Dual-stress-responsive-pathways prevent HCC initiation via p53 suppression of MTHFD1L-RNA m6A-elicited autonomous growth and immune evasion

Dual-stress-responsive-pathways prevent HCC initiation via p53 suppression of MTHFD1L-RNA m6A-elicited autonomous growth and immune evasion | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Dual-stress-responsive-pathways prevent HCC initiation via p53 suppression of MTHFD1L-RNA m 6 A-elicited autonomous growth and immune evasion Hua Lu, Yi-Wei Zhang, Gang peng, Wenjuan Liao, Ji Hoon Jung, Caiyue Li, and 9 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9474831/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 8 You are reading this latest preprint version Abstract Hepatocellular carcinoma (HCC) originates from damaged hepatocytes in chronic liver injury. Though HCC development requires oncoproteins-mediated p53 inhibition, both genetic deletion and continous activation of p53 specifically in mouse hepatocytes are shown to promote liver tumorigenesis, suggesting that chronic liver injury firstly activates p53, and p53 activity needs to be dynamically regulated for preventing hepatocarcinogenesis. Yet, it remains unaddressed how liver injury stresses trigger signaling pathways to rapidly activate p53 prior to its inactivation in a physiological setting. Two pathways, ribosomal proteins (RPs)-MDM2 and 14-3-3-MDMX, are shown to activate p53 upon stresses, and mutations of MDM2 or MDMX that may disrupt these pathways have been found in human cancers including HCC. Using our unique double knock-in (DKI) mice that contain wild-type p53 while harbor defects in these two pathways, we unveiled that basal level of p53 in DKI mice is sufficient for maintaining liver function, however, carcinogen- or unhealthy diet-induced HCC initiation is accelerated in DKI mice. The two p53-pathways are also activated in human cirrhotic livers. We futher identified a one-carbon metabolism (1CM) enzyme methylenetetrahydrofolate dehydrogenase 1 like (MTHFD1L) as the p53-suppressed player in HCC initiation. We found that MTHFD1L is upregulated in human early HCC and correlated with p53 status, and promotes mouse HCC initiation by enhancing autonomous growth and immune evasion of HCC initiating cells (HCICs). With Nanopore RNA-m 6 A-sequencing, we unveiled that MTHFD1L fulfills these bi-functions by fueling methionine cycle that produces methyl groups to maintain the mRNA m 6 A of Snail (an HCC early tumorigenesis driver) and B2m (a light chain of MHC-I required for antigen presentation to CD8+ T) in HCICs. Our findings demonstrate that the two stress-triggered p53-pathways are crucial for preventing liver early tumorigenesis by reversing MTHFD1L upregulation-associated 1CM reprogramming and subsequent Snail and B2m mRNA-m 6 A-mediated cell autonomous growth and immune evasion. Biological sciences/Cancer/Cancer genetics Health sciences/Diseases/Cancer/Tumour-suppressor proteins p53 MDM2 MDMX Mthfdl HCC NASH cirrhosis one-carbon metabolism (1CM) m6A Snail B2m antigen presentation immune evasion HCC initiating cell (HCICs) and liver cancer initiation Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 INTRODUCTION Hepatocellular carcinoma (HCC), the fifth most common cancer worldwide and the third leading cause of cancer-related deaths [ 1 , 2 ], originates from injured hepatocytes that dedifferentiate into progenitor-like cells (or cancer initiating cells, CICs) expressing stem cell markers (e.g., CD44 and EpCAM). These cells expand to acquire the features of fully malignant cells in response to aberrant proliferative stimuli [ 3 – 6 ]. This transformation usually occurs in late-stage liver injury that involves closely intertwining stress responses and pathological processes, including endoplasmic reticulum (ER) and oxidative stresses, chronic inflammation, hepatocyte death, liver regeneration, somatic DNA alterations, fibrosis, and cirrhosis [ 7 , 8 ]. In normal hepatocytes without stress, the tumor suppressor p53 level is kept low by MDM2 and MDMX that work together to inactivate p53 via a negative feedback [ 9 ]. However, during chronic liver injury, p53 activity is dynamically regulated. Early in regeneration, p53 is tightly controlled to enable the hepatocyte de-differentiation and division, but activated later to ensure the genomic integrity, keep the robustness and fidelity of cell proliferation, and prevent neoplasia transformation [ 4 , 10 , 11 ]. Several studies have shown that p53 inhibition by oncoproteins leads to early hepatocarcinogenesis [ 4 , 12 ], suggesting that p53 must be activated prior to its inactivation. However, it remains unaddressed whether and how chornic liver injury-associated stress signals activate p53 under physiological conditions. Two stress responsive signaling pathways can activate p53 by overcoming the MDM2/MDMX-mediated feedback: 1) the ribosomal proteins (RPs)-MDM2-p53 pathway, where ribosomal stress (RS) frees RPs such as RPL11 and RPL5 to bind the MDM2 central domain, and release p53 [ 13 , 14 ]; 2) the 14-3-3-MDMX-p53 pathway, where DNA damage or glucose deprivation-induced MDMX phosphorylation enables 14-3-3 binding to block MDMX suppression of p53 [ 15 , 16 ]. Both the pathways were independently validated by genetic knock-in (KI) mice: MDM2 C305F mice (M2-C305F) contain a point mutation at the central domain C305 position, to disable its binding to RPL5 and RPL11 [ 13 , 17 ], and MDMX 3SA mice (MX-3SA) contain 3 serine mutations in the MDMX gene (Serine 367, 341 and 402) so that this MDMX mutant cannot be phosphorylated in response to stresses [ 15 , 16 ]. Importantly, MDM2 truncations (AA1-302) of its central zinc finger domain (AA305-322) were previously identified in HCC [ 18 ], and our bioinformatic analysis also unveiled an MDM2 somatic point mutation (E316D) in this domain in HCC patients (Figure S24A). However, the role of these MDM2 central domain mutations in HCC has never been addressed. Besides, chronic liver injury-related stresses as mentioned above often induce ribosome dysfunction or activate kinases that phosphorylate MDMX at S342 or S367 [ 13 – 15 , 19 – 22 ]. These studies highly suggest that the two stress responsive pathways may play a preventive role in HCC initiation. To address this, we generated an MDMX/MDM2-double knock-in (DKI) mouse model by crossing the two aforementioned KI mouse lines, and induced HCC in the mice with administration of hepatic procarcinogen diethylnitrosamine (DEN). DEN-induced HCC mimics many of human HCC features [ 23 ]. The DKI mice exhibited significantly higher rates of DEN-induced HCC. HCC initiating cells (HCICs) isolated from these mice displayed enhanced tumorigenesis when introduced into recipient livers. DKI mice under a long-term high-fat diet (HFD) also developed more liver tumors. Moreover, the two p53-pathways were turned on in human livers with cirrhosis, the precancerous condition for HCC [ 24 , 25 ]. We further identified methylenetetrahydrofolate dehydrogenase 1 like (MTHFD1L), a key mitochondrial one-carbon metabolism (1CM) enzyme producing formate [ 26 ], as a p53-suppressed player in HCC initiation. We found that Mthfd1l expression is declined in cirrhotic liver, while upregulated and highly correlated to the p53 mutation status in HCC patients. Mthfd1l depletion in DKI-HCICs significantly delayed HCC onset after intra-splenic transplantation. Furthermore, by fueling the methionine cycle MTHFD1L enhanced the N6-methyladenosine (m 6 A) modification of Snail [ 27 , 28 ] mRNA to boost HCIC growth and stemness, as well as the modification of B2m (β2m) mRNA, a light chain of MHC-I molecules [ 29 ], which significantlly hampered CD8 + T liver tumor infiltration and cytoxicity against HCICs. Hence, our studies reveal a new MDM2/MDMX-p53-Mthfd1l-RNA m 6 A-Snail/B2m axis and its novel role in suppressing HCIC autonomous growth and immune evasion during HCC initiation, providing new insights into the regulation of p53-pathways on metabolic reprogramming, RNA epigenetics, and cancer immunoediting during cacer initiation. Our study also illustrates the bi-functions of MTHFD1L and its potential as a diagnostic biomarker and a therapeutic target for early HCC. RESULTS DEN induced-HCC occurs significantly higher in DKI than in CTL mice To determine if the two p53-activating pathways play a role in preventing HCC initiation, we generated an M2-C305F/MX-3SA DKI mouse line (Figure S1 ). Control (CTL), DKI, M2-C305F and MX-3SA mice were obtained in a normal Mendelian ratio and employed for the experiments. All the mice were without significant defects at both postnatal and adult stages under normal conditions (Figure S1 B). We injected a single dose of DEN (25mg/kg) via i.p. into 2-week-old male and female mice [ 4 ]. All the males and most the females developed malignant multi-nodules in livers 9 months post-DEN (one female CTL had only one nodule). Consistent with previous reports [ 30 , 31 ], the male mice developed significantly more and larger tumors than females (Figs. 1 A and S2). In both sexes, DKI mice exhibited a significantly higher tumor number and/or larger average tumor area per large lobe than other groups (Figs. 1 A-B and S2), indicating that DEN induced-HCC occurrence is enhanced in DKI mice with either sex. Surprisingly, the expression of p53 and its targets p21 and MDM2 at mRNA and protein levels in both tumor and adjacent liver tissues, was comparable among the groups (Figs. 1 C-D). These results implicate that p53 might be activated through the two pathways at the early, but not advanced, stage of HCC to suppress DEN-induced HCC initiation. Pericentral hepatocyte proliferation and HCC initiation induced by DEN are significantly increased in DKI mice To test the speculation above, we examined if p53 is activated by DEN in 2-week-old male mouse livers. Indeed, p53 protein level and/or activity rapidly increased in CTL-livers 6 hours after DEN treatment and remained elevated at 48h, but not in DKI-livers (Figs. 2 A-B and S3). Of note, p53 induction in CTL-livers was mainly localized to the pericentral hepatocytes (zone 3), aligning with reports showing DEN- or NASH-induced HCC originates from this zone [ 4 , 5 ]. Furthermore, the binding of 14-3-3γ to MDMX or L5 to MDM2 was significantly enhanced in DEN-treated CTL-livers, but not in DKI-livers (Figs. 2 C-D), indicating that both pathways contribute to the DEN-induced p53 activation. Also, Ki67 staining revealed significantly higher hepatocyte proliferation in zone 3 of DKI livers at 48h post-DEN (Fig. 2 E), suggesting that p53 activation through these two pathways restrains the compensatory proliferation in CTL livers. We then examined the liver tissues 1 and 4 months post-DEN, when damaged hepatocytes start to transform towards HCICs[ 3 ]. Remarkably, cancer stem cell (CSC) markers (CD44, EpCAM and Nestin [ 4 , 11 ]) and liver regeneration markers were significantly elevated as early as 1 month post-DEN in DKI-livers, along with lower p53 level (Figs. 2 F-G). Four months post-DEN, both CSC markers and malignancy markers (Ly6D and AFP) remained at a higher level in DKI-livers (Figs. 2 H and S4), with more CD44 and AFP double-positive cells in zone 3 (Fig. 2 H). Collectively, our in vivo findings demonstrate that p53 activation through these two pathways prevents DEN-induced HCC initiation in CTL mice. p53 suppresses the stemness and malignancy properties of DKI-HCICs Next, we isolated HCIC-enriched cell clusters from male CTL and DKI-livers 3 months post-DEN as previously described [ 3 , 32 ]. As expected, both fresh (Figs. 3 A-B and S5A) and cultured (Figs. 3 C-D and S5B) DKI-HCICs exhibited significantly higher levels of CSC markers (CD44, EpCAM, KLF4) and malignancy markers, but lower p53 and p21, compared to CTL-HCICs. All HCICs were HNF4α positive (Fig. 3 C), confirming their hepatocyte origin. Furthermore, significantly more spheroids were formed with DKI-HCICs (Fig. 3 E), and greater cell viability was obsereved in DKI-HCICs as detected by cell growth and colony formation assays (Fig. 3 F), primarily attributed to the higher proliferation of DKI-HCICs (Fig. 3 G). However, ectopic p53 markedly reduced their cell viability/growth and spheroid formation as well as the expression of CSC markers (EpCAM, CD44, KLF4 and SOX9) and malignancy marker AFP (Figure S6). These in vitro results demonstrate that p53 activation via the two pathways blocks the hepatocyte transformation towards HCICs. Elevated tumorigenic capacity of transplanted DKI-HCICs in vivo We then transplanted fresh CTL- and DKI-HCICs into male C57BL/6 WT mice via intra-splenic injection (Figure S7) as previously described [ 3 , 32 ]. All DKI-HCIC recipient mice (3 out of 3) developed a tumor nodule in each liver 8 months post-transplantation, and multiple tumor nodules 12 months later (7 out of 7) (Fig. 3 H). Nevertheless, none of the CTL-HCIC recipients developed palpable liver tumors by 8 months after injection. Though 12 months later 7 out of 8 CTL-HCIC recipients developed liver tumors (Fig. 3 H), these tumors were much less and significantly smaller than those in DKI-HCIC recipients (Figs. 3 I and 3 K). H&E staining further confirmed those nodules in both groups as HCC (Fig. 3 J). Again, these results consolidate that impairment of the two p53-pathways significantly accelerates liver tumorigenesis. Mthfd1l is crucial for HCC initiation in DKI mice To identify p53 downstream players in HCC initiation of DKI mice, we performed RNA-seq with DEN-induced liver tumors and adjacent liver tissues of CTL and DKI mice. PCA and heatmap analyses revealed that the overall gene expression pattern of DKI-livers is closer to that of CTL- or DKI- tumors (Figure S8A). Pathway enrichment highlighted significant alterations in metabolic pathways, especially the monocarboxylic acid metabolism that was among the top changed metabolic pathways in DKI-liver versus CTL-liver, or CTL-tumor versus CTL-liver (Figures S8B-C). Notably, Mthfd1l, an enzyme involved in 1CM to produce formate (a type of monocarboxylic acid), was significantly upregulated in DKI-livers, CTL-tumors, DKI-tumors, and all stages of human HCC (Figures S8D-E). Its higher levels were also associated with poor survival in patients with early-stage HCC (stage I) (Figure S8F). Mthfd1l expression in 15-day-old CTL-livers, but not in DKI-livers, were drastically reduced after DEN treatment (Figs. 4 A and B). Four months post-DEN, DKI-livers presented a significantly higher formate levels (Figure S9). Furthermore, fresh or cultured DKI-HCICs showed significantly higher Mthfd1l expression or formate levels compared to CTL and/or SKI counterparts (Figure S10). These findings suggest that MTHFD1L-associated 1CM reprogramming might play a role in the accelerated HCC initiation in DKI mice. To test this idea, we knocked out the Mthfd1l gene in CTL and DKI-HCICs using CRISPR/cas9 system. Mthfd1l was depleted in three cell lines (one CTL- and two DKI-HCIC lines from three individual mice) (Figure S11A). Nanopore direct RNA-seq showed a sharp drop of the read depth in MTHFD1L Exon 6 in HCIC-1 (CTL) and − 3 (DKI) cells, indicating Mthfd1l deficiency (MTH-D), however only 1 nucleotide deletion in MTH-D HCIC-2 (DKI) (Figure S11B), so we employed MTH-D HCIC-1 and − 3 cells for the following experiments. Mthfd1l deletion significantly diminished CSC marker expression and cell viability/growth, more so in DKI-HCICs (Figs. 4 C, S11C-D and S12), and impaired spheroid formation in both the HCICs (Fig. 4 D). We further tested whether formate supplementation could resuce these effects caused by Mthfd1l depletion. After formate treatment for 48h, CD44 expression in MTH-D HCICs was almost recovered to the level in Mthfd1l-proficient (MTH-WT) cells (Figs. 4 C, 4 E and S11C-D). The size, but not the number, of the spheroids was also rescued by formate (Fig. 4 D). However, cell proliferation (Ki67 + signals) and growth showed partial or minimal recovery after formate treatment (Figs. 4 E and S12B). These results implicate that while formate improves some Mthfd1l deletion-caused defects, Mthfd1l might also utilize a non-enzymatic mechanism to promote growth and self-renewal of HCICs. We re-expressed Mthfd1l in MTH-D DKI-HCICs to test if these defects can be rescued. Remarkably, replenishment of Mthfd1l restored CSC marker expression (CD44 and SOX9) (Figs. 4 E), spheroid formation (Fig. 4 F), cell viability (Fig. 4 G), and proliferation (Ki67 signals in Fig. 4 E) of MTH-D HCICs to near-normal levels. Furthermore, 10 months after intra-splenic injection, MTH-WT DKI-HCICs formed multiple tumors in recipient livers, while MTH-D cells produced significantly fewer palpable tumors, along with significantly lower liver/body weight ratio of MTH-D HCIC recipients (Figs. 4 H-I). H&E and IHC staining confirmed that these tumors are AFP-positive HCC (Figs. 4 J and S13) with significantly stronger Ki67 signals in MTH-WT DKI-HCIC-derived tumors (Fig. 4 J). These results demonstrate that Mthfd1l activation owing to impairing the two p53 pathways in DKI mice is responsible for the accelerated HCC initiation. p53 transcriptionally suppresses Mthfd1l by cooperating with HDAC1/2 Our bioinformatic analysis showed that though Mthfd1l is upregulated in HCC patients, its expression is significantly lower in p53-wild type than in p53-mutant HCCs (Figure S14A). Indeed, Mthfd1l expression at mRNA and protein levels was p53-dependent, comparing p53-containing cells (HepG2 liver cancer and H460 lung cancer cells) with p53-null /-mutant cells (Figures S14B-F). p53 overexpression in DKI-HCICs reduced Mthfd1l expression (Figure S14G). These data indicate that p53 transcriptionally suppresses Mthfd1l. Luciferase reporter assays using a reporter vector containing two putative p53 binding sites (BS) in the human MTHFD1L promoter (Mthfd1l-BS1/2) (Figure S15A left) showed that p53 activators reduces the luciferase signals by 50 ~ 60% only in HepG2 (Figures S15A right). This repression was abolished by BS1 mutation (Figure S15B), indicating that p53 