Fumarate hydratase (FH) and cancer: a paradigm of oncometabolism

Abstract

Fumarate hydratase (FH) is an enzyme of the Tricarboxylic Acid (TCA) cycle whose mutations lead to hereditary and sporadic forms of cancer. Although more than twenty years have passed since its discovery as the leading cause of the cancer syndrome Hereditary leiomyomatosis and Renal Cell Carcinoma (HLRCC), it is still unclear how the loss of FH causes cancer in a tissue-specific manner and with such aggressive behaviour. It has been shown that FH loss, via the accumulation of FH substrate fumarate, activates a series of oncogenic cascades whose contribution to transformation is still under investigation. In this review, we will summarise these recent findings in an integrated fashion and put forward the case that understanding the biology of FH and how its mutations promote transformation will be vital to establish novel paradigms of oncometabolism.

Fig. 1. Metabolic adaptations in FH-deficient cells.

FH loss leads to the truncation of the TCA cycle and subsequent accumulation of fumarate (highlighted in orange). Mitochondrial respiration is significantly reduced due to the loss of expression of mitochondrial DNA (mtDNA)-encoded subunits of the electron transport chain (ETC) and the inhibition of Succinate Dehydrogenase (Complex II of the ETC). Because of this, cells increase the glycolytic flux by taking up glucose and producing lactate to obtain energy. Furthermore, part of the carbons from glucose is diverted towards the pentose phosphate pathway (PPP) to maintain the redox homeostasis producing NADPH (purple arrows). Moreover, to maintain the remaining TCA activity FH-deficient cells increase glutamine uptake and oxidation (green arrows). On the one hand, glutamine-derived carbons are metabolized to fumarate and and bilirubin (secreted to the media) produced via the haem pathway (red arrows). On the other hand, glutamine carbons are used through reductive carboxylation to increase fatty acid (FA) synthesis (yellow arrows). Due to the accumulation of fumarate up to millimolar levels, FH-deficient cells activate multiple strategies to buffer the potential toxicity. For example, fumarate can permeate to the nucleus and also be secreted extracellularly. In the cytoplasm, fumarate accumulation leads to aberrant production of argininosuccinate via the reverse reaction of argininosuccinate lyase (ASL) in the urea cycle (grey arrows). In this context, it is important to remark that FH-deficient cells depend on a constant uptake of extracellular arginine, which becomes essential for the viability of these cells, to maintain this buffering system. Finally, fumarate can also alter PNC (Purine Nucleotide Cycle), where the increase of fumarate causes the reversal of adenylosuccinate lyase (ADSL) to form adenylosuccinate, altering de novo purine biosynthesis and making cells reliant on the salvage pathway to support purine synthesis. ATP Adenosine Triphosphate, SDH Succinate Dehydrogenase, ACO2 Aconitase2, OAA Oxaloacetate, NADH Nicotinamide adenine dinucleotide, IMP Inosine Monophosphate, AMP Adenosine Monophosphate, CI-V Electron transport chain Complex I–V), SAICAR succinyl-5-aminoimidazole-4-carboxamide-1-ribose-5′-phosphate, AICAR 5-Amino-1-(5-Phospho-D-ribosyl)imidazole-4-carboxamide.

Fig. 2. FH loss-associated oncogenic signalling.

