Novel Pyrrolo[3,2-d]pyrimidine Compounds Target Mitochondrial and Cytosolic One-carbon Metabolism with Broad-spectrum Antitumor Efficacy

Folate-dependent one-carbon (C1) metabolism is compartmentalized into the mitochondria and cytosol and supports cell growth through nucleotide and amino acid biosynthesis. Mitochondrial C1 metabolism, including serine hydroxymethyltransferase (SHMT) 2, provides glycine, NAD(P)H, ATP, and C1 units for cytosolic biosynthetic reactions, and is implicated in the oncogenic phenotype across a wide range of cancers. Whereas multitargeted inhibitors of cytosolic C1 metabolism, such as pemetrexed, are used clinically, there are currently no anticancer drugs that specifically target mitochondrial C1 metabolism. We used molecular modeling to design novel small-molecule pyrrolo[3,2-d]pyrimidine inhibitors targeting mitochondrial C1 metabolism at SHMT2. In vitro antitumor efficacy was established with the lead compounds (AGF291, AGF320, AGF347) toward lung, colon, and pancreatic cancer cells. Intracellular targets were identified by metabolic rescue with glycine and nucleosides, and by targeted metabolomics using a stable isotope tracer, with confirmation by in vitro assays with purified enzymes. In addition to targeting SHMT2, inhibition of the cytosolic purine biosynthetic enzymes, β-glycinamide ribonucleotide formyltransferase and/or 5-aminoimidazole-4-carboxamide ribonucleotide formyltransferase, and SHMT1 was also established. AGF347 generated significant in vivo antitumor efficacy with potential for complete responses against both early-stage and upstage MIA PaCa-2 pancreatic tumor xenografts, providing compelling proof-of-concept for therapeutic targeting of SHMT2 and cytosolic C1 enzymes by this series. Our results establish structure-activity relationships and identify exciting new drug prototypes for further development as multitargeted antitumor agents.

Therapeutic Targeting of Mitochondrial One-Carbon Metabolism in Cancer

One-carbon (1C) metabolism encompasses folate-mediated 1C transfer reactions and related processes, including nucleotide and amino acid biosynthesis, antioxidant regeneration, and epigenetic regulation. 1C pathways are compartmentalized in the cytosol, mitochondria, and nucleus. 1C metabolism in the cytosol has been an important therapeutic target for cancer since the inception of modern chemotherapy, and "antifolates" targeting cytosolic 1C pathways continue to be a mainstay of the chemotherapy armamentarium for cancer. Recent insights into the complexities of 1C metabolism in cancer cells, including the critical role of the mitochondrial 1C pathway as a source of 1C units, glycine, reducing equivalents, and ATP, have spurred the discovery of novel compounds that target these reactions, with particular focus on 5,10-methylene tetrahydrofolate dehydrogenase 2 and serine hydroxymethyltransferase 2. In this review, we discuss key aspects of 1C metabolism, with emphasis on the importance of mitochondrial 1C metabolism to metabolic homeostasis, its relationship with the oncogenic phenotype, and its therapeutic potential for cancer.

Cellular Pharmacodynamics of a Novel Pyrrolo[3,2-d]pyrimidine Inhibitor Targeting Mitochondrial and Cytosolic One-Carbon Metabolism

Folate-dependent one-carbon (C1) metabolism is compartmentalized in the mitochondria and cytosol and is a source of critical metabolites for proliferating tumors. Mitochondrial C1 metabolism including serine hydroxymethyltransferase 2 (SHMT2) generates glycine for de novo purine nucleotide and glutathione biosynthesis and is an important source of NADPH, ATP, and formate, which affords C1 units as 10-formyl-tetrahydrofolate and 5,10-methylene-tetrahydrofolate for nucleotide biosynthesis in the cytosol. We previously discovered novel first-in-class multitargeted pyrrolo[3,2-d]pyrimidine inhibitors of SHMT2 and de novo purine biosynthesis at glycinamide ribonucleotide formyltransferase and 5-aminoimidazole-4-carboxamide ribonucleotide formyltransferase with potent in vitro and in vivo antitumor efficacy toward pancreatic adenocarcinoma cells. In this report, we extend our findings to an expanded panel of pancreatic cancer models. We used our lead analog AGF347 [(4-(4-(2-amino-4-oxo-3,4-dihydro-5H-pyrrolo[3,2-d]pyrimidin-5-yl)butyl)-2-fluorobenzoyl)-l-glutamic acid] to characterize pharmacodynamic determinants of antitumor efficacy for this series and demonstrated plasma membrane transport into the cytosol, uptake from cytosol into mitochondria, and metabolism to AGF347 polyglutamates in both cytosol and mitochondria. Antitumor effects of AGF347 downstream of SHMT2 and purine biosynthesis included suppression of mammalian target of rapamycin signaling, and glutathione depletion with increased levels of reactive oxygen species. Our results provide important insights into the cellular pharmacology of novel pyrrolo[3,2-d]pyrimidine inhibitors as antitumor compounds and establish AGF347 as a unique agent for potential clinical application for pancreatic cancer, as well as other malignancies. SIGNIFICANCE STATEMENT: This study establishes the antitumor efficacies of novel inhibitors of serine hydroxymethyltransferase 2 and of cytosolic targets toward a panel of clinically relevant pancreatic cancer cells and demonstrates the important roles of plasma membrane transport, mitochondrial accumulation, and metabolism to polyglutamates of the lead compound AGF347 to drug activity. We also establish that loss of serine catabolism and purine biosynthesis resulting from AGF347 treatment impacts mammalian target of rapamycin signaling, glutathione pools, and reactive oxygen species, contributing to antitumor efficacy.

