Search

Cancers harboring homozygous deletion of the glycolytic enzyme enolase 1 (ENO1) are selectively vulnerable to inhibition of the paralogous isoform, enolase 2 (ENO2). We previously identified and characterized a competitive, small molecule phosphonate inhibitor of ENO2, 1-hydroxy-2-oxopiperidin-3-yl phosphonate (HEX), and its lipophilic bis-ester prodrug (POMHEX) in an ENO1-deleted intracranial orthotopic xenograft model of glioblastoma. Treatment with either HEX (150 mg/kg IV/IP) or POMHEX (20 mg/kg IV/IP) yielded tumor regression even after drug discontinuation. However, due to the poor pharmacokinetics of esterase-labile POMHEX, we synthesized a library of novel phosphonate prodrugs with distinct mechanisms of bioactivation and assessed their potency in D423 (ENO1-/-) cells. By conducting a prodrug structure activity relationship (SAR) study, we found that phosphonoamidate esters were efficiently bioactivated in ENO1-deleted glioma cells, while canonical McGuigan (ProTide) prodrugs were not. Other strategies, including salicylic alcohol (cycloSal) and lipid prodrugs of HEX, exhibited low micromolar IC50 values in ENO1-deleted glioma cells and improved stability in human serum over POMHEX. En route, we developed a novel class of aliphatic amine/ester prodrugs that can be broadly applied to efficiently deliver phosph(on)ate pharmacophores in cells. The activity of select prodrugs was also probed using the NCI-60 cell line screen, supporting its use to examine the relationship between prodrugs and cell line-dependent bioactivation. In sum, we have developed a novel class of phosph(on)ate prodrugs that is efficiently bioactivated in cells in vitro. Our prodrug SAR study disputes the common notion that ProTides are universally advantageous promoieties on phosph(on)ate pharmacophores and we provide mechanistic rationale for this observation with HEX. Finally, we show that the cycloSal prodrug yields efficient intracellular delivery of HEX in vitro, with a mechanism of bioactivation consistent with the GBM microenvironment, making this promoiety promising for further evaluation in vivo. Citation Format: Victoria C. Yan, Cong-Dat Pham, Elliot S. Ballato, Kristine L. Yang, Sunada Khadka, Yasaman Barektain, Prakriti Shrestha, Theresa Tran, Anton H. Poral, Mykia Washington, Sudhir Raghavan, Barbara Czako, Federica Pisaneschi, Yu-Hsi Lin, Nikunj Satani, Naima Hammoudi, Jeffrey J. Ackroyd, Dimitra K. Georgiou, Steven W. Millward, Florian L. Muller. Prodrugs of a 1-hydroxy-2-oxopiperidin-3-yl phosphonate enolase inhibitor for the treatment of ENO1-deleted cancers [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 3097.

Background: Precision oncology is currently restricted to activated oncogenes as drug targets, with tumor suppressor deletions remaining largely in-actionable. We developed a novel method that targets tumor suppressor deletions by exploiting vulnerabilities generated by the co-deletion of neighboring metabolic housekeeping genes with critical but normally redundant functions. Proof-of-principle studies were conducted for the glycolytic enzyme Enolase. Cancers harboring the homozygous deletion of 1p36-localized ENO1 remain metabolically active and viable through redundant action of its paralogue, ENO2. Inhibition of ENO2 in cancer cells harboring ENO1-homozygous deletion discriminately kills ENO1-deleted cells. In accordance, our synthetic efforts have focused on phosphonate chemistry to generate clinically-useful substrate mimics of 2-phosphoglycerate, the natural substrate for Enolase. We have generated a phosphonohydroxamate ENO2-inhibitor termed HEX and a pro-drug of which, POMHEX, kills ENO1-homozygously deleted glioma cells in culture with strong specificity. However, the pharmacokinetics of POMHEX are poor, due to the presence of carboxyesterase in extracellular fluids, including plasma. Results: We have synthesized mechanistically different pro-drugs of HEX, utilizing nitro-aromatics as protecting groups. Nitroaromatic groups are bioactivated by nitroreductases intracellularly, in a manner that is inversely related to oxygen concentration. As such, removal of nitroaromatic pro-drug moieties and release of active drug is favored under low oxygen (hypoxic) conditions. We synthesized a series of nitroaromatic pro-drugs of HEX (the structures will be disclosed at AACR presentation, pending patent filings) and demonstrate that 1) these exhibit selective toxicity to ENO1-deleted glioma cells in culture and 2) this killing is dramatically enhanced by lowering oxygen concentrations as compared to POMHEX. We further find that redox potential of the nitroaromatic moiety correlates with the degree of hypoxia activation with the lower redox potential being associated with great oxygen dependence. Importantly, unlike POMHEX, nitroaromatic HEX pro-drugs are stable in human plasma ex-vivo. We have thus successfully synthesized a pro-drug inhibitor of Enolase that is hypoxia-activated. Conclusion and future directions: We are gearing up synthesis for pre-clinical evaluation of nitroaromatic-HEX in intracranial tumor xenograft models. While the primary purpose of Nitroaromatic HEX is for the precision oncology treatment of ENO1-homozygously deleted tumors, it has not escaped our notice that such drugs have wider applicability within the context of targeting treatment-resistant tumors cells hiding in hypoxic niches. Indeed, we have observed that virtually all glioma cell lines can be killed by nitroaromatic HEX under hypoxia. Note: This abstract was not presented at the meeting. Citation Format: Florian Muller, Victoria C. Yan, Elliot S. Ballato, Kristine L. Yang, Dimitra K. Georgiou, Kenisha Arthur, Pakriti Shrestha, Sunada Khadka, Jeffrey Ackroyd. Synthesis of mixed, hypoxia-activated phosphoramidate esters for the inhibition of Enolase in ENO1-deleted glioblastoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 984.

