Leveraging Structure-Based Drug Design to Identify Next-Generation MAT2A Inhibitors, Including Brain-Penetrant and Peripherally Efficacious Leads.

Inhibition of the S-adenosyl methionine (SAM)-producing metabolic enzyme, methionine adenosyltransferase 2A (MAT2A), has received significant interest in the field of medicinal chemistry due to its implication as a synthetic lethal target in cancers with the deletion of the methylthioadenosine phosphorylase (MTAP) gene. Here, we report the identification of novel MAT2A inhibitors with distinct in vivo properties that may enhance their utility in treating patients. Following a high-throughput screening, we successfully applied the structure-based design lessons from our first-in-class MAT2A inhibitor, AG-270, to rapidly redesign and optimize our initial hit into two new lead compounds: a brain-penetrant compound, AGI-41998, and a potent, but limited brain-penetrant compound, AGI-43192. We hope that the identification and first disclosure of brain-penetrant MAT2A inhibitors will create new opportunities to explore the potential therapeutic effects of SAM modulation in the central nervous system (CNS).

Discovery of AG-270, a First-in-Class Oral MAT2A Inhibitor for the Treatment of Tumors with Homozygous MTAP Deletion.

The metabolic enzyme methionine adenosyltransferase 2A (MAT2A) was recently implicated as a synthetic lethal target in cancers with deletion of the methylthioadenosine phosphorylase (MTAP) gene, which is adjacent to the CDKN2A tumor suppressor and codeleted with CDKN2A in approximately 15% of all cancers. Previous attempts to target MAT2A with small-molecule inhibitors identified cellular adaptations that blunted their efficacy. Here, we report the discovery of highly potent, selective, orally bioavailable MAT2A inhibitors that overcome these challenges. Fragment screening followed by iterative structure-guided design enabled >10 000-fold improvement in potency of a family of allosteric MAT2A inhibitors that are substrate noncompetitive and inhibit release of the product, S-adenosyl methionine (SAM), from the enzyme's active site. We demonstrate that potent MAT2A inhibitors substantially reduce SAM levels in cancer cells and selectively block proliferation of MTAP-null cells both in tissue culture and xenograft tumors. These data supported progressing AG-270 into current clinical studies (ClinicalTrials.gov NCT03435250).

MAT2A Inhibition Blocks the Growth of MTAP-Deleted Cancer Cells by Reducing PRMT5-Dependent mRNA Splicing and Inducing DNA Damage.

The methylthioadenosine phosphorylase (MTAP) gene is located adjacent to the cyclin-dependent kinase inhibitor 2A (CDKN2A) tumor-suppressor gene and is co-deleted with CDKN2A in approximately 15% of all cancers. This co-deletion leads to aggressive tumors with poor prognosis that lack effective, molecularly targeted therapies. The metabolic enzyme methionine adenosyltransferase 2α (MAT2A) was identified as a synthetic lethal target in MTAP-deleted cancers. We report the characterization of potent MAT2A inhibitors that substantially reduce levels of S-adenosylmethionine (SAM) and demonstrate antiproliferative activity in MTAP-deleted cancer cells and tumors. Using RNA sequencing and proteomics, we demonstrate that MAT2A inhibition is mechanistically linked to reduced protein arginine methyltransferase 5 (PRMT5) activity and splicing perturbations. We further show that DNA damage and mitotic defects ensue upon MAT2A inhibition in HCT116 MTAP-/- cells, providing a rationale for combining the MAT2A clinical candidate AG-270 with antimitotic taxanes.

Selective Vulnerability to Pyrimidine Starvation in Hematologic Malignancies Revealed by AG-636, a Novel Clinical-Stage Inhibitor of Dihydroorotate Dehydrogenase.

Agents targeting metabolic pathways form the backbone of standard oncology treatments, though a better understanding of differential metabolic dependencies could instruct more rationale-based therapeutic approaches. We performed a chemical biology screen that revealed a strong enrichment in sensitivity to a novel dihydroorotate dehydrogenase (DHODH) inhibitor, AG-636, in cancer cell lines of hematologic versus solid tumor origin. Differential AG-636 activity translated to the in vivo setting, with complete tumor regression observed in a lymphoma model. Dissection of the relationship between uridine availability and response to AG-636 revealed a divergent ability of lymphoma and solid tumor cell lines to survive and grow in the setting of depleted extracellular uridine and DHODH inhibition. Metabolic characterization paired with unbiased functional genomic and proteomic screens pointed to adaptive mechanisms to cope with nucleotide stress as contributing to response to AG-636. These findings support targeting of DHODH in lymphoma and other hematologic malignancies and suggest combination strategies aimed at interfering with DNA-damage response pathways.

