Abstract

Rewiring of cellular metabolism is a well-established hallmark of cancer biology. Cancer cells alter metabolism to support the bioenergetic, biosynthetic, and redox demands of malignant transformation. Cancers vary widely in the substrates and pathways utilized for biomass and energy production. These variations can arise from tissue of origin, tumor driving mutations, and tumor microenvironment pressures. However, highly proliferative tumors share the need to neutralize the reactive oxygen species (ROS) generated by their enhanced metabolic demands. Increased demand for ATP production and biomass production in cancer cells drives increased ROS production in the mitochondria, peroxisomes, endoplasmic reticulum, and other cellular compartments. Therefore, elevated ROS burden relative to normal tissues is prominent is a feature of cancer metabolism. While a baseline level of ROS is necessary for cellular signaling, elevated ROS levels that overwhelm the neutralization capacity of cellular antioxidant systems result in oxidative stress including DNA oxidation, deleterious protein oxidation, and oxidation of lipid membranes. To maintain redox homeostasis, cancer cells exhibit alterations in metabolism to support critical cellular antioxidant systems. Elevated ROS burden and compensatory increased dependency on cellular antioxidant systems make targeting redox metabolism a promising strategy for therapeutic development. The glutathione antioxidant system and the thioredoxin antioxidant system are the two primary cellular antioxidant systems necessary for the maintenance of redox homeostasis in cancer cells. Glutathione (GSH) is the most abundant intracellular antioxidant and is synthesized from glutamate, glycine, and cysteine to form the GSH tripeptide. GSH functions as a reductive co-factor for glutathione peroxidases (GPXs) and glutathione transferases (GSTs) to neutralize hydrogen peroxides, lipid peroxides, and other ROS. Utilization of GSH as a co-factor for ROS neutralization results in oxidation of GSH to form GSSG. GSSG is recycled to GSH by glutathione reductase (GR) in an NADPH dependent process. The thioredoxin system uses thioredoxin (Trx), a small protein antioxidant with a conserved cysteine moiety in the catalytic domain, as the central reductive component. Thioredoxins reduce peroxiredoxins and other redox active proteins to maintain their redox functions, resulting in thioredoxin oxidation. Thioredoxins are recycled to a reduced state by thioredoxin reductase (TrxR) in an NADPH dependent process. The glutathione and thioredoxin systems share many ROS substrates, resulting in considerable overlap between the two antioxidant systems. Dependency upon the glutathione antioxidant system and the thioredoxin antioxidant system vary between cancer subtypes and at different stages of tumor progression. However, dependency upon NADPH production and cysteine are shared between the two systems. As a result, targeting ROS homeostasis through inhibition of NADPH production or cysteine metabolism are emerging mechanisms for therapeutic development in cancer. In particular, interest in inhibiting cysteine metabolism in cancer cells has been growing since the discovery of ferroptosis, a distinct ROS mediated form of programed cell death. Although cysteine is a non-essential amino acid in most normal body tissues, cysteine becomes essential in many cancers due to increased dependence upon intracellular cysteine pools to maintain ROS homeostasis. While cells can synthesize cysteine through the transsulfuration pathway, recent studies suggest that most cancers are dependent upon cystine import through system xCT. The cystine-glutamate anti-porter, system xCT, is over expressed in many cancer subtypes. Additionally, inhibition of xCT or extracellular cystine depravation have been shown to induce ferroptosis in a variety of different cancer subtypes in vitro and in vivo including but not limited to lung cancer, pancreatic cancer, and breast cancer. However, not all cancers are sensitized to xCT inhibition at baseline and metabolically targeted drugs frequently fail in clinical trials when given as a single agent due to the expansive metabolic flexibility of cancer cells. Therefore, research characterizing sensitivity and combinatorial treatment strategies is important for clinical translation of pharmacological xCT inhibition. Alterations in gene expression for key enzymes in redox metabolism or transcription factors regulating antioxidant systems have been used as biomarkers to predict sensitivity to redox metabolism-based cancer therapeutics. Therefore, we used unbiased screening to identify alterations in metabolic gene expression in a diverse group of sarcomas. We identified two sarcoma subtypes, synovial sarcoma (a translocation driven soft tissue sarcoma) and osteosarcoma (a bone sarcoma of complex etiology), that exhibit unique metabolic changes that sensitize them to perturbations of redox metabolism. The first research aim focused on the elucidation of redox metabolism in synovial sarcoma (SS). Screening of gene expression in SS patient samples revealed significantly reduced expression of malic enzyme 1 (ME1). ME1 absence was confirmed in SS patient tissue samples by IHC and in human SS cell lines and mouse tumor models of SS by qPCR and protein expression analysis. ME1 is a cytosolic enzyme that catalyzes the reductive carboxylation of malate to pyruvate producing CO2 and NADPH