Large-scale genomic profiling efforts, such as The Cancer Genome Atlas (TCGA) have painted an unprecedentedly detailed picture of the genetic alterations that underlie carcinogenesis, yet the key challenge remains as to how to turn this information into actionable therapies. Genomic deletions are a frequent event in diverse cancers, which inactivate a limited number of tumor suppressor genes (“driver”-events) but frequently include many chromosomal neighbors as “passengers”, some of which play critical but redundant roles in normal cellular housekeeping. The overall hypothesis we propose to test is that collateral deletions of such “passenger” genes can be utilized as novel targets of synthetic lethality, an idea which we term “collateral lethality." The large number of passenger deleted genes, playing diverse functions in cell homeostasis, offers a rich repertoire of pharmacologically targetable vulnerabilities presenting novel opportunities for the development of personalized anti-neoplastic therapies. We have provided in vitro proof-of-principal of a collaterally deleted glycolytic gene Enolase 1 (ENO1) at the 1p36 tumor suppressor locus in glioblastoma (GBM), that leads to dramatic sensitization to inhibition of the redundant paralogue, ENO2. The next step is to take this concept to the clinic. Our overall goal is to generate a clinical candidate Enolase inhibitor for tumors with ENO1-deletions, like GBM for which there is no other treatment option and the therapeutic benefit could be quite substantial. Citation Format: Yu-Hsi Lin, Naima Hammoudi, Nikunj Satani, Jeffrey Ackroyd, Sunada Khadka, Dimitra Georgiou, Joe Marszalek, Yuting Sun, Marina Protopopova, Maria E. Di Francesco, Barbara Czako, Alan Y. Wang, Ronald A. DePinho, Florian L. Muller. Collateral lethality: A new target for personalized oncology [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 2831.

