Illuminating the Non-Canonical, Pro-Tumorigenic Role of Hippo Tumor Suppressor Kinase STK3 and Its Mechanism of Action in Prostate Cancer.

Prostate cancer (PC) is the most common cancer in men after skin cancers and there are more than 3 million men in the United States that are living with the disease today. Most current therapies target the Androgen Receptor (AR), which is the main driver of PC. AR-targeted therapies prolong PC survival, but resistance emerges and patients eventually succumb to the disease. This highlights the need for new PC molecular targets. Our group puts forth Serine/threonine kinase3 (STK3) as target, which has amplified expression in PC and increased expression from local to metastatic disease suggesting it may play a role in PC development and progression. Here we fully investigate the non-canonical role of STK3 in ferroptosis cell death and how its novel downstream signaling regulates pro-tumorigenic genes including ferroptosis resistance genes. The first three chapters of this dissertation are introductory information outlining the disease and pathways at play. Chapter 1 reviews the landscape of PC including patient populations, pathogenesis of the disease, current treatments and potential for kinases as targets in developing therapies. The development of more advanced PC and resistance to current therapies which causes eventual patient death is particularly emphasized here given that this is the greatest unmet need in the field. Chapter 2 outlines the background of the Hippo/YAP signaling pathway that the kinase, STK3, belongs to and the mechanisms of the canonical tumor suppressor signaling. What is known about the upstream stimuli and downstream pathways is discussed and the large amount unknown about this relatively new pathway is emphasized. In this chapter, we also introduce preliminary public patient data showing that STK3 is amplified in PC and correlates with more aggressive disease and shorter overall survival. This data spurred our work investigating STK3 as a target for PC. Chapter 3 gives an overview of ferroptosis cell death, a novel iron dependent cell death that has been identified as a new cell death pathway to be exploited by cancer therapies. On top of that, ferroptosis has specifically been illuminated as a promising pathway for PC treatment and Hippo signaling has been implicated in ferroptosis sensitivity and resistance. Lastly, this introductory chapter is crucial as later in Chapter 5, ferroptosis is linked to STK3’s non-canonical function in PC. Chapter 4 establishes the non-canonical role of STK3 in PC and proves that it does not act as a tumor suppressor in this context. To explore the role of STK3 in PC, we started by using chemical and genetic inhibition of STK3. For these experiments, we investigated STK3 inhibition in a number of cell lines reflecting a range of PC disease states. We collaborated with medicinal chemists to develop novel chemical inhibitor tools which significantly improved specificity compared to the published STK3/4 inhibitor XMU-MP-1. We also used constitutive and inducible genetic knockdown and knockout systems to inhibit STK3 in vitro and in vivo. We performed PK/PD analysis with our lead STK3i compounds and identified a candidate for in vivo use. We then tested this candidate in an in vivo model to show inhibition of STK3 slowed tumor growth and did not cause off target effects. Lastly, we used matrigel spheroid assays and colony formation assays to show that inhibition of STK3 slowed invasion and decreased colony formation. This set the president for the next chapters which explore the mechanism of STK3 in PC. To determine potential pro-tumorgenic roles of STK3 in PC, we performed unbiased RNA sequencing comparing shNT cells to shSTK3 cells and saw significant downregulation of amino acid transmembrane proteins, oxidative stress, iron binding and ferroptosis pathways. We validated that individual ferroptosis resistance genes SLC7A11 and SLC3A2 had decreased expression with loss of STK3. In Chapter 5, we discuss this transcriptomic data and test the connection between STK3 and these genes using our genetic models, a kinase dead STK3 overexpression model and novel STK3i tools. We showed STK3 corresponds with SLC7A11/SLC3A2 gene and protein expression using knockdown and overexpression. We also test whether STK3 has a functional role in ferroptosis resistance for proliferation and colony formation assays using pharmacologic inducers, extracellular iron overload and increased reactive oxygen species (ROS). Our functional assays show that loss of STK3 sensitizes PC cells to ferroptosis-induced cell death. Conversely, STK3 overexpression confers resistance to ferroptosis inducers but this is dependent upon STK3 kinase activity. This chapter’s data definitively characterizes the link between STK3 and ferroptosis in PC. In the last data chapter, Chapter 6, we explore the mechanism for STK3 regulation of ferroptosis resistance genes. We aimed to determine the STK3 phospho-targets that have a role in ferroptosis resistance in PC and the STK3 non-canonical mechanism(s) of action. We used the transcriptomic data from STK3-depleted PC cells to perform TFBS analysis and determine potential candidates for transcription factors that regulate the STK3 linked genes. Then to determine what STK3 phospho-targets influence ferroptosis genes, we performed phospho-proteomics in PC cells with genetic depletion of STK3. Pathway enrichment analysis from proteomics indicates down regulation of transcription factor and RNA polymerase activity pathways in STK3 depleted cells. Further, integration of our transcriptomic data defined transcription factors and phospho-proteomic data highlighted that STK3 may regulate the CDK9/BRD4/RNA Pol-II transcriptional axis. We tested this hypothesis by validating whether STK3 activity regulated CDK9 phosphorylation using our genetic model for overexpression, kinase dead STK3 and knockdown as well as STK3i. We mined public ChIP databases to investigate whether BRD4/CDK9 had binding peaks at STK3 linked genes. We then validated that BRD4/CDK9 can regulate SLC3A2 and SLC7A11 and binding at their promoter peaks our models. We investigated how STK3 effects the binding of BRD4 at SLC7A11 and SLC3A2 and how it affected the RNA Pol II activity using protein blotting and ChIP analysis. From these experiments, we determine that BRD4 and CDK9 bind at promoter peaks of SLC3A2 and regulate its expression. We also found that loss of STK3 seemed to be causing proximal pausing of the BRD4/RNA Pol II complex which was causing decreased expression of STK3 linked genes. This data lead to development of our model for STK3’s non-canonical, pro-tumorigenic mechanism in PC. This model proposes that STK3 leads to phosphorylation of CDK9 which regulates BRD4/RNA Pol II mediated transcriptional elongation of ferroptosis resistance genes SLC7A11 and SLC3A2 to control PC cell resistance to ferroptosis cell death. Chapter 7 describes our overall conclusions that can be draw from our work investigating STK3. It also explores future directions for this project and addresses the implications of this work for the greater field of PC research and potential translational applications and benefits.