Design, synthesis and biological evaluation of novel pyrrolo[2,3-d]pyrimidine as tumor-targeting agents with selectivity for tumor uptake by high affinity folate receptors over the reduced folate carrier

Tumor-targeted 6-substituted pyrrolo[2,3-d]pyrimidine benzoyl compounds based on 2 were isosterically modified at the 4-carbon bridge by replacing the vicinal (C11) carbon by heteroatoms N (4), O (5) or S (6), or with an N-substituted formyl (7), trifluoroacetyl (8) or acetyl (9). Replacement with sulfur (6) afforded the most potent KB tumor cell inhibitor, ~6-fold better than the parent 2. In addition, 6 retained tumor transport selectivity via folate receptor (FR) α and -β over the ubiquitous reduced folate carrier (RFC). FRα-mediated cell inhibition for 6 was generally equivalent to 2, while the FRβ-mediated activity was improved by 16-fold over 2. N (4) and O (5) substitutions afforded similar tumor cell inhibitions as 2, with selectivity for FRα and -β over RFC. The N-substituted analogs 7-9 also preserved transport selectivity for FRα and -β over RFC. For FRα-expressing CHO cells, potencies were in the order of 8 > 7 > 9. Whereas 8 and 9 showed similar results with FRβ-expressing CHO cells, 7 was ~16-fold more active than 2. By nucleoside rescue experiments, all the compounds inhibited de novo purine biosynthesis, likely at the step catalyzed by glycinamide ribonucleotide formyltransferase. Thus, heteroatom replacements of the CH₂ in the bridge of 2 afford analogs with increased tumor cell inhibition that could provide advantages over 2, as well as tumor transport selectivity over clinically used antifolates including methotrexate and pemetrexed.

Abstract 1656: First-in-class, fluorinated folate receptor specific agents that target tumor cells via inhibition of serine hydroxymethyltransferase 2 (SHMT2) and 5-aminoimidazole-4-carboxamide formyltransferase (AICARFTase)

Tumor-specific folate receptor (FR) targeting over the ubiquitous reduced folate carrier (RFC) allows for specific tumor targeting without the attending dose-limiting toxicity of all clinically used cytotoxic agents. In an attempt to provide selectivity for FRα and/or β, folic acid has been conjugated with a variety of cytotoxic payloads to alleviate dose-limiting toxicities since folic acid is not transported via RFC (the major tissue folate transporter) to any significant extent. These conjugates have been unsuccessful in clinical trials likely in part due to the instability of the linkers used for conjugation. Premature cleavage results in the cytotoxic component being released prior to tumor penetration, thus precluding any possibility of selective tumor targeting. We designed novel pyrrolo[3,2-d]pyrimidine compounds in which both the targeting component and the cytotoxic component are in the same molecule. These compounds do not require any conjugation or linkers and hence do not suffer any premature cleavage. In this report, we synthesized and evaluated a “first-in-class” series of fluorinated analogues that target FRα/β and have limited transport via RFC, thus providing selectivity for tumors expressing FRα and/or FRβ. Two lead molecules, AGF347 and AGF355, were potent inhibitors of KB human tumor cell proliferation in vitro (IC50 = 1.05 nM and 6.25 nM respectively). Based on protection experiments with glycine, thymidine, adenosine and 5-aminoimidazole-4-carboxamide (AICA), AGF347 and AGF355 were identified as dual inhibitors of serine hydroxymethyltransferase 2 (SHMT2), a key enzyme in mitochondrial one-carbon metabolism and reported oncodriver, and AICA ribonucleotide formyltransferase (AICARFTase), the 2nd folate-dependent step in de novo purine biosynthesis. These compounds are currently being evaluated in preclinical studies as potential candidates for clinical trials.

Citation Format: Md Junayed Nayeen, Khushbu Shah, Aleem Gangjee, Aamod Dekhne, Zhanjun Hou, Larry H. Matherly. First-in-class, fluorinated folate receptor specific agents that target tumor cells via inhibition of serine hydroxymethyltransferase 2 (SHMT2) and 5-aminoimidazole-4-carboxamide formyltransferase (AICARFTase) [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 1656.

