The negative effect of G1958A polymorphism on MTHFD1 protein stability and HCC growth

Abstract

Purpose

Methylenetetrahydrofolate dehydrogenase (MTHFD1), a key enzyme on the folate pathway, has been implicated in the tumor development of distinct types of cancers. The single nucleotide polymorphism (SNP) of 1958G > A mutation in the coding region of MTHFD1 (arginine 653 is mutated into glutamine) has been detected in a significant proportion of clinical samples of hepatocellular carcinoma (HCC).

Methods 

Hepatoma cell lines, 97H and Hep3B were used. The expression of MTHFD1 and SNP mutation protein was determined by immunoblotting analysis. The protein ubiquitination of MTHFD1 was detected by immunoprecipitation analysis. The post-translational modification sites and interacting proteins of MTHFD1 in the presence of G1958A SNP were identified by mass spectrometry. Metabolic flux analysis was used to detect the synthesis of relevant metabolites sourced from serine isotope.

Results

The present study showed G1958A SNP of MTHFD1, encoding MTHFD1 R653Q, was associated with the attenuated protein stability caused by ubiquitination-mediated protein degradation. Mechanistically, MTHFD1 R653Q displayed an enhanced binding to the E3 ligase TRIM21, which was responsible for the augmented ubiquitination, and MTHFD1 K504 was identified to be the primary ubiquitination site. The subsequent metabolite analysis revealed MTHFD1 R653Q resulted in the repressed flux of serine-derived methyl group into metabolite precursors for purine synthesis, and the compromised purine synthesis was demonstrated to be responsible for the impeded growth capability in MTHFD1 R653Q-expressing cells. Moreover, the suppressive effect of MTHFD1 R653Q expression in tumorigenesis was verified by xenograft analysis, and the relationship between MTHFD1 G1958A SNP and its protein levels was revealed in clinical human liver cancer specimens.

Conclusion

Our results uncovered an unidentified mechanism underlying of the impact of G1958A SNP on MTHFD1 protein stability and tumor metabolism in HCC. which provides a molecular basis for the according clinical management when considering MTHFD1 as a therapeutic target.

Introduction

Hepatocellular carcinoma (HCC) is one of leading causes of cancer death, which is associated with a fast increasing mortality and a low 5-year survival rate [1]. Although several therapies have been adopted to treat HCC patients such as resection, radiotherapy, liver transplantation and combination therapies [2–4], the lack of miraculous clinical responses for advanced-stage HCC urges us to search more effective drug targets.

Folate-mediated one-carbon metabolism (FOCM) is fundamentally required for fueling methylation reactions and synthesis of dTMP or purine. The physiological significance of dysregulation of FOCM in tumor development has raised more and more attention in recent years, among which the concrete effects of the relevant metabolic enzymes including mitochondrial serine hydroxymethyl transferase (SHMT2) and mitochondrial methylenetetrahydrofolate dehydrogenase (MTHFD2) have been extensively investigated and demonstrated to contribute to tumor progression [5, 6], FOCM as well as the metabolites that engage in the process recently arouses broad concern in cancer target therapy [7]. L-Serine, as the principal one-carbon (1C) source for FOCM, has been to be preferentially to be catabolized through mitochondrial route, while it’s reported that the cytosolic 1C pathway is adequate to provide support to cell growth when SHMT2 or MTHFD2-mediated mitochondrial folate metabolism is interrupted [8, 9]. As one of primary enzymes involved in cytosolic pathway of FOCM, MTHFD1 is recognized a trifunctional enzyme that exhibits methylenetetrahydrofolate dehydrogenase, methenyltetrahydrofolate cyclohydrolase and formyltetrahydrofolate synthetase activities [10]. The previous study utilizing short palindromic repeats (CRISPR) against folate enzymes indicated 10-formyl-THF produced in the cytosol is requisite for de novo purine synthesis, in which MTHFD1 is critically involved [9]. Physiologically, the essential role of synthesis activity of MTHFD1 has been demonstrated in tumorigenesis of colorectal cancer [11], and high expression of MTHFD1 was reported to be correlated with adverse prognosis of HCC [12]. In addition, MTHFD1 single nucleotide polymorphisms (SNPs) has been implicated in human susceptibility to various diseases [13, 14]. Of note, the G1958A SNP of MTHFD1 was reported to be associated with a decreased risk of acute lymphoblastic leukemia (ALL) and other cancers including HCC [15], and there is evidence that the suppressive role of MTHFD1 G1958A in tumorigenesis would be relevant to the reduced activity of encoded protein [14]. However, the mechanism underlying the effect of MTHFD1 G1958A SNP on tumor growth remains to be further clarified.

