Human SHMT inhibitors reveal defective glycine import as a targetable metabolic vulnerability of diffuse large B-cell lymphoma

Significance

Enzymes of the folate cycle are among the most consistently overexpressed proteins in cancer. Whereas multiple clinical agents inhibit thymidylate synthase, no current drugs target the incorporation of one-carbon into folates via serine hydroxymethyltransferase (SHMT). Using genetics, we show that cancer cells require SHMT to generate tumors. We then describe small-molecule SHMT inhibitors, and show that they block the growth of many human cancer cells, with B-cell lymphomas particularly sensitive to SHMT inhibition. We find that this sensitivity arises from the lymphomas’ inability to import the amino acid glycine, which is made as a byproduct of the SHMT reaction. Thus, B-cell lymphomas have an intrinsic defect in amino acid import, which causes a therapeutically targetable metabolic vulnerability.

Abstract

The enzyme serine hydroxymethyltransferse (SHMT) converts serine into glycine and a tetrahydrofolate-bound one-carbon unit. Folate one-carbon units support purine and thymidine synthesis, and thus cell growth. Mammals have both cytosolic SHMT1 and mitochondrial SHMT2, with the mitochondrial isozyme strongly up-regulated in cancer. Here we show genetically that dual SHMT1/2 knockout blocks HCT-116 colon cancer tumor xenograft formation. Building from a pyrazolopyran scaffold that inhibits plant SHMT, we identify small-molecule dual inhibitors of human SHMT1/2 (biochemical IC₅₀ ∼ 10 nM). Metabolomics and isotope tracer studies demonstrate effective cellular target engagement. A cancer cell-line screen revealed that B-cell lines are particularly sensitive to SHMT inhibition. The one-carbon donor formate generally rescues cells from SHMT inhibition, but paradoxically increases the inhibitor’s cytotoxicity in diffuse large B-cell lymphoma (DLBCL). We show that this effect is rooted in defective glycine uptake in DLBCL cell lines, rendering them uniquely dependent upon SHMT enzymatic activity to meet glycine demand. Thus, defective glycine import is a targetable metabolic deficiency of DLBCL.

Cancer growth and proliferation are supported by metabolic changes, including enhanced glucose uptake, aerobic glycolysis (the Warburg effect), and folate-dependent one-carbon (1C) metabolism (1, 2). The predominant source of 1C units in cancer cells is the amino acid serine (3). The enzyme serine hydroxymethyltransferase (SHMT) catalyzes the conversion of serine and tetrahydrofolate (THF) into glycine and 5,10-methylene–THF. Increases in the synthesis and consumption of serine and glycine have been identified in transformed cells and cancers (4–6). Mitochondrial SHMT (SHMT2) and the immediately downstream mitochondrial enzyme 5,10-methylene-tetrahydrofolate dehydrogenase (MTHFD2) are the most consistently overexpressed metabolic enzymes in cancer (7–9) (Fig. 1A). In most rapidly proliferating cells, 1C units generated from serine catabolism in the mitochondria are exported to the cytosol as formate, which is then reassimilated into folates to support nucleotide synthesis (10–12).

Fig. 1.

SHMT is required for tumor formation in vivo. (A) Serine synthesis and catabolism occur in an intercompartmental cycle mediated by cytosolic and mitochondrial SHMT activity. Key enzymes mediating these transformations are highlighted in capital letters. Arrows indicate the directionality of flux in HCT-116 cells, but most reactions are readily reversible. (B) Growth of subcutaneous tumors from HCT-116 WT and ΔSHMT2 cells implanted in opposite flanks of nude mice (mean ± SEM, n = 10, *P < 0.05, paired t test). (C) Tumor growth of subcutaneous tumors from HCT-116 ΔSHMT2 and ΔSHMT1/2 cells implanted in opposite flanks of nude mice (mean ± SEM, n = 10).

While the mitochondrial pathway typically supplies all of the 1C units in proliferating cells in culture, it is not essential in nutrient replete conditions, as evidenced by the viability of SHMT2 and MTHFD2 deletion cell lines (11, 13). In such deletion cells, cytosolic SHMT1 now metabolizes serine to produce 1C units required for purine and thymidine synthesis. However, the flux carried through this enzyme is insufficient to meet glycine demand, and mitochondrial folate-mutant cell lines are glycine auxotrophs (14). Because glycine is abundant in serum, such auxotrophy has not been considered physiologically relevant in mammals. However, recent work has identified functional amino acid shortages in human tumors, suggesting that transport from serum to tumor may be limiting in some contexts, resulting in dependence on intracellular synthesis (15).

One-carbon metabolism is targeted therapeutically by multiple existing drugs, including the common clinical agents pemetrexed, 5-fluorouracil, and methotrexate (16). One mechanism of action common to several of these agents is inhibition of thymidylate synthase, which utilizes 5,10-methylene–THF. While new chemical tools have recently been disclosed that block de novo serine synthesis (17–19), no existing chemotherapies specifically target the production of 1C units from serine, the primary source of 1C units in tumors.

To block the production of 1C units from serine, simultaneous inhibition of both the cytosolic SHMT1 and mitochondrial SHMT2 is necessary. Here we genetically validate that dual SHMT1/2 genetic knockout, in Ras-driven colon cancer cells, prevents xenograft formation. We present the development of a low nanomolar, stereospecific small-molecule inhibitor of human SHMT1/2. Dual SHMT inhibition blocks growth of many cell lines in a manner that is rescued by the soluble 1C donor formate. In diffuse large B-cell lymphoma (DLBCL) cell lines, however, formate does not rescue cell growth but instead paradoxically enhances cancer cell death. We find that this unexpected outcome reflects a previously unappreciated biochemical vulnerability of DLBCL: inability of these cells to take up glycine, which was previously viewed as a nonessential byproduct of the SHMT reaction.

Results

Requirement for SHMT Activity in HCT-116 Xenograft Formation.

We generated clonal deletion cell lines of SHMT1, SHMT2, and SHMT1/2 from the human colorectal carcinoma cell line HCT-116. Paired Cas9 nickase (Cas9n)-containing constructs that encoded single-guide RNA sequences targeting SHMT1 or -2 were transiently transfected into cells, and mutant colonies from single clones were picked as previously described (11). As previously reported, SHMT1 deletion had no effect on cell growth either in cell culture or as subcutaneous xenografts in nude mice. In contrast, SHMT2 deletion cells grew slower in culture and as xenografts (Fig. 1B and Fig. S1A). Liquid chromatography-mass spectrometry (LC-MS) analysis of the soluble metabolites extracted from SHMT2 deletion tumors revealed characteristic signs of defective serine catabolism (Fig. S1B): serine levels were increased ∼twofold and the purine intermediate aminoimidazole carboxamide ribotide (AICAR), whose consumption requires 10-formyl–THF, was elevated ∼25-fold.

Fig. S1.

SHMT1/2 deletion cell lines cannot establish xenografts in nude mice. (A) Weights of mice from experiments shown in Fig. 1 B and C (mean ± SD, n = 10). (B) Intratumor abundance of AICAR and serine from xenografted tumors (mean ± SD, n = 9, ***P < 0.001, paired t test). (C) Western blot showing loss of SHMT1 and SHMT2 expression in SHMT1/2 double-deletion HCT-116 cell lines. (D) Growth of HCT-116 WT and SHMT1/2 double-deletion cells in standard DMEM with and without supplemental 1 mM sodium formate (mean ± SD, n ≥ 4). (E) Representative mice 35 d after injection with HCT-116 SHMT2 deletion cells (right flank) or HCT-116 SHMT1/2 double-deletion cells (left flank).

To generate dual SHMT1/SHMT2 double-deletion cell lines, SHMT2 deletion cells were transfected with Cas9 and guide RNA sequences targeting SHMT1 in the presence of 1 mM sodium formate. Isolated clones cultured in formate grew at rates comparable to WT parental cells; no growth was observed in media without formate (Fig. S1 C and D). To test whether circulating nucleotides and 1C sources in vivo could support the growth of SHMT1/SHMT2 double-deletion cells, we xenografted them into nude mice. No tumors were observed from the SHMT1/SHMT2 double-deletion cells (Fig. 1C and Fig. S1E). Thus, in HCT-116 xenografts, circulating alternative 1C donors (e.g., betaine, sarcosine, formate) and nucleotides are together insufficient to support intracellular 1C metabolism required for tumorigenesis. It remains to be tested whether SHMT activity is essential for tumors derived from other cell lineages, or whether differential requirements might exist for tumorigenesis versus maintenance.

