The role of matrix metalloproteinases (MMP-9) in the proliferation of colorectal cancer metastases to the liver.

2417

Aim: To determine the role of MMP-9 in the proliferation and invasion of colorectal cancer metastases to the liver. Background: MMPs have been shown to play a role in tumorigenesis of primary epithelial tumors such as those of colon and breast origin. The majority of patients who succumb to colorectal cancer, die with a significant burden of metastatic disease in the liver. The role of specific MMPs in the proliferation and growth of metastatic colorectal tumor cells in the liver remains to be elucidated. MMP-9 is known to play a role in the angiogenic switch and in the growth of other tumor types. Through a better understanding of the roles of tumor and stroma derived MMPs at the earliest stages of tumor metastasis, we hope to identify specific, rational targets for future therapies. Methods and Results: We have adapted the murine splenic injection model of colorectal metastases. This orthotopic model is used to recapitulate the latter stages of colorectal metastases. Wild type (WT) C57BL/6 mice and C57BL/6 mice genetically deficient in MMP-9 underwent splenic injection of 1X 10⁵ syngeneic MC38 colon cancer cells. This technique reproducibly resulted in liver metastases in the mice injected. MC38 cells were stably transfected with a luciferase reporter construct to allow in vivo monitoring of the growth of metastatic tumors. RTPCR analysis of MC38 cells shows that these epithelial tumors produce MMP-9. After splenic injection, mice genetically deficient in MMP-9 consistently show minimal metastatic growth in the liver by gross examination as well as bioluminescence imaging when compared to WT controls. After 14 days, the control mice showed more than a 2 fold increase in liver weight with their tumor burden when compared to MMP-9 null mice (p value 0.01). Serial bioluminescence imaging during the 14 day post splenic injection period shows decreased signal intensity in the MMP-9 null mice when compared to WT controls. In addition, we have imaged the same mice at successive time points (days 7 and 14) using MRI. This technique gives spatial and anatomic resolution that corroborates the markedly decreased burden of liver metastases in the MMP-9 null mice as compared to WT controls. Conclusions: To our knowledge this is the first example of bioluminescence imaging of metastatic tumors in an orthotopic model. We have demonstrated that host stromal MMP-9, and not tumor derived MMP-9, is important for the proliferation of colorectal metastases to the liver in this orthotopic murine model. Understanding the role of specific MMPs in colorectal cancer metastases will allow targeted modulation of these enzymes in therapeutic strategies.