In the present study, it was revealed G1958A polymorphism (R653Q) is negatively associated with MTHFD1 protein level in tumor tissues from clinical HCC samples (Fig. 2a). Mechanistically, we showed that mutation of R653Q confers a higher sensitivity of MTHFD1 to be ubiquitinated by TRIM21, in which MTHD1 is identified as a critical modification site. As a consequence, the increased entry of MTHFD1 R653Q into proteasomal degradation in turn compromises purine synthesis and cell growth of HCC cells.

Fig. 2.

G1958A polymorphism is inversely related to MTHFD1 protein level and cell growth. a The cDNA samples acquired from 13 hepatocellular cancer clinical samples indicated that 4 samples carried the 1958G > A SNP. b The expressions of MTHFD1 protein extracted from LIHC cancerous samples were tested by western blotting. c The mRNA level of MTHFD1 in clinical HCC samples was determined by Real-time PCR. d 97H cells expressing with a vector for control shRNA or MTHFD1 shRNA and reconstituted with expression of rMTHFD1 WT and rMTHFD1 R653Q. e and f 1C provider 10-formylTHF and purine synthesis precursor IMP (M1, M2) decreased in rMTHFD1 R653Q-expressing cells. g and h TMP (M2) remarkably increased in rMTHFD1 R653Q-expressing cells compared to WT cells. i Intracellular IMP and TMP levels in 97H rMTHFD1 WT cell and R653Q ells, the level of IMP decreased notably. j Exogenous adenosine (100mM) substantially redeem cell growth arrest in rR653Q-expressing cells

Materials and methods

Cell culture

97H, Hep3B and HEK293T cells were maintained in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum (FBS) and incubated at 37 °C in a humidified atmosphere containing 5% CO₂. All cell lines were routinely tested for mycoplasma to ensure no contamination.

Cell proliferation assay

Cells were incubated with CCK8 (10% vol/vol) that was diluted in normal culture medium at 37 °C for 1–4h until the visual color conversion occurred. The absorbance was measured in a microplate reader at a wavelength of 450 nm.

DNA construction

The DNA constructs and mutagenesis were performed as previously described [16]. The DNA sequence encoding MTHFD1 was cloned into pLenti-puro vector with a 3 × Flag tag. The shRNAs targeting MTHFD1 were generated with primers shRNA #1 AGGATCATTGCACAGAAGA and shRNA #2 TGAAGAAGTGATCAATGCT.

RNA interfering

The siRNAs were synthesized by Sangon Biotech and the target sequences were as follows. Human MTHFD1 siRNA #1 GGACAATTTGGTTGTGGAA3; siRNA #2GGAATGCATCTCTCAGGTT.

Transfection

97H, Hep3B and 293T cells were transfected with various plasmids or siRNA, EZ-Trans (Life iLab) was used to transfect siRNA or plasmids required for lentivirus.

production, respectively.

Gene expression analysis

The total RNA from cells was isolated by using TRIzol (Thermo scientific) according to the manufacturer’s instructions. We used cDNA Reverse Transcription Kit (Takara) to produced cDNA from 1 mg total RNA. The relative level of gene expression was measured by quantitative real-time PCR with Power SYBR Green PCR Master Mix (Applied Biosystems Inc). We calculated relative mRNA levels standardized with human GAPDH level in the same samples.