Abstract 2348: Targeting mitochondrial and cytosolic one-carbon metabolism in epithelial ovarian cancer via folate receptor alpha

Epithelial ovarian cancer (EOC) is the most lethal gynecologic malignancy. Though most patients initially respond to platinum-based therapy, the likelihood of disease reoccurrence is nearly 100%. Thus, new tumor-selective therapies for EOC are urgently needed. One such treatment option involves targeting tumors via folate receptor α (FRα), which is overexpressed in up to 90% of EOCs and shows increasing expression with higher stage and grade of disease. Our laboratory discovered novel 5-substituted pyrrolo[3,2-d]pyrimidine analogs (AGF347, AGF359, AGF362 and AGF363) that inhibit mitochondrial one-carbon (C1) metabolism at serine hydroxymethyltransferase (SHMT) 2, with secondary inhibition at cytosolic enzyme targets including those in de novo purine biosynthesis. Potent inhibition was seen with several FRα-expressing EOC tumor cells. Inhibitory potencies were in order, AGF347 > AGF359 > AGF362 > AGF363. Drug effects were substantially reduced with excess folic acid (FA), confirming FRα-mediated drug uptake. Toward cisplatin resistant SKOV3, TOV112D and A2780 EOC cells, inhibition in the nanomolar range was detected with all compounds. Targeted metabolomics, using L-[2,3,3-2H]serine as a tracer in wild-type or SHMT2 knockdown SKOV3 cells, confirmed all compounds inhibited cytosolic and mitochondrial C1-metabolism (at SHMT2). Apoptosis was detected for all compounds by Annexin V/PI, with partial rescue of apoptosis observed upon addition of glutathione (GSH). Glutathione pools (GSH and total GSH+GSSG) were significantly perturbed by drug treatment with all inhibitors in SKOV3 cells, comparable to GSH levels observed in SHMT2 KD cells. In vivo efficacy studies with SKOV3 xenografts treated with either AGF347 or cisplatin in SCID mice showed cisplatin resistance, while AGF347 demonstrated efficacy and delay in disease progression with a median tumor growth delay of 10 days, with the longest delay being 15 days. Our studies describe a series of novel inhibitors targeting mitochondrial and cytosolic C1-metabolism, selectively delivered via FRα, which show direct cytotoxic effects against cisplatin resistant EOC in vitro and in vivo, and display additional mechanisms of cytotoxicity mediated through glycine depletion.

Citation Format: Adrianne C. Wallace-Povirk, Carrie O'Connor, Xun Bao, Jade Katinas, Jennifer Wong-Roushar, Aamod Dekhne, Zhanjun Hou, Md. Junayed Nayeen, Khushbu Shah, Jose Cardiel Nunez, Jing Li, Seongho Kim, Lisa Polin, Charles E. Dann, Aleem Gangjee, Larry H. Matherly. Targeting mitochondrial and cytosolic one-carbon metabolism in epithelial ovarian cancer via folate receptor alpha [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 2348.

Abstract 3980: Dual targeting mitochondrial and cytosolic one-carbon metabolism via the proton-coupled folate transporter with novel 5-substituted pyrrolo[3,2-d]pyrimidine antifolates

Cellular one-carbon (1C) metabolism generates a host of metabolites especially critical to cancer cells. Classical 1C inhibitors often primarily target a single enzyme, commonly leading to drug resistance and necessitating the development of novel compounds targeting multiple enzymes. We synthesized several novel 5-substituted pyrrolo[3,2-d]pyrimidine compounds (AGF291, AGF320, and AGF347) with transport by the proton-coupled folate transporter (PCFT), a folate transporter with a narrow physiological niche, but commonly expressed in many solid tumors. These analogs showed substantial inhibition in proliferation assays with Chinese hamster ovary sublines engineered to individually express PCFT or the ubiquitously-expressed reduced folate carrier (the major tissue folate transporter), as well as in the PCFT-expressing human cancer cell lines H460 (large cell lung carcinoma), HCT-116 (colorectal carcinoma), and MIA PaCa-2 (pancreatic ductal adenocarcinoma). Full abrogation of inhibitory effects for all compounds required co-treatment with both adenosine and glycine, suggesting dual-targeting of cytosolic de novo purine biosynthesis and mitochondrial 1C/glycine metabolism. For cytosolic de novo purine biosynthesis, lack of protection by 5-aminoimidazole-4-carboxamide (AICA) suggested AICA ribonucleotide formyltransferase (AICARFTase), the second folate-dependent enzyme, as the likely enzyme target. AICARFTase inhibition was confirmed in intact H460 and HCT-116 cells by measuring accumulation of AICAR by targeted metabolomics. Targeting of mitochondrial enzymes in H460 and HCT-116 cells was suggested by selective inhibition of incorporation of [3-14C]serine over [14C]formate into the purine intermediate [14C]formyl glycinamide ribonucleotide. [2,3,3-2H]Serine isotope label scrambling analysis confirmed the mitochondrial target to be serine hydroxymethyltransferase2 (SHMT2). SHMT2 is a potentially promising anticancer drug target as its expression is highly correlated with the malignant phenotype across a broad spectrum of cancers beyond lung and colon, including breast, glioma, and liver. Indeed, our initial in-vivo studies with AGF291 suggest potent efficacy toward both MIA PaCa-2 and H460 xenograft mouse models. Our studies demonstrate the potential for tumor-selective dual targeting of both mitochondrial and cytosolic 1C metabolism to overcome resistance to classical 1C inhibitors.

Citation Format: Aamod S. Dekhne, Gregory S. Ducker, Josephine Frühauf, Khushbu Shah, Md. Junayed Nayeen, Adrianne Wallace-Povirk, Carrie O'Connor, Zhanjun Hou, Lisa Polin, Aleem Gangjee, Joshua D. Rabinowitz, Larry H. Matherly. Dual targeting mitochondrial and cytosolic one-carbon metabolism via the proton-coupled folate transporter with novel 5-substituted pyrrolo[3,2-d]pyrimidine antifolates [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 3980.