inhibits MTHFD1L promoter activity via the Mthfd1l-BS1. Consistently, p53 binding to the MTHFD1L, p21, and Puma promoters was confirmed by ChIP assays in flag-p53-overexpressing Hep3B cells or in doxorubicin-treated HepG2 cells (Figure S15C). P53 recruits an inhibitory complex with HDACs (Figure S16A) [ 33 ] to suppress its targets. Accordingly, treatment with the HDAC1/2 inhibitor TSA almost fully rescued Mthfd1l expression inhibited by the p53 activators only in HepG2, but not p53-null Hep3B, cells (Figure S16B). Consistently, the DNA-protein complexes pulled down by HDAC1/2 antibodies were enriched with the DNA fragments containing Mthfd1l-BS1 of p53 (Figure S16C). These results indicate that p53 represses Mthfd1l transcription by working with HDACs in HCC cells. MTHFD1L promotes HCIC autonomous growth and immune evasion via RNA mA methylation Mthfd1l generates formate to supply ~ 75% of 1-C units in the methionine cycle [ 34 ] that produces direct methyl donor SAM for RNA m 6 A modificaiton. Indeed, in MTH-D HCICs, global mRNA m 6 A levels were drastically declined, but recovered by ectopic Mthfd1l (Fig. 5 A). SAM significantly boosted CSC marker expression, spheroid formation, and growth of MTH-D HCICs (Figures S17A-C), while treatment with STM2457 (STM), an inhibitor of the major RNA m 6 A methylation writer methyltransferase like 3 (METTL3) [ 35 ], remarkably suppressed these traits of HCICs (Figures S17D-G). Also, our database analysis showed that a higher METTL3 level is well correlated with poorer survival in HCC patients (Figure S17H). These results suggest that MTHFD1L promotes HCC initiation by enhancing RNA m 6 A methylation. To test this idea, we performed Nanopore RNA m 6 A-seq with paired MTH-WT and MTH-D HCIC cells (HCIC-1 and HCIC-3). It showed a global reduction of RNA m 6 A in MTH-D cells (Figure S18A) and identified ~ 46 protein-coding genes with an average modification difference ≥ 5%. As one of the top 10 genes (Figures S18B-D), Socs2 was used as a positive control. Its decreased protein level due to the elevation of its mRNA m 6 A is linked to HCC progression [ 36 ]. Then the rest top candidates (Figure S18B) were screened by their response to SAM treatment in MTH-D HCICs (Figs. 5 B and 6 A). Snail, a well-known EMT player and newly identified HCC early tumorigenesis driver [ 27 , 28 ], was upregulated by SAM (Fig. 5 B). Conversely, in HCICs when knocking down Mthfd1l or Mettl3 (Fig. 5 C), or in MTH-D HCICs when comparing with MTH-WT cells (Fig. 5 D), Snail was dramatically reduced. These effects were reversed by re-expression of Mthfd1l in MTH-D HCICs (Figs. 5 D-E), which was further withdrawn by STM treatment (Fig. 5 E). P53 also suppressed Snail in both HCICs and Hep3B cells, but these effects were negated by Mthfd1l co-expression (Figs. 5 F-G). However, knockdown of Mettl3 or Mthfd1l, re-expressing Mthfd1l in MTH-D HCICs, or treatment of STM exerted no effect on Snail mRNA level (Figures S20A-B). Besides, we found that the m 6 A mainly exists in the 3’-UTR region of Snail mRNA (Figure S18C). We further identified several m 6 A sites within that region. Those sites displayed m 6 A level > = 15% in MTH-WT HCICs and the modification decline > = 5% in MTH-D vs. MTH-WT cells (Figures S18C-D). Taken together, these findings demonstrate that the p53-MTHFD1L axis controls cell growth and stemness during HCC initiation, by modulating mRNA m 6 A to translationally upregulate Snail. Hepatocytes express abundant cell surface MHC-I, and can rapidly load and present antigens[ 37 , 38 ]. In MTH-D HCICs, the mRNA m 6 A level of β2-microglobulin (β2m, B2m), the light chain of MHC-I molecules, was reduced (Figures S18B-D). The nanopore m 6 A-seq analysis indicated a major m 6 A site in the coding region (CDS) of B2m mRNA in MTH-WT HCICs, which was decreased by 10% in MTH-D cells (Figures S18C-D). In contrast to the response of SNAIL to SAM treatment, the B2M protein level of MTH-D HCICs was dramatically decreased by SAM (Fig. 6 A). Downregulation of B2m or MHC-I molecules dampens antigen presentation, impairs CD8 + T cell cytotoxicity and contributes to cancer immunotherapy resistance [ 29 , 39 ]. Consistently, B2m is positively correlated with HCC patient survival (Figure S19). Also, Mthfd1l knockdown remarkably increased (Fig. 6 B), while its overexpression suppressed, β2m expression in MTH-WT HCICs, which was rescued by Mettl3 knockdown (Fig. 6 C). β2m protein level was doubled in MTH-D HCICs compared to MTH-WT cells, wheras Mthfd1l re-expression counteracted the higher β2m expression in MTH-D cells, which was recovered by STM treatment (Fig. 6 D). P53 upregulated β2m in both HCIC and Hep3B cells (Figs. 6 E-F), but this effect was eliminated by Mthfd1l overexpression (Fig. 6 F). Notably, B2m mRNA levels were unchanged after Mthfd1l re-expression, STM treatment, Mthfd1l or Mettl3 knockdown in MTH-D or MTH-WT HCICs (Figures S20B-C), indicating a translational silence of m 6 A modified B2m mRNA. To validate the role of β2m in HCIC antigen presentation and subsequent CD8 + T cell activation, we co-cultured HCICs stably expressing chicken ovalbumin antigen (HCIC_OVA) (Figure S21A) with CD8 + T cells freshly isolated from C57 BL/6 OT-1 mouse spleen. OT-1 mice contain transgenic T cell receptor that recognizes OVA peptides presented by MHC-I. After co-culturing for 24h, CD8 + T cytotoxicity against HCIC, indicated by the caspase 3/7 signal, was enhanced by 1.5 times by Mthfd1l or Mettl3 knockdown in HCICs, while significantly weakened by B2m knockdown, which was mostly rescued by STM pre-treament on HCICs (Figures S21B and 6G). Furthermore, Mthfd1l overexpression in HCICs significantly inhibited p53-induced CD8 + T cytotoxicity, whereas STM pre-treatment reversed Mthfd1l-caused decline of CD8 + T cytotoxicity against HCICs (Figures S21B and 6H). Consistently, MTH-deficient (MTH-D) HCICs transplantation-derived liver tumors showed significantly increased CD8 + T cell infiltration in vivo (Fig. 6 I). Together, these in vitro and in vivo results indicate that MTHFD1L promotes the immune evasion of HCICs by suppressing β2m-dependent antigen presentation and CD8 + T cytotoxicity via m 6 A-mediated translational control of B2m. Two p53-pathways play a preventative role in NASH- and cirrhosis-associated HCC initiation To determine if these two p53-pathways are triggered to prevent the NASH-associated HCC initiation, we fed CTL and DKI mice with a long-term HFD (LHF) for 15 months to mimic the human NASH-driven HCC. Both groups comparably developed severe steatosis and mild interstitial fibrosis (Figure S22), with tumors in ~ 75% of CTL and ~ 85% of DKI livers (Fig. 7 A). Notably, all DKI mice developed multiple (≥3) tumor nodules, alongside with significantly higher liver/body weight ratio and more small nodules (diameter ≤ 2mm) in livers, whereas none of the CTL livers had more than two nodules (Figs. 7 A-C). The HCC tumor size usually indicates tumor progression, while tumor number is an index of early tumorigenesis. Expectedly, the tumor nodules in both groups expressed a high level of AFP (Fig. 7 E). However, p53 (and p21) was remarkably induced only in CTL-tumors compared to the adjacent liver tissues, but not in DKI-tumors, correlating with down-regulated MTHFD1L in CTL-tumors (Figs. 7 D-E). Together, these data show that impairment of the two p53-pathways enhances NASH-driven HCC initiation. We further assessed the p53-MTHFD1L signaling in human liver cirrhosis (Figure S23), the HCC precancerous condition [ 24 , 25 ]. As shown in Fig. 7 F, human alcoholic cirrhotic livers presented strong p53 induction, increased MDM2 and downregulated MTHFD1L. IP-WB analysis showed remarkably elevated formation of MDM2-RPL5 and MDMX-14-3-3γ complexes in cirrhotic versus normal livers (Figs. 7 G-H). These results support an active role of the two p53-pathways in preventing human hepatocyte malignant transformation in cirrhosis by suppressing Mthfd1l. DISCUSSION Using a DKI animal model, we have addressed the remaining crucial question: how is p53 rapidly activated prior to its inactivation during chronic injury to block HCIC transformation and HCC initiation? We for the first time identified a dual mode of p53 activation by inactivating both MDM2 and MDMX in response to various stresses to prevent HCC initiation. We showed that imparing the two p53-activating pathways markedly accelerates DEN- or NASH-driven HCC initiation in DKI mice (Figs. 1 – 2 , S4, 7 A-E and 8 ), DKI-HCICs exhibit higher malignancy and stemness (Figs. 3 A-G) which can be reversed by p53 (Figure S6), and DKI-HCICs develop into HCC more severely (Figs. 3 H-K). Correlatedly, p53 was activated through these two pathways in cirrhotic patient livers (Figs. 7 F-H). About 80% of HCCs are developed from cirrhosis that is recognized as a precancerous state [ 24 , 25 ]. In line with the literature, our results consolidate the biological significance of p53 activation through the two pathways in preventing HCC initiation. While hepatocyte specific deletion of p53 spontaneously and consistently induces liver carcinoma formation [ 40 ], continuous p53 activation in hepatocytes (e.g., via MDM2 deletion) also promotes liver tumorigenesis [ 41 ], indicating that tight regulation on p53 activity is needed for preventing hepatocarcinogenesis. Different from other genetic models, our unique DKI mice retain WT p53, providing a special tool for studying p53’s function in real time. Indeed, p53 activation is observed mainly in the end-stage of liver disease, i.e. cirrhotic liver (Fig. 7 F), or in early HCC stages, but blunted in DKI mice (Figs. 2 F and 7 D-E), suggesting a dynamic regulation of p53 activity by the two pathways during chronic liver injury and HCC initiation. Interestingly, similar to their SKI parents [ 16 , 17 ], DKI mice are developmentally normal (Figure S1 ). Thus, the basal level of p53 in DKI mice is sufficient for maintaining normal liver function, but insufficient for protecting the animals from stresses-associated HCC tumorigenesis, highlighting the surveillance role of the two p53-pathways in early hepatocarcinogenesis long before somatic p53 mutations occur [ 4 , 42 , 43 ]. As aforementioned, mutations (Figure S24A) and truncations [ 18 ] of MDM2 central zinc finger domain that is critical for the RP binding were identified in HCC patients. In addition to HCC, other types of cancers also contain MDM2 somatic mutations in its central domain (Figure S24B). Over a dozen RPs and other proteins such as ARF, NPM, and Rb activate p53 under stresses by binding the MDM2 central domain, the same mode demonstrated in our current study, disrupting the MDM2-p53 feedback loop [ 14 , 18 , 44 ]. Similarly, somatic point mutations of MDMX at S367, S342 and other serine sites exist in human cancers (Figure S24C). Our study partially explains why the MDM2 central domain and the MDMX serine sites are mutated in various cancer patients including HCC, and reinforces the clinical significance of the dual-mode activation of p53 in preventing liver tumorigenesis. One-carbon metabolism (1CM) is frequently reprogrammed in many cancers. Various enzymes for 1CM have been linked to different cancers including HCC, rendering the 1CM enzyme networks as very promising therapeutic targets of cancers [ 45 ]. Here, we identified the 1CM enzyme MTHFD1L as a new p53-suppressed downstream player in HCC initiation (Figs. 4 , S 8 -S17, and 8 ). MTHFD1L could accelerate liver cancer progression by encountering oxidative stress [ 46 ]. Our bioinformatic analysis (Figure S8F) suggests that MTHFD1L plays a crucial role in HCC initiation. Consistently, Mthfd1l deletion drastically retarded HCIC tumorigenesis in vivo (Fig. 4 ). Interestingly, Mthfd1l is downregulated and well correlated with increased p53 in human cirrhotic livers (Fig. 7 F). Given Mthfd1l is usually overexpressed in all stages of HCC (Figure S8E), Mthfd1l might serve as a diagnostic marker and a therapeutic target for early HCC. Besides enhanced autonomous growth mechanisms[ 4 ], initiaiting-stage cancers gradually acquire an immune evasion ability to progress to advanced stages [ 39 ]. We demonstrate that impairing the p53-pathways leads to higher expression of Mthfd1l that utilizes both the mechanisms to accelerate HCC initiation (Figures S14-16, 4–6 and 8). By producing mitochondrial formate, MTHFD1L fuels the methionine cycle to maintain the high level of RNA m 6 A (Figs. 5 A and S18A) that is observed in human HCC[ 36 ] and critical for HCIC stemness and proliferation (Figure S17). Specifically, MTHFD1L supports the m 6 A modification of Snail mRNA (Figures S18C-D), resulting in translational upregulation of Snail (Figs. 5 and S20A-B). Studies have established that Snail plays an essential role in HCC onset by inhibiting hepatocyte-differentiation transcriptional factors and promoting cell proliferation [ 27 ], and disrupting bile acid homeostasis [ 28 ]. Interestingly, overexpression of Snail in hepatocytes can independently lead to steatohepatitis and HCC onset [ 28 ]. As aforementioned, hepatocytes can express abundant cell surface MHC-I molecules for antigen presentation to CD8 + T cells[ 37 , 38 ]. In a study, loss of heterozygosity of HLA alleles is identified in 17% of multifocal HCC cases to dampen the neoantigen presentation and enhance the tumor cell immune escape[ 47 ]. Here, we found that B2m is translationally repressed by the MTHFD1L-mediated mRNA m 6 A modification in HCICs (Figs. 6 and S20B-C). In line with the critical role of B2m in MHC-I-mediated antigen presentation to CD8 + T cells and subsequent T cell recognition [ 29 ], MTHFD1L restrained the CD8 + T liver tumor infilatration and cytotoxicity against HCICs (Figs. 6 G-I), leading to reinforced immune evasion of HCICs, while p53 inhibition of Mthfd1l prevented this cancer immunoediting during HCC initiation (Figs. 6 E, F, and H), providing new insights into the p53’s function of reversing cancer metabolic reprogramming to block cancer immunoediting. RNA m 6 A modification is a reversible process regulated by writers (such as METTL3/14, WTAP, etc.), erasers (such as FTO and ALKBH5), and readers (such as IGF2BPs, YTHDFs, etc.), affecting RNA splicing, nuclear export, degradation, stability, and translation [ 48 ]. The fate of m 6 A modified mRNAs largely depends on the specific readers and modification regions. For instance, YTHDF2 promotes mRNA degradation, IGF2BPs enhances mRNA stability or translation [ 48 ], and m 6 A in the mRNA coding region can delay translation elongation [ 49 ]. We found the differential regulation of MTHFD1L on the m 6 A-mediated Snail and B2m expression (Figs. 5 B-G, 6 A-F, S20, and 8 ). It remains to identify the specific m 6 A readers and erasers modulating these targets during HCC initiation. In summary, our study for the first time unveils the surveillance role of the two p53-pathways in averting MTHFD1L-accelerated HCC initiation upon pathologival signals, e.g., carcinogen, NASH, or alcoholic cirrhosis, and the novel bi-functions of MTHFD1L in promoting HCIC proliferation/stemness while suppressing immune elimination during HCC initiation, by fueling the methionine cycle to support RNA m 6 A modifications (Fig. 8 ). It also positions MTHFD1L and its downstream m 6 A targets as promising candidates for early diagnosis, prevention or therapy of HCC. METHODS Sex as biological variants Both male and female mice were recruited for the DEN-induced HCC model, and similar findings were reported for both sexes, with significantly higher HCC incidence in the male mice, which is consistent with other animal studies and human HCC epidemiologic studies [ 30 , 31 , 50 ]. Male mice were used in other animal experiments due to their higher sensitivity and stability to HCC induction, and liver samples from female patients were purchased, as detailed below and in the supplemental information . Animals Mouse lines. M2-C305F and MX-3SA KI mouse lines were generously provided by Jeoff Wahl at SALK institute and Yanping Zhang at UNC at Chapel Hill, respectively [ 16 , 17 ]. M2-C305F/MX-3SA-DKI mouse line was generated by crossing the two single-KI (SKI) mice as detailed in the supplemental information. Both male and female animals were used in this study and maintained in a C57BL/6 genetic background. DEN-induced HCC mouse model. To induce HCC, 15-day-old males and females were injected intraperitoneally (i.p.) with a single dose of DEN (25 mg/kg) and livers were collected 9 months later, following previous studies [ 4 ]. For some experiments involved in acute DEN treatment, 15-day-old male mice were i.p. injected with a single dose of DEN (25 mg/kg), and livers were collected at indicated time points for further analysis. High-fat diet (HFD)-induced non-alcoholic steatohepatitis (NASH) and HCC mouse model . Male control (CTL) and DKI mice were weaned at 3 weeks and fed with HFD (with 60 kcal% fat) (Research Diets, Inc.) and water ad libitum for 15 months. Intra-splenic transplantation of HCICs. C57BL/6 wild-type (WT) mice recruited for HCIC transplantation were from the Jackson Laboratory (Stock# 000664). Briefly, by following previous studies [ 3 , 32 , 51 ], 4-week-old male C57BL/6J WT mice were i.p. injected with retrosine twice with a 2-week interval. 