Immunoprecipitation and immunoblotting

Proteins were extracted with NP-40 lysis buffer (50 mM Tris-HCl (pH 7.4), 150 mM NaCl, 1 mM EDTA, 10% glycerol and 1% NP-40) and followed by immunoprecipitation or immunoblotting. For exogenous immunoprecipitation, cell lysates were centrifuged to remove the cell debris and incubated with flag-M2 beads (Sigma) at 4 °C overnight. For endogenous immunoprecipitation, cell lysates were incubated with the indicated antibodies at 4 °C overnight, and the immunoprecipitate was incubated with protein A/G agarose beads (Santa Cruz) for 2–3h. The beads finally were boiled after extensive washing. For protein extraction from human hepatocellular carcinoma tissue samples, the tissue blocks were homogenized using a grinder, and followed by incubation with NP-40 lysis buffer. The protein concentration was determined using BCA Protein Assay Kit (Thermo Scientific). Proteins from cell lysates were separated by SDS-PAGE, transferred onto PVDF membrane (Millipore) and assessed with the indicated antibodies.

Ubiquitination assay

Cellular assays to measure MTHFD1 polyubiquitination were accomplished as previously described [17]. Cells were transfected with vectors encoding MTHFD1 and HA-Ub and treated with or without 20 µM MG132 for 6 h before the cells were collected. After lysed with NP-40 lysis buffer, the supernatant was incubated with flag-M2 beads at 4 °C overnight and followed by extensive washing. The precipitate was then released in SDS loading buffer and subjected to immunoblotting analysis.

Animal study

All animal experiments conformed to the guidelines of the Institutional Animal Care and Use Committee of Shanghai Jiaotong University. Five-week-old male nu/nu mice were randomly divided into indicated groups, and subcutaneously injected with 5 × 10⁶ gene-modified 97H cells in 100 µL of PBS. Tumor volume was measured by using length (a) and width (b) and calculated using the equation: V = ab²/2.

Statistics and reproducibility

Statistical testing was performed using the two-tailed Student’s t-test. P value < 0.05 were considered statistically significant (*P < 0.05, **P < 0.01, ***P < 0.001). All experiments were performed at least three times unless otherwise indicated.

Results

MTHFD1 supports purine synthesis and growth in HCC cells

To investigate the functional implication of MTHFD1 in hepatocellular carcinoma, MTHFD1 was depleted by the validated shRNAs in 97H and Hep3B human hepatocellular carcinoma cell lines respectively (Fig. 1a and b). Consequently, either long-term colony-formation (Fig. 1c and d) or CCK-8 assay indicated MTHFD1 depletion led to a dramatic suppression of cell viability (Fig. 1e and f). In line with these results, xenograft analysis with subcutaneous injection of 97H cells further demonstrated the tumor growth was largely inhibited by MTHFD1 depletion, as compared with that of control group (Fig. 1g and h). These results reveal the indispensable role of MTHFD1 in the growth of HCC cells.

Fig. 1.

