Human mutations in methylenetetrahydrofolate dehydrogenase 1 impair nuclear de novo thymidylate biosynthesis

Abstract

An inborn error of metabolism associated with mutations in the human methylenetetrahydrofolate dehydrogenase 1 (MTHFD1) gene has been identified. The proband presented with SCID, megaloblastic anemia, and neurologic abnormalities, but the causal metabolic impairment is unknown. SCID has been associated with impaired purine nucleotide metabolism, whereas megaloblastic anemia has been associated with impaired de novo thymidylate (dTMP) biosynthesis. MTHFD1 functions to condense formate with tetrahydrofolate and serves as the primary entry point of single carbons into folate-dependent one-carbon metabolism in the cytosol. In this study, we examined the impact of MTHFD1 loss of function on folate-dependent purine, dTMP, and methionine biosynthesis in fibroblasts from the proband with MTHFD1 deficiency. The flux of formate incorporation into methionine and dTMP was decreased by 90% and 50%, respectively, whereas formate flux through de novo purine biosynthesis was unaffected. Patient fibroblasts exhibited enriched MTHFD1 in the nucleus, elevated uracil in DNA, lower rates of de novo dTMP synthesis, and increased salvage pathway dTMP biosynthesis relative to control fibroblasts. These results provide evidence that impaired nuclear de novo dTMP biosynthesis can lead to both megaloblastic anemia and SCID in MTHFD1 deficiency.

Keywords: MTHFD1; SCID; folate; megaloblastic anemia; thymidylate.

Fig. 1.

De novo and salvage pathway synthesis of dTMP in the nucleus. 1C metabolism is required for the synthesis of purines, dTMP, and methionine. Formate is a major source of 1C units, which are generated in the mitochondria. Mitochondrial-derived formate can enter the cytoplasm and function as a 1C unit for folate metabolism through the activity of MTHFD1. At S phase, the enzymes of the dTMP synthesis pathway undergo SUMO-dependent translocation to the nucleus. The 1C is labeled in bold. Inset shows the thymidylate synthesis cycle, which involves the enzymes MTHFD1, SHMT1, SHMT2α, TYMS, and DHFR as well as the salvage pathway. AdoHcy, S-adenosylhomocysteine; AICAR Tfase, aminoimidazolecarboxamide ribonucleotide transformylase; AdoMet, S-adenosylmethionine; DHF, dihydrofolate; DHFR, dihydrofolate reductase; dUMP, deoxyuridine monophosphate; GAR Tfase, glycinamide ribonucleotide transformylase; MTHFD1 (C), MTHFD1 C activity; MTHFD1 (D), MTHFD1 D activity; MTHFD1 (S), MTHFD1 synthase activity; MTR, methionine synthase; MTHFR, 5,10-methylenetetrahydrofolate reductase; SUMO, small ubiquitin-like modifier.

Fig. 2.

MTHFD1 protein level is decreased in patient fibroblasts but preferentially enriched in the nucleus. (A) MTHFD1 protein levels in whole-cell lysates from control (MCH058 and MCH064) and patient (WG3607) fibroblasts. (B) MTHFD1 nuclear localization in control (MCH064) and patient (WG3607) fibroblasts (green) and DNA stain Draq5 (blue) using confocal microscopy. (C and D) Effect of S-phase arrest [1 mM hydroxyurea (HU)] and impaired MTHFD1 nuclear export [20 nM Leptomycin B (LpmB)] on MTHFD1 nuclear localization in control (MCH058) and patient (WG3607) fibroblasts. The ratio of the cytosolic to nuclear MTHFD1 signal intensity (number of cells per condition was n > 30) is shown as mean and SD. (E) FACS analysis of control (MCH058) and patient (WG3607) fibroblasts after HU treatment. (F) Histogram showing distribution of nuclear-to-cytosolic ratios in control (MCH064 and MCH058) and patient (WG3607) fibroblasts. (G and H) Effect of MTHFD1 R173C mutation on MTHFD1 subcellular localization. MTHFD1-GFP and MTHFD1-R173C-GFP fusion proteins were expressed in HeLa cells with or without HU and imaged using confocal microscopy. The cytosolic-to-nuclear MTHFD1-GFP and MTHFD1-R173C-GFP mutant signal intensity ratios are shown as means and SDs (number of cells per condition was n > 30). The statistical significance is represented as follows: P > 0.05 was nonsignificant (NS) and ***0.001 > P. async, Asynchronous cells; Nuc/Cyt, nuclear to cytosolic; NT, no treatment.

Fig. 3.

dTMP synthesis and DNA damage in patient and matched control fibroblasts. (A) dU suppression assay in control (MCH058 and MCH064) and adenosine deaminase-deficient fibroblasts (WG0549 and WG0290). Results are shown as the ratio of decays per minute (DPM) from [¹⁴C]-deoxyuridine to DPM from [³H]-thymidine in DNA, with n = 3 biological replicates per cell line. Data are shown as means ± SDs and were analyzed using Student’s t test with Bonferroni correction for multiple comparisons. There were no significant differences. (B) [³H]-thymidine incorporation (DPM) in DNA as a function of exogenous glycine concentration. Patient and control fibroblasts were cultured for four doublings in DMEM containing either 2 or 5 mM glycine. Data are shown as means ± SDs of three biological replicates and analyzed by Student’s t test with Bonferroni correction for multiple comparisons. (C) TK1 levels as a function of exogenous glycine in patient and matched control fibroblast cell lysates. Bands were quantified using the ratio of TK1-to-GAPDH expression, and densitometry was performed using ImageJ software. (D and E) MTHFD1-deficient patient fibroblasts (WG3607) exhibit elevated γH2Ax-positive foci per nuclei compared with (D) control (MCH058) fibroblasts and (E) control (MCH058 and MCH064) and adenosine deaminase-deficient fibroblasts (WG0549 and WG0290). DNA damage was quantified as γH2AX-positive area percentage for each cell imaged above a set threshold and presented as a mean of all cells imaged per cell line ± SEM. Number of cells imaged: MCH058, n = 914; MCH064, n = 758; WG0290, n = 653; WG0549, n = 633; WG3607, n = 1,049. The statistical significance was assessed by t test with Bonferroni correction for multiple comparisons and is represented as follows: P > 0.05 was nonsignificant (NS), *0.05 > P > 0.01, and ***0.001 > P. DIC, differential image contrast. (F) MTHFD1-deficient patient fibroblasts display more DNA damage when folate-depleted (P = 0.02). MTHFD1-deficient patient fibroblasts were cultured for five doublings in defined media supplemented (n = 286) or not (n = 400) with folic acid.