4 weeks after the 2nd- retrosine treatment, each mouse received 1.5⋅10 5 freshly isolated viable cells that were enriched with CTL- or DKI-HCICs, or 1⋅10 6 cultured DKI-HCICs (passage 6, P6) with or without Mthfd1l gene deletion depending on experiments, in 100µl PBS via intra-splenic injection. One week after transplantation, CCl4 was given to the recipient mice intraperitoneally once a week for 3 weeks. Liver tissues of the recipient mice were harvested 8 ~ 12 months later for further analysis. All mice were housed under specific pathogen-free conditions with normal diet and water ad libitum . OT-1 mice and CD8 + T cell isolation Two-month-old male C57BL/6 OT-1 mice were purchased from the Jackson Laboratory (Stock# 003831). Briefly, following previous study [ 52 ], OT-1 mouse spleen was homogenized and cell suspension was filtered through a 70µm cell strainer (Fisher), and erythrocytes were lysed. Then the single-cell suspension was centrifuged and counted, and untouched CD8 + T cells were purified using a mouse CD8a + T Cell Isolation Kit according to the manufacturer's instructions. Primary cell isolation and culture Primary HCICs and hepatocytes were isolated from male CTL and DKI mice treated with DEN or Vehicle, by following previous studies [ 3 , 32 , 51 ]. To ensure that enough HCICs can be obtained from mice for experiments and transplantation, male mice were used for DEN treatment. Fifteen-day-old mice were injected with one dose of DEN (25 mg/kg), and 3 months later mouse livers were perfused via inferior vena cava with warmed S&M buffer (0.24% Hepes, 0.019% NaOH, 0.05% KCl, 0.83% NaCl, pH7.4) and digest solution for 5min each, and minced in cold PBS. The cell aggregates containing HCICs were then isolated by sequential filtration with 70µm and 40µm cell-strainers, and subjected to gentle pipetting in cold PBS for intra-splenic injection. For some experiments, HCICs were maintained in collagen-coated Petri dishes in DMEM medium (Gibco) supplemented with 20% fetal bovine serum (FBS) (Sigma), 1⋅ PSN antibiotic mixture (10 ml/L) (Gibco), 20µg/L EGF, 0.01g/L Insulin, 1µm Dexamethasone and 1% L-glutamine. HCICs of passage 3–8 (P3-P8) were used for experiments to ensure the cell stemness. CRISPR/Cas9 mediated gene modification HCIC cells with stable depletion/deletion of Mthfd1l (MTH-D) were generated using lentiviral CRISPR-Cas9 system. In short, three single guide RNAs (sgRNAs) ( Table S2 ) respectively targeting exon 1, 6 and 12 of Mthfd1l gene were designed using online software (IDT), and inserted in plasmid LentiCRISPRv2 expressing wild-type Cas9, then co-transfected into HEK 293T cells with packaging plasmids following the ZhangLab protocol ( https://media.addgene.org/cms/files/Zhang_lab_LentiCRISPR_library_protocol.pdf ). CTL or DKI-HCICs (P3) were infected with lentivirus, and puromycin was added 48h later for selection of HCIC cells stably expressing Mthfd1l sgRNAs and Cas9. Single cell colonies were selected and cultured, followed by WB analysis and RNA sequencing to confirm the deletion. HCIC single colonies expressing Cas9 and WT-Mthfd1l (MTH-WT) were also simultaneously screened after infected with empty LentiCRISPRv2, to serve as paired control cells. To establish MTH-D HCIC cell lines re-expressing Mthfd1l or HCICs expressing chicken ovalbumin (cOVA) (HCIC_OVA), the cells were infected with lentivirus expressing GFP-flag-Mthfd1l or cOVA, respectively. G418 or puromycin was added 48h later for Mthfd1l selection for 10 days or cOVA selection for 5 days, respectively. In vitro CD8 + T cytotoxicity assay Briefly, ~ 4,000 HCIC_OVA cells/well were seeded in triplicate into a 96-well flat-bottom plate. The next day, freshly isolated C57BL/6 OT-1 CD8 + T cells were added to the wells (T cell: HCIC ratio is about 5:1) and co-cultured in complete DMEM growth medium for 24h in an Incucyte (Sartorius) culture chamber. NucView 488 Caspase 3/7 substrate (Biotium) was simultaneously added to each well, and caspase 3/7 positive signal (indicating the amount of T-cell killing) were measured following previous studies[ 53 ]. Five images per well were captured every hour using a 10× objective lens. Caspase 3/7 fluorescence intensity was quantified using the Incucyte built-in software combined with Image J . In those Control wells, HCIC_OVA cells were cultured without CD8 + T. In some experiments, HCIC_OVA cells were co-cultured with OT-1 CD8 + T for 24h, followed by the measurement of released lactate dehydrogenase (LDH) with a LDH Cytotoxicity Assay kit (ThermoFisher) according to the manufacturer’s instructions. Statistics All in vitro experiments were performed in at least triplicate. Data were presented as means ± SEM with N being the sample size. Comparisons among different groups were conducted by using One-way ANOVA or Student’s two-tailed t-test. Probability values of P ≤ 0.05 were considered statistically significant. Supplementary information is available at Cell Death and Differentiation ’s website. Declarations ACKNOWLEDGEMENTS We thank Dr. Geoffrey M Wahlfor offering MDMX 3SA mice (The Salk Institute for Biological Studies), Dr. Jinqiang Zhang (Department of Pathology, Tulane University) for giving suggestion on HCIC transplantation, Drs. Xiaoming Yin, Gang Liu and Bilon Khambu (Department of Pathology, Tulane University), Dr. James Jackson (Department of Biochemistry and Molecular Biology, Tulane University), and Dr. Hongbing Liu (Department of Pediatrics, Tulane University) for providing some antibodies, Drs. Jieqiong Wang and Hyemin Lee (Department of Biochemistry and Molecular Biology, Tulane University) for offering CRISPR-Cas9 empty vectors and some technique support, and Elise Reinosofor assisting in in vitro experiments. AUTHOR CONTRIBUTIONS Y.W.Z. independently conceived and designed the project, conducted the experiments, analyzed the results, and organized them into figures in the Lu laboratory under the supervision of H.L.; G.P. performed the nanopore m 6 A-seq data analyses. W.J.L., J.H.J., J.X.C., C.Y.L., Y.Z., N.P., S.M.M., and E.D.E. assisted Y.W.Z. in some animal works or in vitro experiments; Y.L. and H.G. conducted the nanopore m 6 A sequencing. Y.P.Z. offered MDM2 C305F mice; D.D. offered helps in HCIC isolation and transplantation; S.X.Z. assisted in the initial animal transfer and coordinated RNA-seq analysis. Y.W.Z. wrote and revised and H.L. revised the manuscript. ETHICS APPROVAL AND CONSENT TO PARTICIPATE The frozen smashed human liver tissue samples were deidentified and purchased from XenoTech, BioIVT (Kansas City, KS 66103). All animal experiments were conducted in accordance with the National Institutes Health “Guide for the Care and Use of Laboratory Animals” and were approved by the Institutional Animal Care and Use Committee at Tulane University School of Medicine and our animal protocol numbers are 1097, 2053, and 2266. FUNDING STATEMENT This project was in part supported by NIH-NCI grants R01CA095441, R01CA127724 to H.L. and S.X.Z, who are also in part supported by NIH-NCI grants R01CA234605 and R21CA272890, and the Carol Lavin Bernick Faculty Grant and the Tulane Medical School Pilot fund to Y.W.Z. In addition, Y.W.Z received the NIH-COBRE pilot funds (AP5446G1 and AP5446GG). H.L. was supported by the Reynolds and Ryan Families Endowed Chair fund, Lady Leukemia League Fund, and the NIH-NCI grant U01 CA252965 . D.D. was supported by UC San Diego Altman Clinical and Translational Research Institute (ACTRI)/ NIH grant KL2TR001444 and the San Diego Digestive Diseases Research Center (SDDRC) (NIH grant DK120515). DATA AVAILABILITY GEO accession number for Nanopore RNA m 6 A-seq data is GSE310323. GEO accession number for RNA-seq data will be available upon submission. All raw data of immunoblotting images generated in this study will be deposit and made publicly available upon publication of this manuscript.Any additional information required to reanalyze the data reported in this paper is available from both the corresponding authors upon reasonable request. References Chidambaranathan-Reghupaty S, Fisher PB, Sarkar D. Hepatocellular carcinoma (HCC): Epidemiology, etiology and molecular classification. Adv Cancer Res 2021; 149: 1-61. Vogel A, Meyer T, Sapisochin G, Salem R, Saborowski A. Hepatocellular carcinoma. Lancet 2022; 400: 1345-1362. He G, Dhar D, Nakagawa H, Font-Burgada J, Ogata H, Jiang Y et al . Identification of liver cancer progenitors whose malignant progression depends on autocrine IL-6 signaling. Cell 2013; 155: 384-396. Dhar D, Antonucci L, Nakagawa H, Kim JY, Glitzner E, Caruso S et al . Liver Cancer Initiation Requires p53 Inhibition by CD44-Enhanced Growth Factor Signaling. Cancer Cell 2018; 33: 1061-1077 e1066. Font-Burgada J, Shalapour S, Ramaswamy S, Hsueh B, Rossell D, Umemura A et al . Hybrid Periportal Hepatocytes Regenerate the Injured Liver without Giving Rise to Cancer. Cell 2015; 162: 766-779. Shin S, Wangensteen KJ, Teta-Bissett M, Wang YJ, Mosleh-Shirazi E, Buza EL et al . Genetic lineage tracing analysis of the cell of origin of hepatotoxin-induced liver tumors in mice. Hepatology 2016; 64: 1163-1177. Llovet JM, Kelley RK, Villanueva A, Singal AG, Pikarsky E, Roayaie S et al . Hepatocellular carcinoma. Nat Rev Dis Primers 2021; 7: 6. Tanaka M, Miyajima A. Liver regeneration and fibrosis after inflammation. Inflamm Regen 2016; 36: 19. Huang L, Yan Z, Liao X, Li Y, Yang J, Wang ZG et al . The p53 inhibitors MDM2/MDMX complex is required for control of p53 activity in vivo. Proc Natl Acad Sci U S A 2011; 108: 12001-12006. Charni M, Aloni-Grinstein R, Molchadsky A, Rotter V. p53 on the crossroad between regeneration and cancer. Cell Death Differ 2017; 24: 8-14. Tschaharganeh DF, Xue W, Calvisi DF, Evert M, Michurina TV, Dow LE et al . p53-dependent Nestin regulation links tumor suppression to cellular plasticity in liver cancer. Cell 2014; 158: 579-592. Wang C, Kang H, Yi Y, Ding Y, Wang F, Luo J et al . Rictor mediates p53 deactivation to facilitate the malignant transformation of hepatocytes and promote hepatocarcinogenesis. J Transl Med 2023; 21: 919. Liu Y, He Y, Jin A, Tikunov AP, Zhou L, Tollini LA et al . Ribosomal protein-Mdm2-p53 pathway coordinates nutrient stress with lipid metabolism by regulating MCD and promoting fatty acid oxidation. Proc Natl Acad Sci U S A 2014; 111: E2414-2422. Zhou X, Liao WJ, Liao JM, Liao P, Lu H. Ribosomal proteins: functions beyond the ribosome. J Mol Cell Biol 2015; 7: 92-104. He G, Zhang YW, Lee JH, Zeng SX, Wang YV, Luo Z et al . AMP-activated protein kinase induces p53 by phosphorylating MDMX and inhibiting its activity. Mol Cell Biol 2014; 34: 148-157. Wang YV, Leblanc M, Wade M, Jochemsen AG, Wahl GM. Increased radioresistance and accelerated B cell lymphomas in mice with Mdmx mutations that prevent modifications by DNA-damage-activated kinases. Cancer Cell 2009; 16: 33-43. Macias E, Jin A, Deisenroth C, Bhat K, Mao H, Lindstrom MS et al . An ARF-independent c-MYC-activated tumor suppression pathway mediated by ribosomal protein-Mdm2 Interaction. Cancer Cell 2010; 18: 231-243. Lindstrom MS, Jin A, Deisenroth C, White Wolf G, Zhang Y. Cancer-associated mutations in the MDM2 zinc finger domain disrupt ribosomal protein interaction and attenuate MDM2-induced p53 degradation. Mol Cell Biol 2007; 27: 1056-1068. Zhang F, Hamanaka RB, Bobrovnikova-Marjon E, Gordan JD, Dai MS, Lu H et al . Ribosomal stress couples the unfolded protein response to p53-dependent cell cycle arrest. J Biol Chem 2006; 281: 30036-30045. Brighenti E, Calabrese C, Liguori G, Giannone FA, Trere D, Montanaro L et al . Interleukin 6 downregulates p53 expression and activity by stimulating ribosome biogenesis: a new pathway connecting inflammation to cancer. Oncogene 2014; 33: 4396-4406. Dicks N, Gutierrez K, Michalak M, Bordignon V, Agellon LB. Endoplasmic reticulum stress, genome damage, and cancer. Front Oncol 2015; 5: 11. Chen L, Gilkes DM, Pan Y, Lane WS, Chen J. ATM and Chk2-dependent phosphorylation of MDMX contribute to p53 activation after DNA damage. EMBO J 2005; 24: 3411-3422. Carlessi R, Kohn-Gaone J, Olynyk JK, Tirnitz-Parker JEE. Mouse Models of Hepatocellular Carcinoma. In: Tirnitz-Parker JEE (ed). Hepatocellular Carcinoma : Brisbane (AU), 2019. Coral GP, Branco F, Meurer R, Marcon PDS, Fontes PRO, Mattos AA. Results of Immunohistochemistry in the Differential Diagnosis of Early Hepatocellular Carcinoma and Nodules with High-Grade Dysplasia in Patients with Cirrhosis. Arq Gastroenterol 2021; 58: 82-86. Liao Z, Tang C, Luo R, Gu X, Zhou J, Gao J. Current Concepts of Precancerous Lesions of Hepatocellular Carcinoma: Recent Progress in Diagnosis. Diagnostics (Basel) 2023; 13. Momb J, Lewandowski JP, Bryant JD, Fitch R, Surman DR, Vokes SA et al . Deletion of Mthfd1l causes embryonic lethality and neural tube and craniofacial defects in mice. Proc Natl Acad Sci U S A 2013; 110: 549-554. Moon H, Ju HL, Chung SI, Cho KJ, Eun JW, Nam SW et al . Transforming Growth Factor-beta Promotes Liver Tumorigenesis in Mice via Up-regulation of Snail. Gastroenterology 2017; 153: 1378-1391 e1376. Miura S, Suzuki A. Induction of Steatohepatitis and Liver Tumorigenesis by Enforced Snail Expression in Hepatocytes. Am J Pathol 2020; 190: 1271-1283. Wang H, Liu B, Wei J. Beta2-microglobulin(B2M) in cancer immunotherapies: Biological function, resistance and remedy. Cancer Lett 2021; 517: 96-104. Liu S, Huang F, Ru G, Wang Y, Zhang B, Chen X et al . Mouse Models of Hepatocellular Carcinoma: Classification, Advancement, and Application. Front Oncol 2022; 12: 902820. Kim GY, Kim D, Jeon J, Seo W, Park SH. Navigating liver cancer with mouse models: a comprehensive overview of HCC experimental systems. J Liver Cancer 2025; 25: 239-250. Zhang J, Han C, Ungerleider N, Chen W, Song K, Wang Y et al . A Transforming Growth Factor-beta and H19 Signaling Axis in Tumor-Initiating Hepatocytes That Regulates Hepatic Carcinogenesis. Hepatology 2019; 69: 1549-1563. Murphy M, Ahn J, Walker KK, Hoffman WH, Evans RM, Levine AJ et al . Transcriptional repression by wild-type p53 utilizes histone deacetylases, mediated by interaction with mSin3a. Genes Dev 1999; 13: 2490-2501. Pike ST, Rajendra R, Artzt K, Appling DR. Mitochondrial C1-tetrahydrofolate synthase (MTHFD1L) supports the flow of mitochondrial one-carbon units into the methyl cycle in embryos. J Biol Chem 2010; 285: 4612-4620. Yankova E, Blackaby W, Albertella M, Rak J, De Braekeleer E, Tsagkogeorga G et al . Small-molecule inhibition of METTL3 as a strategy against myeloid leukaemia. Nature 2021; 593: 597-601. Chen M, Wei L, Law CT, Tsang FH, Shen J, Cheng CL et al . RNA N6-methyladenosine methyltransferase-like 3 promotes liver cancer progression through YTHDF2-dependent posttranscriptional silencing of SOCS2. Hepatology 2018; 67: 2254-2270. Chen M, Tabaczewski P, Truscott SM, Van Kaer L, Stroynowski I. Hepatocytes express abundant surface class I MHC and efficiently use transporter associated with antigen processing, tapasin, and low molecular weight polypeptide proteasome subunit components of antigen processing and presentation pathway. J Immunol 2005; 175: 1047-1055. Bertolino P, Trescol-Biemont MC, Rabourdin-Combe C. Hepatocytes induce functional activation of naive CD8+ T lymphocytes but fail to promote survival. Eur J Immunol 1998; 28: 221-236. Gubin MM, Vesely MD. Cancer Immunoediting in the Era of Immuno-oncology. Clin Cancer Res 2022; 28: 3917-3928. Katz SF, Lechel A, Obenauf AC, Begus-Nahrmann Y, Kraus JM, Hoffmann EM et al . Disruption of Trp53 in livers of mice induces formation of carcinomas with bilineal differentiation. Gastroenterology 2012; 142: 1229-1239 e1223. Makino Y, Hikita H, Fukumoto K, Sung JH, Sakano Y, Murai K et al . Constitutive Activation of the Tumor Suppressor p53 in Hepatocytes Paradoxically Promotes Non-Cell Autonomous Liver Carcinogenesis. Cancer Res 2022; 82: 2860-2873. Levine AJ, Jenkins NA, Copeland NG. The Roles of Initiating Truncal Mutations in Human Cancers: The Order of Mutations and Tumor Cell Type Matters. Cancer Cell 2019; 35: 10-15. Schulze K, Imbeaud S, Letouze E, Alexandrov LB, Calderaro J, Rebouissou S et al . Exome sequencing of hepatocellular carcinomas identifies new mutational signatures and potential therapeutic targets. Nat Genet 2015; 47: 505-511. Kurki S, Peltonen K, Latonen L, Kiviharju TM, Ojala PM, Meek D et al . Nucleolar protein NPM interacts with HDM2 and protects tumor suppressor protein p53 from HDM2-mediated degradation. Cancer Cell 2004; 5: 465-475. Cuthbertson CR, Arabzada Z, Bankhead A, 3rd, Kyani A, Neamati N. A Review of Small-Molecule Inhibitors of One-Carbon Enzymes: SHMT2 and MTHFD2 in the Spotlight. ACS Pharmacol Transl Sci 2021; 4: 624-646. Lee D, Xu IM, Chiu DK, Lai RK, Tse AP, Lan Li L et al . Folate cycle enzyme MTHFD1L confers metabolic advantages in hepatocellular carcinoma. J Clin Invest 2017; 127: 1856-1872. Dong LQ, Peng LH, Ma LJ, Liu DB, Zhang S, Luo SZ et al . Heterogeneous immunogenomic features and distinct escape mechanisms in multifocal hepatocellular carcinoma. J Hepatol 2020; 72: 896-908. Ma W, Wu T. RNA m6A modification in liver biology and its implication in hepatic diseases and carcinogenesis. Am J Physiol Cell Physiol 2022; 323: C1190-C1205. Jain S, Koziej L, Poulis P, Kaczmarczyk I, Gaik M, Rawski M et al . Modulation of translational decoding by m(6)A modification of mRNA. Nat Commun 2023; 14: 4784. Abboud Y, Ismail M, Khan H, Medina-Morales E, Alsakarneh S, Jaber F et al . Hepatocellular Carcinoma Incidence and Mortality in the USA by Sex, Age, and Race: A Nationwide Analysis of Two Decades. J Clin Transl Hepatol 2024; 12: 172-181. He G, Yu GY, Temkin V, Ogata H, Kuntzen C, Sakurai T et al . Hepatocyte IKKbeta/NF-kappaB inhibits tumor promotion and progression by preventing oxidative stress-driven STAT3 activation. Cancer Cell 2010; 17: 286-297. Chen X, Lu Q, Zhou H, Liu J, Nadorp B, Lasry A et al . A membrane-associated MHC-I inhibitory axis for cancer immune evasion. Cell 2023; 186: 3903-3920 e3921. Duhen T, Duhen R, Montler R, Moses J, Moudgil T, de Miranda NF et al . Co-expression of CD39 and CD103 identifies tumor-reactive CD8 T cells in human solid tumors. Nat Commun 2018; 9: 2724. Additional Declarations There is no duality of interest Supplementary Files RawimagesforWBDKIHCC02222026.pdf raw WB and data images for CDD supplinfoforCDD.pdf zhang et al suppl infor Cite Share Download PDF Status: Under Review Version 1 posted Review # 1 received at journal 13 May, 2026 Reviewer # 3 agreed at journal 30 Apr, 2026 Reviewer # 2 agreed at journal 30 Apr, 2026 Reviewer # 1 agreed at journal 29 Apr, 2026 Reviewers invited by journal 21 Apr, 2026 Submission checks completed at journal 21 Apr, 2026 First submitted to journal 20 Apr, 2026 Editor assigned by journal 20 