Development of Potent and Selective Inhibitors of Methylenetetrahydrofolate Dehydrogenase 2 for Targeting Acute Myeloid Leukemia: SAR, Structural Insights, and Biological Characterization

Abstract

Methylenetetrahydrofolate dehydrogenase/cyclohydrolase 2 (MTHFD2), a pivotal mitochondrial enzyme in one-carbon metabolism, is significantly upregulated in various cancers but minimally expressed in normal proliferating cells. In contrast, MTHFD1, which performs similar functions, is predominantly expressed in normal cells. Therefore, targeting MTHFD2 with selective inhibitors holds promise for a broader therapeutic window with reduced toxicity and fewer side effects. This study identified selective 2,4-diamino-6-oxo-1,6-dihydropyrimidin-5-yl ureido-based derivatives through systematic chemical modifications and SAR studies. Structural biology investigations revealed substitutions in the phenyl ring and tail region modulate potency and selectivity toward MTHFD2. Additionally, a comprehensive cell screening platform revealed acute myeloid leukemia cells with FLT3 internal tandem duplication mutations are particularly sensitive to these inhibitors. Furthermore, synergistic effects were observed when combining potential compounds with Alimta. Compound 16e emerged as a leading candidate, demonstrating superior inhibition and selectivity for MTHFD2, favorable pharmacokinetics, and potent antitumor efficacy in MOLM-14 xenograft models.

Figure 1

(A) Catalytic reaction of MTHFD2. (B) Structures of MTHFD2 inhibitors.

Scheme 1. Reagents and Conditions: (a) 1 N NaOH, CH₃CN, 25 °C; (b) 1 N NaOH, H₂O, 25 °C for Ester Compounds 12, 14a, 14b and 15a; (c) EDC HCl, HOSu, Dry DMSO, 25 °C

Scheme 2. Reagents and Conditions: (a) bis(Trichloromethyl)carbonate, CH₂Cl₂, DIPEA, 15 °C; (b) Compound 1, 1 N NaOH, CH₃CN, 25 °C; (c) TFA, 25 °C; (d) SOCl₂, Pyridine, CH₂Cl₂, 55 °C; (e) 1 N NaOH, H₂O, 25 °C; (f) l-Glutamic Acid Diethyl Ester Hydrochloride, EDC HCl, HOSu, dry DMSO, 25 °C

Figure 2

Crystal structure of MTHFD2 (light teal) in complex with 16a (yellow) and NAD⁺ (omitted for clarity) (PDB: 9ISC). Hydrogen-bond with protein, water -bridged and π–π interactions are shown in red, pale-cyan and green dashed lines, respectively. Water molecules are shown as spheres (cyan). For clarity, the NAD⁺ is omitted in the figure.

Figure 3

Superimposition of the crystal structure of MTHFD2/16a (PDB 9ISC, light teal/yellow) on MTHFD2/LY374571 (PDB 9IS9, gray/pink). For clarity, the NAD⁺ is omitted in the figure. (A) Interactions of the 1H-tetrazole moiety of 16a with MTHFD2. Hydrogen-bond with protein and water -bridged interactions are shown in yellow and pale-cyan dashed lines, respectively. The water molecule is shown as a sphere (cyan). The hydrogen bond interaction of the γ-carboxylic acid moiety of LY374571 with MTHFD2 is shown in violet. The electron density map for the side chain of R278 is unclear; therefore, the side chain of R278 has been excluded from the MTHFD2/LY374571 structure. (B) The surface view around the tail part of 16a.