A chemical biology screen identifies a vulnerability of neuroendocrine cancer cells to SQLE inhibition.

Aberrant metabolism of cancer cells is well appreciated, but the identification of cancer subsets with specific metabolic vulnerabilities remains challenging. We conducted a chemical biology screen and identified a subset of neuroendocrine tumors displaying a striking pattern of sensitivity to inhibition of the cholesterol biosynthetic pathway enzyme squalene epoxidase (SQLE). Using a variety of orthogonal approaches, we demonstrate that sensitivity to SQLE inhibition results not from cholesterol biosynthesis pathway inhibition, but rather surprisingly from the specific and toxic accumulation of the SQLE substrate, squalene. These findings highlight SQLE as a potential therapeutic target in a subset of neuroendocrine tumors, particularly small cell lung cancers.

MTAP Deletions in Cancer Create Vulnerability to Targeting of the MAT2A/PRMT5/RIOK1 Axis.

Homozygous deletions of p16/CDKN2A are prevalent in cancer, and these mutations commonly involve co-deletion of adjacent genes, including methylthioadenosine phosphorylase (MTAP). Here, we used shRNA screening and identified the metabolic enzyme, methionine adenosyltransferase II alpha (MAT2A), and the arginine methyltransferase, PRMT5, as vulnerable enzymes in cells with MTAP deletion. Metabolomic and biochemical studies revealed a mechanistic basis for this synthetic lethality. The MTAP substrate methylthioadenosine (MTA) accumulates upon MTAP loss. Biochemical profiling of a methyltransferase enzyme panel revealed that MTA is a potent and selective inhibitor of PRMT5. MTAP-deleted cells have reduced PRMT5 methylation activity and increased sensitivity to PRMT5 depletion. MAT2A produces the PRMT5 substrate S-adenosylmethionine (SAM), and MAT2A depletion reduces growth and PRMT5 methylation activity selectively in MTAP-deleted cells. Furthermore, this vulnerability extends to PRMT5 co-complex proteins such as RIOK1. Thus, the unique biochemical features of PRMT5 create an axis of targets vulnerable in CDKN2A/MTAP-deleted cancers.

Functional genomics reveal that the serine synthesis pathway is essential in breast cancer.

Cancer cells adapt their metabolic processes to drive macromolecular biosynthesis for rapid cell growth and proliferation. RNA interference (RNAi)-based loss-of-function screening has proven powerful for the identification of new and interesting cancer targets, and recent studies have used this technology in vivo to identify novel tumour suppressor genes. Here we developed a method for identifying novel cancer targets via negative-selection RNAi screening using a human breast cancer xenograft model at an orthotopic site in the mouse. Using this method, we screened a set of metabolic genes associated with aggressive breast cancer and stemness to identify those required for in vivo tumorigenesis. Among the genes identified, phosphoglycerate dehydrogenase (PHGDH) is in a genomic region of recurrent copy number gain in breast cancer and PHGDH protein levels are elevated in 70% of oestrogen receptor (ER)-negative breast cancers. PHGDH catalyses the first step in the serine biosynthesis pathway, and breast cancer cells with high PHGDH expression have increased serine synthesis flux. Suppression of PHGDH in cell lines with elevated PHGDH expression, but not in those without, causes a strong decrease in cell proliferation and a reduction in serine synthesis. We find that PHGDH suppression does not affect intracellular serine levels, but causes a drop in the levels of α-ketoglutarate, another output of the pathway and a tricarboxylic acid (TCA) cycle intermediate. In cells with high PHGDH expression, the serine synthesis pathway contributes approximately 50% of the total anaplerotic flux of glutamine into the TCA cycle. These results reveal that certain breast cancers are dependent upon increased serine pathway flux caused by PHGDH overexpression and demonstrate the utility of in vivo negative-selection RNAi screens for finding potential anticancer targets.

The selectivity of austocystin D arises from cell-line-specific drug activation by cytochrome P450 enzymes.

The natural product austocystin D was identified as a potent cytotoxic agent with in vivo antitumor activity and selectivity for cells expressing the multidrug resistance transporter MDR1. We sought to elucidate the mechanism of austocystin D's selective cytotoxic activity. Here we show that the selective cytotoxic action of austocystin D arises from its selective activation by cytochrome P450 (CYP) enzymes in specific cancer cell lines, leading to induction of DNA damage in cells and in vitro. The potency and selectivity of austocystin D is lost upon inhibition of CYP activation and does not require MDR1 expression or activity. Furthermore, the pattern of cytotoxicity of austocystin D was distinct from doxorubicin and etoposide and unlike aflatoxin B(1), a compound that resembles austocystin D and is also activated by CYP enzymes to induce DNA damage. Theses results suggest that austocystin D may be of clinical benefit for targeting or overcoming chemoresistance.

