Breast cancer is one of the leading causes of cancer death in women worldwide. Dysregulation of the insulin and insulin-like growth factor (IGF) signaling (IIS) pathway, a crucial signaling axis necessary for regulating normal organismal growth and metabolism, promotes many aspects of breast cancer progression. Insulin receptor substrate (IRS) proteins are enzymatically inactive, cytoplasmic adaptors that act downstream of insulin receptor (IR) and IGF-1 receptor (IGF-1R) activation. Previous studies have shown that IRS2, rather than its homologous family member IRS1, enhances tumor invasion, survival, and metastasis. IRS2 represents an unexplored therapeutic target that, if inhibited, could provide clinical benefit for individuals with aggressive tumors. Development of an efficacious therapy that inhibits IRS2 requires a thorough understanding of the mechanisms by which IRS2 mediates tumor progression and must overcome challenges in inhibiting cytoplasmic proteins that lack enzymatic activity. Thus, the focus of this thesis was to 1) determine the molecular basis by which IRS2 promotes tumor cell invasion, a hallmark of breast cancer, and to 2) test the in vivo efficacy of silencing IRS2 expression as a therapeutic approach to inhibit breast cancer progression.

I investigated how the ability of IRS2 to recruit phosphatidylinositol-3 kinase (PI3K) and the IRS2 C-terminus contributed to breast cancer invasion using in vitro models of breast cancer migration and invasion. I found that the IRS2-PI3K interactions and the IRS2 C-terminus enhanced epithelial-to-mesenchymal transition (EMT) and chemotaxis, two key processes that enable tumor cell invasion. PI3K recruitment, but not the C-terminus, was also needed to suppress random migration, suggesting dynamic regulation of IRS2-dependent cell migration. IRS2 promotes the formation of actin-based protrusions and the phosphorylation of key actin regulatory proteins after IIS pathway activation, providing mechanistic evidence for the role of IRS2 in regulating cell motility. Collectively, these findings give novel insight into IRS2 functions in breast cancer invasion.

Using a syngeneic model of triple negative breast cancer (TNBC), I showed that downregulating IRS2 expression using fully chemically modified small interfering RNAs (siRNA) reduced mammary tumor growth without causing hyperglycemia. I designed and validated siRNAs that selectively silenced IRS2, but not IRS1, in vitro in breast carcinoma cell lines and in vivo in mice. I then systematically identified optimal siRNA chemical modifications and administration route for robust mammary tumor delivery, finding that subcutaneous administration of siRNA conjugated to an albumin-binding dendrimer structure leads to robust delivery to both neoplastic cells and stromal/immune cell populations within primary tumors. Suppression of Irs2 mRNA expression in mammary tumors reduced vimentin expression in primary tumor cells, tumor vascularization, and frequency of M2-like tumor-associated macrophages, suggesting that siRNA treatment modulated tumor cell-intrinsic and -extrinsic processes. Together, these results support the feasibility of using therapeutic siRNAs to target IRS2 in vivo and provide rationale for the development IRS2-targeting therapeutics for the treatment of breast cancer.