Oncogenic β-catenin-driven liver cancer is susceptible to methotrexate-mediated disruption of nucleotide synthesis

Abstract

Background:

Liver cancer is largely resistant to chemotherapy. This study aimed to identify the effective chemotherapeutics for β-catenin-activated liver cancer which is caused by gain-of-function mutation of catenin beta 1 (CTNNB1), the most frequently altered proto-oncogene in hepatic neoplasms.

Methods:

Constitutive β-catenin-activated mouse embryonic fibroblasts (MEFs) were established by deleting exon 3 (β-catenin^Δ(ex3)/+), the most common mutation site in CTNNB1 gene. A screening of 12 widely used chemotherapy drugs was conducted for the ones that selectively inhibited β-catenin^Δ(ex3)/+ but not for wild-type MEFs. Untargeted metabolomics was carried out to examine the alterations of metabolites in nucleotide synthesis. The efficacy and selectivity of methotrexate (MTX) on β-catenin-activated human liver cancer cells were determined in vitro. Immuno-deficient nude mice subcutaneously inoculated with β-catenin wild-type or mutant liver cancer cells and hepatitis B virus (HBV); β-catenin^lox(ex3)/+ mice were used, respectively, to evaluate the efficacy of MTX in the treatment of β-catenin mutant liver cancer.

Results:

MTX was identified and validated as a preferential agent against the proliferation and tumor formation of β-catenin-activated cells. Boosted nucleotide synthesis was the major metabolic aberration in β-catenin-active cells, and this alteration was also the target of MTX. Moreover, MTX abrogated hepatocarcinogenesis of HBV; β-catenin^lox(ex3)/+ mice, which stimulated concurrent Ctnnb1-activated mutation and HBV infection in liver cancer.

Conclusion:

MTX is a promising chemotherapeutic agent for β-catenin hyperactive liver cancer. Since repurposing MTX has the advantages of lower risk, shorter timelines, and less investment in drug discovery and development, a clinical trial is warranted to test its efficacy in the treatment of β-catenin mutant liver cancer.

Introduction

Liver cancer is the sixth most prevalent cancer and the third leading cause of cancer death worldwide.^[1] The high mortality of liver cancer is largely due to that most patients are diagnosed at advanced stage with limited therapeutic options.^[2] Hepatocellular carcinoma (HCC) is the most common type of liver cancer. The first-line drugs approved for advanced HCC are sorafenib and lenvatinib, which have the modest clinical benefits for HCC patients.^[3] Because HCC is featured with numerous genetic abnormalities and complex molecular pathogenesis, investigating the underlying molecular mechanism and targeted therapy is essential for its treatment.

Catenin beta 1 (CTNNB1) is the most frequently mutated proto-oncogene in liver cancer.^[4] Over 50% of hepatoblastoma and more than 20% of HCC contain CTNNB1 mutations.^[5,6] The true mutation frequency of CTNNB1 could be underestimated due to the difficulty in detecting intragenic CTNNB1 deletions through next-generation sequencing analysis.^[7] As more than half of HCC patients are hepatitis B virus (HBV) carriers, chronic hepatitis caused by HBV is a major cause of HCC. CTNNB1 mutations have been detected in 13.4%–19.0% of HBV-associated liver cancer.^[8,9] The causality of hepatocarcinogenesis and CTNNB1 mutation has been established in mice with hepatic β-catenin active mutation.^[10,11]

As a downstream effector of Wnt pathway, β-catenin plays critical roles in the regulation of cell metabolism, genetic stability, proliferation, and survival.^[12–17] After escaping from its degradation complex, β-catenin translocates into the nucleus and interacts with transcription factors to drive target gene expression.^[18] By stimulation of protein kinase B serine/threonine kinase 2 (AKT2)-mediated pyrimidine synthesis, β-catenin with gain-of-function mutation promotes cell proliferation and liver cancer development.^[10] The importance of β-catenin in liver cancer has provided the rationale for developing therapeutics that preferentially inhibit β-catenin-activated tumors. However, no target therapy has been approved yet for β-catenin-activated tumors. Repurposing old drugs is thus an attractive strategy because it lowers overall drug development costs and shortens development timelines.^[19]

By screening 12 widely used chemotherapy drugs, we identified methotrexate (MTX) as a selective inhibitor of β-catenin-driven cell proliferation. MTX blocked β-catenin-stimulated nucleotide synthesis. Furthermore, MTX was efficacious for the treatment of β-catenin-activated and HBV-associated spontaneous HCC in mice. These preclinical findings suggested that MTX might be a promising therapeutic agent for β-catenin-activated liver cancer.