Cancer genomes possess many deletion events targeting tumor suppressor genes (TSG) and neighboring genes in these loci. These deletion patterns prompted us to consider a systematic approach, termed “collateral lethality”, designed to identify cancer-specific vulnerabilities resulting from the deletion of neighboring genes. These bystander genes do not appear to be involved in cancer pathogenesis, yet encode cell-essential functions and are members of multi-gene families that are functionally redundant and co-expressed. Homozygous deletion of SMAD4 is a frequent event in pancreas cancer and other cancer types, totaling >30,000 cases in US annually. SMAD4 deletion often results in co-deletion of the neighboring mitochondrial malic enzyme 2 (ME2) gene. In mammalian cells, two genes (ME2 and ME3) encode redundant cell-essential mitochondrial ME activity. Together, ME2 and ME3 function to generate pyruvate to fuel the TCA cycle, and NADPH to maintain ROS homeostasis. These observations prompted us to hypothesize that the genetic or pharmacological extinction of ME3 activity in a ME2 null cell would specifically compromise cancer cells yet be tolerated in normal host cells possessing ME2 activity. Inducible shRNA strategies were employed to genetically deplete ME3 in ME2-null versus ME2-intact cells followed by apoptosis measurements, integrated metabolomics, and molecular investigations. In collaboration with IONIS Pharmaceuticals, we tested and validated anti-sense oligonucleotide (ASO) to target ME3. Genetic depletion of ME3 in ME2 null, but not ME2 intact, cells resulted in apoptosis and blocked tumorigenic potential. Mechanistically, integrated metabolomic and molecular investigation of mitochondrial ME-deficient cells revealed diminished NADPH production and consequent high ROS, which activates AMP activated protein kinase (AMPK) and which in turn directly suppresses sterol regulatory element-binding protein 1 (SREBP1)-directed transcription of its direct targets including the BCAT2 (Branched chain amino acid transaminase 2) gene. We also determined that mitochondrial MEs regulate the utilization of branched chain amino acid (BCAA) via BCAT2, a transaminase required for BCAA catabolism. Notably, enforced expression of BCAT2 can restore tumorigenic potential of ME2/3 deficiency, and free nucleotides can restore proliferation in cell culture. Specific targeting of ME3 using ASOs could dramatically reduce tumor burden in a subQ tumor model of pancreatic cancer. Thus, a key mechanism driving cancer cell lethality involves BCAAs as crucial metabolites under the critical regulation of the mitochondrial MEs. These studies reveal a collateral lethal vulnerability in pancreas and other cancers that can be targeted pharmacologically in genotype-defined patient populations. We propose that highly specific ME3 inhibitors could provide an effective therapy across a substantial number of cancer patients. Citation Format: Prasenjit Dey, Joelle Baddour, Youngsoo Kim, Robert Macloed, Florian Muller, Chia Chin Wu, Huamin Wang, Andrea Viale, Haoqiang Ying, Giulio Draetta, Anirban Maitra, Alan Wang, Deepak Nagrath, Ronald DePinho. Malic enzyme 3 as a collateral lethality target in pancreatic cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 777.

Glycolysis inhibition is an active area of investigation for the treatment of cancer. However, few compounds have progressed beyond the cell culture stage. We have recently demonstrated that genomic passenger deletion of the glycolytic enzyme Enolase 1 (ENO1) leaves gliomas harboring such deletions solely reliant on ENO2, rendering them exquisitely sensitive to enolase inhibitors Collateral Lethality. However, the tool compound that we employed for these in vitro studies, Phosphonoacetohydroxamate (PhAH), has very poor pharmacological properties and was ineffective in vivo. We recently reported that a structural analogue of PhAH, the natural phosphonate antibiotic SF2312, is a high potency inhibitor of Enolase. While more potent than PhAH, SF2312 remains poorly cell permeable. Here, we generated a Pivaloyloxymethyl (POM) ester pro-drug derivative of SF2312, termed POMSF, which increased the potency in cell based systems by ~50-fold. POMSF is selectively active against ENO1-deleted glioma cells in culture at ~19 nM, versus μM for SF2312. However, POMSF displayed poor aqueous stability. A derivative of POMSF, termed POMHEX, showed greater stability and its active form, HEX, showed 4-fold preference for ENO1 over ENO2. Labeled 13C-glucose tracing shows that POMHEX inhibits glycolysis at the Enolase step in all cell lines tested, but with ~100-fold greater potency in ENO1-deleted lines. POMHEX selectively killed ENO1-deleted glioma cells with an IC50 Citation Format: Yu-Hsi Lin, Nikunj Satani, Naima Hammoudi, Federica Pisaneschi, Paul Leonard, David Maxwell, Zhenghong Peng, Todd Link, Lee IV R. Gilbert, Ananth Bosajou, Duoli Sun, Joe Marszalek, Yuting Sun, John S. McMurray, Pijus K. Mandal, Maria E. Di Francesco, Barbara Czako, Alan Wang, William Bornmann, Ronald A. DePinho, Florian Muller. Pomhex, a cell-permeable high potency enolase inhibitor with utility for collateral lethality treatment of cancer [abstract]. In: Proceedings of the AACR Precision Medicine Series: Opportunities and Challenges of Exploiting Synthetic Lethality in Cancer; Jan 4-7, 2017; San Diego, CA. Philadelphia (PA): AACR; Mol Cancer Ther 2017;16(10 Suppl):Abstract nr A39.