Small-Molecule Inhibitors of Human SHMT1/2.

Compounds with the pyrazolopyran scaffold represented by compound 1 (Fig. 2A) were described as inhibitors of plant SHMT and showed efficacy as herbicides (20). Derivatives with a meta-thiophene substitution were recently published as inhibitors of Plasmodium SHMT (21). When these compounds were tested in human cell culture, potency was poor (22). We optimized compounds of this class for human SHMT1 and 2 (23). Compounds of this class were modestly more potent in vitro against SHMT1 than SHMT2. Changes that improve potency against both human isoforms include introduction of an isopropyl group at the chiral four-carbon of the pyrano ring and adding steric bulk to the metasubstitutions on the phenyl ring (compound 2). Aromatic substitution at this position further increased potency, yielding compound 3, which inhibits T cell proliferation (24). We term this inhibitor serine hydroxymethyltranferase inhibitor 1, or SHIN1.

Fig. 2.

A folate-competitive cell-permeable inhibitor of human SHMT1/2. (A) Structure of pyrazolopyran inhibitor of plant SHMT (compound 1) and two optimized inhibitors for human SHMT1/2 (compounds 2 and 3; compound 3 = SHIN1). IC₅₀s shown are for human SHMT1 and -2 in an in vitro assay. (B, Left) Compound 2 in complex with human SHMT2 as solved in a 2.5-Å resolution X-ray crystal structure. The electron density of the compound is shown as the 2F_o–F_c map contoured at 0.5 σ and generated with compound 2 omitted. (Right) An overlay of the SHMT2/compound 2 structure with the structure of 5-formyl–THF-triglutamate in complex with rabbit SHMT1. (C) Growth of HCT-116 WT ±1 mM formate and ΔSHMT1 and ΔSHMT2 cell lines in the presence of increasing concentrations of SHIN1 (n ≥ 3). (D) Cellular IC₅₀ values for growth inhibition by compound 2 and SHIN1.

To understand the binding mode of these inhibitors, we solved a 2.47-Å structure of human SHMT2 as a dimer in complex with glycine, pyridoxal 5′-phosphate (PLP), and racemic compound 2 (Fig. 2B and Table S1) (PDB ID code 5V7I). Electron density was identified in both binding pockets of the protein dimer, but in only one active site was it well resolved. Similar to the solved structure of a pyrazolopyran inhibitor in complex with Plasmodium vivax SHMT (21), hydrogen binding contacts with the exocyclic amine are made with the amide backbone of L166 and between the pyrazole and H171. Overlaying our inhibitor-bound structure with a previously solved structure of rabbit SHMT1 bound to 5-formyl–THF triglutamate (PDB ID code 1LS3) revealed that the bicyclic ring system of compound 2 and pteridine moiety of folate occupy the same space, but at a different angle (Fig. 2B). However, hydrogen bond contacts are preserved, and engage the inhibitor at several core positions, including the exocyclic amine and the pyrazole nitrogens. The substituted phenyl ring and associated pyrrolidine of compound 2 trace along the para-aminobenzoic acid moiety of folate as it exits the pteridine binding pocket toward the solvent-exposed folate polyglutamate side chain. Directly adjacent to the pyrrolidine lies a tyrosine residue that is well positioned to form a π-stacking interaction with the phenyl of SHIN1, potentially contributing the improved potency of this compound. Given the conserved nature of the SHMT active site, these compounds are likely to inhibit SHMT enzymes not only of humans, but also other mammals.

Both compound 2 and SHIN1 contain a single chiral center. Although crystallization was performed with racemic compound 2, the electron density was consistent with only a single enantiomer binding to the enzyme. Using chiral chromatography, we separated compound 2 and confirmed enantioselective enzyme inhibition (Fig. S2A).

Fig. S2.

A folate competitive SHMT inhibitor. (A) Enzymatic inhibition of human SHMT1 and SHMT2 with enantiomerically resolved fractions of compound 2. (B) (−)-SHIN1 does not inhibit growth of HCT-116 cells at concentrations up to 30 µM (mean ± SD, n = 3). (C) Formate rescue of cell growth in SHIN1 treated cells requires glycine. Normalized growth of HCT-116 cells treated with 10 µM (+)-SHIN1 in DMEM with and without standard glycine and supplemented with formate (mean ± SD, n = 3). (D) Growth of human pancreatic cell line 8988T with indicated concentrations of SHIN1 (mean ± SD, n = 6).

Cell Growth Inhibition.

We next sought to investigate the activity of compound 2 and SHIN1 against cytosolic and mitochondrial SHMT isoforms in cultured cells. The inactive (−) enantiomer of SHIN1 had no significant effect on growth in HCT-116 cells at doses up to 30 µM (Fig. S2B), whereas the active (+) enantiomer blocked growth with half-maximal inhibitory constants (IC₅₀) of 870 nM (Fig. 2 C and D). To analyze the effects of inhibition on each isoform independently, we used the SHMT1 and SHMT2 HCT-116 deletion clones. The active enantiomers of both compounds, (+)-2 and (+)-SHIN1, were potent against cytosolic SHMT1, as evidenced by IC₅₀ for growth of less than 50 nM in SHMT2 deletion cells (Fig. 2 C and D). In contrast, SHMT1 deletion cells showed indistinguishable sensitivity from WT, confirming that mitochondrial SHMT inhibition is limiting for compound efficacy (Fig. 2D).

As compound 2 and SHIN1 both have similar biochemical activities against SHMT1 and SHMT2, the much higher doses required for functional inhibition of cellular SHMT2 likely reflects a combination of imperfect mitochondrial penetration and greater intrinsic cellular SHMT2 activity (i.e., a substantial functional reserve due to high SHMT2 expression). Importantly, the effects on cell growth of compound 2 and SHIN1 could be rescued by addition of formate, indicating that they inhibit cell growth through on-target depletion of cellular 1C pools (Fig. 2D). However, because glycine is also a product of the SHMT reaction, formate can only rescue cell growth when this amino acid is present in the media (Fig. S2C).

Notably, while most cancers have high mitochondrial 1C pathway activity, certain cancer cells, such as the pancreatic cancer cell line 8988T, harbor genetic lesions in the mitochondrial folate pathway activity and therefore rely on SHMT1 to generate 1C units (11). In such cells, SHIN1 impairs cell growth at concentrations <100 nm due to its potent engagement of cellular SHMT1 (Fig. S2D).

SHMT Target Engagement.

Inhibition of cellular SHMT activity can be monitored by isotope tracers and LC-MS. U-¹³C serine is catabolized in the mitochondria by SHMT2 into U-¹³C-glycine and a ¹³C-5,10-methylene–THF. Glycine is further incorporated into downstream metabolites, such as glutathione and purines, whereas the folate 1C unit can be exported to the cytosol for incorporation into purines and thymidine. In addition, glycine and a 1C unit can recombine to make partially labeled serine via SHMT1 or SHMT2 (Fig. S3A). To assess target engagement, we compared the effects of SHMT genetic manipulations to pharmacological treatment with SHIN1. Serine media consumption was inhibited in both HCT-116 SHMT1/2 double-deletion cells and WT cells treated with (+)-SHIN1 (Fig. S3B). Glycine production from serine and subsequent incorporation into glutathione or ADP was completely blocked in SHMT1/2 double-deletion cells, as evidenced by the missing M+2 labeling fraction (Fig. 3A). Nearly complete blockade was observed in WT cells treated with (+)-SHIN1 but not the inactive enantiomer (−)-SHIN1. Drug treatment also blocked recombination of glycine and 10-formyl–THF to reform serine (Fig. S3B). Genetic deletion of SHMT1/2, and to a lesser extent SHMT2, results in a build-up of purine biosynthetic intermediates upstream of steps requiring 10-formyl–THF as a substrate (Fig. 3B). Such build-up is also seen with (+)-SHIN1. Thus, SHIN1 phenocopies—in an enantioselective manner—the metabolic consequences of SHMT genetic deletion.

