Nuclear PTEN Regulates Thymidylate Biosynthesis and Cellular Sensitivity to Antifolate Treatment

Metabolic reprogramming contributes to tumorigenesis and holds significant promise for cancer therapy. The PTEN tumor suppressor governs a variety of biological processes, including metabolism, by acting on distinct molecular targets in different subcellular compartments. In the cytoplasm, PTEN regulates a plethora of metabolic processes through antagonizing the PI3K/AKT/mTORC1 pathway. However, the metabolic regulation of PTEN in the nucleus remain undefined. Using a gain-of-function approach to examine the metabolic consequences of PTEN targeted to different sub-cellular compartments in human prostate cancer cell lines, we reveal a nuclear function for PTEN in controlling de novo thymidylate biosynthesis and may also open novel therapeutic avenues for targeting nuclear-excluded PTEN prostate cancer cells with anti-folate cancer treatment. The first four chapters of this dissertation are introductory information that outlines the role of PTEN in cancer, the importance of metabolic compartmentalization, the fundamentals of pyrimidine biosynthesis pathways, and the novelty of anti-folate cancer treatments. Chapter 1 explains the role of PTEN as a tumor suppressor and continues on to discuss its role at the cell membrane as a metabolic regulator. The gap in knowledge in the field is understanding the role nuclear PTEN plays as a metabolic regulator. This is important as nuclear PTEN has been shown to be associated with more aggressive cancer phenotypes. Chapter 2 focuses on the background of metabolic compartmentalization and how, by strategically placing genes and metabolites spatiotemporally, the cell is able to execute key molecular mechanisms more efficiently. In this chapter, we highlight where the current field is with metabolic compartmentalization, the advantages of compartmentalization and how utilization of this knowledge can be used for definitive therapeutics. Chapter 3 focuses on pyrimidine biosynthesis and thymidylate biosynthesis. Thymidylate is synthesized de novo by thymidylate synthase (TYMS), with the enzymes dihydrofolate reductase (DHFR) and methylenetetrahydrofolate dehydrogenase 1 (MTHFD1) or serine hydroxymethyltransferase (SHMT) which are required to regenerate 5, 10-methylenetetrahydrofolate. MTHFD1 is the primary source of 5,10-methylenetetrhydrofolate generation, and therefore its proper function in the nucleus ensures the functioning of de novo thymidylate biosynthesis. Using Mass-Spectrometry, we discovered that MTHFD1 is a top candidate protein interacting with PTEN in human prostate cancer cells. This is the key focus of this paper and will be important background for Chapter 7 and 8 which delve into the thymidylate pathway and the potential role of nuclear PTEN. A deeper understanding of nuclear PTEN’s regulation of MTHFD1 may, in turn, open new therapeutic avenues for anti-folate cancer treatment tailored to PTEN sub-cellular localization. This brings us to Chapter 4, which focuses on the current market of anti-folate treatment and the importance of precision medicine and combinatorial treatment. Chapter 5 is the start of our project, it is the basis of the remainder chapters and what allowed us to elucidate the importance of nuclear PTEN. To explore the role of PTEN as not just a tumor suppressor, but as a metabolic regulator, we first sought to generate overexpression cell lines with PTEN localized to various subcellular compartments. As explained in chapter 1 and 2, we discuss how in the cytosol inactive PTEN can be recruited to the plasma membrane where it functions as a lipid phosphatase to suppress the activation of the proto-oncogenic phosphoinositide 3-kinase (PI3K)–AKT-mTOR signaling pathway. In the nucleus, PTEN acts to induce cell cycle arrest and maintain genomic stability. However, the role of PTEN in metabolism is incompletely understood. It is clearly understood that each subcellular compartment harbors specific metabolic activities and PTEN is present in different subcellular locations where it performs distinct functions acting on specific effectors. To explore this, we used a gain-of-function approach to examine the metabolic consequences of PTEN targeted to different sub-cellular compartments. Plasmids were generated using site-directed mutagenesis and PCR. We used the vector pTRIPZ, which is an inducible TET-ON system. This was necessary as overexpression of a tumor suppressor in cancer cell lines, if left permanently on, leads to slow cell growth and expulsion of the plasmid by inherent cancer cell mechanisms. We generated a vector plasmid, which was used as the baseline for all experiments, a wildtype PTEN plasmid, which was “normal” PTEN and PTEN had the ability to localize to the membrane or nucleus as it pleased, a cytoplasmic membrane plasmid, which localized PTEN permanently to the membrane, nuclear PTEN, which localized PTEN permanently to the nucleus, and mutant C124S PTEN, which generates a catalytically silent PTEN variant preventing its role at the membrane and its involvement in the PI3K-AKT-mTOR pathway. All of the plasmids were transfected as a lentivirus into PC3 and C4-2 prostate cancer cells. Both cell lines are PTEN-null and C4-2 represents