Abstract 1658: 5N-substituted pyrrolo[3,2-d]pyrimidines: Tumor-targeted agents with first-in-class dual inhibition of serine hydroxymethyl transferase 2 and 5-amino-4-imidazolecarboxamide ribonucleotide formyl transferase enzymes

Design, modeling and synthesis of pyrrolo[3,2-d]pyrimidine analogs were carried out based on the concept of dual mechanisms-of-action for achieving maximal antitumor efficacy. Tumor-targeted delivery via agents that are selectively transported by the folate receptor (FRα/β) and/or PCFT, overexpressed in tumor cells such as epithelial ovarian cancer and non-small cell lung cancer, over the ubiquitously expressed reduced folate carrier (RFC) would afford tumor selectivity. Gangjee, Matherly and coworkers previously reported 5-substituted pyrrolo[2,3-d]pyrimidines as inhibitors of de novo purine nucleotide biosynthesis glycinamide ribonucleotide formyltransferase and 5-aminoimidazole-4-carboxamide (AICA) ribonucleotide formyltransferase (AICARFTase). To synthesize potential dual inhibitors of de novo purine biosynthesis at cytosolic AICARFTase and mitochondrial one-carbon metabolism at serine hydroxymethyl transferase (SHMT) 2, we generated structural hybrids of these inhibitors and 5,10-methylenetetrahydrofolate (the SHMT2 cofactor), 5-substituted pyrrolo[3,2-d]pyrimidine analogs AGF291, AGF299, AGF300, AGF318, AGF320, and AGF331. Molecular modeling of these compounds utilizing X-ray crystal structures of the transporters (FRα/β) and enzyme targets (AICARFTase, SHMT2) predicted excellent activities. The influence of bridge lengths and the aromatic moieties of these novel compounds were evaluated by testing their anti-proliferative activities in Chinese hamster ovary cells expressing human FRα or PCFT, and in FRα- and/or PCFT expressing nasopharyngeal carcinoma (KB) and lung cancer (H460) cells. All the pyrrolo[3,2-d]pyrimidine analogs were inhibitory toward FR- and/or PCFT-expressing cells. Compared to its pyrrolo[2,3-d]pyrimidine analog, AGF300 increased inhibition of cells via FR uptake from IC50 8.6 nM to 2.1 nM (by 4-fold) and decreased uptake through RFC from 56.5 nM to 516 nM (by 10-fold), suggesting a FR-targeted delivery of the agent. Toward H460 xenografts in SCID mice, AGF291 was efficacious. Protection studies with thymidine, glycine, adenosine and AICA and radiotracer studies with [3-14C]serine identified the likely intracellular targets of the pyrrolo[3,2-d]pyrimidine analogs as SHMT2 and AICARFTase. Mitochondrial one-carbon metabolism originating with SHMT2 is a critical source of reducing equivalents and one-carbon units for cytosolic biosynthesis and SHMT2 has been implicated as an oncodriver in several tumor types. While inhibition of SHMT2 was the subject of exploratory studies, our novel pyrrolo[3,2-d]pyrimidine series are “first-in-class” in terms of their in vitro and in vivo antitumor efficacies attributable to dual targeting SHMT2 and AICARFTase.

Citation Format: Khushbu Shah, Aamod Dekhne, Zhanjun Hou, Larry Matherly, Aleem Gangjee. 5N-substituted pyrrolo[3,2-d]pyrimidines: Tumor-targeted agents with first-in-class dual inhibition of serine hydroxymethyl transferase 2 and 5-amino-4-imidazolecarboxamide ribonucleotide formyl transferase enzymes [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 1658.

Mitochondrial and Cytosolic One-Carbon Metabolism Is a Targetable Metabolic Vulnerability in Cisplatin-Resistant Ovarian Cancer

One-carbon (C1) metabolism is compartmentalized between the cytosol and mitochondria with the mitochondrial C1 pathway as the major source of glycine and C1 units for cellular biosynthesis. Expression of mitochondrial C1 genes including SLC25A32, serine hydroxymethyl transferase (SHMT) 2, 5,10-methylene tetrahydrofolate dehydrogenase 2, and 5,10-methylene tetrahydrofolate dehydrogenase 1-like was significantly elevated in primary epithelial ovarian cancer (EOC) specimens compared with normal ovaries. 5-Substituted pyrrolo[3,2-d]pyrimidine antifolates (AGF347, AGF359, AGF362) inhibited proliferation of cisplatin-sensitive (A2780, CaOV3, IGROV1) and cisplatin-resistant (A2780-E80, SKOV3) EOC cells. In SKOV3 and A2780-E80 cells, colony formation was inhibited. AGF347 induced apoptosis in SKOV3 cells. In IGROV1 cells, AGF347 was transported by folate receptor (FR) α. AGF347 was also transported into IGROV1 and SKOV3 cells by the proton-coupled folate transporter (SLC46A1) and the reduced folate carrier (SLC19A1). AGF347 accumulated to high levels in the cytosol and mitochondria of SKOV3 cells. By targeted metabolomics with [2,3,3-2H]L-serine, AGF347, AGF359, and AGF362 inhibited SHMT2 in the mitochondria. In the cytosol, SHMT1 and de novo purine biosynthesis (i.e., glycinamide ribonucleotide formyltransferase, 5-aminoimidazole-4-carboxamide ribonucleotide formyltransferase) were targeted; AGF359 also inhibited thymidylate synthase. Antifolate treatments of SKOV3 cells depleted cellular glycine, mitochondrial NADH and glutathione, and showed synergistic in vitro inhibition toward SKOV3 and A2780-E80 cells when combined with cisplatin. In vivo studies with subcutaneous SKOV3 EOC xenografts in SCID mice confirmed significant antitumor efficacy of AGF347. Collectively, our studies demonstrate a unique metabolic vulnerability in EOC involving mitochondrial and cytosolic C1 metabolism, which offers a promising new platform for therapy.