in the process. ME1 contributes to the cytoplasmic NADPH pool along with enzymes in the oxidative pentose phosphate pathway (oxPPP) and IDH1. Targeted tracing of glucose metabolism in ME1-null WT SS cell lines and ME1 overexpressing (OE) control cell lines revealed increased flux through the pentose phosphate pathway (PPP) in ME1-null SS. Inhibition of the oxPPP with G6PD-i revealed that flux through the oxPPP was necessary in ME1-null WT SS to maintain intracellular NADPH levels for cell survival. Consistently, ME1-null WT SS exhibited reduced GSH/GSSG ratios compared to ME1-OE SS controls. Small molecule inhibitors of the glutathione antioxidant pathway (BSO) and the thioredoxin antioxidant pathway (D9) demonstrated a shift in dependence from the GSH pathway to the thioredoxin pathway in ME1-null SS. Metabolomics revealed reduced intracellular cysteine levels in ME1-null WT SS relative to ME1-OE SS controls, suggesting increased cysteine demand in the context of ME1 absence. To target the intrinsic redox metabolism of ME1-null SS we utilized small molecule inhibitors of cystine import through system xCT (erastin and ACXT-3012) in ME1-null SS and ME1-OE controls. ME1 null SS were sensitized to xCT inhibition both in vitro and in vivo. xCT inhibition in vitro with erastin resulted in significant increases in lipid peroxidation and induction of cell death via ferroptosis. In vivo inhibition of xCT with ACXT-3012 resulted in tumor control in ME1-null SS xenografts but had no effect on tumor growth in ME1-OE control xenografts. Collectively these data demonstrate that absence of ME1 in SS results in significant alterations to redox metabolism that potentiate cell death with inhibition of cystine import through system xCT. Similar unbiased screening methods in osteosarcoma recently revealed that PHGDH overexpression in osteosarcoma patient samples is a recurrent metabolic feature that correlates with significantly worse patient prognosis. PHGDH is the rate limiting enzyme in the de novo serine synthesis pathway, which has been linked to mitochondrial NADPH production and redox homeostasis in cancers and endothelial tissues. Therefore, the second research aim sought to evaluate the contribution of de novo serine synthesis to redox homeostasis in osteosarcoma. Unbiased screening of whole metabolomic data revealed significant reductions in GSH/GSSG ratios upon inhibition of PHGDH in osteosarcoma cell lines. Characterization of ROS levels in osteosarcoma cell lines treated with NCT-503, a small molecular inhibitor of PHGDH, revealed significant accumulation of mitochondrial ROS. Collectively, these data suggest that de novo serine synthesis is necessary for redox homeostasis in osteosarcoma. Serine can contribute to GSH synthesis by functioning as a substrate for cysteine synthesis through the transsulfuration pathway. Upon PHGDH inhibition intracellular serine levels were reduced while other substrates of the transsulfuration pathway accumulated and cysteine levels were reduced. These data collectively suggest reduced activity of the transsulfuration pathway. Consistently, U13C6-glucose tracing demonstrated that PHGDH inhibition halted glucose contribution to GSH synthesis in osteosarcoma cell lines. While PHGDH inhibition in osteosarcoma results in oxidative stress and reduced proliferation it does not induce cell death. To identify compensatory mechanisms promoting cell survival with PHGDH inhibition in osteosarcoma, mRNA screening of a panel of metabolism genes was conducted. Inhibition of PHGDH resulted in significantly increased expression of SLC7A11, the cystine-glutamate antiporter of system xCT. Consistently, treatment with PHGDH induced increased cystine import in osteosarcoma and increased xCT plasma membrane expression. We hypothesized that PHGDH inhibition increased dependency upon exogenous cystine import through xCT. Consistent with this hypothesis, treatment with NCT-503 significantly increased sensitivity to xCT inhibition in vitro resulting in significant cell death in physiologic conditions. Taken together these research aims highlight a powerful approach for the development of targeted therapeutics based on redox metabolism in sarcoma. In SS and osteosarcoma, two genetically diverse sarcoma subtypes that share few genetic features, unbiased screening revealed distinct metabolic changes that induced sensitivity to redox perturbations. Through detailed metabolic characterization it was demonstrated that ME1 absence in SS and PHGDH inhibition in osteosarcoma function through desperate mechanisms to sensitize both cancers to inhibition of cystine import. This work identifies novel opportunities for translational development in both SS and osteosarcoma through xCT inhibition. Additionally, this work suggests a pipe-line approach of unbiased screening, metabolic characterization, and pharmacological targeting to be applied to other sarcoma subtypes for further exploration of redox metabolism and therapeutic development in sarcoma.

Committee Chair

Brian Van Tine

Committee Members

Daniel Link; Jason Held; Julie Schwarz; Kian-Huat Lim

Degree

Doctor of Philosophy (PhD)

Author's Department

Biology & Biomedical Sciences (Molecular Cell Biology)

Author's School

Graduate School of Arts and Sciences

Document Type

Dissertation

Date of Award

4-3-2025

Language

English (en)

Download

Available for download on Friday, April 02, 2027

Included in

Biology Commons

Share

COinS