SAT-141 Glycoprotein NMB (GPNMB) Is Pro-Tumorigenic in TSC2-Null Cancer Cells and Is a Potential Drug Target and Biomarker for Lymphangioleiomyomatosis (LAM)

Abstract

Lymphangioleiomyomatosis (LAM) is an estrogen-sensitive lung disease found almost exclusively in women that is characterized by hyperproliferation of smooth muscle cells forming small tumors, or LAM lesions throughout the lungs of patients. Growth of these tumors leads to progressive loss of pulmonary function, and sometimes subsequent lung transplantation. LAM tumor cells contain mutations in either the TSC1 or TSC2 genes, leading to activation of the mTORC1 pathway. In fact, mTOR inhibitors such as sirolimus are commonly used to treat LAM; however, these drugs are not always effective and have significant side effects, suggesting the need for new therapeutic targets. Interestingly, another important feature of LAM cells is that they express melanocytic markers that are normally found in melanocytes or melanoma cells. From RNASeq analysis of a mouse model for LAM that we designed, we discovered significant upregulation of the melanocytic marker Glycoprotein Non-Metastatic Melanoma Protein B (GPNMB), a type I transmembrane protein. GPNMB was not only highly expressed in our mouse model (a uterine specific TSC2-null mouse), it was also expressed in TSC2-null cell lines, and human LAM patient lung samples. In our hands, knocking down GPNMB expression by siRNA directed against GPNMB mRNA decreased migration and proliferation in TSC2-null cells. Additionally, we found that GPNMB’s large ectodomain is shed by TSC2-null cells and can be detected in the blood of human patients with LAM. Finally, MMP 2 and 9 can be secreted as a result of ectodomain shedding and its interaction with integrins. Accordingly, we did indeed see a decrease in MMP 2/9 expression in TSC2-null cells with reduced GPNMB expression from treatment with siRNA directed against GPNMB mRNA. Overall, our results demonstrate the potential importance of GPNMB in LAM tumor progression, and suggest that GPNMB may be a possible LAM biomarker and target for its treatment.