Large scale genomic characterization efforts such as TCGA have painted an unprecedentedly detailed picture of the genetic alterations that underlie tumorigenesis. Yet, the majority of genetic alterations are passenger rather than driver events and are considered unactionable. We have previously proposed that passenger or collateral deletions could serve as pharmacologically targetable vulnerabilities Collateral Lethality, in case passenger genes are homozygously deleted and member of a paralogous gene family carrying out an essential housekeeping function. We have presented proof-of-principal, whereby passenger deletion of the glycolytic gene Enolase 1 (ENO1) as part of the 1p36-tumor suppressor locus, renders glioma cells harboring such deletions highly sensitive to ablation of its redundant paralogue, ENO2. While our original analysis identified ENO1-homozygous deletions in Glioblastoma (GBM), recent bioinformatics analyzes, backed by immunohistochemistry, show that ENO1-homozygous deletions also occur in Hepatocellular Carcinoma and Cholangiocarcinoma. In GBM, multisector analysis of primary and recurrent tumors, indicate that ENO1 deletion is an early event which is homogenously distributed through the primary tumor and persist during recurrence. To pharmacologically exploit ENO1-deletion, we have pursued two approaches. First, we have synthesized cell-permeable prodrug derivatives of the natural Enolase inhibitor SF2312. The lead compound, POMHEX, shows potent killing of ENO1-deleted glioma cells in the low nM range while ENO1-restored isogenic or normal cells can tolerate μM doses. POMHEX has a short half-life yet can eradicate intracranial xenografted ENO1-deleted tumors, provided extensive breakdown of the blood-brain barrier. Our second approach to targeting ENO1-deletion consisted of chemical biology screening of drug libraries for the ability to kill ENO1-deleted but not isogenic rescued cells. We find that ENO1-deleted cells show a dramatic sensitization to inhibitors of the mitochondrial electron transport chain. These include tool compounds such as rotenone as well as compounds not previously associated with mitochondria, such as Mubritinib and an experimental anti-neoplastic agent previously described as a HIF1-inhibitor, now known to inhibit mitochondrial Complex I. The latter agent shows potent activity against ENO1-deleted intracranial xenografts. The likely cause for this sensitivity is the inability of ENO1-deleted cells to compensatory upregulate glycolysis in response mitochondrial inhibition, the typical response of ENO1-intact glioma cells and normal cells. Together, these data indicate that passenger deletion of ENO1 is an encouraging drug-target and provide support for collateral lethality as a viable therapeutic strategy, which, given the large number of passenger deletions in the cancer genome, may be broadly applicable. Citation Format: Yu-Hsi Lin, Nikunj Satani, Naima Hammoudi, Joe Marszalek, Yuting Sun, Marina Protopopova, Maria E. Di Francesco, Barbara Czako, Alan Wang, Ronald A. DePinho, Florian L. Muller. Passenger deletion of ENO1 as a collateral lethality target in cancer. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 3837.

Cancer cells display metabolic properties distinct from normal cells and are therefore believed to be dependent on key metabolic enzymes. The effectiveness of targeting metabolism for cancer therapy has been largely restricted due to metabolic adaptability. We postulate that tumors with metabolic vulnerabilities will be limited in such adaptation, thus providing a unique opportunity for therapeutic intervention. Hereditary leiomyomatosis renal cell carcinoma (HLRCC) is a metabolically vulnerable cancer, due to mutations in a key TCA cycle enzyme - fumarate hydratase (FH), thus rendering them deficient in oxidative phosphorylation (OXPHOS). We performed a loss-of-function genetic screen in a FH-null HLRCC model (UOK262 cells) and identified 6-phosphogluconate dehydrogenase (6PGD) - a rate-limiting enzyme in the non-oxidative arm of pentose phosphate pathway - as a critical regulator of cell growth. Inhibition of 6PGD using either dox-inducible shRNA or a small molecule inhibitor dramatically blocks the growth of UOK262 cells, and in contrast, has no effect on 6PGD-deficient NB1 cells. The growth inhibition from knockdown is fully rescued by expressing shRNA-resistant 6PGD. Conversely, ectopic expression of 6PGD in NB1 cells results in a gain of function phenotype. Importantly, in vivo, 6PGD knockdown causes robust regression of established UOK262 xenografts. Mechanistically, 6PGD inhibition causes the accumulation of 6-phosphogluconate, blocks glycolysis, and induces ROS production. Furthermore, in tumors with intact TCA cycle, 6PGD inhibition shows synergism with OXPHOS Complex I inhibitor IACS-10759. Together, our data suggest that 6PGD is a bona fide oncology target and provide a strong rationale for developing small molecule inhibitors that could be used to treat OXPHOS-deficient tumors as single agent, or combined with inhibitors of OXPHOS. Citation Format: Yuting Sun, Madhavi Bandi, Timothy Lofton, Melinda Smith, Christopher Bristow, Norma Rogers, Chang Edward, Mary Geck Do, Yongying Jiang, Pietro Morlacchi, Florian Muller, Faika Mseeh, Barbara Czako, Wylie Palmer, Carlo Toniatti, Philip Jones, Giulio Draetta, Timothy Heffernan, Joseph Marszalek. Functional genomics reveals dependency on 6-phosphogluconate dehydrogenase in OXPHOS-deficient tumors. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 1023.