MTHFD1 supports purine synthesis and growth in HCC cells. a and b 97H and Hep3B cells were stably transfected with two independent MTHFD1 shRNAs. The pLKO.1 vector was used in control group (shCtrl). The effect of knockdown was tested by western blotting (WB). c and d Long term colony forming test of 97H and Hep3B cells. Cells grew for 14 days and were then fixed and stained. e and f Cell viability were detected by CCK-8 assay in 97H and Hep3B cells. g and h A total of 1 × 10⁷ 97H cells stably transfected with shCtrl and shRNA #1 were subcutaneously injected into athymic nude mice. Knockdown of MTHFD1 inhibits cell growth, thus, inhibits the tumor growth. (i) [2,3,3-²H] serine was utilized to label into thymidine (dTMP) and purines. Double ²H-labeled dTMP and purines was produced from 5,10-methyleneTHF and 10-formylTHF directly from serine (light red circles). 1C units generated from mitochondrial pathway produced single ²H-labeled dTMP and purines. (j and k) 1C provider 10-formylTHF and purine synthesis precursor IMP (M1, M2) decreased in MTHFD1 knockdown cells. (l and m) dTMP (M2) remarkably increased as a result of mitochondrial methyl transfer blocked. (n) Intracellular IMP and TMP levels in 97H MTHFD1 shCtrl or knockdown cells. (o) Exogenous adenosine (100mM) rather than uridine (100mM) or thymidine (100mM) significantly reversed cell growth restriction caused by depletion of MTHFD1

In principle, loss of MTHFD1 tends to block the sequential reactions in the cytosolic branch of one-carbon metabolism, during which the metabolite intermediates 10-formylTHF and 5,10-methyleneTHF can act as the methyl-donor contributing to purine and dTMP synthesis respectively (Fig. 1i). Thus, we wondered whether the repressive effect of MTHFD1 depletion on cell growth was attributed to its potential impact on nucleotides synthesis. Given on the pivotal role of serine catabolism in one-carbon metabolism, [2,3,3-²H] serine was utilized to examine the influence of MTHFD1 deficiency on the relevant carbon flux (Fig. 1i). As a result, MTHFD1 depletion blocked the formation of ²H-labeled 10-formyl THF, and both M+1 (the fraction with one ² H) and M+2 fraction of IMP, which could be bidirectionally derived from [2,3,3-²H] serine mediated by cytosol-localized SHMT1 and mitochondria-localized SHMT2 (Fig. 1j and k). In contrast, although the flux of ²H of serine into M+1 fraction of 5,10-methylene THF and dTMP was notably decreased upon MTHFD1 depletion (Fig. 1l and m), M+2 fraction of dTMP was compensatively upregulated (Fig. 1m), supporting the conception that serine catabolism for dTMP production is preferentially started from mitochondria, while would be alternatively consumed by SHMT1 in the cytosol once the former cycle is blocked [8]. In line with the result from Metabolic flux analysis, the cellular levels of IMP instead of dTMP was remarkably decreased upon MTHFD1 depletion (Fig. 1n). Importantly, the repression of cell viability by MTHFD1 depletion was well restored by supplementation of exogenous adenosine rather than exogenous uridine and thymidine, in which the levels of IMP was efficiently elevated (Fig. 1o), suggesting the substitutive effect of salvage purine synthesis from adenosine. These results suggest the enzymatic activity of MTHFD1 for purine synthesis is critical for cell growth of HCC cells.

G1958A polymorphism is inversely related to MTHFD1 protein level and cell growth

To better define the clinical relevance of metabolic effect of MTHFD1, here we focused our attention on the physiological effect of the common single nucleotide polymorphism (SNP) of MTHFD1 gene, MTHFD1 G1958A (R653Q, dbSNP ID: rs2236225) variant, which was reported to be linked with the reduced risk for multiple types of cancers including hepatocellular carcinoma (HCC) [15]. Reverse transcription-PCR analysis of freshly excised tumor tissues from 13 clinical HCC samples indicated that 4 of 13 samples possess the MTHFD1 G1958A SNP (Fig. 2a). Intriguingly, immunoblotting analysis showed the protein (Fig. 2b) rather than mRNA (Fig. 2c) level of MTHFD1 in R653Q-tissues was shown to be remarkably lower compared with that of the wild-type (WT)-tissues, raising the possibility that the mutation of R653Q is implicated in the regulation of MTHFD1 protein levels. To determine this, the shRNA-resistant WT MTHFD1 (rMTHFD1 WT) or shRNA-resistant MTHFD1 R653Q (rMTHFD1 R653Q) was reconstitutively expressed in 97H and Hep3B cells with depletion of endogenous MTHFD1. Of note, no endogenous MTHFD1 G1958A SNP was detected in these cell lines. As a result, the protein amount of MTHFD1 was shown to be much lower in rMTHFD1 R653Q-expressing cells compared with the WT counterpart (Fig. 2d). In line with the observations shown above (Fig. 1j-n), metabolic analysis showed rMTHFD1 R653Q expression resulted in a similar effect on the flux pattern of [2,3,3-²H] serine (Fig. 2e-h) and IMP production to that upon MTHFD1 depletion (Fig. 2i). Meanwhile, functional analysis indicated expression of rMTHFD1 R653Q notably attenuated cell viability, which was exclusively restored by exogenous adenosine supplementation (Fig. 2j). These data suggest G1958A polymorphism is negatively associated with the protein level of MTHFD1 and the related purine synthesis in HCC cells.