Fig. S3.

Isotope tracing and metabolomics demonstrate that SHIN1 inhibits the SHMT reaction. (A) Schematic of isotope labeling from U-13C-serine into downstream metabolites. Heavy (¹³C) atoms are represented by filled in circles. (B, Upper) Fraction of original serine remaining in media after 24 h incubation U-¹³C serine (±)-SHIN1 (5 µM). (Lower) M+1 ¹³C-labeling fraction of media serine after 24 h (mean ± SD, n = 3). (C) Metabolite abundances (total ion count) in HCT-116 WT cells (Upper) and HCT-116 cells treated with (+)-SHIN1 (10 µM) (Lower) compared with those in ∆SHMT1/∆SHMT2 double-deletion cells (mean, n = 3).

Fig. 3.

(+)-SHIN1 inhibits SHMT1/2 in HCT-116 cells. (A) M+2 ¹³C-labeling fraction of intracellular ADP and glutathione after 24 h ¹³C-serine coincubation with DMSO, 5 µM (+)-SHIN1, or 5 μM (–)-SHIN1. ΔSHMT1/2 cells were cultured without formate for the duration of labeling (mean ± SD, n = 3). (B) Normalized (to DMSO HCT-116 WT cells) levels of purine biosynthetic pathway intermediates after 24-h incubation ±SHIN1 (mean ± SD, n = 3). (C) Total metabolite abundances in HCT-116 cells treated with DMSO vs. (+)-SHIN1 (10 µM) for 48 h. Metabolites whose abundances differ by more than fourfold between conditions are highlighted in red (mean, n = 3). (D) Metabolite abundance in HCT-116 cells treated with DMSO or SHIN1 in the presence of 1 mM sodium formate. The same metabolites whose abundances were different in C are highlighted in red (mean, n = 3).

To assess the selectivity of the metabolic effects of SHIN1, we performed untargeted LC-MS analysis on soluble metabolites from drug-treated cells (Fig. 3C). In addition to purine intermediates, we saw build-up of purine salvage products (xanthosine, guanosine), whose increase is consistent with purine insufficiency. We further saw build-up of homocysteine, a classic marker of 1C deficiency. We also observed depletion of the pyrimidine intermediate N-carbamoyl-aspartate, likely reflecting feedback inhibition of aspartate transcarbamoylase by excess pyrimidines in the purine-starved cells (25). Importantly, there were no other large changes in metabolism, suggestive of off-target effects. Moreover, the changes in abundances were rescued by formate (Fig. 3D) and metabolite abundances in SHMT1/2 double-deletion cells closely matched those from WT cells treated with (+)-SHIN1 (Fig. S3C). Thus, at doses sufficient to robustly inhibit SHMT1 and SHMT2 in cell culture, (+)-SHIN1 selectively targets 1C metabolism.

Cancer Cell Line Sensitivity to SHMT Inhibition.

Unfortunately, SHIN1 and related pyrazolopyrans are unstable in liver microsome assays and have poor in vivo half-lives, precluding their immediate use in animal models. Accordingly, we focused on their in vitro application across a wide range of cancer cell lines. Specifically, we screened a panel of nearly 300 human cancer cell lines for growth in the presence of the (+)-enantiomer of compound 2 (Fig. 4A and Table S2). The median IC₅₀ was 4 µM. Cell lines of B-cell lymphoma origin were enriched in the more sensitive half of cells (P < 0.001, Fisher’s exact test). This effect was driven by a pronounced sensitivity of Burkitt’s and DLBCL lymphomas (Fig. 4A). We then rescreened a set of hematological cancer lines with (+)-SHIN1, supplemented with and without formate to test for rescue (Fig. 4B). Like HCT-116 cells, cell lines of T cell origin, such as acute lymphocytic leukemia (ALL) cells, were largely rescued from the antigrowth effects of (+)-SHIN1 by formate (Fig. 4B, gray bars). In contrast, formate failed to rescue the growth of B-cell lymphoma lines.

Fig. 4.

SHMT inhibitors are particularly active against B-cell malignancies. (A) Ranked IC₅₀, in units of molarity, of compound (+)-2 for growth inhibition of 298 human cancer cell lines. Lines of B-cell origin are highlighted in red and are enriched among the more sensitive cells (IC₅₀ < 4 µM). (B) IC₅₀ of (+)-SHIN1, with and without 1 mM formate, for growth inhibition of select hematological cell lines. (C) Fraction of Jurkat and Su-DHL-4 cells that are apoptotic after 24 h (+)-SHIN1 treatment (2.5 and 5 µM respectively) as indicated by flow cytometry using FITC-Annexin V staining (mean ± SD, n ≥ 3, *P < 0.05, ***P < 0.001, unpaired t test).

To explore this surprising lack of rescue further, we analyzed by flow cytometry the effect of (+)-SHIN1, with and without formate, on the DLBCL cell line Su-DHL-4. SHIN1 itself induced apoptosis as measured by Annexin V surface staining (Fig. 4C and Fig. S4A). Apoptosis was enhanced by cotreatment with formate. In contrast, as expected, formate rescued Jurkat E6-1 leukemia cells from apoptosis (Fig. 4C and Fig. S4B).

Fig. S4.

Flow cytometry histograms of (+)-SHIN1 treated cells. (A) Representative flow cytometry histograms of B-cell line Su-DHL-4 treated with (+)-SHIN1 (5 µM) and formate (1 mM). Etoposide (0.33 µM) was used as a positive control. Cells were stained with propidium iodide and FITC-conjugated Annexin V; 10,000 events shown. (B) Representative flow cytometry histograms of 2.5 µM (+)-SHIN1 treated Jurkat cells. Fraction of events in apoptotic quadrant shown. Cells were stained with propidium iodide and FITC-conjugated Annexin V; 10,000 events shown.

To account for these observations, we hypothesize that the failure of formate to rescue growth in the DLBCL cell lines is due to a requirement for both glycine and 1C units made by SHMT in these cells. When glycine is limiting, formate can enhance the cytotoxicity of SHMT inhibition. For example, formate augments the effect of SHIN1 in HCT-116 cells in glycine-free media (Fig. S2C). Mechanistically, by supplying 5,10-methylene–THF, formate may drive residual SHMT enzymatic function in the glycine-consuming direction. Alternatively, whereas cells may have the machinery to sense 1C deficiency and safely pause growth (e.g., due to AICAR activation of AMPK), they may lack comparable mechanisms for surviving glycine limitation.

DLBCL Cells Require SHMT to Make Glycine for Purine Synthesis.

The inability of formate to rescue the antigrowth effects of SHIN1 in DLBCL cell lines suggested that glycine may be limiting in these cells. To explore this hypothesis, we characterized the metabolic effects of SHIN1 in DLBCL and Jurkat cells treated with (+)-SHIN1 (72 h, 5 µM) with and without formate. In Jurkat and DLBCL cell lines Su-DHL-4 and Su-DHL-2 in the absence of formate, SHIN1 treatment led to a large reduction in nucleotide triphosphates (Fig. 5A and Fig. S5A). This can be rationalized as reflecting impaired purine synthesis, which requires both 1C units and glycine, with pyrimidines also falling due to endogenous mechanisms that balance their levels with those of purines. There is also a component of energy stress, particularly in Su-DHL-4 cells, as nucleotide monophosphates were increased, not decreased (Fig. S5B). Consistent with 1C limitation, dTTP, whose synthesis requires a folate 1C unit, was more depleted than other pyrimidines.

Fig. 5.

