Altered protein nucleocytoplasmic trafficking is increasingly recognized as a bona fide driver of cancer progression. Many tumor suppressor proteins (e.g. p53, Rb) function in the nucleus, but undergo increased nuclear export in many types of cancer resulting from overexpression of their nuclear exporter XPO1/CRM1. Profilin-1 (Pfn1) is an actin-binding protein with well-documented anti-tumor and anti-metastatic activities in various types of cancer. However, its role as a tumor suppressor remains controversial because it is rarely mutated and paradoxically essential for cell growth and survival owing to its facilitation of actin polymerization. Though generally considered a cytoplasmic protein, Pfn1 is present in the nucleus of many mammalian cells with poorly understood function. We have previously found that the in vitro anti-tumor effect of Pfn1 in breast cancer cells requires its ability to enter nucleus. This suggests that Pfn1's tumor suppressor activity may stem from its ?moonlighting? function in the nucleus that is spatially separated from its essential cytoplasmic actin- regulatory functions. Pfn1, in a complex with actin, undergoes active nuclear export by the nuclear exporter exportin-6 (XPO6). Unlike XPO1/CRM1, XPO6 is highly selective with Pfn1/actin being its only known cargo. We found in the TCGA datasets that the mRNA level of XPO6 is significantly upregulated in 65-100% of tumor samples across 12 different types (including breast). We also discovered that XPO6 protein level is significantly increased in a panel of luminal and basal-like breast cancer cell lines as compared to untransformed breast epithelial cells. XPO6 knockdown selectively inhibited the growth and survival of several breast cancer cell lines while having little on the untransformed control cells. Together, these data suggest that the primary mode of Pfn1 deregulation in cancer may be its increased nuclear export resulting from overexpression of its nuclear exporter XPO6. In addition, we have uncovered a novel interaction between nuclear Pfn1 and the multi-protein super elongation complex (SEC) by co-IP and mass spectrometry. SEC positively regulates the transcription elongation of many pro-tumor and pro-EMT/metastasis genes by phosphorylating (through its components CDK9/cyclin T1) and un-pausing promoter proximal RNA polymerase II to trigger productive elongation. We demonstrate that Pfn1 knockdown increases phospho-RNA Pol II level in breast cancer cells and sensitizes them to CDK9 inhibition. Together, our data support the hypothesis that Pfn1 functions in the nucleus as a tumor suppressor by inhibiting SEC-mediated gene transcription and its subcellular localization is frequently deregulated in cancer through overexpression of its nuclear exporter XPO6. In this grant, we will test this hypothesis using various approaches including in vitro and in vivo tumor models, transcriptome and epigenome analysis, and primary tumor sample analysis.
It is an emerging notion that cancer can be driven by altered cellular location of oncogenes and tumor suppressors. The actin-binding protein profilin-1 (Pfn1) has tumor suppressive function in multiple types of cancer but its clinical relevance remains unclear due to its extremely low mutation rate and its paradoxically essential function in cell growth and survival. We hypothesize that Pfn1's tumor suppressor activity stems from its ?moonlighting? function in the nucleus that involves transcription repression but undergoes frequent downregulation in cancer due to increased nuclear export by its nuclear exporter exportin-6 (XPO6). We propose to test this hypothesis with the ultimate goal of developing novel therapeutics to prevent Pfn1 nuclear export by XPO6 and increase its nuclear tumor-inhibitory activity.