GBM remains one of the most difficult cancers to treat. Despite surgery, radiation and chemotherapy, tumors invariably recur. There is a strong correlation between extend of resection and subsequent time to recurrence and ultimate patient survival. A key challenge for the neurosurgeon is to minimize removal normal brain tissue whilst being as aggressive as possible with regards of removal of tumor tissue. Defining tumor boundary during surgery remains challenging. Fluorescent dyes such as 5- Aminolevulinic acid, have been studied as aids for tumor delineation but have failed to gain widespread acceptance in clinical practice. It has long been known that a substantial number of cancers, including Glioblastoma, exhibit a hyperpolarization of the plasma and mitochondrial membrane potential. This hyperpolarization is manifest by increased retention of lipophilic cationic dyes (Nernstian probes), such as the SPECT-CT probes 99Tc-Sestamibi and 99Tc-Tetrofosmin in both glioma cells in culture and primary tumors. We synthesized a series of ester derivatives of Rhodamine B with high fluorescence quantum yield in the red spectrum and a very favorable biological safety profile. Utilizing a series of intracranial xenografted glioblastoma pre-clinical mouse models, we show that a single I.V. injection of 1 mg/kg fluoroethylrhodamine ester (RhoFe) results in dramatic selective fluorescence in tumor but not surrounding normal brain tissue. This was observed even in tumors with minimal breakdown of the blood brain barrier, as determined by T1-contrast enhancing agents. RhoFe-tumor-selective fluorescence can remain visible up to 72 hours after injection. Similar results were obtained with Tetramethylrhodamine ester (TMRE) and rhodamine 123 (Rho123), but unlike RhoFe, TMRE proved highly toxic, while Rho123 fluorescence was both weaker and in the green spectrum, accompanied by higher endogenous auto-fluorescence. While well tolerated in vivo, RhoFe shows strong photoxicity in cell culture, suggesting potential as a photodynamic therapy agent. Taken together, these data suggest that RhoFe may be a promising tumor painting agent, with potential utility in fluorescence guided surgery as well as photodynamic therapy. Citation Format: Yu-Hsi Lin, Steven Millward, Seth Gammon, Nikunj Satani, Naima Hammoudi, Joshua Philip Gray, Lindsay E. Kelderhouse, Argentina Ornelas, Haley Schroeder, Florian L. Muller. Rhodamine esters as fluorescent tumor painting agents for glioblastoma. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 4242.

Glycolysis inhibition is an active area of investigation in cancer. However, few compounds have progressed beyond the cell culture stage. We have recently demonstrated that genomic passenger deletion of the glycolytic enzyme Enolase 1 (ENO1) leaves gliomas harboring such deletions with less than 10% of normal enzymatic activity, rendering them exquisitely sensitive to enolase inhibitors. However, the tool compound that we employed for these in vitro studies, Phosphonoacetohydroxamate (PhAH), has very poor pharmacological properties and was ineffective in vivo. We performed a SAR studies to increase inhibitor specificity towards ENO2 as well as pro-druging to increase cell permeability. The lead compound generated by these efforts, termed POMHEX, is selectively active against ENO1-deleted glioma cells in culture at ∼35nM (versus μM for PhAH). Using an orthotopic intracranial xenografted model where tumor growth and response to therapy are monitored by MRI, we show that POMHEX is capable of eradicating intracranial ENO1-deleted tumors, with mice remaining recurrence-free even after treatment discontinuation. Taken together, these results reinforce that glycolysis is a viable target and provide in vivo proof-of-principal for the concept of using passenger deletions as targetable vulnerabilities in cancer therapy. Citation Format: Yu-Hsi Lin, Joe Marszalek, Yuting Sun, Naima Hammoudi, Paul Leonard, David Maxwell, Nikunj Satani, Peng Zhang, Todd Link, Gilbert Lee, Maria E. Di Francesco, Barbara Czako, Alan Y. Want, Ronald A. DePinho, Florian L. Muller. Pomhex: a cell-permeable high potency Enolase inhibitor with in vivo anti-neoplastic activity. [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2015 Nov 5-9; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2015;14(12 Suppl 2):Abstract nr C183.