Apr, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9474831","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":627010196,"identity":"8c46f2c3-5d63-4423-9bce-362a7d574dec","order_by":0,"name":"Hua Lu","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAs0lEQVRIiWNgGAWjYBACA3YGxgcfGBhkIFw2YrQwMzAbzkhg4CFJC5s0D0lazJl5DKRtfxzm4Z/dfIDhQ9lhwlosm3kMjHMSDvNI3DmWwDjjHBFaDA7zGCSDtDDcyDFg5m0jUsthC6AW+Rv5H5j/EqnFsJkBqMXgRg4DMyNxWtiKGXvS0nkMb6QZHOw5l06EluPN23/8sLGWk7uR/PDBjzJrwlpQwAES1Y+CUTAKRsEowAUAUO425GqiadcAAAAASUVORK5CYII=","orcid":"https://orcid.org/0000-0002-9285-7209","institution":"Tulane University School of Medicine","correspondingAuthor":true,"prefix":"","firstName":"Hua","middleName":"","lastName":"Lu","suffix":""},{"id":627010197,"identity":"559b0008-1d5f-48c5-bf44-9f9cbca8ce8e","order_by":1,"name":"Yi-Wei Zhang","email":"","orcid":"","institution":"Tulane University School of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Yi-Wei","middleName":"","lastName":"Zhang","suffix":""},{"id":627010198,"identity":"9328b9e5-7ef1-46ed-99b1-0962cdb70e72","order_by":2,"name":"Gang peng","email":"","orcid":"","institution":"Indiana University School of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Gang","middleName":"","lastName":"peng","suffix":""},{"id":627010199,"identity":"64e9e5b7-a680-40b2-b803-506cbdd7386f","order_by":3,"name":"Wenjuan Liao","email":"","orcid":"","institution":"Tulane University School of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Wenjuan","middleName":"","lastName":"Liao","suffix":""},{"id":627010200,"identity":"88e1757d-5e9b-4192-87a9-3487a6f1c768","order_by":4,"name":"Ji Hoon Jung","email":"","orcid":"","institution":"Tulane University School of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Ji","middleName":"Hoon","lastName":"Jung","suffix":""},{"id":627010201,"identity":"990f5f2d-8f90-44ec-b7b5-d99554ffdb9a","order_by":5,"name":"Caiyue Li","email":"","orcid":"","institution":"Tulane University School of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Caiyue","middleName":"","lastName":"Li","suffix":""},{"id":627010202,"identity":"13a284a1-810f-4c86-8d84-8d24d910e3d2","order_by":6,"name":"Yunlong Liu","email":"","orcid":"","institution":"IUPUI","correspondingAuthor":false,"prefix":"","firstName":"Yunlong","middleName":"","lastName":"Liu","suffix":""},{"id":627010203,"identity":"36b3fb06-1bc2-4a01-8d23-e10a7f724fe3","order_by":7,"name":"Hong Gao","email":"","orcid":"","institution":"Indiana University School of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Hong","middleName":"","lastName":"Gao","suffix":""},{"id":627010204,"identity":"7f43b765-c11d-4af0-b0df-4710b13c665e","order_by":8,"name":"Yitian Zha","email":"","orcid":"","institution":"Tulane University School of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Yitian","middleName":"","lastName":"Zha","suffix":""},{"id":627010205,"identity":"988b87fa-261f-457b-b6bc-53e866cfbabb","order_by":9,"name":"Nathan Pham","email":"","orcid":"","institution":"Tulane University School of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Nathan","middleName":"","lastName":"Pham","suffix":""},{"id":627010206,"identity":"12095209-df38-4e8d-a0c1-d67e9506578b","order_by":10,"name":"Stephen Meaderis","email":"","orcid":"","institution":"Tulane University School of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Stephen","middleName":"","lastName":"Meaderis","suffix":""},{"id":627010207,"identity":"b98d14ab-f1b9-4b0c-8e31-d48c1f38bdc2","order_by":11,"name":"Evan Eslter","email":"","orcid":"","institution":"Tulane University School of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Evan","middleName":"","lastName":"Eslter","suffix":""},{"id":627010208,"identity":"e6b6db52-0abb-4978-a54b-99910062660c","order_by":12,"name":"Yanping Zhang","email":"","orcid":"","institution":"UNC","correspondingAuthor":false,"prefix":"","firstName":"Yanping","middleName":"","lastName":"Zhang","suffix":""},{"id":627010209,"identity":"d25a710e-d0fa-4af4-8b74-960d23d40c20","order_by":13,"name":"Debanjan Dhar","email":"","orcid":"","institution":"University of California at San Diego","correspondingAuthor":false,"prefix":"","firstName":"Debanjan","middleName":"","lastName":"Dhar","suffix":""},{"id":627010210,"identity":"2665bb75-df64-46a8-9f75-abc8c8a718c8","order_by":14,"name":"Shelya Zeng","email":"","orcid":"https://orcid.org/0000-0001-5195-9420","institution":"Tulane University- school of medicine","correspondingAuthor":false,"prefix":"","firstName":"Shelya","middleName":"","lastName":"Zeng","suffix":""}],"badges":[],"createdAt":"2026-04-20 16:27:15","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9474831/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9474831/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":108157365,"identity":"5e912ee6-41a0-46f0-935c-60d218496c29","added_by":"auto","created_at":"2026-04-30 02:55:53","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":3536883,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eMale DKI mice show increased DEN-induced HCC incidence.\u003c/strong\u003e\u003cbr\u003e\nFifteen-day-old male mice were treated with DEN (25mg/kg), and livers were collected 9 months later.\u003cbr\u003e\n \u003cstrong\u003e(A)\u003c/strong\u003e Gross liver morphology and tumor nodule counts via macroscopy.\u003cbr\u003e\n \u003cstrong\u003e(B)\u003c/strong\u003e H\u0026amp;E staining on liver sections, and tumor area per large lobe (L.L.) quantified via ImageJ.\u003cbr\u003e\n \u003cstrong\u003e(C–D)\u003c/strong\u003e Western blot (C) and qPCR (D) analyses of tumor and adjacent tissues (n = 4 for qPCR).\u003cbr\u003e\nData shown as mean ± SEM. (T, tumor)\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-9474831/v1/e8747e2982f748822df9f264.png"},{"id":108157367,"identity":"bd257209-f6e8-4c7f-a3cb-ad47966bdc30","added_by":"auto","created_at":"2026-04-30 02:55:53","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":7980418,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eDKI mice show impaired p53 activation and accelerated HCC initiation after DEN treatment.\u003c/strong\u003e\u003cbr\u003e\n \u003cstrong\u003e(A)\u003c/strong\u003e Liver sections from DEN (25mg/kg)-treated 15-day-old males were stained for p53. \u003cbr\u003e\n \u003cstrong\u003e(B)\u003c/strong\u003e 48h post-DEN, liver sections were stained with anti-p53, flowed by cyanine 3 (Cy3) TSA-plus incubation. p53+ nuclei in zone 3 per high magnification field (HMF) were quantified. (n=3-4/group)\u003cbr\u003e\n \u003cstrong\u003e(C–D)\u003c/strong\u003e Co-IP was performed on liver tissues 6h post-DEN.\u003cbr\u003e\n \u003cstrong\u003e(E)\u003c/strong\u003e Liver sections were stained for Ki67 to assess cell proliferation 48h post-DEN. (n=3-5/group)\u003cbr\u003e\n \u003cstrong\u003e(F–H)\u003c/strong\u003e Livers were harvested 1 (F-G) or 4 (H) months post-DEN for WB (F), qPCR (G), and immunostaining (H). (n=3-4/group)\u003c/p\u003e\n\u003cp\u003e(C, central vein; ND, not detectable)\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-9474831/v1/0b48312004bca7d34c98fbc6.png"},{"id":108157375,"identity":"7dab1200-e757-4b80-a0d2-5fe0e7f0d307","added_by":"auto","created_at":"2026-04-30 02:55:54","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":6527239,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eDKI-HCICs with reduced p53 show enhanced stemness, malignancy, and tumorigenicity.\u003c/strong\u003e\u003cbr\u003e\n \u003cstrong\u003e(A–B)\u003c/strong\u003e HCIC-enriched clusters were isolated 3 months post-DEN and analyzed by WB (A) and qPCR (B, n=3-5/group).\u003cbr\u003e\n \u003cstrong\u003e(C–D)\u003c/strong\u003e Cultured HCICs were stained for cancer stem cell (CSC)/malignancy markers (C) and subjected to WB (D).\u003cbr\u003e\n \u003cstrong\u003e(E–G)\u003c/strong\u003e Cultured HCICs were assessed for spheroid formation (E), growth and colony formation (F), and proliferation (G). Experiments were repeated 4-5 times.\u003cbr\u003e\n \u003cstrong\u003e(H–K)\u003c/strong\u003e Freshly isolated CTL and DKI-HCICs were transplanted into C57BL/6J mice via intra-splenic injection. Liver morphology (H), liver/body weight ratio and nodule count (I), H\u0026amp;E staining (J), and tumor area (K, using \u003cem\u003eImage J\u003c/em\u003e) were evaluated at 8 and/or 12 months post-transplantation. (n=7-8/group)\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-9474831/v1/a8dcd298dd33987ab876c7fe.png"},{"id":108182943,"identity":"f5053ea6-7ec0-4825-93db-d0f33c8fbf1c","added_by":"auto","created_at":"2026-04-30 08:59:42","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":7622437,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eMthfd1l is essential for accelerated HCC initiation in DKI mice.\u003c/strong\u003e\u003cbr\u003e\n \u003cstrong\u003e(A–B)\u003c/strong\u003e Livers from DEN-treated 15-day-old males were analyzed for Mthfd1l expression by WB and qPCR at 6h or 48h.\u003cbr\u003e\n \u003cstrong\u003e(C–D)\u003c/strong\u003e CTL or DKI-HCICs with wild-type (MTH-WT) or CRISPR-depleted (MTH-D) Mthfd1l were treated with vehicle or formate (1 mM) for 48h and analyzed by immunostaining (C) and spheroid assay (D). Experiments were repeated 3-5 times for each HCICs.\u003cbr\u003e\n \u003cstrong\u003e(E–G)\u003c/strong\u003e MTH-D DKI-HCICs re-expressing Mthfd1l or vector control were subjected to IF staining (E), spheroid formation (F), and growth/colony formation assays (G). Experiments were repeated 3-5 times.\u003cbr\u003e\n \u003cstrong\u003e(H–J)\u003c/strong\u003e MTH-WT or MTH-D DKI-HCICs were transplanted into C57BL/6J mice via intra-splenic injection. After 10 months, liver morphology and tumor nodules were assessed (H–I), followed by IHC analysis (J).\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-9474831/v1/2307accbef31c2cdf371190d.png"},{"id":108157370,"identity":"3021fa71-f418-4bff-884c-78bca4e50cd0","added_by":"auto","created_at":"2026-04-30 02:55:53","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":987985,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eMthfd1l promotes Snail mRNA m\u003c/strong\u003e\u003csup\u003e\u003cstrong\u003e6\u003c/strong\u003e\u003c/sup\u003e\u003cstrong\u003eA methylation to enhance HCIC proliferation.\u003c/strong\u003e\u003cbr\u003e\n\u003cstrong\u003e(A)\u003c/strong\u003e MTH-WT and MTH-D HCICs were transfected with Mthfd1l or vector, followed by m\u003csup\u003e6\u003c/sup\u003eA Dot-blot. Experiments were repeated with 3 different HCIC cell lines, and representative images were shown.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(B-F)\u003c/strong\u003e WB was performed on MTH-WT or MTH-D HCICs with treatments or gene modifications including SAM (15 and 30 mM), Mettl3 or Mthfd1l knockdown, and Mthfd1l or p53 overexpression.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(G)\u003c/strong\u003e p53-null Hep3B cells were transiently transfected with Mthfd1l and/or p53, followed by WB.\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-9474831/v1/ebf41ce53686db05b8e47323.png"},{"id":108182484,"identity":"3f81a256-9fbb-469c-b591-c1790d3a4ccb","added_by":"auto","created_at":"2026-04-30 08:59:23","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":5828120,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eMthfd1l promotes B2m mRNA m\u003c/strong\u003e\u003csup\u003e\u003cstrong\u003e6\u003c/strong\u003e\u003c/sup\u003e\u003cstrong\u003eA methylation in HCICs to suppress CD8+ T cell infiltration and cytotoxicity.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(A-E)\u003c/strong\u003e WB was performed on MTH-WT or MTH-D HCICs with treatments or gene modifications including SAM, Mettl3 or Mthfd1l knockdown, and Mthfd1l or p53 overexpression. Images in (A) were from the same WB of Figure 5B, and the samle loading control images were presented. Images in (E) were from the same WB of Figure 5F. \u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(F)\u003c/strong\u003e Hep3B cells were transiently transfected with Mthfd1l and/or p53, followed by WB.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(G)\u003c/strong\u003e After gene knockdown or STM pre-treatment, HCIC-1_OVA cells were co-cultured with OT-1 CD8+ T cells, and caspase 3/7 activity (green fluorescence) was monitored under Incucyte.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e(H)\u003c/strong\u003e HCIC-1_OVA cells transfected with Mthfd1l and/or p53 or pretreated with STM were co-cultured with OT-1 CD8+ T and tested by LDH cytotoxicity assay.\u003c/p\u003e\n\u003cp\u003e(\u003cstrong\u003eI\u003c/strong\u003e) Sections of mouse liver tissue containing tumors derived from MTH-WT and WTH-D HCIC transplantation were stained with anti-CD8a antibody. Arrows indicate CD8+ T cells. (T, tumor area)\u003c/p\u003e","description":"","filename":"6.png","url":"https://assets-eu.researchsquare.com/files/rs-9474831/v1/46f136facdf7ed4d0dd76539.png"},{"id":108157372,"identity":"e310733c-8daf-4818-b17e-c8f1e155d0eb","added_by":"auto","created_at":"2026-04-30 02:55:53","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":4867245,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eThe two p53-activating pathways suppress HFD-induced HCC initiation and are activated in human cirrhotic livers.\u003c/strong\u003e\u003cbr\u003e\n \u003cstrong\u003e(A–E)\u003c/strong\u003e CTL and DKI mice were fed a high-fat diet (HFD) for 15 months. Tumor incidence (A), liver morphology (B), liver/body weight ratio and tumor count (C), WB (D), and IHC (E) analyses were performed. (CD: chow diet)\u003cbr\u003e\n \u003cstrong\u003e(F–H)\u003c/strong\u003e Human liver samples (normal, steatohepatitis, and cirrhosis) were analyzed by WB (F) and Co-IP (G, H).\u003c/p\u003e","description":"","filename":"7.png","url":"https://assets-eu.researchsquare.com/files/rs-9474831/v1/ee69538de3cefa15ccd1ec0f.png"},{"id":108157373,"identity":"6da38dfd-b4a8-4ff2-b836-8e7ebc6e32b0","added_by":"auto","created_at":"2026-04-30 02:55:53","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":2207675,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eModel of the Stress - RPs-MDM2 / 14-3-3-MDMX -p53-Mthfd1l-RNA m\u003c/strong\u003e\u003csup\u003e\u003cstrong\u003e6\u003c/strong\u003e\u003c/sup\u003e\u003cstrong\u003eA-Snail/B2m axis in HCC initiation.\u003c/strong\u003e\u003cbr\u003e\nIn DKI mice, impaired p53 response to stress from carcinogens, NASH, or cirrhosis leads to reduced suppression of Mthfd1l. Elevated Mthfd1l promotes HCIC proliferation and immune evasion, accelerating transformation into HCC. (Red “X”s indicate disrupted pathway; arrows present activation; bars present inhibition.)\u003c/p\u003e","description":"","filename":"8.png","url":"https://assets-eu.researchsquare.com/files/rs-9474831/v1/d8fcf00bb25537e762713408.png"},{"id":108490931,"identity":"32089aba-4f61-467c-a0a5-316c76cc9a59","added_by":"auto","created_at":"2026-05-05 09:50:09","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":37780746,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9474831/v1/26b8d1ed-fab8-4eb5-b1d6-66409dfebfdf.pdf"},{"id":108182527,"identity":"bee5ce2d-3db9-4b2e-a220-35206e9bf96c","added_by":"auto","created_at":"2026-04-30 08:59:25","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":2110291,"visible":true,"origin":"","legend":"raw WB and data images for CDD","description":"","filename":"RawimagesforWBDKIHCC02222026.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9474831/v1/fb014d52bee97a8abae38373.pdf"},{"id":108182705,"identity":"64f0f475-c0b5-48fd-9963-3423a6687e12","added_by":"auto","created_at":"2026-04-30 08:59:30","extension":"pdf","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":3852074,"visible":true,"origin":"","legend":"zhang et al suppl infor","description":"","filename":"supplinfoforCDD.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9474831/v1/c9cc62cd7efa694b181524f8.pdf"}],"financialInterests":"There is no duality of interest","formattedTitle":"\u003cp\u003eDual-stress-responsive-pathways prevent HCC initiation via p53 suppression of MTHFD1L-RNA m\u003csup\u003e6\u003c/sup\u003eA-elicited autonomous growth and immune evasion\u003c/p\u003e","fulltext":[{"header":"INTRODUCTION","content":"\u003cp\u003eHepatocellular carcinoma (HCC), the fifth most common cancer worldwide and the third leading cause of cancer-related deaths [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e], originates from injured hepatocytes that dedifferentiate into progenitor-like cells (or cancer initiating cells, CICs) expressing stem cell markers (e.g., CD44 and EpCAM). These cells expand to acquire the features of fully malignant cells in response to aberrant proliferative stimuli [\u003cspan additionalcitationids=\"CR4 CR5\" citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]. This transformation usually occurs in late-stage liver injury that involves closely intertwining stress responses and pathological processes, including endoplasmic reticulum (ER) and oxidative stresses, chronic inflammation, hepatocyte death, liver regeneration, somatic DNA alterations, fibrosis, and cirrhosis [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e, \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eIn normal hepatocytes without stress, the tumor suppressor p53 level is kept low by MDM2 and MDMX that work together to inactivate p53 via a negative feedback [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]. However, during chronic liver injury, p53 activity is dynamically regulated. Early in regeneration, p53 is tightly controlled to enable the hepatocyte de-differentiation and division, but activated later to ensure the genomic integrity, keep the robustness and fidelity of cell proliferation, and prevent neoplasia transformation [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]. Several studies have shown that p53 inhibition by oncoproteins leads to early hepatocarcinogenesis [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e], suggesting that p53 must be activated prior to its inactivation. However, it remains unaddressed whether and how chornic liver injury-associated stress signals activate p53 under physiological conditions.