ROLE OF THE ENDOCYTIC PROTEIN EPSIN 3 IN MAMMARY GLAND TRANSFORMATION

Background: Endocytosis is a crucial mechanism regulating cellular signaling and its dysregulation is linked to cancer. We previously showed that the endocytic protein Epsin 3 (EPN3) is overexpressed in 50% of breast cancers (BC) and amplified in approximately 10% of cases. EPN3 overexpression correlates with increased tumor invasiveness and distant metastasis. Interestingly, the well-characterized oncogene ERBB2 is co-amplified with EPN3 in half of the EPN3-amplified cases, suggesting the possible cooperation of the two genes. By overexpressing EPN3 in non-transformed mammary epithelial cells (MCF10A), we showed that it induces a partial epithelial-to-mesenchymal transition (pEMT), a state linked to the expansion of the stem cell pool and to cellular invasion. Mechanistically, EPN3 overexpression induces pEMT by enhancing E-Cadherin (ECAD) endocytosis and turnover, which destabilizes cell-to-cell adherens junctions. Hypothesis and Aims: We hypothesized that EPN3, through its regulation of endocytic programs leading to pEMT, acts as an oncogene and may cooperate with other oncogenes in inducing tumorigenesis and metastasis. Our aim was to test this hypothesis by investigating the role of EPN3 in BC development and progression using murine models. Specifically, we used in vivo models to determine the impact of EPN3 overexpression on mammary gland morphogenesis, tumorigenesis, and metastatic dissemination. In addition, we aimed to determine whether the endocytic function of EPN3 is linked to its oncogenic role. Results: To better understand the role of EPN3 in mammary gland tumorigenesis, we generated a conditional EPN3 knock-in (KI) mouse model. This model revealed that although EPN3 overexpression is insufficient by itself to induce mammary tumors, it alters mammary morphogenesis by increasing ductal branching. Single-cell RNA-seq revealed increased expression of an EMT signature in basal mammary cells and a pro-lactating and proliferative phenotype in all epithelial compartments upon EPN3 overexpression. Isolated primary mammary epithelial cells from EPN3-KI mice exhibited accelerated ECAD internalization and a heightened migratory/invasive phenotype compared to CTRL cells, mirroring results obtained in EPN3-overexpressing MCF10A cells. The mechanistic dissection of EPN3-mediated ECAD endocytosis revealed that it occurs via the glycolipid-lectin (GL-Lect) pathway. Indeed, EPN3-induced ECAD endocytosis and cellular invasion were inhibited using the pan-galectin and galectin 3 (Gal3) inhibitors, lactose and I3, respectively. Additionally, to investigate whether EPN3 and ERBB2 cooperate in tumor progression, we generated an EPN3-KI:MMTV-NeuN mouse overexpressing EPN3 and the ERBB2 rat homologue, NeuN, in the mammary gland. These mice displayed an accelerated tumor onset and increased pulmonary metastatic dissemination compared with MMTV-NeuN mice. Conclusions and Future Work: Our findings in mouse models indicate that EPN3 overexpression in the mammary gland promotes conditions favorable to tumorigenesis and metastasis. These pro-oncogenic effects are associated with increased ductal branching and enhanced migratory/invasive phenotypes mechanistically linked to accelerated ECAD internalization and alterations of molecular pathways related to EMT and lactation. EPN3-mediated ECAD endocytosis occurs via the GL-Lect pathway and can be inhibited by galectin inhibitors. Furthermore, the cooperation of EPN3 and ERBB2 accelerates tumor onset and metastasis, highlighting a potential synergistic role of these two oncogenes in human BC. Future studies will investigate the role of aberrant ECAD endocytosis in EPN3-dependent phenotypes in primary mammary epithelial cells and mammary organoids derived from mice using ex vivo assays. Our findings reveal a novel GL-Lect endocytic pathway with EPN3 as a key player that contributes to BC development and metastasis. A better understanding of this specific endocytic route could lead to the identification of novel molecular targets for therapeutic intervention in BC patients.