Prostate cancer progression and tumor-induced osteolysis: Implications of a tumor-associated MT1-MMP/RANKL/RANK signaling axis
Abstract
Tumor-derived proteases influence normal bone remodeling in Prostate Cancer (PC) growth within the bone microenvironment drastically increasing disease burden and patient morbidity. Elucidating key molecules involved in intraosseous tumor growth is essential for advancing treatment of this incurable and end stage of the disease. Matrix Metalloproteinases are involved in degradation of a vast array of substrates, and increased tumor-associated MMP activity is involved in multiple cancer phenotypes. Specifically, Membrane-Type-1 Matrix Metalloproteinase (MT1-MMP) is the primary collagenase responsible for enzymatic degradation and cell growth within a threedimensional type-I collagen matrix, which is the major component of non-mineralized extracellular matrix (ECM) of bone. The functional significance of MT1-MMP with regard to the bone became obvious when MT1-MMP-deficient mice were found to have marked skeletal defects. MT1-MMP is also known for its ability to cleave non-ECM substrates. Interestingly, recent evidence suggests a potential role of MT1-MMP as a sheddase of a membrane-bound form of Receptor Activator of NF-kappa beta Ligand (RANKL). The importance of RANKL regulation of bone remodeling was demonstrated in RANKL-deficient mice resulting in severe osteopetrosis. Whether the shed form of RANKL retains the ability to bind its cognate receptor, RANK, promoting osteoclast activity is unknown. Briefly, RANKL is a well-known osteoclastogenic factor primarily expressed on the surface of cells of the osteoblast lineage. New evidence suggests that PC bone tumors also express both RANKL and RANK. Additionally, in PC cells RANK activation has been shown to enhance migration and lead to upregulation of osteoclast-related genes. The potential influence of RANKL activity generated from PC in the bone environment has not been sufficiently investigated. Furthermore, MT1-MMP is rarely expressed in normal epithelial tissue; however, we found, via immunohistochemical techniques, that positive staining of tumor-associated MT1-MMP correlates with the progression of primary prostate lesions. We hypothesized that tumor-associated MT1-MMP aids in the expansion of the bone metastatic deposit via direct proteolytic degradation of a collagen I rich bone matrix. Additionally, due to the osteomimicry often displayed by tumor cells in the bone, and the need for tumor-stromal interactions to take place for the metastatic progress of the tumor, we postulated that the tumor-associated MT1-MMP-dependent shedding of tumor-derived RANKL will initiate an MT1-MMP/RANKL/RANK signaling axis working in a paracrine manner with cells endogenous to the bone, and in an autocrine manner with PC cells. We found MT1-MMP expression in real PC bone metastases, obtained postmortem, from patients with PC. To characterize the role cancer cell-associated MT1-MMP has on intraosseous PC growth, we developed stable transfectants with high (LNCaP-MT1) and low (LNCaP-Neo) MT1-MMP expression. We found that LNCaP-MT1 cells, upon intra-tibial injection in immunosuppressed mice, resulted in decreased bone area, and increased tumor growth. Because the osteolytic response observed in vivo was unlikely due exclusively to direct degradation of ECM by tumor-associated MT1-MMP at the tumor-bone interface, we first examined the existence of our hypothesized tumor associated MT1-MMP/RANKL/RANK signaling axis signaling axis on osteoclasts. We found CM from LNCaP-MT1 cells, which endogenously express RANKL, had an increased ability to induce osteoclast maturation. Furthermore, this effect was abrogated in the presence of the RANKL decoy receptor, osteoprotegrin (OPG, or upon MT1-MMP inhibition. These data suggest that tumor-derived MT1-MMP processing of RANKL is an event which promotes RANK activation within the bone microenvironment. After establishing a paracrine function of tumor-associated MT1-MMP/RANKL/RANK signaling axis we investigated a novel tumor-associated MT1-MMP/RANKL/RANK autocrine signaling axis in PC cells. We began by assessing the downstream effect of RANKL on PC cells, finding RANKL to induce Src-dependent PC cell migration. To establish the link between tumor-associated MT1-MMP and our hypothesized autocrine pathway we treated the LNCaP-derived PC cell line, C42b, with CM from our LNCaP-MT1 cells. LNCaP-MT1-derived CM were able to significantly increase C42b cell migration and Src activation. Furthermore, the phenotypic effect of LNCaP-MT1-derived CM could be significantly hindered using selective inhibitors for MT1-MMP and Src. Overall, this study implicated the substantial effect prostate tumor-associated MT1-MMP has on tumor expansion in bone. Additionally, we highlight a the role of a tumor-associated MT1-MMP/RANKL/RANK paracrine and autocrine signaling mechanism acting on cells endogenous to the bone, and the tumor cells themselves. Importantly, this study reveals multiple potential therapeutic targets focused on inhibition of prostate tumor-associated MT1-MMP, RANKL, RANK, or Src for treatment of PC skeletal metastases.
Recommended Citation
Aaron Sabbota,
"Prostate cancer progression and tumor-induced osteolysis: Implications of a tumor-associated MT1-MMP/RANKL/RANK signaling axis"
(January 1, 2009).
ETD Collection for Wayne State University.
Paper AAI3350208.
http://digitalcommons.wayne.edu/dissertations/AAI3350208
View More