\u003c/p\u003e \u003cp\u003eTwo stress responsive signaling pathways can activate p53 by overcoming the MDM2/MDMX-mediated feedback: 1) the ribosomal proteins (RPs)-MDM2-p53 pathway, where ribosomal stress (RS) frees RPs such as RPL11 and RPL5 to bind the MDM2 central domain, and release p53 [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]; 2) the 14-3-3-MDMX-p53 pathway, where DNA damage or glucose deprivation-induced MDMX phosphorylation enables 14-3-3 binding to block MDMX suppression of p53 [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]. Both the pathways were independently validated by genetic knock-in (KI) mice: MDM2\u003csup\u003eC305F\u003c/sup\u003e mice (M2-C305F) contain a point mutation at the central domain C305 position, to disable its binding to RPL5 and RPL11 [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e], and MDMX\u003csup\u003e3SA\u003c/sup\u003e mice (MX-3SA) contain 3 serine mutations in the MDMX gene (Serine 367, 341 and 402) so that this MDMX mutant cannot be phosphorylated in response to stresses [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]. Importantly, MDM2 truncations (AA1-302) of its central zinc finger domain (AA305-322) were previously identified in HCC [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e], and our bioinformatic analysis also unveiled an MDM2 somatic point mutation (E316D) in this domain in HCC patients (Figure S24A). However, the role of these MDM2 central domain mutations in HCC has never been addressed. Besides, chronic liver injury-related stresses as mentioned above often induce ribosome dysfunction or activate kinases that phosphorylate MDMX at S342 or S367 [\u003cspan additionalcitationids=\"CR14\" citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e, \u003cspan additionalcitationids=\"CR20 CR21\" citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]. These studies highly suggest that the two stress responsive pathways may play a preventive role in HCC initiation.\u003c/p\u003e \u003cp\u003eTo address this, we generated an MDMX/MDM2-double knock-in (DKI) mouse model by crossing the two aforementioned KI mouse lines, and induced HCC in the mice with administration of hepatic procarcinogen diethylnitrosamine (DEN). DEN-induced HCC mimics many of human HCC features [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e]. The DKI mice exhibited significantly higher rates of DEN-induced HCC. HCC initiating cells (HCICs) isolated from these mice displayed enhanced tumorigenesis when introduced into recipient livers. DKI mice under a long-term high-fat diet (HFD) also developed more liver tumors. Moreover, the two p53-pathways were turned on in human livers with cirrhosis, the precancerous condition for HCC [\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e]. We further identified methylenetetrahydrofolate dehydrogenase 1 like (MTHFD1L), a key mitochondrial one-carbon metabolism (1CM) enzyme producing formate [\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e], as a p53-suppressed player in HCC initiation. We found that Mthfd1l expression is declined in cirrhotic liver, while upregulated and highly correlated to the p53 mutation status in HCC patients. Mthfd1l depletion in DKI-HCICs significantly delayed HCC onset after intra-splenic transplantation. Furthermore, by fueling the methionine cycle MTHFD1L enhanced the N6-methyladenosine (m\u003csup\u003e6\u003c/sup\u003eA) modification of Snail [\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e, \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e] mRNA to boost HCIC growth and stemness, as well as the modification of B2m (β2m) mRNA, a light chain of MHC-I molecules [\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e], which significantlly hampered CD8\u0026thinsp;+\u0026thinsp;T liver tumor infiltration and cytoxicity against HCICs. Hence, our studies reveal a new MDM2/MDMX-p53-Mthfd1l-RNA m\u003csup\u003e6\u003c/sup\u003eA-Snail/B2m axis and its novel role in suppressing HCIC autonomous growth and immune evasion during HCC initiation, providing new insights into the regulation of p53-pathways on metabolic reprogramming, RNA epigenetics, and cancer immunoediting during cacer initiation. Our study also illustrates the bi-functions of MTHFD1L and its potential as a diagnostic biomarker and a therapeutic target for early HCC.\u003c/p\u003e"},{"header":"RESULTS","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eDEN induced-HCC occurs significantly higher in DKI than in CTL mice\u003c/h2\u003e \u003cp\u003eTo determine if the two p53-activating pathways play a role in preventing HCC initiation, we generated an M2-C305F/MX-3SA DKI mouse line (Figure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e). Control (CTL), DKI, M2-C305F and MX-3SA mice were obtained in a normal Mendelian ratio and employed for the experiments. All the mice were without significant defects at both postnatal and adult stages under normal conditions (Figure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003eB).\u003c/p\u003e \u003cp\u003eWe injected a single dose of DEN (25mg/kg) via i.p. into 2-week-old male and female mice [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. All the males and most the females developed malignant multi-nodules in livers 9 months post-DEN (one female CTL had only one nodule). Consistent with previous reports [\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e, \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e], the male mice developed significantly more and larger tumors than females (Figs.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eA and S2). In both sexes, DKI mice exhibited a significantly higher tumor number and/or larger average tumor area per large lobe than other groups (Figs.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eA-B and S2), indicating that DEN induced-HCC occurrence is enhanced in DKI mice with either sex. Surprisingly, the expression of p53 and its targets p21 and MDM2 at mRNA and protein levels in both tumor and adjacent liver tissues, was comparable among the groups (Figs.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eC-D). These results implicate that p53 might be activated through the two pathways at the early, but not advanced, stage of HCC to suppress DEN-induced HCC initiation.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003ePericentral hepatocyte proliferation and HCC initiation induced by DEN are significantly increased in DKI mice\u003c/h3\u003e\n\u003cp\u003eTo test the speculation above, we examined if p53 is activated by DEN in 2-week-old male mouse livers. Indeed, p53 protein level and/or activity rapidly increased in CTL-livers 6 hours after DEN treatment and remained elevated at 48h, but not in DKI-livers (Figs.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA-B and S3). Of note, p53 induction in CTL-livers was mainly localized to the pericentral hepatocytes (zone 3), aligning with reports showing DEN- or NASH-induced HCC originates from this zone [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. Furthermore, the binding of 14-3-3γ to MDMX or L5 to MDM2 was significantly enhanced in DEN-treated CTL-livers, but not in DKI-livers (Figs.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eC-D), indicating that both pathways contribute to the DEN-induced p53 activation. Also, Ki67 staining revealed significantly higher hepatocyte proliferation in zone 3 of DKI livers at 48h post-DEN (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eE), suggesting that p53 activation through these two pathways restrains the compensatory proliferation in CTL livers.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eWe then examined the liver tissues 1 and 4 months post-DEN, when damaged hepatocytes start to transform towards HCICs[\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. Remarkably, cancer stem cell (CSC) markers (CD44, EpCAM and Nestin [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]) and liver regeneration markers were significantly elevated as early as 1 month post-DEN in DKI-livers, along with lower p53 level (Figs.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eF-G). Four months post-DEN, both CSC markers and malignancy markers (Ly6D and AFP) remained at a higher level in DKI-livers (Figs.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eH and S4), with more CD44 and AFP double-positive cells in zone 3 (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eH). Collectively, our \u003cem\u003ein vivo\u003c/em\u003e findings demonstrate that p53 activation through these two pathways prevents DEN-induced HCC initiation in CTL mice.\u003c/p\u003e\n\u003ch3\u003ep53 suppresses the stemness and malignancy properties of DKI-HCICs\u003c/h3\u003e\n\u003cp\u003eNext, we isolated HCIC-enriched cell clusters from male CTL and DKI-livers 3 months post-DEN as previously described [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e, \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e]. As expected, both fresh (Figs.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA-B and S5A) and cultured (Figs.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eC-D and S5B) DKI-HCICs exhibited significantly higher levels of CSC markers (CD44, EpCAM, KLF4) and malignancy markers, but lower p53 and p21, compared to CTL-HCICs. All HCICs were HNF4α positive (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eC), confirming their hepatocyte origin. Furthermore, significantly more spheroids were formed with DKI-HCICs (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eE), and greater cell viability was obsereved in DKI-HCICs as detected by cell growth and colony formation assays (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eF), primarily attributed to the higher proliferation of DKI-HCICs (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eG). However, ectopic p53 markedly reduced their cell viability/growth and spheroid formation as well as the expression of CSC markers (EpCAM, CD44, KLF4 and SOX9) and malignancy marker AFP (Figure S6). These \u003cem\u003ein vitro\u003c/em\u003e results demonstrate that p53 activation via the two pathways blocks the hepatocyte transformation towards HCICs.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e\n\u003ch3\u003eElevated tumorigenic capacity of transplanted DKI-HCICs in vivo\u003c/h3\u003e\n\u003cp\u003eWe then transplanted fresh CTL- and DKI-HCICs into male C57BL/6 WT mice via intra-splenic injection (Figure S7) as previously described [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e, \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e]. All DKI-HCIC recipient mice (3 out of 3) developed a tumor nodule in each liver 8 months post-transplantation, and multiple tumor nodules 12 months later (7 out of 7) (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eH). Nevertheless, none of the CTL-HCIC recipients developed palpable liver tumors by 8 months after injection. Though 12 months later 7 out of 8 CTL-HCIC recipients developed liver tumors (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eH), these tumors were much less and significantly smaller than those in DKI-HCIC recipients (Figs.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eI and \u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eK). H\u0026amp;E staining further confirmed those nodules in both groups as HCC (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eJ). Again, these results consolidate that impairment of the two p53-pathways significantly accelerates liver tumorigenesis.\u003c/p\u003e\n\u003ch3\u003eMthfd1l is crucial for HCC initiation in DKI mice\u003c/h3\u003e\n\u003cp\u003eTo identify p53 downstream players in HCC initiation of DKI mice, we performed RNA-seq with DEN-induced liver tumors and adjacent liver tissues of CTL and DKI mice. PCA and heatmap analyses revealed that the overall gene expression pattern of DKI-livers is closer to that of CTL- or DKI- tumors (Figure S8A). Pathway enrichment highlighted significant alterations in metabolic pathways, especially the monocarboxylic acid metabolism that was among the top changed metabolic pathways in DKI-liver versus CTL-liver, or CTL-tumor versus CTL-liver (Figures S8B-C). Notably, Mthfd1l, an enzyme involved in 1CM to produce formate (a type of monocarboxylic acid), was significantly upregulated in DKI-livers, CTL-tumors, DKI-tumors, and all stages of human HCC (Figures S8D-E). Its higher levels were also associated with poor survival in patients with early-stage HCC (stage I) (Figure S8F). Mthfd1l expression in 15-day-old CTL-livers, but not in DKI-livers, were drastically reduced after DEN treatment (Figs.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA and B). Four months post-DEN, DKI-livers presented a significantly higher formate levels (Figure S9). Furthermore, fresh or cultured DKI-HCICs showed significantly higher Mthfd1l expression or formate levels compared to CTL and/or SKI counterparts (Figure S10). These findings suggest that MTHFD1L-associated 1CM reprogramming might play a role in the accelerated HCC initiation in DKI mice.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eTo test this idea, we knocked out the Mthfd1l gene in CTL and DKI-HCICs using CRISPR/cas9 system. Mthfd1l was depleted in three cell lines (one CTL- and two DKI-HCIC lines from three individual mice) (Figure S11A). Nanopore direct RNA-seq showed a sharp drop of the read depth in MTHFD1L Exon 6 in HCIC-1 (CTL) and \u0026minus;\u0026thinsp;3 (DKI) cells, indicating Mthfd1l deficiency (MTH-D), however only 1 nucleotide deletion in MTH-D HCIC-2 (DKI) (Figure S11B), so we employed MTH-D HCIC-1 and \u0026minus;\u0026thinsp;3 cells for the following experiments. Mthfd1l deletion significantly diminished CSC marker expression and cell viability/growth, more so in DKI-HCICs (Figs.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eC, S11C-D and S12), and impaired spheroid formation in both the HCICs (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eD). We further tested whether formate supplementation could resuce these effects caused by Mthfd1l depletion. After formate treatment for 48h, CD44 expression in MTH-D HCICs was almost recovered to the level in Mthfd1l-proficient (MTH-WT) cells (Figs.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eC, \u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eE and S11C-D). The size, but not the number, of the spheroids was also rescued by formate (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eD). However, cell proliferation (Ki67\u0026thinsp;+\u0026thinsp;signals) and growth showed partial or minimal recovery after formate treatment (Figs.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eE and S12B). These results implicate that while formate improves some Mthfd1l deletion-caused defects, Mthfd1l might also utilize a non-enzymatic mechanism to promote growth and self-renewal of HCICs. We re-expressed Mthfd1l in MTH-D DKI-HCICs to test if these defects can be rescued. Remarkably, replenishment of Mthfd1l restored CSC marker expression (CD44 and SOX9) (Figs.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eE), spheroid formation (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eF), cell viability (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eG), and proliferation (Ki67 signals in Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eE) of MTH-D HCICs to near-normal levels. Furthermore, 10 months after intra-splenic injection, MTH-WT DKI-HCICs formed multiple tumors in recipient livers, while MTH-D cells produced significantly fewer palpable tumors, along with significantly lower liver/body weight ratio of MTH-D HCIC recipients (Figs.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eH-I). H\u0026amp;E and IHC staining confirmed that these tumors are AFP-positive HCC (Figs.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eJ and S13) with significantly stronger Ki67 signals in MTH-WT DKI-HCIC-derived tumors (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eJ). These results demonstrate that Mthfd1l activation owing to impairing the two p53 pathways in DKI mice is responsible for the accelerated HCC initiation.\u003c/p\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003ep53 transcriptionally suppresses Mthfd1l by cooperating with HDAC1/2\u003c/h2\u003e \u003cp\u003eOur bioinformatic analysis showed that though Mthfd1l is upregulated in HCC patients, its expression is significantly lower in p53-wild type than in p53-mutant HCCs (Figure S14A). Indeed, Mthfd1l expression at mRNA and protein levels was p53-dependent, comparing p53-containing cells (HepG2 liver cancer and H460 lung cancer cells) with p53-null /-mutant cells (Figures S14B-F). p53 overexpression in DKI-HCICs reduced Mthfd1l expression (Figure S14G). These data indicate that p53 transcriptionally suppresses Mthfd1l.\u003c/p\u003e \u003cp\u003eLuciferase reporter assays using a reporter vector containing two putative p53 binding sites (BS) in the human MTHFD1L promoter (Mthfd1l-BS1/2) (Figure S15A left) showed that p53 activators reduces the luciferase signals by 50\u0026thinsp;~\u0026thinsp;60% only in HepG2 (Figures S15A right). This repression was abolished by BS1 mutation (Figure S15B), indicating that p53 inhibits MTHFD1L promoter activity via the Mthfd1l-BS1. Consistently, p53 binding to the MTHFD1L, p21, and Puma promoters was confirmed by ChIP assays in flag-p53-overexpressing Hep3B cells or in doxorubicin-treated HepG2 cells (Figure S15C).\u003c/p\u003e \u003cp\u003eP53 recruits an inhibitory complex with HDACs (Figure S16A) [\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e] to suppress its targets. Accordingly, treatment with the HDAC1/2 inhibitor TSA almost fully rescued Mthfd1l expression inhibited by the p53 activators only in HepG2, but not p53-null Hep3B, cells (Figure S16B). Consistently, the DNA-protein complexes pulled down by HDAC1/2 antibodies were enriched with the DNA fragments containing Mthfd1l-BS1 of p53 (Figure S16C). These results indicate that p53 represses Mthfd1l transcription by working with HDACs in HCC cells.