Figure 4

Superimposition of the crystal structure of MTHFD1/16e/NADP⁺ (PDB 9ISL, wheat/darksalmon/yellow-orange) on MTHFD1/LY374571/NADP⁺ (PDB 9ISE, forest/cyan/slate) or MTHFD1/16g/NADP⁺ (PDB 9ISR, deep purple/magenta). Hydrogen-bonds are shown in red dashed lines. (A) The presence of the chloride atom on the phenyl ring of compound 16e induced the rotation of the nicotinamide group of NADP⁺. Additionally, the introduction of the chloride atom resulted in rotational adjustments of the phenyl ring and caused the γ-carboxylic acid to move away from β strand h1. (B) The shift of the γ-carboxylic acid in 16e results in the loss or decrease of hydrophobic and van der Waals interactions with Leu51, Tyr52, and Tyr240. (C) The 1H-tetrazole moiety of 16g shifted toward β strand h1 and was hydrogen bonded with Tyr240. For clarity, the NADP⁺ is omitted in the figure.

Figure 5

(A,B) Superimposition of the crystal structures of MTHFD2/16g/NAD⁺ (PDB 9IT6, light blue/light orange) with MTHFD1/16g/NADP⁺ (PDB 9ISR, deep purple/magenta) (C) Superimposition of the crystal structures of MTHFD2/16e/NAD⁺ (PDB 9IT3, light pink/orange) with MTHFD1/16e/NADP⁺ (PDB 9ISL, wheat/darksalmon). NAD⁺ or NADP⁺ is omitted for clarity. Hydrogen bonds between compound and MTHFD2 are shown as red dashed lines, while those with MTHFD1 are depicted as deep teal dashed lines. Residues in MTHFD1 are indicated in parentheses.

Figure 6

Using LY374571 as a reference compound to identify optimal cell models for evaluating MTHFD2 inhibitors. (A) The efficacy of LY374571 on cell growth inhibition in multiple hematological tumor cells and solid tumor cells. Cell viability was examined by MTT assay (hematological tumor cells) and methylene blue staining assay (solid tumor cells), respectively, after drug treatment for 72 h. In the experiment, the number of cells in the control group is set to 100%. The number of cells treated with 20 μM LY374571 is expressed as a percentage of cell viability relative to the control group. (B) Comparison of the efficacy of LY374571 on cell growth inhibition with increasing drug concentration for 72 h in MOLM-13 and MOLM-14 cells (upper panel) and HL-60 and CCRF-CEM (lower panel). Data was from three independent experiments and shown as means ± SD.

Figure 7

Investigation of combination benefits of Alimta with MTHFD2 inhibitors in MOLM-14 cells. MOLM-14 cells were treated with Alimta and MTHFD2 inhibitors (16a, 16e, or LY374571) alone or in combination for 72 h. Cell viability was evaluated by MTT assay and displayed by heatmap (A,D,G). Cell viability is expressed as a percentage relative to the control group, with red indicating higher viability and green indicating lower viability. Bar graphs illustrate cell viability after treatment with Alimta and compound 16a (B), 16e (E), or LY374571 (H). The results are presented as mean ± standard deviation (SD), and statistical significance is indicated by an asterisk (*), p < 0.05. combination index (CI) plot for Alimta with compound 16a (C), 16e (F), or LY374571 (I). The CI values were calculated by CalCusyn software to evaluate the synergistic effect of the two compounds. A CI value less than 1 indicates synergy, a CI value equal to 1 indicates an additive effect, and a CI value greater than 1 indicates antagonism.

Figure 8

In vivo antitumor efficacy of MTHFD2 inhibitors in MOLM-14 xenograft mouse model. MOLM-14 cells were subcutaneously implanted into NOD SCID mice and the tumor-bearing mice were treated with LY374571 (10 mg/kg, intravenously injected, QD, 2 cycles of 5-on/2-off). Tumor volume (A) and body weight (B) were measured twice weekly. MOLM-14 cells were subcutaneously implanted into NOD SCID mice and the tumor-bearing mice were treated with the compound 16e (15 mg/kg, intravenously injected, QD). Tumor volume (C) and body weight (D) were measured twice weekly. (E) Tumor volumes of mice treated with vehicle or compound 16e on day 14 after drug administration. Each point represents the tumor volume of an individual mouse (n = 7 per group). The data are presented as mean ± standard deviation (SD). Statistical significance between the vehicle-treated group and the 16e-treated group is indicated by double asterisks (**), p < 0.01.