Background In a genomic screen for determinants of the tumour vasculature, we identified insulin receptor (INSR) to mark the tumour endothelium. As a functional role for insulin/INSR in cancer has been suggested and markers of the tumour endothelium may be attractive therapeutic targets, we investigated the role of INSR in angiogenesis. Methods In a genomic screen for determinants of the tumour vasculature we identified insulin receptor to mark the tumour endothelium. Results The current report demonstrates the following: (i) the heavy overexpression of INSR on angiogenic vasculature in human tumours and the correlation to short survival, (ii) that INSR expression in the tumour vasculature is mainly representing the short oncofoetal and non-metabolic isoform INSR-A, (iii) the angiogenic activity of insulin on endothelial cells (EC) in vitro and in vivo, (iv) suppression of proliferation and sprouting of EC in vitro after antibody targeting or siRNA knockdown, and (v) inhibition of in vivo angiogenesis in the chicken chorioallantoic membrane (CAM) by anti-INSR antibodies. We additionally show, using preclinical mouse as well as patient data, that treatment with the inhibitor sunitinib significantly reduces the expression of INSR-A. Conclusions The current study underscores the oncogenic impact of INSR and suggests that targeting the INSR-A isoform should be considered in therapeutic settings.

Background: Invasion is a fundamental step in metastasis, the leading cause of breast cancer-related death. We have shown previously that primary breast tumors invade collectively as clusters of cancer cells (Cheung et al. Cell 2013). These clusters are led by a small subpopulation of cancer cells that highly express the basal epithelial protein cytokeratin 14 (K14) and other stem cell genes (Cheung et al. PNAS 2016). K14+ leader cells metastasize as clusters, reverting to a proliferative growth state upon colonization of a secondary site (Cheung and Ewald. Science 2016). The mechanisms regulating their transitions between collective invasion and growth states remains poorly understood. Understanding these mechanisms could produce new insights into tumor cell collectivity and uncover new ways to treat disseminated cell clusters that have already metastasized. Results: RNA-seq analysis of K14+ and K14- tumor cells revealed that K14+ tumor cells have increased expression of genes encoding the hemidesmosome, a major adhesion complex that plays an essential role in anchoring epithelial cells to the basement membrane. The highest upregulated hemidesmosome gene was collagen 17A1 (COL17A1), a transmembrane protein and structural component of the hemidesmosome that is involved in cell migration during wound healing and tissue regeneration. Consistent with these RNA-seq data, we observed by IF staining in vivo that K14 and COL17A1 are typically co-expressed in collective invasive strands at the invasive front of MMTV-PyMT mammary carcinomas. We next applied ex-vivo 3D culture assays to determine the function of COL17A1 in K14+ cell invasion and proliferation. In mammary tumor organoids we observed that COL17A1 is expressed in basal K14+ tumor cells. However, while the COL17A1 endodomain is uniformly expressed, the ectodomain is expressed in patches, indicating active ectodomain cleavage, a process that reduces cellular motility (Franzke et al. The EMBO Journal 2002). Surprisingly, loss of total COL17A1 expression by shRNA-knockdown significantly impairs tumor organoid growth ˜2 fold and prevents invasion in basement membrane rich matrigel. Further studies are underway to investigate the role of COL17A1 during collective invasion in other extracellular matrix environments. Conclusions: Taken together, our data shows that invasive leader cells express COL17A1 in 3D culture and in vivo and that this protein is required for tumor organoid invasion and growth. Therefore, therapeutically targeting COL17A1, a cell surface protein, could prevent both the migratory and proliferative phases of breast cancer metastasis. We are currently investigating how COL17A1 and its shed COL17A1 ectodomain enable leader cells to toggle between invasive and proliferative states. Citation Format: Greenwood E, Cane Gasull P, Moore B, Cheung K. The hemidesmosome protein collagen 17A1 is required for collective invasion and growth of mammary tumor organoids [abstract]. In: Proceedings of the 2017 San Antonio Breast Cancer Symposium; 2017 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2018;78(4 Suppl):Abstract nr P1-01-14.