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eMTHFD1L promotes HCIC autonomous growth and immune evasion via RNA mA methylation\u003c/h3\u003e\n\u003cp\u003eMthfd1l generates formate to supply\u0026thinsp;~\u0026thinsp;75% of 1-C units in the methionine cycle [\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e] that produces direct methyl donor SAM for RNA m\u003csup\u003e6\u003c/sup\u003eA modificaiton. Indeed, in MTH-D HCICs, global mRNA m\u003csup\u003e6\u003c/sup\u003eA levels were drastically declined, but recovered by ectopic Mthfd1l (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA). SAM significantly boosted CSC marker expression, spheroid formation, and growth of MTH-D HCICs (Figures S17A-C), while treatment with STM2457 (STM), an inhibitor of the major RNA m\u003csup\u003e6\u003c/sup\u003eA methylation writer methyltransferase like 3 (METTL3) [\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e], remarkably suppressed these traits of HCICs (Figures S17D-G). Also, our database analysis showed that a higher METTL3 level is well correlated with poorer survival in HCC patients (Figure S17H). These results suggest that MTHFD1L promotes HCC initiation by enhancing RNA m\u003csup\u003e6\u003c/sup\u003eA methylation.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eTo test this idea, we performed Nanopore RNA m\u003csup\u003e6\u003c/sup\u003eA-seq with paired MTH-WT and MTH-D HCIC cells (HCIC-1 and HCIC-3). It showed a global reduction of RNA m\u003csup\u003e6\u003c/sup\u003eA in MTH-D cells (Figure S18A) and identified\u0026thinsp;~\u0026thinsp;46 protein-coding genes with an average modification difference \u0026ge; 5%. As one of the top 10 genes (Figures S18B-D), Socs2 was used as a positive control. Its decreased protein level due to the elevation of its mRNA m\u003csup\u003e6\u003c/sup\u003eA is linked to HCC progression [\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e]. Then the rest top candidates (Figure S18B) were screened by their response to SAM treatment in MTH-D HCICs (Figs.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eB and \u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eA). Snail, a well-known EMT player and newly identified HCC early tumorigenesis driver [\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e, \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e], was upregulated by SAM (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eB). Conversely, in HCICs when knocking down Mthfd1l or Mettl3 (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eC), or in MTH-D HCICs when comparing with MTH-WT cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eD), Snail was dramatically reduced. These effects were reversed by re-expression of Mthfd1l in MTH-D HCICs (Figs.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eD-E), which was further withdrawn by STM treatment (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eE). P53 also suppressed Snail in both HCICs and Hep3B cells, but these effects were negated by Mthfd1l co-expression (Figs.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eF-G). However, knockdown of Mettl3 or Mthfd1l, re-expressing Mthfd1l in MTH-D HCICs, or treatment of STM exerted no effect on Snail mRNA level (Figures S20A-B). Besides, we found that the m\u003csup\u003e6\u003c/sup\u003eA mainly exists in the 3\u0026rsquo;-UTR region of Snail mRNA (Figure S18C). We further identified several m\u003csup\u003e6\u003c/sup\u003eA sites within that region. Those sites displayed m\u003csup\u003e6\u003c/sup\u003eA level\u0026thinsp;\u0026gt;\u0026thinsp;=\u0026thinsp;15% in MTH-WT HCICs and the modification decline\u0026thinsp;\u0026gt;\u0026thinsp;=\u0026thinsp;5% in MTH-D vs. MTH-WT cells (Figures S18C-D). Taken together, these findings demonstrate that the p53-MTHFD1L axis controls cell growth and stemness during HCC initiation, by modulating mRNA m\u003csup\u003e6\u003c/sup\u003eA to translationally upregulate Snail.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eHepatocytes express abundant cell surface MHC-I, and can rapidly load and present antigens[\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e, \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e]. In MTH-D HCICs, the mRNA m\u003csup\u003e6\u003c/sup\u003eA level of β2-microglobulin (β2m, B2m), the light chain of MHC-I molecules, was reduced (Figures S18B-D). The nanopore m\u003csup\u003e6\u003c/sup\u003eA-seq analysis indicated a major m\u003csup\u003e6\u003c/sup\u003eA site in the coding region (CDS) of B2m mRNA in MTH-WT HCICs, which was decreased by 10% in MTH-D cells (Figures S18C-D). In contrast to the response of SNAIL to SAM treatment, the B2M protein level of MTH-D HCICs was dramatically decreased by SAM (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eA). Downregulation of B2m or MHC-I molecules dampens antigen presentation, impairs CD8\u0026thinsp;+\u0026thinsp;T cell cytotoxicity and contributes to cancer immunotherapy resistance [\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e, \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e]. Consistently, B2m is positively correlated with HCC patient survival (Figure S19). Also, Mthfd1l knockdown remarkably increased (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eB), while its overexpression suppressed, β2m expression in MTH-WT HCICs, which was rescued by Mettl3 knockdown (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eC). β2m protein level was doubled in MTH-D HCICs compared to MTH-WT cells, wheras Mthfd1l re-expression counteracted the higher β2m expression in MTH-D cells, which was recovered by STM treatment (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eD). P53 upregulated β2m in both HCIC and Hep3B cells (Figs.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eE-F), but this effect was eliminated by Mthfd1l overexpression (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eF). Notably, B2m mRNA levels were unchanged after Mthfd1l re-expression, STM treatment, Mthfd1l or Mettl3 knockdown in MTH-D or MTH-WT HCICs (Figures S20B-C), indicating a translational silence of m\u003csup\u003e6\u003c/sup\u003eA modified B2m mRNA. To validate the role of β2m in HCIC antigen presentation and subsequent CD8\u0026thinsp;+\u0026thinsp;T cell activation, we co-cultured HCICs stably expressing chicken ovalbumin antigen (HCIC_OVA) (Figure S21A) with CD8\u0026thinsp;+\u0026thinsp;T cells freshly isolated from C57 BL/6 OT-1 mouse spleen. OT-1 mice contain transgenic T cell receptor that recognizes OVA peptides presented by MHC-I. After co-culturing for 24h, CD8\u0026thinsp;+\u0026thinsp;T cytotoxicity against HCIC, indicated by the caspase 3/7 signal, was enhanced by 1.5 times by Mthfd1l or Mettl3 knockdown in HCICs, while significantly weakened by B2m knockdown, which was mostly rescued by STM pre-treament on HCICs (Figures S21B and 6G). Furthermore, Mthfd1l overexpression in HCICs significantly inhibited p53-induced CD8\u0026thinsp;+\u0026thinsp;T cytotoxicity, whereas STM pre-treatment reversed Mthfd1l-caused decline of CD8\u0026thinsp;+\u0026thinsp;T cytotoxicity against HCICs (Figures S21B and 6H). Consistently, MTH-deficient (MTH-D) HCICs transplantation-derived liver tumors showed significantly increased CD8\u0026thinsp;+\u0026thinsp;T cell infiltration \u003cem\u003ein vivo\u003c/em\u003e (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eI). Together, these \u003cem\u003ein vitro\u003c/em\u003e and \u003cem\u003ein vivo\u003c/em\u003e results indicate that MTHFD1L promotes the immune evasion of HCICs by suppressing β2m-dependent antigen presentation and CD8\u0026thinsp;+\u0026thinsp;T cytotoxicity via m\u003csup\u003e6\u003c/sup\u003eA-mediated translational control of B2m.\u003c/p\u003e\n\u003ch3\u003eTwo p53-pathways play a preventative role in NASH- and cirrhosis-associated HCC initiation\u003c/h3\u003e\n\u003cp\u003eTo determine if these two p53-pathways are triggered to prevent the NASH-associated HCC initiation, we fed CTL and DKI mice with a long-term HFD (LHF) for 15 months to mimic the human NASH-driven HCC. Both groups comparably developed severe steatosis and mild interstitial fibrosis (Figure S22), with tumors in ~\u0026thinsp;75% of CTL and ~\u0026thinsp;85% of DKI livers (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eA). Notably, all DKI mice developed multiple (\u0026ge;3) tumor nodules, alongside with significantly higher liver/body weight ratio and more small nodules (diameter\u0026thinsp;\u0026le;\u0026thinsp;2mm) in livers, whereas none of the CTL livers had more than two nodules (Figs.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eA-C). The HCC tumor size usually indicates tumor progression, while tumor number is an index of early tumorigenesis. Expectedly, the tumor nodules in both groups expressed a high level of AFP (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eE). However, p53 (and p21) was remarkably induced only in CTL-tumors compared to the adjacent liver tissues, but not in DKI-tumors, correlating with down-regulated MTHFD1L in CTL-tumors (Figs.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eD-E). Together, these data show that impairment of the two p53-pathways enhances NASH-driven HCC initiation.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eWe further assessed the p53-MTHFD1L signaling in human liver cirrhosis (Figure S23), the HCC precancerous condition [\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e]. As shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eF, human alcoholic cirrhotic livers presented strong p53 induction, increased MDM2 and downregulated MTHFD1L. IP-WB analysis showed remarkably elevated formation of MDM2-RPL5 and MDMX-14-3-3γ complexes in cirrhotic versus normal livers (Figs.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003eG-H). These results support an active role of the two p53-pathways in preventing human hepatocyte malignant transformation in cirrhosis by suppressing Mthfd1l.\u003c/p\u003e"},{"header":"DISCUSSION","content":"\u003cp\u003eUsing a DKI animal model, we have addressed the remaining crucial question: how is p53 rapidly activated prior to its inactivation during chronic injury to block HCIC transformation and HCC initiation? We for the first time identified a dual mode of p53 activation by inactivating both MDM2 and MDMX in response to various stresses to prevent HCC initiation. We showed that imparing the two p53-activating pathways markedly accelerates DEN- or NASH-driven HCC initiation in DKI mice (Figs.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e–\u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e, S4, \u003cspan class=\"InternalRef\"\u003e7\u003c/span\u003eA-E and \u003cspan class=\"InternalRef\"\u003e8\u003c/span\u003e), DKI-HCICs exhibit higher malignancy and stemness (Figs.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003eA-G) which can be reversed by p53 (Figure S6), and DKI-HCICs develop into HCC more severely (Figs.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003eH-K). Correlatedly, p53 was activated through these two pathways in cirrhotic patient livers (Figs.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e7\u003c/span\u003eF-H). About 80% of HCCs are developed from cirrhosis that is recognized as a precancerous state [\u003cspan class=\"CitationRef\"\u003e24\u003c/span\u003e, \u003cspan class=\"CitationRef\"\u003e25\u003c/span\u003e]. In line with the literature, our results consolidate the biological significance of p53 activation through the two pathways in preventing HCC initiation.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eWhile hepatocyte specific deletion of p53 spontaneously and consistently induces liver carcinoma formation [\u003cspan class=\"CitationRef\"\u003e40\u003c/span\u003e], continuous p53 activation in hepatocytes (e.g., via MDM2 deletion) also promotes liver tumorigenesis [\u003cspan class=\"CitationRef\"\u003e41\u003c/span\u003e], indicating that tight regulation on p53 activity is needed for preventing hepatocarcinogenesis. Different from other genetic models, our unique DKI mice retain WT p53, providing a special tool for studying p53’s function in real time. Indeed, p53 activation is observed mainly in the end-stage of liver disease, i.e. cirrhotic liver (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e7\u003c/span\u003eF), or in early HCC stages, but blunted in DKI mice (Figs.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003eF and \u003cspan class=\"InternalRef\"\u003e7\u003c/span\u003eD-E), suggesting a dynamic regulation of p53 activity by the two pathways during chronic liver injury and HCC initiation. Interestingly, similar to their SKI parents [\u003cspan class=\"CitationRef\"\u003e16\u003c/span\u003e, \u003cspan class=\"CitationRef\"\u003e17\u003c/span\u003e], DKI mice are developmentally normal (Figure \u003cspan class=\"InternalRef\"\u003eS1\u003c/span\u003e). Thus, the basal level of p53 in DKI mice is sufficient for maintaining normal liver function, but insufficient for protecting the animals from stresses-associated HCC tumorigenesis, highlighting the surveillance role of the two p53-pathways in early hepatocarcinogenesis long before somatic p53 mutations occur [\u003cspan class=\"CitationRef\"\u003e4\u003c/span\u003e, \u003cspan class=\"CitationRef\"\u003e42\u003c/span\u003e, \u003cspan class=\"CitationRef\"\u003e43\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eAs aforementioned, mutations (Figure S24A) and truncations [\u003cspan class=\"CitationRef\"\u003e18\u003c/span\u003e] of MDM2 central zinc finger domain that is critical for the RP binding were identified in HCC patients. In addition to HCC, other types of cancers also contain MDM2 somatic mutations in its central domain (Figure S24B). Over a dozen RPs and other proteins such as ARF, NPM, and Rb activate p53 under stresses by binding the MDM2 central domain, the same mode demonstrated in our current study, disrupting the MDM2-p53 feedback loop [\u003cspan class=\"CitationRef\"\u003e14\u003c/span\u003e, \u003cspan class=\"CitationRef\"\u003e18\u003c/span\u003e, \u003cspan class=\"CitationRef\"\u003e44\u003c/span\u003e]. Similarly, somatic point mutations of MDMX at S367, S342 and other serine sites exist in human cancers (Figure S24C). Our study partially explains why the MDM2 central domain and the MDMX serine sites are mutated in various cancer patients including HCC, and reinforces the clinical significance of the dual-mode activation of p53 in preventing liver tumorigenesis.\u003c/p\u003e \u003cp\u003eOne-carbon metabolism (1CM) is frequently reprogrammed in many cancers. Various enzymes for 1CM have been linked to different cancers including HCC, rendering the 1CM enzyme networks as very promising therapeutic targets of cancers [\u003cspan class=\"CitationRef\"\u003e45\u003c/span\u003e]. Here, we identified the 1CM enzyme MTHFD1L as a new p53-suppressed downstream player in HCC initiation (Figs.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003e, S\u003cspan class=\"InternalRef\"\u003e8\u003c/span\u003e-S17, and \u003cspan class=\"InternalRef\"\u003e8\u003c/span\u003e). MTHFD1L could accelerate liver cancer progression by encountering oxidative stress [\u003cspan class=\"CitationRef\"\u003e46\u003c/span\u003e]. Our bioinformatic analysis (Figure S8F) suggests that MTHFD1L plays a crucial role in HCC initiation. Consistently, Mthfd1l deletion drastically retarded HCIC tumorigenesis \u003cem\u003ein vivo\u003c/em\u003e (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e4\u003c/span\u003e). Interestingly, Mthfd1l is downregulated and well correlated with increased p53 in human cirrhotic livers (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e7\u003c/span\u003eF). Given Mthfd1l is usually overexpressed in all stages of HCC (Figure S8E), Mthfd1l might serve as a diagnostic marker and a therapeutic target for early HCC.\u003c/p\u003e \u003cp\u003eBesides enhanced autonomous growth mechanisms[\u003cspan class=\"CitationRef\"\u003e4\u003c/span\u003e], initiaiting-stage cancers gradually acquire an immune evasion ability to progress to advanced stages [\u003cspan class=\"CitationRef\"\u003e39\u003c/span\u003e]. We demonstrate that impairing the p53-pathways leads to higher expression of Mthfd1l that utilizes both the mechanisms to accelerate HCC initiation (Figures S14-16, 4–6 and 8). By producing mitochondrial formate, MTHFD1L fuels the methionine cycle to maintain the high level of RNA m\u003csup\u003e6\u003c/sup\u003eA (Figs.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003eA and S18A) that is observed in human HCC[\u003cspan class=\"CitationRef\"\u003e36\u003c/span\u003e] and critical for HCIC stemness and proliferation (Figure S17). Specifically, MTHFD1L supports the m\u003csup\u003e6\u003c/sup\u003eA modification of Snail mRNA (Figures S18C-D), resulting in translational upregulation of Snail (Figs.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003e and S20A-B). Studies have established that Snail plays an essential role in HCC onset by inhibiting hepatocyte-differentiation transcriptional factors and promoting cell proliferation [\u003cspan class=\"CitationRef\"\u003e27\u003c/span\u003e], and disrupting bile acid homeostasis [\u003cspan class=\"CitationRef\"\u003e28\u003c/span\u003e]. Interestingly, overexpression of Snail in hepatocytes can independently lead to steatohepatitis and HCC onset [\u003cspan class=\"CitationRef\"\u003e28\u003c/span\u003e]. As aforementioned, hepatocytes can express abundant cell surface MHC-I molecules for antigen presentation to CD8 + T cells[\u003cspan class=\"CitationRef\"\u003e37\u003c/span\u003e, \u003cspan class=\"CitationRef\"\u003e38\u003c/span\u003e]. In a study, loss of heterozygosity of HLA alleles is identified in 17% of multifocal HCC cases to dampen the neoantigen presentation and enhance the tumor cell immune escape[\u003cspan class=\"CitationRef\"\u003e47\u003c/span\u003e]. Here, we found that B2m is translationally repressed by the MTHFD1L-mediated mRNA m\u003csup\u003e6\u003c/sup\u003eA modification in HCICs (Figs.