Cancer-associated IDH1 mutations produce 2-hydroxyglutarate.

Mutations in the enzyme cytosolic isocitrate dehydrogenase 1 (IDH1) are a common feature of a major subset of primary human brain cancers. These mutations occur at a single amino acid residue of the IDH1 active site, resulting in loss of the enzyme's ability to catalyse conversion of isocitrate to alpha-ketoglutarate. However, only a single copy of the gene is mutated in tumours, raising the possibility that the mutations do not result in a simple loss of function. Here we show that cancer-associated IDH1 mutations result in a new ability of the enzyme to catalyse the NADPH-dependent reduction of alpha-ketoglutarate to R(-)-2-hydroxyglutarate (2HG). Structural studies demonstrate that when arginine 132 is mutated to histidine, residues in the active site are shifted to produce structural changes consistent with reduced oxidative decarboxylation of isocitrate and acquisition of the ability to convert alpha-ketoglutarate to 2HG. Excess accumulation of 2HG has been shown to lead to an elevated risk of malignant brain tumours in patients with inborn errors of 2HG metabolism. Similarly, in human malignant gliomas harbouring IDH1 mutations, we find markedly elevated levels of 2HG. These data demonstrate that the IDH1 mutations result in production of the onco-metabolite 2HG, and indicate that the excess 2HG which accumulates in vivo contributes to the formation and malignant progression of gliomas.

Chemical labeling strategies for cell biology.

Methods to visualize, track, measure and perturb proteins in living cells are central to biomedicine's efforts to characterize and understand the spatial and temporal underpinnings of life inside cells. Although fluorescent proteins have revolutionized such studies, they have shortcomings, which have spurred the creation of alternative approaches to chemically label proteins in live cells. In this review we highlight research questions that can be addressed using site-specific chemical labeling and present a comparison of the various labeling techniques that have been developed. We also provide a 'roadmap' for selection of appropriate labeling techniques(s) and outline generalized strategies to validate and troubleshoot chemical labeling experiments.

A general approach for chemical labeling and rapid, spatially controlled protein inactivation.

Chemical labeling of proteins inside of living cells can enable studies of the location, movement, and function of proteins in vivo. Here we demonstrate an approach for chemical labeling of proteins that uses the high-affinity interaction between an FKBP12 mutant (F36V) and a synthetic, engineered ligand (SLF'). A fluorescein conjugate to the engineered ligand (FL-SLF') retained binding to FKBP12(F36V) and possessed similar fluorescence properties as parental fluorescein. FL-SLF' labeled FKBP12(F36V) fusion proteins in live mammalian cells, and was used to monitor the subcellular localization of a membrane targeted FKBP12(F36V) construct. Chemical labeling of FKBP12(F36V) fusion proteins with FL-SLF' was readily detectable at low expression levels of the FKBP12(F36V) fusion, and the level of fluorescent staining with FL-SLF' was proportional to the FKBP12(F36V) expression level. This FL-SLF'-FKBP12(F36V) labeling technique was tested in fluorophore assisted laser inactivation (FALI), a light-mediated technique to rapidly inactivate fluorophore-labeled target proteins. FL-SLF' mediated FALI of a beta-galactosidase-FKBP12(F36V) fusion protein, causing rapid inactivation of >90% of enzyme activity upon irradiation in vitro. FL-SLF' also mediated FALI of a beta-galactosidase fusion expressed in living NIH 3T3 cells, where beta-galactosidase activity was reduced in 15 s. Thus, FL-SLF' can be used to monitor proteins in vivo and to target rapid, spatially and temporally defined inactivation of target proteins in living cells in a process that we call FK-FALI.

In vivo targeting of organic calcium sensors via genetically selected peptides.

A library of constrained peptides that form stable folded structures was screened for aptamers that bind with high affinity to the fluorescent dye Texas red. Two selected clones had binding constants to Texas red of 25 and 80 pM as phage and binding had minimal effects on the fluorescence of Texas red. The peptides interact with distinct but overlapping regions of Texas red. One peptide bound to X-rhod calcium sensors, which share the same core fluorophore as Texas red. These dyes retained calcium sensitivity when bound to the peptide. This peptide was used to label a fusion protein with X-rhod-5F in vivo, and X-rhod sensed changes in calcium locally. Thus, minimal, constrained peptides can functionally bind to environmentally sensitive dyes or other organic agents in biological contexts, suggesting tools for in vivo imaging and analysis.