone, and in neural cells. The liver from day 15 embryos contained gelatinase activity at 105 kDa consistent with MMP-9. Thus, MMP-9 expression appears to be expressed in specific organs in a precise temporal sequence during development.

Expression of MMP-9 mRNA, a type IV collagenase gene product, was followed during embryonic development of the mouse brain using in situ hybridization. Murine embryos from 7.5 to 15 days after fertilization were sectioned and evaluated for MMP-9 expression. During early development, from day 7.5 to day 9, no signal was detected in the cells of the neuroepithelium or in cells of the cephalic mesenchyme of the neural tube. At day 11, gene expression was localized to the Rathke's pouch and the germinal zone of the primitive ventricular system. At day 13, but most notably at day 15, high levels of MMP-9 were expressed by progenitor cells in close association with the development of structures, such as the hypophysis, the choroid plexus, the ganglion cell layer of the retina and the uveal tract. High MMP-9 mRNA levels were also associated with dense cellular aggregates destined to form the highly vascular grey matter of the brain. The presence of MMP-9 mRNA was confirmed using a ribonuclease protection assay. A 105 kDa gelatinase, consistent with the expected molecular mass for the murine MMP-9, was detected in embryonic brain extracts by substrate gel electrophoresis. To our knowledge, this is the first report on the localization of MMP-9 in developing neural tissues. Our results suggest that MMP-9 expression may have a previously unsuspected role in neural development.

The expression of MMP-9 in rat embryo fibroblasts (REF) transformed with Ha-ras or with Ha-ras and v-myc is associated with metastatic behavior. In contrast, REF transformed with Ha-ras and the adenovirus E1A genes (E1A) are tumorigenic, do not release MMP-9 and are rarely metastatic. In this report, we establish that E1A expression results in decreased levels of MMP-9 mRNA in an Ha-ras and v-myc transformed cell line and examine which of the functional domains of E1A participate in the inhibition of MMP-9 expression and which contribute to the suppression of metastasis. The metastatic 2.10 REF line, derived by co-transfection with v-myc and Ha-ras, constitutively expresses high levels of MMP-9 (92 kDa gelatinase). Transfection of E1A wild-type plasmids into this cell line eliminates detectable MMP-9 mRNA expression and greatly reduces MMP-9 activity. Transfection of 2.10 with E1A plasmids encoding mutations in conserved region 2 (CR2) retained inhibition of MMP-9 similar to the inhibition seen with wild-type E1A. Transfection with E1A containing mutations in CRI or the amino terminal region diminished, but did not fully inhibit, MMP-9 expression. In contrast, inhibition of MMP-9 was lost in with E1A mutations in CR3. Cells transfected with E1A mutants in CR1, the amino terminal region or CR3 retained metastatic behavior. Our findings delineate the regions of E1A responsible for MMP-9 inhibition and further define the domains of E1A responsible for inhibition of metastasis.

To determine whether specific collagenolytic enzymes are expressed in human fetal membranes with labor, we examined gelatinase activity in extracts of amniochorion by zymography. The 92-kd gelatinase (MMP-9) was barely detectable in extracts of fetal membranes before the onset of labor but was readily demonstrable in extracts prepared from membranes isolated from laboring women or membranes collected immediately after delivery. In contrast, the 72-kd gelatinase (MMP-2) was detectable in extracts from pre- and post-labor membranes. Ethylenediaminetetracetic acid and the tissue inhibitor of metalloproteinases, TIMP-1, inhibited the gelatinase activities detected by zymography, confirming that the enzymes are metalloproteinase. Assay of amniochorion gelatinase activity using a radiolabeled denatured collagen substrate revealed a more than twofold increase in activity comparing pre-labor with post-labor fetal membrane extracts. A function-blocking anti-MMP-9 monoclonal antibody inhibited pre-labor membrane gelatinase activity by approximately 11.5%, which was only slightly greater inhibition than observed with irrelevant monoclonal antibodies. However, post-labor membrane gelatinase activity was reduced by 53% by the function-blocking antibody, indicating that MMP-9 is a major contributor to the increased gelatinase activity extractable from post-labor membranes. Western blot analyses demonstrated increased MMP-9 protein in amniochorion extracts after onset of labor. MMP-9 protein and mRNA were co-localized in amnion epithelium, underlying macrophages and chorion laeve trophoblast and decidual cells after labor. We conclude that 1) MMP-9 activity and protein in human amniochorion increases with labor and 2) MMP-9 is expressed by amnion epithelium, macrophages and chorion laeve trophoblast and decidual cells. The increased expression of MMP-9 may result in degradation of the extracellular matrix of the fetal membranes and facilitate their rupture under both physiological and pathological conditions.