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e6\u003c/span\u003e and S20B-C). In line with the critical role of B2m in MHC-I-mediated antigen presentation to CD8 + T cells and subsequent T cell recognition [\u003cspan class=\"CitationRef\"\u003e29\u003c/span\u003e], MTHFD1L restrained the CD8 + T liver tumor infilatration and cytotoxicity against HCICs (Figs.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e6\u003c/span\u003eG-I), leading to reinforced immune evasion of HCICs, while p53 inhibition of Mthfd1l prevented this cancer immunoediting during HCC initiation (Figs.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e6\u003c/span\u003eE, F, and H), providing new insights into the p53’s function of reversing cancer metabolic reprogramming to block cancer immunoediting.\u003c/p\u003e \u003cp\u003eRNA m\u003csup\u003e6\u003c/sup\u003eA modification is a reversible process regulated by writers (such as METTL3/14, WTAP, etc.), erasers (such as FTO and ALKBH5), and readers (such as IGF2BPs, YTHDFs, etc.), affecting RNA splicing, nuclear export, degradation, stability, and translation [\u003cspan class=\"CitationRef\"\u003e48\u003c/span\u003e]. The fate of m\u003csup\u003e6\u003c/sup\u003eA modified mRNAs largely depends on the specific readers and modification regions. For instance, YTHDF2 promotes mRNA degradation, IGF2BPs enhances mRNA stability or translation [\u003cspan class=\"CitationRef\"\u003e48\u003c/span\u003e], and m\u003csup\u003e6\u003c/sup\u003eA in the mRNA coding region can delay translation elongation [\u003cspan class=\"CitationRef\"\u003e49\u003c/span\u003e]. We found the differential regulation of MTHFD1L on the m\u003csup\u003e6\u003c/sup\u003eA-mediated Snail and B2m expression (Figs.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e5\u003c/span\u003eB-G, \u003cspan class=\"InternalRef\"\u003e6\u003c/span\u003eA-F, S20, and \u003cspan class=\"InternalRef\"\u003e8\u003c/span\u003e). It remains to identify the specific m\u003csup\u003e6\u003c/sup\u003eA readers and erasers modulating these targets during HCC initiation.\u003c/p\u003e \u003cp\u003eIn summary, our study for the first time unveils the surveillance role of the two p53-pathways in averting MTHFD1L-accelerated HCC initiation upon pathologival signals, e.g., carcinogen, NASH, or alcoholic cirrhosis, and the novel bi-functions of MTHFD1L in promoting HCIC proliferation/stemness while suppressing immune elimination during HCC initiation, by fueling the methionine cycle to support RNA m\u003csup\u003e6\u003c/sup\u003eA modifications (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e8\u003c/span\u003e). It also positions MTHFD1L and its downstream m\u003csup\u003e6\u003c/sup\u003eA targets as promising candidates for early diagnosis, prevention or therapy of HCC.\u003c/p\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003cdiv id=\"Sec13\" class=\"Section3\"\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003cp\u003e\u003c/p\u003e \u003c/div\u003e "},{"header":"METHODS","content":"\u003ch2\u003eSex as biological variants\u003c/h2\u003e\u003cp\u003eBoth male and female mice were recruited for the DEN-induced HCC model, and similar findings were reported for both sexes, with significantly higher HCC incidence in the male mice, which is consistent with other animal studies and human HCC epidemiologic studies [\u003cspan class=\"CitationRef\"\u003e30\u003c/span\u003e, \u003cspan class=\"CitationRef\"\u003e31\u003c/span\u003e, \u003cspan class=\"CitationRef\"\u003e50\u003c/span\u003e]. Male mice were used in other animal experiments due to their higher sensitivity and stability to HCC induction, and liver samples from female patients were purchased, as detailed below and in the \u003cem\u003esupplemental information\u003c/em\u003e.\u003c/p\u003e\u003ch2\u003eAnimals\u003c/h2\u003e\u003cp\u003e \u003cem\u003eMouse lines.\u003c/em\u003e M2-C305F and MX-3SA KI mouse lines were generously provided by Jeoff Wahl at SALK institute and Yanping Zhang at UNC at Chapel Hill, respectively [\u003cspan class=\"CitationRef\"\u003e16\u003c/span\u003e, \u003cspan class=\"CitationRef\"\u003e17\u003c/span\u003e]. M2-C305F/MX-3SA-DKI mouse line was generated by crossing the two single-KI (SKI) mice as detailed in the supplemental information. Both male and female animals were used in this study and maintained in a C57BL/6 genetic background.\u003c/p\u003e\u003cp\u003e \u003cem\u003eDEN-induced HCC mouse model.\u003c/em\u003e To induce HCC, 15-day-old males and females were injected intraperitoneally (i.p.) with a single dose of DEN (25 mg/kg) and livers were collected 9 months later, following previous studies [\u003cspan class=\"CitationRef\"\u003e4\u003c/span\u003e]. For some experiments involved in acute DEN treatment, 15-day-old male mice were i.p. injected with a single dose of DEN (25 mg/kg), and livers were collected at indicated time points for further analysis.\u003c/p\u003e\u003cp\u003e \u003cem\u003eHigh-fat diet (HFD)-induced non-alcoholic steatohepatitis (NASH) and HCC mouse model\u003c/em\u003e. Male control (CTL) and DKI mice were weaned at 3 weeks and fed with HFD (with 60 kcal% fat) (Research Diets, Inc.) and water \u003cem\u003ead libitum\u003c/em\u003e for 15 months.\u003c/p\u003e\u003cp\u003e \u003cem\u003eIntra-splenic transplantation of HCICs.\u003c/em\u003e C57BL/6 wild-type (WT) mice recruited for HCIC transplantation were from the Jackson Laboratory (Stock# 000664). Briefly, by following previous studies [\u003cspan class=\"CitationRef\"\u003e3\u003c/span\u003e, \u003cspan class=\"CitationRef\"\u003e32\u003c/span\u003e, \u003cspan class=\"CitationRef\"\u003e51\u003c/span\u003e], 4-week-old male C57BL/6J WT mice were i.p. injected with retrosine twice with a 2-week interval. 4 weeks after the 2nd- retrosine treatment, each mouse received 1.5⋅10\u003csup\u003e5\u003c/sup\u003e freshly isolated viable cells that were enriched with CTL- or DKI-HCICs, or 1⋅10\u003csup\u003e6\u003c/sup\u003e cultured DKI-HCICs (passage 6, P6) with or without Mthfd1l gene deletion depending on experiments, in 100µl PBS via intra-splenic injection. One week after transplantation, CCl4 was given to the recipient mice intraperitoneally once a week for 3 weeks. Liver tissues of the recipient mice were harvested 8 ~ 12 months later for further analysis. All mice were housed under specific pathogen-free conditions with normal diet and water \u003cem\u003ead libitum\u003c/em\u003e.\u003c/p\u003e\u003cp\u003e \u003cstrong\u003eOT-1 mice and CD8 + T cell isolation\u003c/strong\u003e \u003c/p\u003e\u003cp\u003eTwo-month-old male C57BL/6 OT-1 mice were purchased from the Jackson Laboratory (Stock# 003831). Briefly, following previous study [\u003cspan class=\"CitationRef\"\u003e52\u003c/span\u003e], OT-1 mouse spleen was homogenized and cell suspension was filtered through a 70µm cell strainer (Fisher), and erythrocytes were lysed. Then the single-cell suspension was centrifuged and counted, and untouched CD8 + T cells were purified using a mouse CD8a + T Cell Isolation Kit according to the manufacturer's instructions.\u003c/p\u003e\u003ch2\u003ePrimary cell isolation and culture\u003c/h2\u003e\u003cp\u003ePrimary HCICs and hepatocytes were isolated from male CTL and DKI mice treated with DEN or Vehicle, by following previous studies [\u003cspan class=\"CitationRef\"\u003e3\u003c/span\u003e, \u003cspan class=\"CitationRef\"\u003e32\u003c/span\u003e, \u003cspan class=\"CitationRef\"\u003e51\u003c/span\u003e]. To ensure that enough HCICs can be obtained from mice for experiments and transplantation, male mice were used for DEN treatment. Fifteen-day-old mice were injected with one dose of DEN (25 mg/kg), and 3 months later mouse livers were perfused via inferior vena cava with warmed S\u0026amp;M buffer (0.24% Hepes, 0.019% NaOH, 0.05% KCl, 0.83% NaCl, pH7.4) and digest solution for 5min each, and minced in cold PBS. The cell aggregates containing HCICs were then isolated by sequential filtration with 70µm and 40µm cell-strainers, and subjected to gentle pipetting in cold PBS for intra-splenic injection.\u003c/p\u003e\u003cp\u003eFor some experiments, HCICs were maintained in collagen-coated Petri dishes in DMEM medium (Gibco) supplemented with 20% fetal bovine serum (FBS) (Sigma), 1⋅ PSN antibiotic mixture (10 ml/L) (Gibco), 20µg/L EGF, 0.01g/L Insulin, 1µm Dexamethasone and 1% L-glutamine. HCICs of passage 3–8 (P3-P8) were used for experiments to ensure the cell stemness.\u003c/p\u003e\u003ch2\u003eCRISPR/Cas9 mediated gene modification\u003c/h2\u003e\u003cp\u003eHCIC cells with stable depletion/deletion of Mthfd1l (MTH-D) were generated using lentiviral CRISPR-Cas9 system. In short, three single guide RNAs (sgRNAs) (\u003cem\u003eTable \u003cspan class=\"InternalRef\"\u003eS2\u003c/span\u003e)\u003c/em\u003e respectively targeting exon 1, 6 and 12 of Mthfd1l gene were designed using online software (IDT), and inserted in plasmid LentiCRISPRv2 expressing wild-type Cas9, then co-transfected into HEK 293T cells with packaging plasmids following the \u003cem\u003eZhangLab\u003c/em\u003e protocol (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://media.addgene.org/cms/files/Zhang_lab_LentiCRISPR_library_protocol.pdf\u003c/span\u003e\u003cspan class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003cem\u003e).\u003c/em\u003e CTL or DKI-HCICs (P3) were infected with lentivirus, and puromycin was added 48h later for selection of HCIC cells stably expressing Mthfd1l sgRNAs and Cas9. Single cell colonies were selected and cultured, followed by WB analysis and RNA sequencing to confirm the deletion. HCIC single colonies expressing Cas9 and WT-Mthfd1l (MTH-WT) were also simultaneously screened after infected with empty LentiCRISPRv2, to serve as paired control cells.\u003c/p\u003e\u003cp\u003eTo establish MTH-D HCIC cell lines re-expressing Mthfd1l or HCICs expressing chicken ovalbumin (cOVA) (HCIC_OVA), the cells were infected with lentivirus expressing GFP-flag-Mthfd1l or cOVA, respectively. G418 or puromycin was added 48h later for Mthfd1l selection for 10 days or cOVA selection for 5 days, respectively.\u003c/p\u003e\u003ch2\u003eIn vitro CD8 + T cytotoxicity assay\u003c/h2\u003e\u003cp\u003eBriefly, ~ 4,000 HCIC_OVA cells/well were seeded in triplicate into a 96-well flat-bottom plate. The next day, freshly isolated C57BL/6 OT-1 CD8 + T cells were added to the wells (T cell: HCIC ratio is about 5:1) and co-cultured in complete DMEM growth medium for 24h in an Incucyte (Sartorius) culture chamber. NucView 488 Caspase 3/7 substrate (Biotium) was simultaneously added to each well, and caspase 3/7 positive signal (indicating the amount of T-cell killing) were measured following previous studies[\u003cspan class=\"CitationRef\"\u003e53\u003c/span\u003e]. Five images per well were captured every hour using a 10× objective lens. Caspase 3/7 fluorescence intensity was quantified using the Incucyte built-in software combined with \u003cem\u003eImage J\u003c/em\u003e. In those Control wells, HCIC_OVA cells were cultured without CD8 + T.\u003c/p\u003e\u003cp\u003eIn some experiments, HCIC_OVA cells were co-cultured with OT-1 CD8 + T for 24h, followed by the measurement of released lactate dehydrogenase (LDH) with a LDH Cytotoxicity Assay kit (ThermoFisher) according to the manufacturer’s instructions.\u003c/p\u003e\u003ch2\u003eStatistics\u003c/h2\u003e\u003cp\u003eAll in vitro experiments were performed in at least triplicate. Data were presented as means ± SEM with N being the sample size. Comparisons among different groups were conducted by using One-way ANOVA or Student’s two-tailed t-test. Probability values of \u003cem\u003eP\u003c/em\u003e ≤ 0.05 were considered statistically significant.\u003c/p\u003e\u003cp\u003eSupplementary information is available at \u003cem\u003eCell Death and Differentiation\u003c/em\u003e’s website.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eACKNOWLEDGEMENTS\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe thank Dr. Geoffrey M Wahlfor offering MDMX\u003csup\u003e3SA\u003c/sup\u003e mice (The Salk Institute for Biological Studies), Dr. Jinqiang Zhang (Department of Pathology, Tulane University) for giving suggestion on HCIC transplantation, Drs. Xiaoming Yin, Gang Liu and Bilon Khambu (Department of Pathology, Tulane University), Dr. James Jackson (Department of Biochemistry and Molecular Biology, Tulane University), and Dr. Hongbing Liu (Department of Pediatrics, Tulane University) for providing some antibodies, Drs. Jieqiong Wang and Hyemin Lee (Department of Biochemistry and Molecular Biology, Tulane University) for offering CRISPR-Cas9 empty vectors and some technique support, and Elise Reinosofor assisting in \u003cem\u003ein vitro\u003c/em\u003e experiments.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAUTHOR CONTRIBUTIONS\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eY.W.Z. independently conceived and designed the project, conducted the experiments, analyzed the results, and organized them into figures in the Lu laboratory under the supervision of H.L.; G.P. performed the nanopore m\u003csup\u003e6\u003c/sup\u003eA-seq data analyses. W.J.L., J.H.J., J.X.C., C.Y.L., Y.Z., N.P., S.M.M., and E.D.E.\u0026nbsp;assisted Y.W.Z. in some animal works or \u003cem\u003ein vitro\u003c/em\u003e experiments; Y.L. and H.G. conducted the nanopore m\u003csup\u003e6\u003c/sup\u003eA sequencing. Y.P.Z. offered MDM2\u003csup\u003eC305F\u003c/sup\u003e mice; D.D. offered helps in HCIC isolation and transplantation; S.X.Z. assisted in the initial animal transfer and coordinated RNA-seq analysis. Y.W.Z. wrote and revised and H.L. revised the manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eETHICS APPROVAL AND CONSENT TO PARTICIPATE\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe frozen smashed human liver tissue samples were deidentified and purchased from XenoTech, BioIVT (Kansas City, KS 66103). All animal experiments were conducted in accordance with the National Institutes Health “Guide for the Care and Use of Laboratory Animals” and were approved by the Institutional Animal Care and Use Committee at Tulane University School of Medicine and our animal protocol numbers are 1097, 2053, and 2266.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFUNDING STATEMENT\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis project was in part supported by NIH-NCI grants R01CA095441, R01CA127724 to H.L. and S.X.Z, who are also in part supported by NIH-NCI grants R01CA234605 and R21CA272890, and the Carol Lavin Bernick Faculty Grant and the Tulane Medical School Pilot fund to Y.W.Z. In addition, Y.W.Z received the NIH-COBRE pilot funds (AP5446G1 and AP5446GG). H.L. was supported by the Reynolds and Ryan Families Endowed Chair fund, Lady Leukemia League Fund, and the NIH-NCI grant U01\u003cstrong\u003eCA252965\u003c/strong\u003e. D.D. was supported by UC San Diego Altman Clinical and Translational Research Institute (ACTRI)/ NIH grant KL2TR001444 and the San Diego Digestive Diseases Research Center (SDDRC) (NIH grant DK120515).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDATA AVAILABILITY\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eGEO accession number for Nanopore RNA m\u003csup\u003e6\u003c/sup\u003eA-seq data is GSE310323. GEO accession number for RNA-seq data will be available upon submission.\u0026nbsp;All raw data of immunoblotting images generated in this study will be deposit and made publicly available upon publication of this manuscript.Any additional information required to reanalyze the data reported in this paper is available from both the corresponding authors upon reasonable request.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eChidambaranathan-Reghupaty S, Fisher PB, Sarkar D. Hepatocellular carcinoma (HCC): Epidemiology, etiology and molecular classification. \u003cem\u003eAdv Cancer Res\u003c/em\u003e 2021; 149: 1-61.\u003c/li\u003e\n\u003cli\u003eVogel A, Meyer T, Sapisochin G, Salem R, Saborowski A. Hepatocellular carcinoma. \u003cem\u003eLancet\u003c/em\u003e 2022; 400: 1345-1362.\u003c/li\u003e\n\u003cli\u003eHe G, Dhar D, Nakagawa H, Font-Burgada J, Ogata H, Jiang Y\u003cem\u003e et al\u003c/em\u003e. Identification of liver cancer progenitors whose malignant progression depends on autocrine IL-6 signaling. \u003cem\u003eCell\u003c/em\u003e 2013; 155: 384-396.\u003c/li\u003e\n\u003cli\u003eDhar D, Antonucci L, Nakagawa H, Kim JY, Glitzner E, Caruso S\u003cem\u003e et al\u003c/em\u003e. Liver Cancer Initiation Requires p53 Inhibition by CD44-Enhanced Growth Factor Signaling. \u003cem\u003eCancer Cell\u003c/em\u003e 2018; 33: 1061-1077 e1066.\u003c/li\u003e\n\u003cli\u003eFont-Burgada J, Shalapour S, Ramaswamy S, Hsueh B, Rossell D, Umemura A\u003cem\u003e et al\u003c/em\u003e. Hybrid Periportal Hepatocytes Regenerate the Injured Liver without Giving Rise to Cancer. \u003cem\u003eCell\u003c/em\u003e 2015; 162: 766-779.\u003c/li\u003e\n\u003cli\u003eShin S, Wangensteen KJ, Teta-Bissett M, Wang YJ, Mosleh-Shirazi E, Buza EL\u003cem\u003e et al\u003c/em\u003e. Genetic lineage tracing analysis of the cell of origin of hepatotoxin-induced liver tumors in mice. \u003cem\u003eHepatology\u003c/em\u003e 2016; 64: 1163-1177.\u003c/li\u003e\n\u003cli\u003eLlovet JM, Kelley RK, Villanueva A, Singal AG, Pikarsky E, Roayaie S\u003cem\u003e et al\u003c/em\u003e. Hepatocellular carcinoma. \u003cem\u003eNat Rev Dis Primers\u003c/em\u003e 2021; 7: 6.\u003c/li\u003e\n\u003cli\u003eTanaka M, Miyajima A. Liver regeneration and fibrosis after inflammation. \u003cem\u003eInflamm Regen\u003c/em\u003e 2016; 36: 19.\u003c/li\u003e\n\u003cli\u003eHuang L, Yan Z, Liao X, Li Y, Yang J, Wang ZG\u003cem\u003e et al\u003c/em\u003e. The p53 inhibitors MDM2/MDMX complex is required for control of p53 activity in vivo. \u003cem\u003eProc Natl Acad Sci U S A\u003c/em\u003e 2011; 108: 12001-12006.