Glycine made by SHMT is required for B-lymphoma cell line growth. (A) Normalized total ion counts of nucleotide triphosphates in Jurkat ALL cells and Su-DHL-4 DLBCL cells after 72-h treatment with (+)-SHIN1 (5 µM). Coculture with 1 mM formate restores nucleotide levels selectively in Jurkat cells (mean ± SD, n = 3–6). (B) Normalized glutathione levels from Jurkat and Su-DHL-4 cells treated as in A (mean ± SD, n = 3–6). (C) Growth of Su-DHL-4 cells treated with (+)-SHIN1 and hypoxanthine (100 µM) or thymidine (16 µM; mean ± SD, n = 3). (D) Intracellular U-¹³C-glycine assimilation kinetics in Jurkat and Su-DHL-4 (gly, glycine; GSH, glutathione; mean ± SD, n = 3). (E) The steady-state labeling fraction of intracellular metabolites synthesized from glycine in cancer cell lines cultured in RPMI containing U-¹³C-glycine (mean ± SD, n = 3). (F) Cell growth (normalized to DMSO) of DLBCL and other hematopoietic cancer lines with 2.5 µM SHIN1, in RPMI with or without 1 mM formate and 10× physiological glycine (100 mg/L); all conditions included at least normal media glycine (10 mg/L; mean ± SD, n = 3). (G) Cell growth (or death) as measured by log₂-fold change in cell number over 48 h in Su-DHL-4 cells cultured in RPMI with and without glycine (10 mg/L), formate (1 mM), the glycine transporter inhibitor RG1678 (300 nM), and/or (+)-SHIN1 (5 µM) (mean ± SD, n = 3). (H) Schematic illustrating the proposed glycine vulnerability in B cells. The SHMT reaction makes two products, 5,10-methylene–THF and glycine. When SHMT is inhibited, exogenous formate can be incorporated into the 1C cycle, whereas in B cells poor glycine uptake limits the ability of extracellular glycine to rescue.

Fig. S5.

SHMT inhibition in DLBCL cell lines results in depletion of glycine-derived metabolic products. (A) Normalized total ion counts of nucleotide triphosphates in Su-DHL-2 cells after 72 h treatment with (+)-SHIN1 (5 µM) and formate (1 mM; mean ± SD, n = 3). (B) Normalized total ion counts of nucleotide mono- and diphosphates in Jurkat and Su-DHL-4 cells after 72 h treatment with (+)-SHIN1 (5 µM) and formate (1 mM; mean ± SD, n = 3). (C) Formate (1 mM) rescues 1C shortage induced by SHIN1 (5 µM, 72 h) as evidenced by normalization of the levels of the purine intermediate AICAR (mean ± SD, n = 3). (D) Normalized glutathione levels from Jurkat and Su-DHL-2 cells treated as in A (mean ± SD, n = 3). (E) Normalized Su-DHL-4 growth after 48 h treatment with 2.5 µM (+)-SHIN1 and 250 µM glutathione (GSH; mean ± SD, n = 2). (F) Sensitivity of Su-DHL-4 cells to SHIN1 is dependent upon media glycine concentration. Cells were cultured in either normal RPMI (10 mg/L glycine) or RPMI containing 10× or 0.1× glycine and (+)-SHIN1 for 48 h, and growth was measured by resazurin assay (mean ± SD, n = 3). (G) Sensitivity of Jurkat cell growth to (+)-SHIN1 as a function of glycine concentration in RPMI in the same conditions as in E (mean ± SD, n ≥ 3).

Formate supplementation restored nucleotide levels in Jurkat but not DLBCL cell lines. We confirmed that formate rescues folate 1C levels in DLBCL cells, as the AICAR accumulation induced by (+)-SHIN1 is fully reversed (Fig. S5C). Thus, while nucleotide synthesis in SHIN1-treated Jurkat cells is solely limited by 1C units, an additional factor is lacking in DLBCL cells. Consistent with glycine being the second factor missing in DLBCL cells, (+)-SHIN1 treatment depleted the glycine-containing redox defense tripeptide glutathione (Fig. 5B and Fig. S5D). Strikingly, while SHIN1 alone did not alter glutathione in Jurkat cells, formate addition caused glutathione depletion. This further validates that, when SHMT is inhibited, provision of excess 1C units can cause glycine stress. Glutathione supplementation did not rescue growth (Fig. S5E). Based on these results, we predicted that growth in SHMT-inhibited DLBCL cells might be restored with purine supplementation, which would simultaneously alleviate 1C and glycine metabolic stress. Growth was partially rescued in Su-DHL-4 cells treated with hypoxanthine (Fig. 5C). Thymidine, which rescues the effects of the classic antifolate pemetrexed but does not contain glycine, had no benefit in SHIN1-treated DLBCL cells. Thus, SHIN1 blocks cell growth through a progressive depletion of purines, leading to loss of nucleotide triphosphates. Restoration of purines levels restores growth.

The depletion of glycine-derived metabolites in DLBCL cells led us to examine whether glycine shortage might also impact protein synthesis. Severe amino acid shortages lead to loss of cognate tRNA charging and thus ribosome stalling, which can be measured using ribosome profiling (15). We performed ribosome profiling on Su-DHL-4 cells treated with (+)-SHIN1 (Fig. S6 A and B). Untreated Su-DHL-4 cells growing in RPMI did not show evidence of glycyl-tRNA insufficiency; no enrichment for these codons was observed (Fig. S6C). Furthermore, we did not observe any difference in glycine codon occupancy between treated and control cells (Fig. S6D). Collectively, these results suggest a hierarchy in the sensitivity of different intracellular metabolic products to glycine levels: glutathione synthesis is most sensitive, followed by purine synthesis, with protein synthesis most resistant. This hierarchy is consistent with biochemical measurements of the K_m values of the relevant enzymes: the glycyl-tRNA amino acid synthase has a lower K_m for glycine (15 µM) than that found in glycinamide ribonucleotide synthetase (45 µM) or glutathione synthetase (452 µM) (26–28).

Fig. S6.

SHIN1 treated Su-DHL-4 cells have sufficient glycine for tRNA charging. (A and B) Su-DHL-4 cells were treated with 5 µM (+)-SHIN1 or DMSO for 48 h before cell lysis, ribosome footprinting, and library purification. Abundance and phasing of sequenced reads shown. Highlighted reads (28–30 nt) were used for subsequent codon occupancy analysis. (C) Codon occupancy plotted against total codon frequency from ribosome profiling of Su-DHL-4 cells treated with DMSO (control). Shaded region is ±1 SD (σ) of the mean codon occupancy. Glycine codons (red) are highlighted. (D) Ratio of codon occupancies for SHIN1 treated Su-DHL-4 cells compared with DMSO control. Shaded region is ±1 SD (σ) of the mean ratio of codon occupancy. Glycine codons (red) are highlighted. Refer to SI Methods for details.

Defective Glycine Uptake in DLBCL.

SHIN1 induced glycine deficiency in DLBCL cells, even though they were cultured in complete media with glycine (RPMI, 10 mg/L glycine = 130 µM). This suggested that glycine uptake is intrinsically impaired in these cells. Using U-¹³C-glycine, we monitored the kinetics of extracellular glycine incorporation into cells and downstream metabolic products (Fig. 5D). Labeling of intracellular glycine products, such as glutathione and ADP, was markedly less in Su-DHL-4 cells than Jurkat cells. In a larger set of cell lines, composed of both other hematological cancer and adherent cell lines, steady-state labeling of intracellular metabolites from glycine was significantly lower in B-cell lymphoma cell lines (Fig. 5E).

Given the apparent glycine shortage in these B cells upon SHIN1 treatment, we next sought to augment extracellular glycine levels and evaluate response to drug. We first altered the concentration of glycine in RPMI and observed response to drug. A reduction of glycine in the media modestly improved the potency of SHIN1, indicating that the cells were sensitive to extracellular glycine. More strikingly, increasing the media glycine by 10-fold substantially rescued the cells from SHIN1 (Fig. S5F). In contrast, in Jurkat cells, a small amount of extracellular glycine was sufficient and more did not further rescue the cells from SHIN1 (Fig. S5G). Across a set of DLBCL cell lines, representing both ABC and GBC subtypes, supplying both formate and supraphysiologic glycine (100 mg/L, 1.3 mM) generally rescued cell growth (Fig. 5F). These results indicate the importance of both products of the SHMT reaction, glycine and folate 1C units, for the proliferation of DLBCL cell lines.