Members of the matrix metalloproteinase (MMP) family have been implicated in the metastasis of tumor cells, but no direct evidence linking any given member of the MMP family to metastatic behavior has been presented. Rat embryo cells transformed by the Ha-ras and v-myc oncogenes or by Ha-ras alone are metastatic in nude mice and release the 92-kDa gelatinase/collagenase (MMP-9), whereas those transformed by Ha-ras plus the adenovirus E1A gene are not metastatic and do not release MMP-9. Here we demonstrate that MMP-9 expression can be induced in these tumorigenic but nonmetastatic rat cells by transfection with an MMP-9 expression vector. Transfection of a MMP-9 expression vector, but not control DNAs, conferred metastatic capacity on the nonmetastatic cells. The majority of colonies isolated after continued passage either in vivo or in vitro had lost the MMP-9 expression vector. However, occasional cells were isolated from metastases which retained MMP-9 expression after passage. These cells retained metastatic capacity. In contrast, cells isolated after losing MMP-9 expression also lost the ability to metastasize. These results provide direct evidence that MMP-9 has a role in tumor metastasis.

Previous studies have correlated release of the 92 kDa type IV collagenase/gelatinase by tumor cells in culture with metastatic potential. We have now demonstrated that the ability of tumor cells that do not express the 92 kDa gelatinase to induce release of this metalloproteinase from normal fibroblasts may also be associated with the metastatic phenotype. A transformed rat embryo cell line, 2.8, failed to release the 92 kDa gelatinase alone in culture, but gave rise to metastatic tumors whose explants contained the 92 kDa gelatinase. In contrast, a non-metastatic transformed cell line, RA3, did not express the 92 kDa gelatinase alone in culture or in tumor explants. To explore the mechanisms that might govern host-tumor cell interactions in this system, we have studied the effects of co-culture of these transformed cell lines with rat embryo fibroblasts (REF) in culture. 92 kDa gelatinase expression was induced by co-culture of 2.8 with REF, but co-culture of the non-metastatic line RA3 with REF did not result in induction of the 92 kDa gelatinase. The 92 kDa gelatinase in these co-cultures was released by the fibroblasts; methanol-fixed 2.8 cells induced 92 kDa gelatinase expression in REF, but fixed REF cells did not induce enzyme expression in 2.8 cells. This suggested that cell contact was required for induction, which was confirmed by showing that 92 kDa gelatinase induction in co-culture was abolished by separating REF from 2.8 by solute-permissive membranes. In addition, REF could not be stimulated to produce the 92 kDa gelatinase by 2.8-derived conditioned medium, by 2.8-derived extracellular matrix, or by isolated matrix components. These data indicate that metastatic tumor cells can induce 92 kDa gelatinase expression in fibroblasts through a mechanism dependent upon cell contact. In situ hybridization of nude mouse tumors derived from these transformed cell lines revealed 92 kDa gelatinase expression in the stroma of tumors from 2.8, but not in tumors from RA3. Therefore, the experiments based on in vitro co-culture of tumor cells and fibroblasts, together with the in situ localization of mRNA to host cells, suggest that host production of the 92 kDa gelatinase may occur in response to direct contact with metastatic tumor cells.

Many enzymes capable of proteolytic degradation of extracellular matrix and basement membranes have been implicated in tumor progression, including the matrix metalloproteinases, cathepsins, plasminogen activators, and heparanase. Matrix metalloproteinases, a family of zinc-dependent proteases, participate in several steps in tumor progression, including invasion, metastasis, and angiogenesis. In this review, we will give a brief overview of this protease family, and we will review in vitro and in vivo evidence implicating a particular metalloproteinase, the 92-kD type IV collagenase/gelatinase (MMP-9 or gelatinase B), as well as other metalloproteinases, in cancer progression. Finally, using recent studies from our laboratory, we will demonstrate the importance of both tumor cell and host stromal cell production of MMP-9 in tumor progression.