Abstract 2992: Cellular pharmacodynamics of mitochondrial one-carbon metabolism-targeting 5-substituted pyrrolo[3,2-d]pyrimidine antifolates

Folate-dependent one-carbon metabolism (1CM) is compartmentalized in the mitochondria and cytosol and generates a number of metabolites critical to tumor propagation. Folates are taken up by the plasma membrane facilitative transporters, reduced folate carrier (RFC; major tissue transporter) and proton-coupled folate transporter (PCFT; narrow physiological niche, but commonly expressed in solid tumors), and then transported into mitochondria via the mitochondrial folate transporter (MFT; SLC25A32). Although drug-targeting of cytosolic 1CM remains a clinical mainstay for a variety of cancers, development of clinically-useful agents targeting mitochondrial 1CM remains elusive. Of particular pharmacological interest is the mitochondrial 1CM enzyme, serine hydroxymethyltransferase2 (SHMT2). SHMT2 expression correlates with the oncogenic phenotype in a host of different cancers and, overall, SHMT2 is the fifth-most differentially expressed metabolic enzyme in cancer versus normal tissues. Despite its unequivocal oncogenic importance and therapeutic potential, there are no clinically relevant inhibitors of SHMT2. In this study, we characterized cellular pharmacodynamics of novel 5-substituted pyrrolo[3,2-d]pyrimidine antifolates (AGF291, AGF320, and AGF347) which show in vitroantitumor efficacy toward H460 lung, HCT-116 colon, and MIA PaCa-2 pancreatic cancer cells. Inhibition of mitochondrial SHMT2 and cytosolic 1CM at the purine nucleotide biosynthesis enzymes glycinamide ribonucleotide formyltransferase (GARFTase) and/or 5-aminoimidazole-4-carboxamide ribonucleotide formyltransferase (AICARFTase) by this series was established by in vitro targeted metabolomics in H460, HCT-116, and MIA PaCa-2 cells and in vitro cell-free assays with purified enzymes. By depleting SHMT2-derived formate, these compounds potentiate their own direct inhibition of GARFTase and AICARFTase. Depletion of adenine nucleotide pools in vitro by all compounds led to inhibition of mTOR signaling to S6K1 in HCT116 cells. Subcellular fractionation of MIA PaCa-2 and MFT-null and human MFT-transfected glyB Chinese hamster ovary cells confirmed synthesis of polyglutamyl forms of AGF347 in both cytosol and mitochondria with mitochondrial uptake of AGF347 in part mediated by MFT. Treatment by all compounds decreased the cellular GSH/GSSG ratio, indicating depleted ability to combat oxidative stress. In vivo, AGF347 demonstrated potent antitumor efficacy against MIA PaCa-2 xenografts in SCID mice (n=5) with median tumor growth delay (T-C) in 4 mice >38 days and 1 of 5 tumor-free survivors (cured). In vivo metabolomics on tumor xenografts confirmed inhibition of serine catabolism and purine biosynthesis. Collectively, our studies establish the exceptional therapeutic potential of inhibitors dual-targeting mitochondrial and cytosolic 1CM.

Citation Format: Aamod Dekhne, Khushbu Shah, Gregory S. Ducker, Md. Junayed Nayeen, Jade M. Katinas, Jennifer Wong, Arpit Doshi, Xun Bao, Hasini Kalpage, Jenney Liu, Seongho Kim, Adrianne Wallace-Povirk, Changwen Ning, Carrie O'Connor, Zhanjun Hou, Lisa Polin, Jing Li, Maik Hüttemann, Joshua D. Rabinowitz, Charles E. Dann, Aleem Gangjee, Larry Matherly. Cellular pharmacodynamics of mitochondrial one-carbon metabolism-targeting 5-substituted pyrrolo[3,2-d]pyrimidine antifolates [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 2992.