Fig. 3.

The attenuated protein level of MTHFD1 R653Q is linked to protein ubiquitination. a and b 97H and Hep3B cells-expressing rMTHFD1 WT and R653Q were treated with MG132 (20 µM) or NH₄Cl (20 mM) for 12h. c and d 97 H and Hep3B cells were treated with cycloheximide (CHX; 100 µg/mL) for the indicated length of time. Immunoblotting analysis (left panel) and relative intensity of MTHFD1 to GAPDH (right panel) were performed. e 97H and Hep3B cells were treated with cycloheximide (CHX; 100 µg/mL) for the indicated length of time in the presence or absence of MG132 (20 µM) or NH₄Cl (20 mM). f Cells transfected with indicated plasmids were collected to perform immunoprecipitation in the presence of MG132 (20 µM)

The attenuated protein level of MTHFD1 R653Q is linked to protein ubiquitination

To further investigate the mechanism underlying the negative effect of R653Q on MTHFD1 protein level, cells reconstituted with expression of WT rMTHFD1 and R653Q rMTHFD1 were treated with proteasome inhibitor MG132 and autophagy-lysosome pathway inhibitor ammonium chloride (NH₄Cl) respectively. As a result, the decreased protein level of MTHFD1 R653Q was efficiently reversed by MG132 rather than NH₄Cl treatment, indicating MTHFD1 R653Q was more prone to being targeted to proteasomal degradation (Fig. 2a and b). In accordance, immunoblotting analysis showed MTHFD1 R653Q displayed a notable impairment of protein stability compared with that of the WT counterpart (Fig. 2c and d), and the MTHFD1 R653Q protein levels post CHX addition was able to be specifically restored by MG132 treatment (Fig. 2e). Additionally, the subsequent immunoprecipitation analysis indicated MTHFD1 R653Q underwent a remarkably higher ubiquitination than WT counterpart (Fig. 2f). These data suggest G1958A polymorphism (R653Q) confers the vulnerability to ubiquitin-proteasome-mediated protein degradation on MTHFD1. (Fig. 3)

MTHD1 R653Q is primarily ubiquitinated at K504

To further verify the effect of G1958A polymorphism on MTHFD1 ubiquitination, the mass spectrometry analysis was performed following precipitation of Flag-tagged MTHFD1 in its-stably expressing 97H cells. As shown in Fig. 4a, a highly evolutionarily conserved residue Lys 504 (K504) (Fig. 4a, left panel), was identified to be the unique site undergoing ubiquitination in MTHFD1 R653Q (Fig. 4a, right panel), revealing its susceptibility to be modified, although it could not be excluded that the other residues could be potentially ubiquitinated. Further immunoprecipitation analysis in HEK293T cells indicated the ectopically expressed double mutant MTHFD1 R653Q/K504R as well as WT MTHFD1 were resistant to ubiquitination in a great level, in contrast with that of MTHFD1 R653Q (Fig. 4b). Then the shRNA-resistant MTHFD1 R653Q/K504R (rMTHFD1 R653Q/K504R) was reconstitutively expressed in MTHFD1-depleted 97H and Hep3B cells (Fig. 4c and d), in which a comparable protein level of rMTHFD1 R653Q/K504R to the WT counterpart was observed. Functionally, a higher cell viability of rMTHFD1 R653Q/K504R-expressing cells was observed compared with that of rMTHFD1 R653Q-expressing cells (Fig. 4e-h). In line with this, rMTHFD1 R653Q/K504R displayed a notably stronger protein stability than that of rMTHFD1 R653Q (Fig. 4i-l). Furthermore, xenograft analysis showed the tumor growth post subcutaneous injection of 97H cells was intensively inhibited by rMTHFD1 R653Q expression, with which the growth-suppressive effect was largely compromised upon rMTHFD1 R653Q/K504R expression by contrast (Fig. 4m and n).