Knowing that manipulating glycine could augment the efficacy of SHIN1, we tested different mechanisms to decrease glycine. As observed previously, when formate was added, SHIN1 was transformed from being a drug that slowed cell growth to one that was fully cytostatic (Fig. 5G). Further removing glycine caused significant cell death. Interestingly, combining the glycine reuptake transporter 1 (GlyT1; SLC6A9) inhibitor RG1678 with SHIN1 further increased cell death, even in the presence of media glycine (29) (Fig. 5G). These results suggest that glycine uptake in these cells is mediated by GLYT1 and that combinations of formate, SHMT inhibitor, and GLYT1 inhibitor may selectively target these cells.

Discussion

Targeting folate metabolism has been employed clinically to treat cancer for over 70 y (30). Despite the use of antifolates and other antimetabolites in many important chemotherapy regimens, their clinical effectiveness is limited by side effects in normal proliferating tissue. Identifying metabolic processes that can be targeted in a more tumor-selective manner remains a major challenge.

In this study, we targeted the SHMT reaction, which uses serine to generate a folate-bound 1C unit and glycine. Consistent with prior reports (22), we found that pyrazolopyrans have detectable activity against human SHMT. Through substantial chemistry efforts, we enhanced the potency for human SHMT by over 100-fold, resulting in inhibitors such as SHIN1 with on-target dual SHMT1/2 cellular inhibition at nanomolar to low micromolar concentrations. We extensively validated the on-target activity of these compounds using metabolomics in combination with genetics. While these compounds have appropriate stability for cell culture studies, including of primary T cells (24), they are not currently usable in vivo due to rapid clearance.

Screening of cancer cell lines for sensitivity to small-molecule SHMT1/2 inhibitors revealed specific metabolic vulnerabilities of certain cancers. One mode of sensitization, exemplified by the pancreatic cancer cell line 8988T, results from defects in mitochondrial folate metabolism. Such cells are dependent upon SHMT1 for production of 1C units, and functionally have low reserve SHMT activity, rendering them sensitive to low concentrations of SHIN1.

By a different mechanism, B-cell lymphomas are also uniquely sensitive to SHMT inhibition. We show that these cells are intrinsically deficient in glycine uptake and thus require glycine made by SHMT to grow. When combined with formate, SHMT inhibitors do not function as classic antifolates by disrupting 1C metabolism, but rather, in cells with impaired glycine uptake, as cell-type–specific glycine depletion agents (Fig. 5H). As a fundamental precursor to many essential biomolecules, glycine is in high demand. Indeed, the quantitative demand for glycine to support protein, nucleotide, and glutathione synthesis exceeds the cellular requirement for 1C units (11). Using metabolomics and ribosome profiling, we characterized the susceptibility of these processes to glycine stress. Consistent with the reported enzymatic K_m values, glutathione and purine synthesis were more sensitive than protein synthesis to glycine depletion.

Targeting an amino acid vulnerability is a well-established therapeutic strategy in cancer. A useful comparison with the intrinsic defect in glycine uptake in DLBCL is the defect in asparagine synthesis in ALL, which creates a dependence upon external sources of asparagine (31). This dependence is targeted by asparaginase, a core medicine in pediatric ALL therapy (32). In DLBCL, because the defect is in glycine transport rather than synthesis, the therapeutic strategy rests on inhibiting intracellular glycine synthesis. The resulting efficacy can be increased either by decreasing extracellular glycine, or more promisingly therapeutically, by further depleting intracellular glycine by formate addition or glycine uptake inhibition. Formate is attractive because it can rescue the effects of SHMT inhibition in normal tissues with strong glycine uptake. Both approaches however, may exacerbate toxicity in tissues with naturally low glycine transport. While glycine transport is poorly characterized in vivo in most tissues, existing data suggest immune and neurological tissues may be potentially sensitive to modulation of glycine synthesis. Going forward, a careful assessment of amino acid transport in vivo will be required to understand how to best exploit glycine transport defects for therapy.

Methods

All mouse work was approved by the Princeton University Institutional Animal Care and Use Committee. For metabolite measurements, cultured cells were incubated in media containing dialyzed FBS and the isotopically labeled metabolite of interest. Cells were quenched with cold methanol and metabolites analyzed by LCMS. Full-length human SHMT1 and -2 protein was isolated from Escherichia coli using nickel capture followed by cleavage of the HIS tag using tobacco etch virus (TEV) protease. Complete chemical synthesis details and compound characterizations are provided in Chemical Synthesis Methods. All experimental procedures are described in detail in SI Methods.

SI Methods

Cell Lines, Reagents, Constructs, and Antibodies.

Fresh aliquots of HCT-116 (CCL-247, lot# 60506215) were purchased from ATCC before beginning experiments. SHMT2 was knocked out in HCT-116 lines using CRISPR/Cas9 nickase method using the following PAM sequences in exon 2: GGACAGGCAGTGTCGTGGCCTGG, TCTCAGGATCACTGTCCGACAGG (33). Jurkat (clone E6.1), Farage, Su-DHL-2, Su-DHL-4, Su-DHL-6, Daudi, Toledo, REH, and LN-229 cells were from ATCC; 8988T was from DSMZ. Adherent cell lines were subcultured in 5% CO₂ at 37 °C using DMEM (CellGro 10–017; Mediatech) supplemented with 10% FBS (F2442; Sigma-Aldrich); suspension cell lines were subcultured in 5% CO₂ at 37 °C in RPMI-1640 (11875; Gibco) with 10% serum. For all experiments, media supplemented with 10% dialyzed FBS was used (F0392; Sigma-Aldrich). Cas9 nickase and guide RNA expression plasmid pSpCasn(BB)-2A-Puro (48141) was purchased from Addgene. All primers were synthesized by IDT. RG1678 was purchased from MedChem Express (HY-10809). Antibodies were used according to their manufacturers’ directions. Anti-SHMT1 (12612) and SHMT2 (12762) were obtained from Cell Signaling Technologies. Anti-MTHFD2 (ab151447) and MTHFD1 (ab70203) were obtained from Abcam.

Mouse Xenografts.

All animal studies were approved and conducted under the supervision of the Princeton University Institutional Animal Care and Use Committee. For tumor growth studies, 6- to 7-wk-old female CD1/nude mice were bilaterally injected in the rear flanks with either HCT-116 control or mutant cells (1 × 10⁶ cells in 100 mL 1:1 PBS:Matrigel). Tumor growth measurements were taken biweekly using two caliper measurements [volume = 1∕2 (length × width²)]. Animals were killed when tumors reached 1,000 mm³ or if they displayed any signs of distress or morbidity. Upon termination of the study, tumors were removed and immediately frozen in liquid nitrogen for LC-MS analysis. Isolated tumors were weighed, then 50-mg tissue was disrupted using a cryomill and lysed in 1 mL ice-cold 40:40:20 acetonitrile:methanol:water. Solids were precipitated, spun down, and reextracted with 1 mL lysis buffer. Combined supernatants were dried down and resuspended in water to a concentration of 50 mg/mL (original tumor mass) before analysis by LC-MS.

In Vitro SHMT Assay.

Full-length human cytosolic SHMT1 (residues 1–483, Uniprot ID P34896) was expressed as an N-terminal His₆-tagged protein with an integrated TEV protease site and purified in Escherichia coli using nickel capture followed by size-exclusion chromatography. Human mitochondrial SHMT2 (residues 30–504, Uniprot ID P34897) with mitochondrial leader sequence deleted was expressed as an N-terminal His₆-tagged protein and purified in E. coli using nickel capture followed by size-exclusion chromatography. In vitro activity was assayed by direct MS-based measurement of serine synthesis. Inhibitor (DMSO, 1% final concentration) was added to 384-well plates. Fifteen microliters of substrate solution [25 mM TEA (pH 7.5), 500 μM meTHF, 1,800 μM glycine] was added to each inhibitor-containing well. Reaction was started upon addition of 15-μL enzyme solution [25 mM TEA (pH 7.5) and 10 nM SHMT] for a final volume of 30 μL and incubated at room temperature for 60 min before quenching with 30 μL of 10% trichloroacetic acid in acetonitrile (ACN). Serine production was quantified by MS (Rapidfire 365; Agilent Technologies).