Abstract 4794: Targeting mitochondrial and cytosolic one carbon metabolism of pancreatic adenocarcinoma via the proton-coupled folate transporter with novel 5-substituted pyrrolo[3,2-d]pyrimidine analogs

Pancreatic cancer (PaC) represents the 4th leading cause of cancer-related deaths in the US with a mortality rate of 99%. The 5-year overall survival rate for PaC is currently 8%. One-carbon (C1) metabolism is frequently altered in cancer. For PaC, TCGA data sets show that elevated expression of key enzymes involved in cytosolic [5-aminoimidazole-4-carboxamide ribonucleotide formyltransferase (AICARFTase) and serine hydroxymethyltransferase (SHMT)1] and mitochondrial [SHMT2 and methylene tetrahydrofolate dehydrogenase 2 (MTHFD2)] C1 metabolism is associated with poor survival. Antifolate therapeutics disrupt cytosolic C1 pathways required for syntheses of thymidylate, purines, and certain amino acids, and are a mainstay for therapy of several cancers. Antifolate uptake into tumors and tissues involves the reduced folate carrier, the major tissue folate transporter, and the proton-coupled folate transporter (PCFT), which shows a more limited tissue distribution but is widely expressed in human solid tumors and is active only at acidic pHs characterizing the tumor microenvironment. We discovered novel 5-substituted pyrrolo[3,2-d]pyrimidine analogs (AGF347, AGF359) with PCFT transport that potently inhibited proliferation of PaC cell lines (AsPC-1, BxPC-3, CFPAC-1, HPAC, MIA PaCa-2 and PANC-1), of which HPAC (KRAS mutant) and BxPC-3 (KRAS wild-type) cells were most sensitive. The PaC cell lines all expressed PCFT transcripts and proteins that were active for PCFT transport with 3H-AGF347 at acid pH. When HPAC cells were incubated with 3H-AGF347 over 48 h, drug accumulated in both cytosol and mitochondria. 3H-AGF347 was extensively metabolized to polyglutamates. Treatment of PaC cells with AGF347 and AGF359 inhibited proliferation by inducing glycine and adenosine auxotrophy that was rescued by excess glycine and adenosine. This implied that both mitochondria and cytosolic C1 metabolism was inhibited. Inhibition of mitochondrial SHMT2 and cytosolic SHMT1, glycinamide ribonucleotide formyltransferase and/or AICARFTase was confirmed by in vitro targeted metabolomics and assays with purified enzymes. Tumor cell killing was confirmed (with HPAC and BxPC-3) by colony-forming assays with AGF347 and AGF359 and drug-induced apoptosis with AGF347 was demonstrated (with HPAC) by annexin V-PI staining and flow cytometry. AGF347 and AGF359 depleted purine nucleotides and inhibited mTOR signaling via S6K1 at least in part (for BxPC-3) via activation of AMPK, likely due to elevated ZMP accompanying suppression of AICARFTase. Collectively, our studies identify first-in-class inhibitors and establish the considerable therapeutic potential of dual-targeting mitochondrial and cytosolic C1 metabolism in PaC independent of KRAS mutation status and reflecting cellular uptake by PCFT.

Citation Format: Changwen Ning, Aamod Dekhne, Md. Junayed Nayeen, Jade M. Katinas, Jennifer Wong, Josephine Frühauf, Xun Bao, Carrie O’Connor, Adrianne Wallace-Povirk, Jing Li, Charles E. Dann, Aleem Gangjee, Larry H. Matherly, Zhanjun Hou. Targeting mitochondrial and cytosolic one carbon metabolism of pancreatic adenocarcinoma via the proton-coupled folate transporter with novel 5-substituted pyrrolo[3,2-d]pyrimidine analogs [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 4794.

Abstract 4800: Targeting mitochondrial and cytosolic one-carbon metabolism in epithelial ovarian cancer via folate receptor alpha

Epithelial ovarian cancer (EOC) is the most lethal gynecologic malignancy. Even though most patients initially respond to platinum-based therapy, the likelihood of disease reoccurrence is virtually 100%. Thus, there is an urgent need for new tumor-selective therapies for EOC. One such treatment option involves targeting tumors via folate receptor alpha (FRα) which is overexpressed in up to 90% of EOCs and shows increasing expression with stage and grade of disease. Our laboratory discovered novel 5-substituted pyrrolo[3,2-d]pyrimidine analogs (AGF291, AGF320, AGF347, AGF359 and AGF362) which inhibit mitochondrial one-carbon (C1) metabolism at serine hydroxymethyltransferase (SHMT) 2, with secondary inhibitions at cytosolic enzyme targets including those in de novo purine biosynthesis and SHMT1. Potent inhibition was seen toward isogenic Chinese hamster ovary (CHO) cell lines individually expressing FRα, the reduced folate carrier (RFC, ubiquitously expressed tissue folate transporter) and the proton-coupled folate transporter (PCFT, expressed in a limited number of normal tissues and inactive at normal pH, and several solid tumors at acidic pH including EOC), and with FRα-expressing tumor cells, including KB and IGROV1, a human EOC cell line. Inhibitory potencies were in order, AGF347 > AGF362 >> AGF291 = AGF320 = AGF359. Drug effects were substantially reduced with excess folic acid, confirming FRα-mediated drug uptake. Toward cisplatin resistant SKOV3, TOV112D and A2780 EOC cells, inhibition in the nanomolar range was detected with all compounds. Excess folic acid abrogated drug effects to varying degrees, suggesting significant uptake by the PCFT and/or the RFC, in addition to FRα. Short-term (5 minutes) cell uptake assays with the EOC cell lines and [3H]AGF347 confirmed transport by PCFT and RFC. With sustained exposures resulting in steady state AGF347 accumulations under both physiologic (pH 7.2) and acidic (pH 6.8, approximating the tumor microenvironment) conditions, FRα uptake predominated. [3H]AGF347 treatment of IGROV1 EOC resulted in substantial drug accumulation in both cytosol and mitochondria. Inhibition of cell proliferation for all analogs was reversed by addition of both glycine and adenosine, implicating C1 metabolism in mitochondria including SHMT2 and de novo purine biosynthesis in the cytosol as the targeted pathways; protection by 5-aminoimidazole-4-carboxamide (AICA) with or without glycine was incomplete, implying direct targeting of AICA ribonucleotide formyltransferase (AICARFTase), the second folate-dependent enzyme in purine biosynthesis. Our studies describe first-in-class FRα-selective compounds targeting mitochondrial and cytosolic C1 metabolism, with potent activity against EOC, including cisplatin resistant EOC.

