Mst1r supports prostate cancer invasion, dissemination and formation of bone metastases

The receptor tyrosine kinase, MST1R has been implicated in prostate cancer growth and angiogenesis and, in this context; an important role has been attributed to MST1R expressed on tumor-associated macrophages. Here, we observed that MST1R expression in human prostate cancer cell lines correlates with androgen-independency and metastatic capacity. We expressed MST1R shRNAs in androgen-independent, metastatic PC3 cells. Bulk selected stable populations showed ~50% reduction in MST1R mRNA and protein surface expression. As a consequence, cell scattering in 2D and invasive outgrowth in 3D collagen matrices was attenuated, which could be restored by HGF stimulation of the MST1R-related MET receptor. PC3shMST1R cells were also blocked in their ability to disseminate in a zebrafish embryo xenotransplantation model. Furthermore, RNAi targeting MST1R prevented the formation of bone metastases following intracardiac inoculation in mice. Together, these findings demonstrate that down regulation of MST1R can inhibit prostate cancer invasion, dissemination and metastatic colonization. INTRODUCTION The macrophage stimulating 1 receptor (MST1R; RON; Met-related tyrosine kinase) is a receptor tyrosine kinase that has been implicated in various epithelial malignancies including breast, colon, ovarian, liver, and head and neck cancers (1-11). MST1R is also overexpressed in primary prostate cancer and at metastatic prostate cancer sites (12). Experiments using MST1R kinase-deficient mice have established a role for MST1R in prostate tumor growth and angiogenesis (13). Interestingly, this can be explained at least in part by a critical role for MST1R expression on macrophages in the tumor microenvironment (14). MST1R transgene expression in prostate epithelium and findings from in vitro experiments indicate that MST1R expressed on prostate cancer cells could contribute to tumor cell survival, production of angiogenic chemokines, and tumor growth (12, 15). Importantly, while localized prostate cancer has an almost 100% survival rate, the occurrence of metastases in bone and other distant sites lowers 5-year survival to ~30%(16). MST1R has not been implicated in prostate cancer progression. The MST1R-related MET receptor stimulates tumor growth and lymph node metastasis in a xenograft model using human PC3 prostate cancer cells (17) and HGF-MET signaling is an important target in prostate cancer progression (18). In this study, we took an RNAi-based approach to investigate the role of MST1R in several aspects of the prostate cancer metastatic cascade including cell migration, invasion, dissemination, and formation of bone metastases. MATERIALS AND METHODS Prostate cancer cell lines were obtained from ATCC and cultured according to the standard protocol. PC3M-Pro4luc cells have been described previously (19). Lentiviral shRNA constructs targeting MST1R (TRCN000012148; TRCN000012150) were obtained from the MISSION library (Sigma-Aldrich). Stable shRNA expressing PC3 and PC3-M-Pro4luc cells were bulk selected by puromycin. Generation of extracellular matrix (ECM)-embedded tumor cell spheroids was performed as previously described (20). Spheroid outgrowth and collagen invasion was quantified using an automated Image pro 7-based plugin to calculate surface area of the spheroid, number of cells migrating into the collagen, and mean cumulative distance (MCD) travelled by these cells. Recombinant hepatocyte growth factor (HGF; Sigma) was used at 5 ng/ml. For zebrafish xenograft assays, CMDiI-labeled PC3 cells were injected in the yolk of 2-day-old fli-EGFP-casper em97 bryos and fixed 6 days post-implantation as described previously (21). The automated process for collection of confocal image stacks, generation of in-focus composite images, alignment and orientation of the images, and subsequent quantification of tumor cell dissemination was done as described previously (21). Dissemination is described as mean cumulative distance (MCD) reflecting cumulative distance from the primary injection site of all tumor cells in each embryo, averaged over all embryos. Experimental bone metastasis in mice was analyzed by ~weekly whole body bioluminescent imaging (BLI) of nude mice following intracardiac injection with PC3-M-Pro4luc cells as described (19). Metastatic lesions in bone were identified by immunohistochemistry. FACS and Western blot experiments were performed as described previously (22) using MET phospho-Tyr1349 (Cell Signaling) and tubulin (Sigma) antibodies. Data for all experiments are presented as mean ± SEM of at least 2 independent biological replicates. Student's t test (two-tailed) was used to compare groups. RESULTS AND DISCUSSION We evaluated MST1R mRNA expression in a panel of prostate cancer cell lines. MST1R levels in cells reported to be androgen-independent and metastatic in mice, including PC3 cells were 2-8 fold higher than levels found in androgen-dependent non-metastatic cells (Fig 1A). MST1R protein expression was also detected in PC3 cells (Fig 1B). Likewise, in a series of human prostate cancer xenografts representing various aspects of human prostate cancer progression (23), MST1R mRNA levels were higher in most androgenindependent xenografts as compared to androgen-dependent xenografts (Fig 1C). Based on these expression data, we asked if MST1R plays a role in aspects of prostate cancer progression. To address this question, MST1R expression in PC3 cells was silenced using lentiviral shRNAs. In bulk puromycin-selected stable PC3 shRNA populations, two distinct shRNAs caused ~50% silencing MST1R expression (Fig 2A). 98 FIGURE 1. MST1R expression on human prostate cancer cells. A: MST1R mRNA expression determined by qPCR in panel of human prostate cancer cell lines. AD, androgen-dependent; AI, androgen independent. B: Immuno-histochemical staining showing MST1R expression in PC3 cells. Right panel shows zoom-in of area marked by dotted line in left panel. Adapted from www.proteinatlas.org. C: MST1R RNA expression determined by qPCR in series of human prostate cancer xenograft models. VC aP LN C aP 22 R V1 LA PC 4 M D A PC a2 B D U 14 5 PC 3 0 50 100 150 200 250 B