Fig. 4.

MTHFD1 is ubiquitinated at K504 site. a Lys504 of MTHFD1 is evolutionarily conserved in the indicated species (left panel), which has been found ubiquitination in mass spectrometry (right panel). b Cells transfected with indicated plasmids were subjected to immunoprecipitation in the presence of MG132 (20 µM). c and d 97H (c) and Hep3B (d) cells expressing with a vector for control shRNA or MTHFD1 shRNA and redeemed with expression of rMTHFD1-WT, R653Q and R653Q/K504R. e and f Long term colony forming test of 97H (e) and Hep3B (f) cells. Cells grew for 14 days and were then fixed and stained. g and h 97H (g) and Hep3B (h) cells expressing with a vector for control shRNA or MTHFD1 shRNA and redeemed with expression of rMTHFD1-WT, R653Q and R653Q/K504R. Cell viability was detected by CCK-8 assay. i-l 97H (i and j) and Hep3B (k and l) cells were treated with cycloheximide (CHX; 100 µg/mL) for the indicated length of time in the presence or absence of MG132 (20 µM) or NH₄Cl (20 mM). Representative tumor xenografts (m) and quantification of tumor weight (n) are shown

TRIM21 is responsible for MTHFD1 ubiquitination

The E3 ligase that mediates MTHFD1 ubiquitination remains unknown. Mass spectrometry analysis simultaneously revealed TRIM21 (RING-Type E3 Ubiquitin Transferase) was included in the list of MTHFD1 R653Q-associated proteins with a considerable score (Fig. 5a), and the immunoprecipitation analysis in HEK293T cells indicated MTHFD1 R653Q rather than WT MTHFD1 or MTHFD1 R653Q/K504R interacted with TRIM21 in a prominent level (Fig. 5b). Importantly, it was found that the protein level (Fig. 5c) as well as the protein stability of rMTHFD1 R653Q (Fig. 5d and e) was greatly restored by TRIM21 depletion in 97H and Hep3B cells. In accordance to the distinct endogenous protein level of TRIM21 between Hep3B, 97H and HEK293T cells (Fig. 5f), TRIM21 depletion was shown to largely abolished the MTHFD1 ubiquitination in 97H cells regardless of the form of MTHFD1 expressed (Fig. 5g); in contrast, overexpression of TRIM21 tremendously increased MTHFD1 R653Q ubiquitination in HEK293T cells (Fig. 5h). These results demonstrate TRIM21 acts the primary E3 ligase for MTHFD1 ubiquitination, and importantly regulates MTHFD1 protein stability.

Fig. 5.