Citation Format: Adrianne Wallace-Povirk, Carrie O’Connor, Aamod Dekhne, Zhanjun Hou, Md. Junayed Nayeen, Khushbu Shah, Aleem Gangjee, Larry Matherly. Targeting mitochondrial and cytosolic one-carbon metabolism in epithelial ovarian cancer via folate receptor alpha [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 4800.

Abstract 3239: Selective tumor targeting with 5-substituted pyrrolo[2,3-d]pyrimidines with heteroatom bridge substitution

Three specialized systems exist for membrane transport of folates and antifolates across biological membranes. The reduced folate carrier (RFC) is ubiquitously expressed in tumors and normal tissues. Folate receptors (FR) α and β, and the proton-coupled folate receptor (PCFT) exhibit narrow patterns of tissue expression and serve specialized physiologic roles. FRs are expressed in a number of cancer cells (e.g., FRα in ovarian cancer and non-small cell lung cancer; FRβ in acute myelogenous leukemia), whereas PCFT is expressed in a large number of human solid tumors but not leukemias. FRs are either inaccessible to the circulation (FRα) or are nonfunctional (FRβ) in normal tissues. PCFT transport is limited in most normal tissues, given the requirement of acidic pH for optimal activity. These properties facilitate tumor targeting of cytotoxic compounds with specificities for PCFT or/and FR transport. Clinically used antifolates such as methotrexate and pemetrexed are substrates for RFC and their uptake by both normal tissues and tumors confers dose-limiting toxicity due to limited tumor selectivity. We previously reported a 5-substituted pyrrolo[2,3-d]pyrimidine antifolate with a 4-carbon bridge and a phenyl glutamate side chain (AGF127). Previous studies of related 6-substituted pyrrolo[2,3-d]pyrimidines established that the nature and length of the bridge region plays an important role in determining tumor cell potency and transport selectivity. Thus, it was of interest to vary the length and insert heteroatoms into the side chain -(CH2)4- of AGF127, including -(CH2)n-S- (compound 1, n= 3 and compound 2, n = 4) and -(CH2)n-O- (compound 3, n= 3 and compound 4, n = 4), to determine the impact on drug potency and transport selectivity. The novel analogs were tested as growth inhibitors against engineered Chinese hamster ovary (CHO) cells expressing human FRα (RT16), RFC (PC43-10), and PCFT (R2/PCFT4). AGF127 showed potent inhibition of CHO cells expressing FRα (IC50 = 8.6 nM), and reduced activity toward CHO cells expressing RFC (57 nM) or PCFT (840 nM). Compounds 1, 2, and 3, showed reduced inhibitions toward FRα-expressing RT16 cells (4-35-fold) compared to AGF127; however, compound 4 was equipotent to AGF127. Like AGF127, none of the series were effective PCFT substrates. Compound 3, like AGF127, inhibited RFC-expressing PC43-10 cells (IC50 = 37 nM); RFC activity was reduced 6- to >20-fold for compounds 1, 2, and 4. Toward FRα-, RFC-, and PCFT-expressing KB human tumor cells, the order of potency was 4 = AGF127 > 2 = 3 = 1. Compound 4 was also more active (15-fold) than pemetrexed with KB cells. Collectively, the potent and selective activity of compound 4 toward FRα-expressing tumor cells and limited activity toward RFC-expressing cells offers significant advantages over AGF127 and clinically used non-targeted antifolates, suggesting that further preclinical evaluation is warranted.

Citation Format: Aleem Gangjee, Rishabh Mohan, Manasa Ravindra, Adrianne Wallace-Povirk, Carrie O’Connor, Aamod Dekhne, Zhanjun Hou, Larry H. Matherly. Selective tumor targeting with 5-substituted pyrrolo[2,3-d]pyrimidines with heteroatom bridge substitution [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 3239. doi:10.1158/1538-7445.AM2017-3239

Abstract 789: Multi-targeted novel 5-substituted pyrrolo[3,2-d]pyrimidines with tumor-selective targeting and inhibition of cytosolic de novo purine biosynthesis and mitochondrial one-carbon metabolism