Upon FH loss, and consequently fumarate accumulation, several oncogenic pathways are altered. For example, fumarate can inhibit the activity of α-ketoglutarate-dependent dioxygenases (αKGDDs), including prolyl hydroxylases (PHDs), Jumonji C-domain lysine demethylases (JmjC-KDMs), and 10–11 translocation (TET) DNA cytosine-oxidizing enzymes. The inhibition of PHDs lead to the stabilization of hypoxia-inducible factor (HIF1A) even in normoxic conditions, known as pseudohypoxia. This phenomenon leads to the activation of signalling cascades associated with tumorigeneses, such as angiogenesis (VEGF), proliferation (TGFα) and glycolytic flux activation (activation of LDHA and GLUT1, inhibition of PDH). In the nucleus, fumarate inhibits the function of JmjC-KDMs and TETs affecting DNA and histones demethylation respectively. Specifically, the inhibition and demethylation of miR200 and CDKN2A (p16) has been shown to trigger an epithelial-to-mesenchymal transition and to inhibit senescence respectively in HLRCC patients. In line with this, FH modulates chromatin accessibility and the activation of FOXA2-mediated antioxidant response. Beyond αKGDDs inhibition, FH loss modulates the energy sensing in the cells. For example, it has been shown to inhibit and activate AMPK function. The inhibition of AMPK can lead to the activation of lipid biosynthesis through acetyl-CoA carboxylase (ACC) and the activation of mTOR signalling. AMPK activation, instead, was shown to protect cells from apoptosis. Further evidence supports the activation of mTOR through the inactivation of ABL1, modulated by the protein-tyrosine phosphatase PTPN12. Additionally, cyclic AMP (cAMP) levels increase upon FH loss, affecting cellular energy metabolism. Beyond energy sensing, well-known oncogenic pathways have been shown to be altered in FH-deficient models. For instance, the tumour suppressor PTEN can be inhibited by fumarate through succination (2SC), activating the phosphatidylinositol-3-kinase (PI3K) cascade. Moreover, FH loss has been associated with the activation of the integrated stress response (ISR) through ATF4. Given the regulation of PI3K pathway, mTOR and ATF4 by FH loss, it is tempting to speculate a potential regulation node between them (red line). In addition, HIRA loss has been recently found to increase the tumorigenic potential of FH-deficient cells through the MYC proto-oncogene and E2F transcriptional programs. Furthermore, FH loss can also regulate and increase DNA damage response and repair upon ionising radiation (IR). Finally, it has been recently discovered that FH loss can trigger the activation of the innate immune response activating the cGAS/STING/TBK1 pathway upon mitochondrial DNA (mtDNA) release to the cytosol. CDKN2A Cyclin-Dependent Kinase Inhibitor 2A, LDHA Lactate Dehydrogenase A, GLUT1 Glucose Transporter 1, PHD Pyruvate Dehydrogenase Complex, VEGF Vascular Endothelial Factor, TGFα Transforming Growth Factor alpha, FOXA2 Forkhead Box A2, AMPK AMP-activated Protein Kinase, PTEN Phosphatase and tensin homolog, ATF4 Activating Transcription Factor 4, mTOR mammalian target of rapamycin, cGAS cyclic GMP–AMP synthase, STING Stimulator Of Interferon Response CGAMP Interactor 1, TBK1 TANK-binding kinase 1, OX Oxidation.

Fig. 3. Therapeutic intervention for HLRCC patients.

Given the control that FH loss and fumarate accumulation exert on several oncogenic factors, multiple inhibitors are being or could expect to be used in clinical interventions. For instance, PI3K cascade activation can be targeted using Sunitinib and the downstream effects of it using AKT inhibitors. Moreover, AMPK is re-activated through metformin treatment, indirectly inhibiting mTOR activation. Still, metformin could affect the complex I in the electron transport chain. As FH deficiency leads to decreased oxidative phosphorylation (OXPHOS), it is unclear whether this treatment could be beneficial. Beyond metformin, mTOR activation could be targeted by ABL1 inhibitors. Metabolically, these tumours could benefit from HMOX1 inhibition using zinc protophorphyrin (ZnPP) or an imidazole-based inhibitor SLV-11199, and LDHA inhibitors affecting the glycolytic flux. In addition, the inhibition of the purine salvage pathway using 6-mercaptopurine (6-MP) has been shown to affect the viability of FH-deficient cells. Furthermore, several anti-angiogenic therapies based on VEGF and EGFR inhibition are currently being used in the clinic. Interestingly, the new results highlighting the role of MYC activation in these tumours could open new therapeutic strategies based on a novel MYC inhibitor, omomyc. Finally, given the alterations occurring in the epigenetic machinery in FH-deficient tumours, inhibitors of DNA methyltransferases (DNMTs) are being tested in clinical trials. PI3K Phosphatidylinositol-3-kinase, AKT AKT Serine/Threonine Kinase 1, ABL1 ABL Proto-Oncogene 1, mTOR mammalian target of rapamycin, AMPK AMP-activated Protein Kinase, PTEN Phosphatase and tensin homolog, HMOX1 Heme Oxygenase 1, VEGF Vascular Endothelial Factor, LDHA Lactate Dehydrogenase A, EGFR Epidermal Growth Factor Receptor, 2SC 2-succinic-cysteine, CI-V Electron transport chain Complex I–V, HIF1A Hypoxia Inducible Factor A, PHD Prolyl hydroxylase.