Phosphatidylinositol 3,4,5-trisphosphate (PI(3,4,5)P3)-dependent RAC exchanger 2 (PREX2) is a guanine nucleotide exchange factor (GEF) for the Ras-related C3 botulinum toxin substrate 1 (RAC1) GTPase. As a GEF, PREX2 facilitates the exchange of GDP for GTP on RAC1. GTP bound RAC1 then activates its downstream effectors, including p21-activated kinases (PAK). PREX2, RAC1, and PAK kinases all have key roles within the insulin signaling pathway. The insulin receptor is a tyrosine kinase that phosphorylates the insulin receptor substrate (IRS) family of adaptor proteins, leading to the activation of phosphatidylinositide 3-kinase (PI3K) and the generation of PI(3,4,5)P3. PI(3,4,5)P3 then activates numerous downstream signaling proteins, including AKT and RAC1, to regulate several important cellular processes, such as glucose metabolism and cell proliferation. In addition to being a RAC1 GEF, PREX2 affects the insulin signaling pathway by inhibiting the lipid phosphatase activity of phosphatase and tensin homolog (PTEN), which dephosphorylates PI(3,4,5)P3 to antagonize PI3K. PREX2 is also important in cancer, which is likely a consequence of both its role as a RAC1 GEF and as a PTEN inhibitor. PREX2 GEF activity is activated by PI(3,4,5)P3 and by Gβγ, which is a heterodimer that is released after GPCR activation. However, PREX2 regulation within specific signaling pathways is poorly understood. This thesis aims to understand the regulation of PREX2 downstream of ligand binding to receptors on the cell surface, with a focus on insulin. This is achieved by studying the phosphorylation of PREX2 after insulin stimulation and by characterizing protein-protein interactions involving PREX2 and key proteins in the insulin signaling pathway. Herein, we identified PI(3,4,5)P3-dependent phosphorylation events on PREX2 that occur downstream of insulin stimulation. Phosphorylation of PREX2 also occurred downstream of Gβγ, suggesting that phosphorylation was associated with the activation of PREX2 GEF activity. Interestingly, phosphorylation of PREX2 reduced GEF activity towards RAC1 and a phospho-mimicking mutation of PREX2 at an insulin-mediated phosphorylation site reduced cancer cell invasion. Phosphorylation of PREX2 also decreased PREX2 binding to the cellular membrane, PI(3,4,5)P3, and Gβγ, providing a mechanism for reduced GEF activity. These data suggested that phosphorylation was part of a negative feedback circuit to decrease the RAC1 signal, which led to the identification of the PAK kinases as mediators of PREX2 phosphorylation. Importantly, insulin-induced phosphorylation of PREX2 was delayed compared to AKT, which is consistent with a model where PREX2 phosphorylation by PAK occurs after activation of PREX2 to attenuate its function. Altogether, we propose that second messengers activate the PREX2-RAC1 signal, which sets in motion a cascade whereby PAK kinases phosphorylate and negatively regulate PREX2 to decrease RAC1 activation. This type of regulation would allow for transient activation of the PREX2-RAC1 signal. We then asked whether PAK phosphorylation of PREX2 was altered in cancer. To do this, we analyzed four recurrent somatic PREX2 tumor mutations, R155W, R297C, R299Q, and R363Q. Interestingly, all four mutants had reduced insulin and PAK1 dependent phosphorylation, and R297C had lower levels of phosphorylation induced by PI3K activating tumor mutants. This suggests that tumors might be mutating PREX2 in order to avoid PAK mediated negative regulation of RAC1. Lastly, we characterized PREX2 interactions with proteins that are critical for insulin signaling, with a focus on the interaction between the PREX2 pleckstrin homology (PH) domain and PTEN. PREX2 inhibition of PTEN is mediated by the PH domain, and we discovered that the β3β4 loop of the PH domain was required for binding of the isolated PH domain to PTEN. We also found that PREX2 co-immunoprecipitates with other insulin related proteins, including the p85 regulatory subunit of PI3K, IRS4, and the insulin receptor. Taken together, the studies in this thesis solidify the role of PREX2 in insulin signaling by showing that PREX2 GEF activity is tightly regulated by insulin and PAK-induced phosphorylation and also by characterizing PREX2 interactions with critical insulin related proteins. Further, this PAK dependent negative regulatory circuit downstream of both PI(3,4,5)P3 and Gβγ activation of PREX2 could have impacts in many aspects of biology given the roles that PREX2 and RAC1 have in critical cellular functions such as cell motility and glucose metabolism, and in diseases such as cancer and diabetes.

Indiana University-Purdue University Indianapolis (IUPUI) / The islet beta cell is unique in its ability to synthesize and secrete insulin for use in the body. A number of factors including proinflammatory cytokines, free fatty acids, and islet amyloid are known to cause beta cell stress. These factors lead to lipotoxic, inflammatory, and ER stress in the beta cell, contributing to beta cell dysfunction and death, and diabetes. While transcriptional responses to beta cell stress are well appreciated, relatively little is known regarding translational responses in the stressed beta cell. To study translation, I established conditions in vitro with MIN6 cells and mouse islets that mimicked UPR conditions seen in diabetes. Cell extracts were then subjected to polyribosome profiling to monitor changes to mRNA occupancy by ribosomes. Chronic exposure of beta cells to proinflammatory cytokines (IL-1 beta, TNF-alpha, IFN-gamma), or to the saturated free fatty acid palmitate, led to changes in global beta cell translation consistent with attenuation of translation initiation, which is a hallmark of ER stress. In addition to changes in global translation, I observed transcript specific regulation of ribosomal occupancy in beta cells. Similar to other privileged mRNAs (Atf4, Chop), Pdx1 mRNA remained partitioned in actively translating polyribosomes during the UPR, whereas the mRNA encoding a proinsulin processing enzyme (Cpe) partitioned into inactively translating monoribosomes. Bicistronic luciferase reporter analyses revealed that the distal portion of the 5’ untranslated region of mouse Pdx1 (between bp –105 to –280) contained elements that promoted translation under both normal and UPR conditions. In contrast to regulation of translation initiation, deoxyhypusine synthase (DHS) and eukaryotic translation initiation factor 5A (eIF5A) are required for efficient translation elongation of specific stress relevant messages in the beta cell including Nos2. Further, p38 signaling appears to promote translational elongation via DHS in the islet beta cell. Together, these data represent new insights into stress induced translational regulation in the beta cell. Mechanisms of differential mRNA translation in response to beta cell stress may play a key role in maintenance of islet beta cell function in the setting of diabetes.