One-carbon (C1) metabolism supports a number of physiological and pathophysiological processes ranging from stem cell renewal to cancer progression. Clinically used antifolates are transported into both tumor and normal cells by the ubiquitously expressed reduced folate carrier (RFC). Uptake of targeted agents via tumor-specific folate receptors (FRs) over RFC would permit tumor-selectivity, while limiting dose-limiting toxicities associated with standard chemotherapy. Serine catabolism in mitochondria is the major source of glycine and C1 units for cytosolic biosynthesis, preserves redox balance and minimizes reactive oxygen species, and is an important source of ATP. Among the mitochondrial C1 enzymes, serine hydroxymethyltransferase 2 (SHMT2) and 5,10-methylene tetrahydrofolate (me-THF) dehydrogenase 2 (MTHFD2) are highly expressed in tumors versus normal tissues. SHMT2 has been suggested to be an important oncodriver. However, there are no clinically relevant inhibitors of these enzymes. To generate potential inhibitors of these enzymes, we synthesized 5-substituted pyrrolo[3,2-d]pyrimidine analogs as structural hybrids of cytotoxic 5-substituted pyrrolo[2,3-d]pyrimidines and me-THF. The 5-substituted pyrrolo[3,2-d] pyrimidine with a four carbon bridged phenyl side chain AGF300 afforded selective uptake via FRα over RFC, with inhibition of mitochondrial C1 metabolism and de novo purine biosynthesis, resulting in inhibition of KB human tumor cell proliferation. Inhibition of KB cells by AGF300 was reversed by glycine and adenosine. As previous studies of related 5-substituted pyrrolo[2,3-d]pyrimidines established that the nature and length of the bridge plays an important role in determining tumor cell potency and transport selectivity, we replaced the carbon adjacent to the phenyl ring in AGF300 with heteroatoms, including O (AGF323), S (AGF346) or NH (AGF350). These compounds were tested as growth inhibitors against engineered Chinese hamster ovary (CHO) cells singly expressing human FRα (RT16) or RFC (PC43-10). Incorporation of the O, S and NH in the pyrrolo[3,2-d]pyrimidine analogs preserved excellent inhibition of FRα-containing CHO cells (IC50s of 57 nM, 77 nM and 50 nM, respectively); there was no inhibition of cells with RFC uptake up to 1000 nM. AGF323, AGF346 and AGF350 inhibited proliferation of KB cells which was reversed by excess glycine and adenosine. This establishes that for AGF323, AGF346 and AGF350, both mitochondrial and cytosolic C1 metabolism were inhibited. The development of novel compounds targeting mitochondrial and cytosolic C1 pathways with tumor-selective uptake is highly significant in that this would overcome the drawbacks of currently used cytotoxic agents for cancer.

Citation Format: Nian Tong, Khushbu Shah, Aleem Gangjee, Carrie O’Connor, Adrianne W. Porvirk, Aamod Dekhne, Zhanjun Hou, Larry H. Matherly. Multi-targeted novel 5-substituted pyrrolo[3,2-d]pyrimidines with tumor-selective targeting and inhibition of cytosolic de novo purine biosynthesis and mitochondrial one-carbon metabolism [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 789.

Relationship between volume of bone irradiated during pelvic radiation therapy (RT) for rectal cancer and complete blood counts (CBC)

551

Background: CBCs decrease during standard 3D-conformal pelvic RT for rectal cancer. We evaluated RT plans and CBCs to determine if this decrease is due to RT or only concurrent chemotherapy. Methods: Pelvic bones were contoured for 69 patients who had CT-based treatment plans and CBCs 3 months pre and during RT. CBCs included white blood cells (WBC), hemoglobin (Hgb) and platelets (PLT). The pelvic bones included the bilateral femoral heads and innominate bones, sacrum and L5. Dose volume histograms (DVH) for the pelvic bones were encoded as volumes (Vol) receiving 10 (V10) to 50 Gy (V50) in cc, and percent (Perc) (P10 to P50). The paired t-test was used to compare CBCs pre and during RT. CBCs were analyzed with a repeated measures general linear model for uneven time intervals (R 2.15.1). Results: Median age was 60 y, with 40 men (58%) and preoperative RT in 32 (46%). The median RT dose was 50.4 Gy (range 25.2-60.4 Gy), with all but 2 patients receiving ≥45 Gy. Among 677 CBCs, there were 675 WBC, 677 Hgb, and 672 PLT. The average WBC pre and during RT were 7.7 vs 5.4 bil/L (p<0.001), Hgb 12.9 vs 12.5 g/dL (p=0.005), and PLT 271 vs 227 bil/L (p<0.001). A statistical model with variables for time, the 3-week chemotherapy cycle, the pre-RT average CBC, and 1st and 2nd-order interactions of time with Vol and Perc was constructed. All Vol and Perc for all bones as one structure were significant predictors of the PLT timecourse during RT, with the first-order time-dose interaction encoding a decrease over time, and the second-order time-dose interaction a recovery towards the end of RT. For WBC, V40 and P40 were most predictive (p<0.001), with P40 more so. There were no time-dependent interactions for Hgb. All models had a significant effect of pre RT average CBC and the 3-week chemo cycle. Conclusions: Whereas chemotherapy acts systemically, RT affects a circumscribed area. As such, bone marrow outside the RT field can compensate for lost hematopoiesis over time. Pelvic bone RT dose correlates with PLT and WBC (circulation half-life 1-3 d) but not Hgb (half-life 120 d) supporting an effect of RT in decreased CBC during treatment. Research of approaches that minimize bone marrow dose during pelvic radiotherapy is warranted.