\u003c/li\u003e\n\u003cli\u003eCharni M, Aloni-Grinstein R, Molchadsky A, Rotter V. p53 on the crossroad between regeneration and cancer. \u003cem\u003eCell Death Differ\u003c/em\u003e 2017; 24: 8-14.\u003c/li\u003e\n\u003cli\u003eTschaharganeh DF, Xue W, Calvisi DF, Evert M, Michurina TV, Dow LE\u003cem\u003e et al\u003c/em\u003e. p53-dependent Nestin regulation links tumor suppression to cellular plasticity in liver cancer. \u003cem\u003eCell\u003c/em\u003e 2014; 158: 579-592.\u003c/li\u003e\n\u003cli\u003eWang C, Kang H, Yi Y, Ding Y, Wang F, Luo J\u003cem\u003e et al\u003c/em\u003e. Rictor mediates p53 deactivation to facilitate the malignant transformation of hepatocytes and promote hepatocarcinogenesis. \u003cem\u003eJ Transl Med\u003c/em\u003e 2023; 21: 919.\u003c/li\u003e\n\u003cli\u003eLiu Y, He Y, Jin A, Tikunov AP, Zhou L, Tollini LA\u003cem\u003e et al\u003c/em\u003e. Ribosomal protein-Mdm2-p53 pathway coordinates nutrient stress with lipid metabolism by regulating MCD and promoting fatty acid oxidation. \u003cem\u003eProc Natl Acad Sci U S A\u003c/em\u003e 2014; 111: E2414-2422.\u003c/li\u003e\n\u003cli\u003eZhou X, Liao WJ, Liao JM, Liao P, Lu H. Ribosomal proteins: functions beyond the ribosome. \u003cem\u003eJ Mol Cell Biol\u003c/em\u003e 2015; 7: 92-104.\u003c/li\u003e\n\u003cli\u003eHe G, Zhang YW, Lee JH, Zeng SX, Wang YV, Luo Z\u003cem\u003e et al\u003c/em\u003e. AMP-activated protein kinase induces p53 by phosphorylating MDMX and inhibiting its activity. \u003cem\u003eMol Cell Biol\u003c/em\u003e 2014; 34: 148-157.\u003c/li\u003e\n\u003cli\u003eWang YV, Leblanc M, Wade M, Jochemsen AG, Wahl GM. Increased radioresistance and accelerated B cell lymphomas in mice with Mdmx mutations that prevent modifications by DNA-damage-activated kinases. \u003cem\u003eCancer Cell\u003c/em\u003e 2009; 16: 33-43.\u003c/li\u003e\n\u003cli\u003eMacias E, Jin A, Deisenroth C, Bhat K, Mao H, Lindstrom MS\u003cem\u003e et al\u003c/em\u003e. An ARF-independent c-MYC-activated tumor suppression pathway mediated by ribosomal protein-Mdm2 Interaction. \u003cem\u003eCancer Cell\u003c/em\u003e 2010; 18: 231-243.\u003c/li\u003e\n\u003cli\u003eLindstrom MS, Jin A, Deisenroth C, White Wolf G, Zhang Y. Cancer-associated mutations in the MDM2 zinc finger domain disrupt ribosomal protein interaction and attenuate MDM2-induced p53 degradation. \u003cem\u003eMol Cell Biol\u003c/em\u003e 2007; 27: 1056-1068.\u003c/li\u003e\n\u003cli\u003eZhang F, Hamanaka RB, Bobrovnikova-Marjon E, Gordan JD, Dai MS, Lu H\u003cem\u003e et al\u003c/em\u003e. Ribosomal stress couples the unfolded protein response to p53-dependent cell cycle arrest. \u003cem\u003eJ Biol Chem\u003c/em\u003e 2006; 281: 30036-30045.\u003c/li\u003e\n\u003cli\u003eBrighenti E, Calabrese C, Liguori G, Giannone FA, Trere D, Montanaro L\u003cem\u003e et al\u003c/em\u003e. Interleukin 6 downregulates p53 expression and activity by stimulating ribosome biogenesis: a new pathway connecting inflammation to cancer. \u003cem\u003eOncogene\u003c/em\u003e 2014; 33: 4396-4406.\u003c/li\u003e\n\u003cli\u003eDicks N, Gutierrez K, Michalak M, Bordignon V, Agellon LB. Endoplasmic reticulum stress, genome damage, and cancer. \u003cem\u003eFront Oncol\u003c/em\u003e 2015; 5: 11.\u003c/li\u003e\n\u003cli\u003eChen L, Gilkes DM, Pan Y, Lane WS, Chen J. ATM and Chk2-dependent phosphorylation of MDMX contribute to p53 activation after DNA damage. \u003cem\u003eEMBO J\u003c/em\u003e 2005; 24: 3411-3422.\u003c/li\u003e\n\u003cli\u003eCarlessi R, Kohn-Gaone J, Olynyk JK, Tirnitz-Parker JEE. Mouse Models of Hepatocellular Carcinoma. In: Tirnitz-Parker JEE (ed). \u003cem\u003eHepatocellular Carcinoma\u003c/em\u003e: Brisbane (AU), 2019.\u003c/li\u003e\n\u003cli\u003eCoral GP, Branco F, Meurer R, Marcon PDS, Fontes PRO, Mattos AA. Results of Immunohistochemistry in the Differential Diagnosis of Early Hepatocellular Carcinoma and Nodules with High-Grade Dysplasia in Patients with Cirrhosis. \u003cem\u003eArq Gastroenterol\u003c/em\u003e 2021; 58: 82-86.\u003c/li\u003e\n\u003cli\u003eLiao Z, Tang C, Luo R, Gu X, Zhou J, Gao J. Current Concepts of Precancerous Lesions of Hepatocellular Carcinoma: Recent Progress in Diagnosis. \u003cem\u003eDiagnostics (Basel)\u003c/em\u003e 2023; 13.\u003c/li\u003e\n\u003cli\u003eMomb J, Lewandowski JP, Bryant JD, Fitch R, Surman DR, Vokes SA\u003cem\u003e et al\u003c/em\u003e. Deletion of Mthfd1l causes embryonic lethality and neural tube and craniofacial defects in mice. \u003cem\u003eProc Natl Acad Sci U S A\u003c/em\u003e 2013; 110: 549-554.\u003c/li\u003e\n\u003cli\u003eMoon H, Ju HL, Chung SI, Cho KJ, Eun JW, Nam SW\u003cem\u003e et al\u003c/em\u003e. Transforming Growth Factor-beta Promotes Liver Tumorigenesis in Mice via Up-regulation of Snail. \u003cem\u003eGastroenterology\u003c/em\u003e 2017; 153: 1378-1391 e1376.\u003c/li\u003e\n\u003cli\u003eMiura S, Suzuki A. Induction of Steatohepatitis and Liver Tumorigenesis by Enforced Snail Expression in Hepatocytes. \u003cem\u003eAm J Pathol\u003c/em\u003e 2020; 190: 1271-1283.\u003c/li\u003e\n\u003cli\u003eWang H, Liu B, Wei J. Beta2-microglobulin(B2M) in cancer immunotherapies: Biological function, resistance and remedy. \u003cem\u003eCancer Lett\u003c/em\u003e 2021; 517: 96-104.\u003c/li\u003e\n\u003cli\u003eLiu S, Huang F, Ru G, Wang Y, Zhang B, Chen X\u003cem\u003e et al\u003c/em\u003e. Mouse Models of Hepatocellular Carcinoma: Classification, Advancement, and Application. \u003cem\u003eFront Oncol\u003c/em\u003e 2022; 12: 902820.\u003c/li\u003e\n\u003cli\u003eKim GY, Kim D, Jeon J, Seo W, Park SH. Navigating liver cancer with mouse models: a comprehensive overview of HCC experimental systems. \u003cem\u003eJ Liver Cancer\u003c/em\u003e 2025; 25: 239-250.\u003c/li\u003e\n\u003cli\u003eZhang J, Han C, Ungerleider N, Chen W, Song K, Wang Y\u003cem\u003e et al\u003c/em\u003e. A Transforming Growth Factor-beta and H19 Signaling Axis in Tumor-Initiating Hepatocytes That Regulates Hepatic Carcinogenesis. \u003cem\u003eHepatology\u003c/em\u003e 2019; 69: 1549-1563.\u003c/li\u003e\n\u003cli\u003eMurphy M, Ahn J, Walker KK, Hoffman WH, Evans RM, Levine AJ\u003cem\u003e et al\u003c/em\u003e. Transcriptional repression by wild-type p53 utilizes histone deacetylases, mediated by interaction with mSin3a. \u003cem\u003eGenes Dev\u003c/em\u003e 1999; 13: 2490-2501.\u003c/li\u003e\n\u003cli\u003ePike ST, Rajendra R, Artzt K, Appling DR. Mitochondrial C1-tetrahydrofolate synthase (MTHFD1L) supports the flow of mitochondrial one-carbon units into the methyl cycle in embryos. \u003cem\u003eJ Biol Chem\u003c/em\u003e 2010; 285: 4612-4620.\u003c/li\u003e\n\u003cli\u003eYankova E, Blackaby W, Albertella M, Rak J, De Braekeleer E, Tsagkogeorga G\u003cem\u003e et al\u003c/em\u003e. Small-molecule inhibition of METTL3 as a strategy against myeloid leukaemia. \u003cem\u003eNature\u003c/em\u003e 2021; 593: 597-601.\u003c/li\u003e\n\u003cli\u003eChen M, Wei L, Law CT, Tsang FH, Shen J, Cheng CL\u003cem\u003e et al\u003c/em\u003e. RNA N6-methyladenosine methyltransferase-like 3 promotes liver cancer progression through YTHDF2-dependent posttranscriptional silencing of SOCS2. \u003cem\u003eHepatology\u003c/em\u003e 2018; 67: 2254-2270.\u003c/li\u003e\n\u003cli\u003eChen M, Tabaczewski P, Truscott SM, Van Kaer L, Stroynowski I. Hepatocytes express abundant surface class I MHC and efficiently use transporter associated with antigen processing, tapasin, and low molecular weight polypeptide proteasome subunit components of antigen processing and presentation pathway. \u003cem\u003eJ Immunol\u003c/em\u003e 2005; 175: 1047-1055.\u003c/li\u003e\n\u003cli\u003eBertolino P, Trescol-Biemont MC, Rabourdin-Combe C. Hepatocytes induce functional activation of naive CD8+ T lymphocytes but fail to promote survival. \u003cem\u003eEur J Immunol\u003c/em\u003e 1998; 28: 221-236.\u003c/li\u003e\n\u003cli\u003eGubin MM, Vesely MD. Cancer Immunoediting in the Era of Immuno-oncology. \u003cem\u003eClin Cancer Res\u003c/em\u003e 2022; 28: 3917-3928.\u003c/li\u003e\n\u003cli\u003eKatz SF, Lechel A, Obenauf AC, Begus-Nahrmann Y, Kraus JM, Hoffmann EM\u003cem\u003e et al\u003c/em\u003e. Disruption of Trp53 in livers of mice induces formation of carcinomas with bilineal differentiation. \u003cem\u003eGastroenterology\u003c/em\u003e 2012; 142: 1229-1239 e1223.\u003c/li\u003e\n\u003cli\u003eMakino Y, Hikita H, Fukumoto K, Sung JH, Sakano Y, Murai K\u003cem\u003e et al\u003c/em\u003e. Constitutive Activation of the Tumor Suppressor p53 in Hepatocytes Paradoxically Promotes Non-Cell Autonomous Liver Carcinogenesis. \u003cem\u003eCancer Res\u003c/em\u003e 2022; 82: 2860-2873.\u003c/li\u003e\n\u003cli\u003eLevine AJ, Jenkins NA, Copeland NG. The Roles of Initiating Truncal Mutations in Human Cancers: The Order of Mutations and Tumor Cell Type Matters. \u003cem\u003eCancer Cell\u003c/em\u003e 2019; 35: 10-15.\u003c/li\u003e\n\u003cli\u003eSchulze K, Imbeaud S, Letouze E, Alexandrov LB, Calderaro J, Rebouissou S\u003cem\u003e et al\u003c/em\u003e. Exome sequencing of hepatocellular carcinomas identifies new mutational signatures and potential therapeutic targets. \u003cem\u003eNat Genet\u003c/em\u003e 2015; 47: 505-511.\u003c/li\u003e\n\u003cli\u003eKurki S, Peltonen K, Latonen L, Kiviharju TM, Ojala PM, Meek D\u003cem\u003e et al\u003c/em\u003e. Nucleolar protein NPM interacts with HDM2 and protects tumor suppressor protein p53 from HDM2-mediated degradation. \u003cem\u003eCancer Cell\u003c/em\u003e 2004; 5: 465-475.\u003c/li\u003e\n\u003cli\u003eCuthbertson CR, Arabzada Z, Bankhead A, 3rd, Kyani A, Neamati N. A Review of Small-Molecule Inhibitors of One-Carbon Enzymes: SHMT2 and MTHFD2 in the Spotlight. \u003cem\u003eACS Pharmacol Transl Sci\u003c/em\u003e 2021; 4: 624-646.\u003c/li\u003e\n\u003cli\u003eLee D, Xu IM, Chiu DK, Lai RK, Tse AP, Lan Li L\u003cem\u003e et al\u003c/em\u003e. Folate cycle enzyme MTHFD1L confers metabolic advantages in hepatocellular carcinoma. \u003cem\u003eJ Clin Invest\u003c/em\u003e 2017; 127: 1856-1872.\u003c/li\u003e\n\u003cli\u003eDong LQ, Peng LH, Ma LJ, Liu DB, Zhang S, Luo SZ\u003cem\u003e et al\u003c/em\u003e. Heterogeneous immunogenomic features and distinct escape mechanisms in multifocal hepatocellular carcinoma. \u003cem\u003eJ Hepatol\u003c/em\u003e 2020; 72: 896-908.\u003c/li\u003e\n\u003cli\u003eMa W, Wu T. RNA m6A modification in liver biology and its implication in hepatic diseases and carcinogenesis. \u003cem\u003eAm J Physiol Cell Physiol\u003c/em\u003e 2022; 323: C1190-C1205.\u003c/li\u003e\n\u003cli\u003eJain S, Koziej L, Poulis P, Kaczmarczyk I, Gaik M, Rawski M\u003cem\u003e et al\u003c/em\u003e. Modulation of translational decoding by m(6)A modification of mRNA. \u003cem\u003eNat Commun\u003c/em\u003e 2023; 14: 4784.\u003c/li\u003e\n\u003cli\u003eAbboud Y, Ismail M, Khan H, Medina-Morales E, Alsakarneh S, Jaber F\u003cem\u003e et al\u003c/em\u003e. Hepatocellular Carcinoma Incidence and Mortality in the USA by Sex, Age, and Race: A Nationwide Analysis of Two Decades. \u003cem\u003eJ Clin Transl Hepatol\u003c/em\u003e 2024; 12: 172-181.\u003c/li\u003e\n\u003cli\u003eHe G, Yu GY, Temkin V, Ogata H, Kuntzen C, Sakurai T\u003cem\u003e et al\u003c/em\u003e. Hepatocyte IKKbeta/NF-kappaB inhibits tumor promotion and progression by preventing oxidative stress-driven STAT3 activation. \u003cem\u003eCancer Cell\u003c/em\u003e 2010; 17: 286-297.\u003c/li\u003e\n\u003cli\u003eChen X, Lu Q, Zhou H, Liu J, Nadorp B, Lasry A\u003cem\u003e et al\u003c/em\u003e. A membrane-associated MHC-I inhibitory axis for cancer immune evasion. \u003cem\u003eCell\u003c/em\u003e 2023; 186: 3903-3920 e3921.\u003c/li\u003e\n\u003cli\u003eDuhen T, Duhen R, Montler R, Moses J, Moudgil T, de Miranda NF\u003cem\u003e et al\u003c/em\u003e. Co-expression of CD39 and CD103 identifies tumor-reactive CD8 T cells in human solid tumors. \u003cem\u003eNat Commun\u003c/em\u003e 2018; 9: 2724.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"cell-death-and-differentiation","isNatureJournal":false,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"cdd","sideBox":"Learn more about [Cell Death \u0026 Differentiation](http://www.nature.com/cdd/)","snPcode":"41418","submissionUrl":"https://mts-cdd.nature.com/cgi-bin/main.plex","title":"Cell Death \u0026 Differentiation","twitterHandle":"@cddpress","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"p53, MDM2, MDMX, Mthfdl, HCC, NASH, cirrhosis, one-carbon metabolism (1CM), m6A, Snail, B2m, antigen presentation, immune evasion, HCC initiating cell (HCICs), and liver cancer initiation","lastPublishedDoi":"10.21203/rs.3.rs-9474831/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9474831/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eHepatocellular carcinoma (HCC) originates from damaged hepatocytes in chronic liver injury. Though HCC development requires oncoproteins-mediated p53 inhibition, both genetic deletion and continous activation of p53 specifically in mouse hepatocytes are shown to promote liver tumorigenesis, suggesting that chronic liver injury firstly activates p53, and p53 activity needs to be dynamically regulated for preventing hepatocarcinogenesis. Yet, it remains unaddressed how liver injury stresses trigger signaling pathways to rapidly activate p53 prior to its inactivation in a physiological setting. Two pathways, ribosomal proteins (RPs)-MDM2 and 14-3-3-MDMX, are shown to activate p53 upon stresses, and mutations of MDM2 or MDMX that may disrupt these pathways have been found in human cancers including HCC. Using our unique double knock-in (DKI) mice that contain wild-type p53 while harbor defects in these two pathways, we unveiled that basal level of p53 in DKI mice is sufficient for maintaining liver function, however, carcinogen- or unhealthy diet-induced HCC initiation is accelerated in DKI mice. The two p53-pathways are also activated in human cirrhotic livers. We futher identified a one-carbon metabolism (1CM) enzyme methylenetetrahydrofolate dehydrogenase 1 like (MTHFD1L) as the p53-suppressed player in HCC initiation. We found that MTHFD1L is upregulated in human early HCC and correlated with p53 status, and promotes mouse HCC initiation by enhancing autonomous growth and immune evasion of HCC initiating cells\u0026nbsp; (HCICs). With Nanopore RNA-m\u003csup\u003e6\u003c/sup\u003eA-sequencing, we unveiled that MTHFD1L fulfills these bi-functions by fueling methionine cycle that produces methyl groups to maintain the mRNA m\u003csup\u003e6\u003c/sup\u003eA of Snail (an HCC early tumorigenesis driver) and B2m (a light chain of MHC-I required for antigen presentation to CD8+ T) in HCICs. Our findings demonstrate that the two stress-triggered p53-pathways are crucial for preventing liver early tumorigenesis by reversing MTHFD1L upregulation-associated 1CM reprogramming and subsequent Snail and B2m mRNA-m\u003csup\u003e6\u003c/sup\u003eA-mediated cell autonomous growth and immune evasion.\u003c/p\u003e","manuscriptTitle":"Dual-stress-responsive-pathways prevent HCC initiation via p53 suppression of MTHFD1L-RNA m6A-elicited autonomous growth and immune evasion","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-04-30 02:55:46","doi":"10.21203/rs.3.rs-9474831/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"editorInvitedReview","content":"This content is not available.","date":"2026-05-13T07:46:15+00:00","index":1,"fulltext":"This content is not available."},{"type":"reviewerAgreed","content":"This content is not available.","date":"2026-04-30T12:02:39+00:00","index":3,"fulltext":"This content is not available."},{"type":"reviewerAgreed","content":"This content is not available.","date":"2026-04-30T07:57:24+00:00","index":2,"fulltext":"This content is not available."},{"type":"reviewerAgreed","content":"This content is not available.","date":"2026-04-30T03:15:08+00:00","index":1,"fulltext":"This content is not available."},{"type":"reviewersInvited","content":"","date":"2026-04-21T11:49:11+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-04-21T11:41:29+00:00","index":"","fulltext":""},{"type":"submitted","content":"Cell Death \u0026 Differentiation","date":"2026-04-20T16:23:11+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-04-20T16:23:11+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"cell-death-and-differentiation","isNatureJournal":false,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"cdd","sideBox":"Learn more about [Cell Death \u0026 Differentiation](http://www.nature.com/cdd/)","snPcode":"41418","submissionUrl":"https://mts-cdd.nature.com/cgi-bin/main.plex","title":"Cell Death \u0026 Differentiation","twitterHandle":"@cddpress","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"c54c34f9-9568-4423-8cd2-9343e57e0d3b","owner":[],"postedDate":"April 30th, 2026","published":true,"recentEditorialEvents":[{"type":"editorInvitedReview","content":"This content is not available.","date":"2026-05-13T07:46:15+00:00","index":1,"fulltext":"This content is not available."},{"type":"reviewerAgreed","content":"This content is not available.","date":"2026-04-30T12:02:39+00:00","index":3,"fulltext":"This content is not available."},{"type":"reviewerAgreed","content":"This content is not available.","date":"2026-04-30T07:57:24+00:00","index":2,"fulltext":"This content is not available."},{"type":"reviewerAgreed","content":"This content is not available.","date":"2026-04-30T03:15:08+00:00","index":1,"fulltext":"This content is not available."}],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":67278899,"name":"Biological sciences/Cancer/Cancer genetics"},{"id":67278900,"name":"Health sciences/Diseases/Cancer/Tumour-suppressor proteins"}],"tags":[],"updatedAt":"2026-04-30T02:55:46+00:00","versionOfRecord":[],"versionCreatedAt":"2026-04-30 02:55:46","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9474831","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9474831","identity":"rs-9474831","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}