X-Ray Crystallography and Structure Determination.

For crystallography studies, SHMT2 was expressed in E. coli by isfopropyl-β-d-thiogalactopyranoside (IPTG) induction at 16 °C overnight. Protein was first purified from crude lysate by Ni-affinity chromatography before dialysis in lysis buffer followed by TEV protease digestion (Invitrogen) overnight. The lysate was loaded onto a second Ni-affinity column, eluted, and finally purified by size-exclusion chromatography (Superdex 200 column; GE Healthcare) and assessed for purity (>95%) by SDS/PAGE analysis with Coomassie staining. SHMT2 was cocrystalized with 1.5 mM racemic 2 at 4 °C in 0.1 M Tris pH 8.5, 17% (wt/vol) PEG 20000 using the hanging-drop method. Diffraction data were collected at beamline 22-ID-D of SER-CAT at the Advanced Photon Source, where crystals diffracted to 2.5 Å, and were processed using XDS (Table S1). The structure was solved by molecular replacement using Phaser with SHMT2 structure (PDB ID code 4PVF) as an initial search model (34). Repeated cycles of manual building/rebuilding using Coot were alternated with rounds of refinement using Refmac5. Coordinates, parameter files, and molecular topology of 2 were generated by PRODRG.

Isotopic Labeling.

Isotopic tracers were purchased from the following sources: [U-¹³C]serine, [U-¹³C]glycine (Cambridge Isotope Laboratories); [¹³C]formate (Sigma-Aldrich). Serine and glycine isotopically labeled media was prepared from amino acid free DMEM or RPMI base (US Biological 9800–13/9010–10) and supplemented with dialyzed FBS.

Inhibitor Metabolism Studies.

HCT-116 cells were plated in 6-cm tissue culture dishes in DMEM supplemented with 10% dialyzed FBS. For inhibitor assays, cells were cotreated with inhibitor and labeled media for 24 h. Serine isotope label scrambling and consumption was assayed 24 h after media change by media sampling. For all other cellular metabolites, fresh media with inhibitor was added 2 h before extraction (24-h total exposure). Metabolism was quenched and metabolites were extracted by aspirating media and immediately adding 800 μL −80 °C 80:20 methanol:water extraction solution. The resulting mixture was incubated on dry ice, scraped, collected into a microfuge tube, and centrifuged at 5,000 × g for 5 min. Insoluble pellets were reextracted with 250-μL extraction solution. The supernatants were combined and dried under N₂, and finally resuspended in water to a dilution of 50 times the packed cell volume. Samples were analyzed by reversed-phase ion-pairing chromatography coupled with negative-mode electrospray-ionization high-resolution MS on a stand-alone Orbitrap (ThermoFisher Exactive) (35). For glycine labeling into suspension cell lines, 2 × 10⁶ cells were cultured with appropriate labeled RPMI for 24 h at a starting density of 5 × 10⁵/mL. To harvest, cells were pelleted by centrifugation (1,500 × g for 5 min), washed with 5 mL PBS, pelleted, and then quenched with 500 μL −80 °C 80:20 methanol:water extraction solution before downstream processing. For glycine labeling studies, HCT-116 and LN-229 cells were cultured in RPMI. For drug treatment experiments in suspension cells, SHIN1 was added to 2 mL of media containing 2 × 10⁶ cells and incubated for 72 h before harvesting as above.

Flow Cytometry.

Apoptosis was detected using FITC-conjugated Annexin V staining followed by quantification by flow cytometry (BMS500FI-300; eBioscience). Su-DHL-4 cells were seeded in 24-well plates at a density of 4 × 10⁵ cells/mL in RPMI and treated with inhibitor [(+)-SHIN1, 5 µM], or etoposide (0.33 µM) for 24 h. Cells were then harvested by centrifugation, washed with 1× PBS, 1× 200 µL binding buffer, and incubated with 10 µL of FITC-conjugated Annexin V. Cells were then washed again with 200 µL binding buffer and incubated with 5 μL of propidium iodide, and analyzed with an LSRII Flow Cytometer (BD Biosciences) at an excitation of 488 for both FITC-Annexin V.

Proliferation Assays.

For proliferation assays, 2.5 × 10³ cells were plated in each well of a 96-well plate in 100 µL DMEM supplemented with 10% dialyzed FBS. For drug treatment assays, cells were plated in media and allowed to adhere overnight. Fresh media containing inhibitor (0.1% DMSO final concentration) was added the next day and cells growth measured thereafter as fluorescence intensity using a Synergy HT plate reader (BioTek Instruments). For suspension cell lines, cell number was counted directly using Trypan blue and the Countess system (Invitrogen). For growth in dropout media, cells were pelleted by centrifugation, rinsed once with PBS, pelleted, and brought up in the drop-out media.

Cell Screen.

Cell screening was performed by Shanghai Chempartners. Complete screen data are reported in Table S2. The 298 human cancer cell lines were originally obtained from international repositories [American Type Culture Collection (ATCC), Deutsche Sammlung von Mikroorganismem und Zellkulturnen (DSMZ), Japanese Collection of Research Bioresources (JCRB), RIKEN]. Cells were subcultured in their specific recommended growth medium. For each cell line tested, cells were seeded in 96-well opaque-walled clear-bottomed plates at a previously established density (1,500–18,000 per well) in 100 μL growth media and incubated overnight. The following day, compound 2 was dosed in duplicate as a twofold serial dilution from 50 µM to 98 nM in 0.5% DMSO (all concentrations final). Cells were incubated with compound for 96 h. Growth was assayed at the end of the experiment by quantification of cellular ATP content using the CellTiter-Glo assay (Promega). For the targeted formate rescue screen, 37 cell lines of hematological origin were plated as described above. Compound 3 was dosed in a twofold dilution series from 50 µM to 98 nM with and without 1 mM sodium formate. All measurements were performed in duplicate. After 96 h, cells were treated with 1 µM Calcein AM for 30 min at room temperature and imaged by cytometry using the Acumen eX3 laser scanning image cytometer (TTP Labtech).

Sample Preparation and Library Generation for Ribosome Profiling.

Ribosome profiling libraries were generated using the Illumina TruSeq Ribo profile (Mammalian) kit (RPHMR12126; Illumina Inc.). Su-DHL-4 cells were treated with (+)-SHIN1 (5 µM) or DMSO for 48 h before lysis. 1.5–2 × 10⁷ cells were harvested by rapid centrifugation (1 min at 3,000 × g). Pellets were immediately placed into liquid N₂ and allowed to thaw in 500 µL lysis buffer containing 0.1 mg/mL cyclohexamide. After nuclease treatment, samples were purified by sucrose cushion ultracentrifugation. Ribosome footprints were purified and depleted of rRNA with Illumina RiboZero Gold kit (MRZG12324). The mRNA fragments were then end repaired, ligated to a 3′-linker, reverse-transcribed, and amplified following the manufacturer’s instructions. Purified libraries were sequenced on a HiSeq2500 (Illumina) in rapid mode.

Sequence alignment and ribosome A-site mapping.

The sequencing results were uploaded to GALAXY for data processing and mapping. The Illumina adaptor sequence was first trimmed from the raw reads and processed reads were aligned using BWA against the Homo sapiens hg38 assembly. The python package plastid was used to assign each aligned read a codon corresponding to the ribosomal A-site on the transcript using a 15-nt offset from the 5′ end (36). Only reads with lengths ranging from 28 to 30 nt were used.

Calculation of codon occupancy.

Only transcripts longer than 100 codons with transcripts per million higher than 15 were used. For each transcript, 10 codons from each end were discarded to remove signal from initiating and terminating ribosomes. Codon occupancy is defined as the ratio between the measured and the expected counts for a given codon. Occupancy is first generated for each codon in every gene and the average of all genes is reported.