early stage prostate cancer and is AR positive, while PC3 cells represent later stage metastatic prostate cancer and are AR negative. Prostate cancer is one of the leading causes of morbidity and mortality in the world. Identification of novel therapeutics to combat this deadly disease is still needed today, despite advances in PSA testing, molecular diagnostics, and androgen-deprivation therapy treatments. Inactivation of PTEN by deletion or mutation is found in 20% of primary prostate tumor samples and over 50% of castration-resistant tumors. This highlights the importance of using prostate cancer cell lines as a model system. Cell lines were confirmed for PTEN localization by 3 methods: Western blotting, subcellular fractionation assays, and immunofluorescence assays. Confirmation of all 5 cell lines in both PC3 and C4-2 systems sets up the system used in the next couple chapters. Chapter 6 focuses on the metabolic profiling of our cell lines and highlights the big data generated opening up multiple avenues to be explored. It emphasizes the potential role of PTEN as not only a tumor suppressor, but as a metabolic regulator. Unbiased metabolic profiling was performed on PC3 and C4-2 overexpression cell lines. A polar metabolomics profiling platform using selected reaction monitoring with the 5500 QTRAP hybrid triple quadrupole mass spectrometer, created by Dr John Asara, was utilized for metabolomics data generation. Using LC MS/MS for polar metabolite profiling the platform is a single normal phase hydrophilic interaction liquid chromatographic run (HILIC) and has a short mass spectra acquisition time of only 15 minutes. The platform allows for over 250 metabolic compounds to be targeted without chromatographically scheduled selected reaction monitoring. Lastly, all analysis of data is done through Metaboanalyst, and output files are generated in R command. Large data sets were generated for each cell type and allowed for multiple avenues of research for this dissertation. Looking at a heat map of all 5 cell lines, it was clear nuclear PTEN stood out and followed a different and opposite pattern to that of wildtype and cytoplasmic membrane PTEN. This led us to compare nuclear PTEN to vector PTEN null cell lines and determine which pathways were enriched. Pyrimidine biosynthesis was at the top of the list, which leads us to chapter 7. Chapter 7 focuses on our analysis of pyrimidine biosynthesis and the potential role nuclear PTEN plays in it. The absence of nuclear PTEN is associate with more aggressive disease in patients and therefore can serve as a useful biomarker and potential therapeutic avenue. While purine metabolism happens exclusively in the cytoplasm, pyrimidine metabolism occurs in the cytoplasm, mitochondria, and nucleus. Most notable from our data, in relation to pyrimidine pathways, was the upregulation of dTMP levels. dTMP, as mentioned in chapter 3, is produced from dUMP and both metabolites participate in thymidine biosynthesis. Nuclear localization of the dTMP biosynthesis pathway is a critical factor for allowing the pathway to play its role in DNA synthesis and proper cell division. This pathway utilizes folate one-carbon metabolism of which the enzyme MTHFD1 is an important facilitator of the pathway. Previous data shows, through mass spectrometric analysis, that MTHFD1 was identified as a PTEN interacting protein. This information suggests a potential role that PTEN plays in thymidylate biosynthesis. This chapter aims to highlight the potential roles PTEN plays in thymidylate biosynthesis and pave the way for novel therapeutic treatment. Chapter 8 discusses the findings in chapter 7 and their ability to contribute to the metabolic therapeutic field. We utilized crystal violet assays to measure cell growth over a 6 day time period and treated cells with both an anti-folate and anti-PI3K drug to maximize efficacy against nuclear PTEN-null cancer patients. Pharmaceuticals that target enzymes in folate-dependent dTMP biosynthesis are developed as anti-cancer drugs. 5-fluorouracil (5-FU) is the most widely used anti-folate cancer drug and focuses on inhibition of TYMS. A full discussion of 5-FU’s pathway is discussed in chapter 4. The information and literature lead us to hypothesize that presence of nuclear-PTEN would affect 5-FU efficacy and combinatorial therapeutics would be a novel treatment method. In combination with 5-FU we used Buparlisib, an oral PI3K inhibitor, to inhibit tumor growth via the PI3K pathway which has been shown to promote castration and chemotherapy resistance. We sought to determine if BKM used in combination with 5-FU could be a potential therapeutic option for patients with nuclear PTEN loss. Previous data shows evidence of a protective effect of nuclear PTEN on drug treatment suggesting novel precision therapeutic treatment for loss of nuclear PTEN patients. Chapter 9 concludes our findings and provides combinatorial therapeutic pathways for metastatic cancer cell lines that have PTEN deletions and or mutations. The chapter also explores future directions for the work, highlights the contributions to the field of PTEN and metabolism, and recognizes PTEN as a metabolic regulator and gene that is vital in cancer tumorigenesis progression, applicable pan-cancer.