Unexpected Effect of Matrix Metalloproteinase Down-Regulation on Vascular Intravasation and Metastasis of Human Fibrosarcoma Cells Selected In vivo for High Rates of Dissemination

Abstract

The human tumor/chick embryo model involving grafting of human HT-1080 fibrosarcoma cells on the chorioallantoic membrane was used in conjunction with quantitative real-time Alu PCR to select in vivo a pair of isogenic cell lines (HT-hi/diss and HT-lo/diss), dramatically differing in their ability to disseminate from the primary tumor (i.e., intravasate into the chorioallantoic membrane vasculature and metastasize to the lungs). During an immunohistochemical time course study, HT-hi/diss cells were sequentially visualized having escaped from the primary tumors, engaged with the blood vessels, and eventually observed inside the chorioallantoic membrane capillaries, thus reflecting early intravasating events. In contrast, HT-lo/diss cells seemed restricted to their primary tumor. Importantly, after i.v. inoculation, both variants arrested, extravasated, and proliferated in host tissues with similar efficiencies, highlighting that the observed earlier events at the periphery of the primary tumor could account for their differential dissemination. In a mechanistic probing of these events, we determined that HT-hi/diss intravasation was sensitive to a broad-range matrix metalloproteinase (MMP) inhibitor. To analyze the possible role of individual MMPs, membrane-bound MMP-14 and secreted MMP-9 were individually down-regulated in HT-hi/diss cells with their corresponding small interfering RNAs. Despite efficient down-regulation of MMP-14, neither intravasation nor metastasis of HT-hi/diss cells was affected significantly. However, a substantial down-regulation of MMP-9 was accompanied by a surprising 3-fold increase in intravasation and metastasis. The results emphasize a rising awareness that targeting certain MMPs might result in an enhanced malignancy, exemplified herein at the intravasation level as this step of the metastatic cascade is dissected and quantified.