Chemical Synthesis Methods

Commercially available starting materials were used without further purification. ¹H and ¹³C NMR spectra were recorded on Varian or Bruker 500-, 400-, or 300-MHz spectrometers and referenced to solvent peaks. Coupling constants (J) are reported in hertz, chemical shifts are reported in δ (ppm) as either s (singlet), d (doublet), t (triplet), q (quartet), quin (quintet), dd (doublet of doublets), or m (multiplet).

Cyclobutyl(3,5-dichlorophenyl)methanone (5): To a mixture of N,O-dimethylhydroxylammonium chloride (19.2 g, 200 mmol, 1.00 eq) and pyridine (31.6 g, 400 mmol, 2.00 eq) in DCM (800 mL) at 0 °C was added cyclobutanecarobnyl chloride (23.6 g, 200 mmol, 1.00 eq). The mixture was stirred at 20 °C for 12 h. Then the mixture was diluted with H₂O (400 mL) and separated. The organic phase was washed with HCl (200 mL ×2, 2N), brine (200 mL ×2), dried over Na₂SO₄, filtered, and concentrated to give 60 g of crude product. The crude product was purified by distillation (oil pump, 50 °C) to give compound 4 (20 g, 70%) as colorless oil and used as purified in the next step. Chloro(isopropyl)magnesium (771 mg, 7.50 mmol, 0.75 eq) was added n-BuLi (961 mg, 15.00 mmol, 1.50 eq) dropwise at −25 °C. After a white solid precipitated, the reaction mixture was diluted with 10 mL dry THF at −25 °C and turned yellow. Then 1-bromo-3,5-dichlorobenzene (2.25 g, 10.00 mmol, 1.00 eq) was added in the reaction mixture at −25 °C and the mixture was stirred for 30 min. To the above reaction mixture was added a solution of compound 4 (1.43 g, 10.00 mmol, 1.00 eq) in 20 mL dry THF at −25 °C then stirred at 16 °C for 20 min to give yellow liquid. The reaction mixture was poured into 30 mL water and was extracted with ethyl acetate (30 mL 2×). The combined organic layers were washed with brine (30 mL 3×), dried over Na₂SO₄, filtered, and concentrated in vacuum. The residue was purified by MPLC chromatography (petroleum ether:ethyl acetate = 1/0 ∼ 20/1) to give compound 5 (1.00 g, 4.38 mmol, 43.82% yield) as yellow liquid.

6-amino-4-cyclobutyl-4-(3,5-dichlorophenyl)-3-methyl-1,4-dihydropyrano[2,3-c]pyrazole-5-carbonitrile (1): To a mixture of compound 5 (500.00 mg, 2.19 mmol, 1.00 eq), propanedinitrile (289.47 mg, 4.38 mmol, 2.00 eq) and pyridine (693.22 mg, 8.76 mmol, 4.00 eq) in CHCl₃ (15.00 mL) were added TiCl₄ (2.40 g, 12.64 mmol, 5.77 eq) dropwise at 5 °C, then the mixture was stirred at 18 °C for 1 h until completion as assayed by thin-layer chromatography (TLC). The mixture was poured into 2N HCl (20 mL) and extracted with DCM (30 mL 2×). The organic phase was washed with brine (30 mL 3×), dried over Na₂SO₄, and concentrated. The residue was used in the next step directly. To a mixture of 6 (200.00 mg, 721.63 µmol, 1.00 eq) and 3-methyl-1,4-dihydropyrazol-5-one (70.79 mg, 721.63 µmol, 1.00 eq) in 1:1 EtOH (1.20 mL) and dioxane (1.20 mL) was added piperdine (6.14 mg, 72.16 µmol, 0.10 eq) under N₂. The mixture was stirred under microwave at 65 °C for 3 h until completion by TLC and concentrated under vacuum before purification by prep TLC to give compound 1 as a yellow solid (37.30 mg, 96 µmol, 13.3% yield). MS (ES+) calc m/z for C₁₈H₁₆C_l2N₄O (M+H)⁺¹: 375.078, found: 375.2; ¹H NMR (DMSO-d₆, 400 MHz) δ 1.58 (d, J = 8.38 Hz, 2H), 1.65 (s, 3H), 1.80 (d, J = 5.29 Hz, 3H), 1.96 (d, J = 5.73 Hz, 1H), 3.22 (m, 1H), 7.02 (s, 2H), 7.19 (d, J = 1.76 Hz, 2H), 7.50 (t, J = 1.76 Hz, 1H), 12.21 (s, 1H); ¹³C NMR (DMSO-d₆, 500 MHz) δ 10.54, 16.79, 23.80, 24.15, 43.31, 45.15, 58.99, 98.86, 120.51, 126.34, 126.63, 133.82, 135.06, 149.91, 155.20, 162.00.

2-methyl-1-(3-(pyrrolidin-1-yl)-5-(trifluoromethyl)phenyl)propan-1-one (8): Into a 50-mL three-necked round-bottom flask was placed 1,3-dibromo-5-(trifluoromethyl)benzene (1.2 g, 3.95 mmol, 1.00 equivalent), oxolane (10 mL). This was followed by the addition of butyllithium (1.6 mL, 1.10 eq, 2.5 M n-BuLi in hexane). To this was added N-methoxy-N,2-dimethylpropanamide (420 mg, 3.20 mmol, 1.00 eq). The resulting solution was stirred for 1 h at −78 °C in a liquid nitrogen bath. The reaction was then quenched by the addition of 25 mL of saturated NH₄Cl aq. The resulting solution was extracted with 3 × 25 mL of ethyl acetate and the organic layers combined. The resulting mixture was washed with 1 × 25 mL of brine before drying over anhydrous sodium sulfate before concentrating under vacuum to give 0.45 g (39%) of 7 as a yellow solid. Together in a 50-mL flask, 7 (450 mg, 1.52 mmol, 1.00 eq), toluene (20 mL), pyrrolidine (140 mg, 1.97 mmol, 1.00 eq), t-BuONa (237 mg, 2.47 mmol, 2.00 eq), Pd₂(dba)₃ (50 mg, 0.05 mmol, 0.10 eq), and Xantphos (0.1 g, 0.20 eq) were stirred for 16 h at 110 °C. The reaction was quenched by the addition of 50 mL of water. The resulting solution was extracted with 3 × 25 mL of ethyl acetate and the organic layers combined. The resulting mixture was washed with 4 × 50 mL of brine and dried over anhydrous sodium sulfate before concentrating under vacuum. The residue was applied onto a silica gel column with ethyl acetate/petroleum ether (1:20). This resulted in 0.27 g (62%) of 8 as a white solid. ¹H NMR (Chloroform-d, 300 MHz) δ 1.25 (d, J = 6.8 Hz, 6H), 2.02–2.16 (m, 4H), 3.33–3.45 (m, 4H), 3.56 (p, J = 6.9 Hz, 1H), 6.86–7.00 (m, 1H), 7.29 (d, J = 2.0 Hz, 1H), 7.43 (s, 1H).