TRIM21 is responsible for MTHFD1 ubiquitination. a Some of the proteins with high score interacted with MTHFD1 R653Q have been found in mass spectrometry. b Cells transfected with indicated plasmids were subjected to immunoprecipitation in the presence of MG132 (20 µM). c Cells with stably expression of indicated rMTHFD1 including WT, R653Q and R653Q/K504R were transfected with or without TRIM21 siRNA. The whole cell lysates were collected for immunoblotting (d and e). 97H and Hep3B Cells were treated with cycloheximide (CHX; 100 µg/mL) for the indicated length of time before immunoblotting. The relative intensity of MTHFD1 to GAPDH were performed. f The endogenous protein levels of TRIM21 were tested in Hep3B, 97H and HEK293T cells. g and h Cells transfected with indicated plasmids were collected to perform immunoprecipitation in the presence of MG132 (100µM), immunoprecipitation assays were performed to test the ubiquitination level

Discussion

MTHFD1 G1958A SNP (R653Q) has been found to be negatively correlated with human tumor incidence, while the relevant mechanism is largely unclear. In this study, we show expression of MTHFD1 R653Q is negatively associated with HCC cells growth, which is coupled to the impaired cytosolic production of 10-formylTHF and purine de novo synthesis. It was further revealed MTHFD1 R653Q-impeded cytosolic 1C pathway is attributed to its compromised protein stability. In terms of mechanism, biochemical analyses demonstrated MTHFD1 R653Q compared with WT MTHFD1 is more likely to be subjected to ubiquitination and the subsequent protein degradation mediated by the E3 ligase TRIM21 (Fig. 6). These observations underscore the functional role of MTHFD1 in HCC growth, and provide a molecular basis for improving clinical treatment in HCC patients with the specific MTHFD1 polymorphism.

Fig. 6.

Working model. MTHFD1 R653Q displayed an enhanced binding to the E3 ligase TRIM21, which was responsible for the augmented ubiquitination, and MTHFD1 K504 was identified to be the primary ubiquitination site. The subsequent metabolite analysis revealed MTHFD1 R653Q resulted in the repressed flux of serine-derived methyl group into metabolite precursors for purine synthesis, and the compromised purine synthesis was demonstrated to be responsible for the impeded growth capability in MTHFD1 R653Q-expressing cells

Based on the previous findings suggesting the essential role of mitochondrial FOCM enzymes in tumor development, the present study exemplifies the indispensable effect of MTHFD1-mediated cytosolic 1 C pathway in HCC growth. Indeed, the analysis from public database reveals that the liver displays the highest expression level MTHFD1 among all organs [18], implying the specific requirement of liver for role of MTHFD1-mediated cytosolic 1 C pathway. Of note, whether the concrete functional status of MTHFD1 can be distinguished between normal and tumor tissues of liver remains unclear, and the further investigation in this regard would be helpful for a better understanding the relation of cytosolic 1 C pathway to the other reprogrammed metabolic pathways in tumors. Another unsolved issue is why MTHFD1 R653Q possesses a high accessibility for TRIM21 binding, to which the exploration of the impact of R653Q mutation on MTHFD1 protein structure should be essential.

Author contributions

This study was conceived by Q.X. and Y.J., K.R. and K.Z. designed the study. K.R. and K.Z. performed experiments. Q.Z., G.H., N.S., W.W., M.Y. B.Z., Y.Z. and Y.J. provided technical and material support. K.R. and J.H wrote the paper with comments from all authors.

Funding

This work was supported by the Shanghai Yangfan Program (21YF1424900); Shanghai Organ Transplantation Research Center (2022ZZ01016); National Natural Science Foundation of China (81972205 and 92059205).

Data availability

The original contributions presented in the study are included in the article. Further inquiries can be directed to the corresponding authors.

Declarations

Ethical approval

The animal study was reviewed and approved by School of Medicine, Shanghai Jiao Tong University. This study related to patients has been approved by the Institutional Review Board of Renji Hospital affiliated to Shanghai Jiao Tong University School of Medicine. This study related to patients has been approved by the Institutional Review Board of Renji Hospital affiliated to Shanghai Jiao Tong University School of Medicine and was performed in accordance with the ethical standards as laid down in the 1964 Declaration of Helsinki and its later amendments or comparable ethical standards. 

Conflict of interest

The authors have no conflict of interest.