6-amino-4-isopropyl-3-methyl-4-(3-(pyrrolidin-1-yl)-5-(trifluoromethyl)phenyl)-1,4-dihydropyrano[2,3-c]pyrazole-5-carbonitrile (2): Into a 10-mL sealed tube, was placed 8 (120 mg, 0.42 mmol, 1.00 eq), propanedinitrile (150 mg, 2.27 mmol, 6.00 eq), HMDS (180 mg, 1.12 mmol, 3.00 eq), and acetic acid (1 mL). The resulting solution was stirred for 16 h at 65 °C in an oil bath. The reaction was then quenched by the addition of 25 mL of water. The resulting solution was extracted with 3 × 15 mL of ethyl acetate and the organic layers combined. The resulting mixture was washed with 1 × 25 mL of brine, dried over anhydrous sodium sulfate, and concentrated under vacuum. The residue was applied onto a silica gel column with ethyl acetate/petroleum ether (1:5). This resulted in 0.07 g (50%) of 9 as a red solid. Into a 25-mL sealed tube, 9 (1.5 g, 4.50 mmol, 1.00 eq) 3-methyl-4,5-dihydro-1H-pyrazol-5-one (450 mg, 4.59 mmol, 1.00 eq), dioxane (2 mL), EtOH (2.5 mL), and piperidine (400 mg, 4.71 mmol, 1.00 eq) were stirred for 44 h at 65 °C in an oil bath. The reaction was then quenched by the addition of 30 mL of water. The resulting solution was extracted with 3 × 20 mL of ethyl acetate and the organic layers combined. The resulting mixture was washed with 1 × 45 mL of brine, dried over anhydrous sodium sulfate and concentrated under vacuum. The crude product (2.5 g) was purified by Prep-HPLC with the following conditions: Column, Xbridge Prep C18 OBD Column 19 × 15 mm 5 μm C-0013; mobile phase, Phase A:Waters (10 mmol/L NH₄HCO₃) Phase B:ACN; Detector: UV 254/220. 700 mg product 2 was obtained. MS (ES+) calc m/z for C₂₂H₂₅F₃N₅O (M+H)⁺¹: 432.20, found: 432.0; ¹H NMR (400 MHz, Chloroform-d): δ 0.90 (d, J = 6.6 Hz, 3H), 1.00 (d, J = 6.6 Hz, 3H), 1.99–2.05 (m, 4H), 2.79 (p, J = 6.6 Hz, 1H), 3.24–3.36 (m, 4H), 4.72 (s, 2H), 6.64 (s, 1H), 6.75 (s, 1H), 6.82 (s, 1H).

Chiral resolution: 700 mg of racemate 2 was separated by PREP-CHIRAL-HPLC with the following conditions: Column, Chiralpak IB 4.6 × 250 mm, 5 µm HPLC Chiral-A(IB)001IB00CE-LA026, Phase, hexanes (0.3% DEA): EtOH = 70:30, Detector, UV 254/220. This resulted in 81.1 mg (4%) of (+)-2 white solid, and 88.7 mg (5%) of (−)-2 as a white solid.

Methyl 5-isobutyryl-[1,1′-biphenyl]-3-carboxylate (11): Into a 15-mL round-bottom flask, 3-(2-methylpropanoyl)-5-phenylbenzonitrile (1.5 g, 6.02 mmol, 1.00 eq), methanol (15 mL), sodium hydroxide (960 mg, 24.00 mmol, 4 eq), and water (15 mL) were combined and stirred overnight at 90 °C. The mixture was concentrated under vacuum. The resulting solution was diluted with 100 mL of H₂O. The solution was acidified to pH 3 with HCl (2 M). The resulting solution was extracted with 4 × 50 mL of ethyl acetate and the organic layers combined and dried in an oven under reduced pressure and concentrated under vacuum. This resulted in 800 mg (50% yield) of 3-(2-methylpropanoyl)-5-phenylbenzoic acid (10) as a white solid, which was used without further purification. Into a 100-mL round-bottom flask purged and maintained with an inert atmosphere of nitrogen, was placed 10 (600 mg, 2.24 mmol, 1.00 eq), methanol (30 mL), and thionyl chloride (528 mg, 4.45 mmol, 2 eq). The resulting solution was stirred overnight at 80 °C. The resulting mixture was concentrated under vacuum yielding 500 mg (79%) of methyl 3-(2-methylpropanoyl)-5-phenylbenzoate (11) as a yellow oil.

Methyl 5-(6-amino-5-cyano-4-isopropyl-3-methyl-1,4-dihydropyrano[2,3-c]pyrazol-4-yl)-[1,1′-biphenyl]-3-carboxylate (13): Into a 20-mL round-bottom flask, was placed 11 (400 mg, 1.42 mmol, 1.0 eq), propanedinitrile (468.5 mg, 7.09 mmol, 5.0 eq), acetic acid (4 mL), and HMDS (689.3 mg, 4.27 mmol, 3.01 equiv). The resulting solution was stirred overnight at 65 °C. The resulting mixture was concentrated under vacuum. The residue was applied onto a silica gel column with ethyl acetate/petroleum ether (1:20). This resulted in 240 mg (51%) of methyl 3-(1,1-dicyano-3-methylbut-1-en-2-yl)-5-phenylbenzoate (12) as a yellow oil. Into an 8-mL round-bottom flask purged and maintained with an inert atmosphere of nitrogen, was placed 12 (240 mg, 0.73 mmol, 1.00 eq), 3-methyl-4,5-dihydro-1H-pyrazol-5-one (86 mg, 0.88 mmol, 1.21 eq), ethanol (1.5 mL), 1,4-dioxane (1.5 mL), and piperidine (65 L). The solution was stirred overnight at 65 °C. The resulting mixture was concentrated under vacuum. This resulted in 310 mg (crude) of methyl 3-[6-amino-5-cyano-3-methyl-4-(propan-2-yl)-1H,4H-pyrano[2,3-c]pyrazol-4-yl]-5-phenylbenzoate (13) as a brown oil. MS (ES+) calc m/z for C₂₅H₂₅N₄O₃ (M+H)⁺¹: 429.2, found: 429.0; ¹H NMR (MeOH-d₄, 300 MHz) δ 0.89 (d, J = 6.6 Hz, 3H), 1.01 (d, J = 6.6 Hz, 3H), 1.21 (t, J = 7.2 Hz, 1H), 1.80 (s, 3H), 2.85–2.94 (m, 1H), 3.90 (s, 3H), 4.08 (t, J = 7.5, 1H), 7.32–7.41 (m, 1H), 7.45 (d, J = 7.2 Hz, 2H), 7.57 (d, J = 7.5 Hz, 2H), 7.83 (s, 1H), 8.00 (s, 1H), 8.10 (s, 1H).

6-amino-4-(5-(hydroxymethyl)-[1,1′-biphenyl]-3-yl)-4-isopropyl-3-methyl-1,4-dihydropyrano[2,3-c]pyrazole-5-carbonitrile (3): 13 (100 mg, 0.23 mmol, 1.00 eq) and methanol (10 mL) were placed into a 50-mL round-bottom flask purged and maintained with an inert atmosphere of nitrogen. This was followed by the addition of NaBH₄ (30 mg, 0.79 mmol, 3.00 eq) by syringe at 0 °C for 3 min. The resulting solution was stirred for 1 h at room temperature. The reaction was then quenched by the addition of 10 mL of NH₄Cl (aq). The resulting mixture was concentrated under vacuum. The resulting solution was extracted with 3 × 10 mL of ethyl acetate and the organic layers combined and concentrated under vacuum. The residue was applied onto a silica gel column with ethyl acetate/petroleum ether (1:2). This resulted in 25.3 mg (27%) of 3 as a white solid. MS (ES+) calc m/z for C₂₄H₂₅N₄O₂ (M+H)⁺¹: 401.2, found: 401.1; ¹H NMR (DMSO-d₆, 500 MHz) δ 0.80 (d, J = 6.5 Hz, 3H), 0.92 (d, J = 6.6 Hz, 3H), 1.78 (s, 3H), 2.83 (quin, J = 6.6 Hz, 1H), 4.54 (d, J = 5.7 Hz, 2H), 5.25 (t, J = 5.8 Hz,1H), 5.76 (s, 1H), 6.82 (s, 2H), 7.25 (d, J = 1.7 Hz, 1H), 7.33–7.40 (m, 1H), 7.43–7.51 (m, 4H), 7.55–7.63 (m, 2H). ¹³C NMR (DMSO-d₆, 500 MHz): δ 39.54, 39.71, 39.87, 39.95, 40.04, 40.09, 40.20, 40.26, 40.37, 40.43, 40.54, 121.26, 123.38, 125.13, 126.12, 127.14, 127.82, 129.46, 139.85, 141.20, 143.35, 144.91, 162.20.

Chiral resolution: 20.25 mg of racemate 3 was separated by PREP-CHIRAL-HPLC with the following conditions: Column: Chiralpak IA 2 × 250 mm, Gradient: 30% isopropanol/CO₂, 100 bar, 60 mL/min. Detector, UV 220. This resulted in 9.6 mg of (−)-3 (inactive enantiomer, [α]_D^20.3 = −119 (c 0.13, isopropanol)) as a white solid, and 7.9 mg of (+)-3 (active enantiomer) as a white solid.