Methods

Ethical approval

All animal studies were approved by the Animal Center of the Institute of Basic Medical Sciences, Chinese Academy of Medical Sciences and Peking Union Medical College (No. ACUC-A02-2022-057).

Chemicals and reagents

Fludarabine phosphate (T6501), oxaliplatin (T0164), docetaxel (T1034), capecitabine (T1408), doxorubicin (T1456), irinotecan hydrochloride (T0486L), carboplatin (T1058), and calcium folinate (T0148) were purchased from TargetMol Chemicals (Shanghai, China). 5-Fluorouracil (5-FU, S1209), sorafenib (S7397), gemcitabine (S1149), cisplatin (S1166), and MTX (S5097) were purchased from Selleck Biotechnology (Shanghai, China). Dulbecco's Modified Eagle Medium (DMEM, #11995-065), Roswell Park Memorial Institute (RPMI) 1640 (#72400047), fetal bovine serum (#10100147C), non-essential amino acid solution (#11140050), penicillin G and streptomycin (#15140163), glutamax (#35050061), and sodium pyruvate (#11360070) were purchased from Thermo Fisher (Carlsbad, CA, USA). Horse serum was purchased from Procell (#164215, Wuhan, Hubei, China). Puromycin (#P8230), polybrene (#H8761), and 0.1% crystal violet (#G1063) were purchased from Solarbio (Beijing, China). The β-catenin^mut plasmid was purchased from Addgene (#24204, Watertown, MA, USA).

Cell culture

Culture conditions were reported for all the cell lines used in this study.^[20] Wild type (WT) T, β-catenin^Δ(ex3)/+ mouse embryonic fibroblasts (MEFs) were previously constructed by our laboratory^[20], 293FT (Abclonal, Wuhan, China), HepaRG (Thermo fisher), CCC-HEL-1 (China Infrastructure of Cell Line Resource, Beijing, China), NCTC1469 (China Infrastructure of Cell Line Resource), HepG2 (China Infrastructure of Cell Line Resource), Hepa 1-6 (Procell, Wuhan, China), HCCLM3 (China Infrastructure of Cell Line Resource) and Huh7 (Procell) were cultured in DMEM, SNU886 (Cobioer biosciences, Nanjing, China) and SNU398 (Sagen tech, Guangzhou, China) were cultured in DMEM or RPMI-1640. Cells were maintained at 37°C incubator with 5% CO₂.

Metabolic analysis

MEFs were plated onto 6-well plates (5 × 10⁵ cells/well) and cultured in DMEM for 24 h. Cell number was counted in a parallel control dish. Cells were quickly washed with cold phosphate buffered saline (PBS) for three times and were then lysed in pre-cooled extraction solution (1 × 10⁶ cells/mL). The cell lysates were vortexed at 4°C for 5 min and then centrifuged at 16,000×g, 4°C for 15 min. The supernatants were analyzed by liquid chromatography-mass spectrometry (LC-MS) analysis. The protein concentrations of cell pallets were also determined.^[10]

Small interfering RNA (siRNA) transfection

The siRNA oligonucleotides against CTNNB1 were purchased from Hippo Biotechnology (Hangzhou, Zhejiang, China). The cells were transfected with siRNA and lipofectamine 2000 (Thermo Fisher) as per the manufacturer's instructions. The siRNA sequences used in the study were listed in Supplementary Table 1, http://links.lww.com/CM9/B677.

Western blotting analysis

Western blotting was performed as previously described.^[20] The membranes were incubated with primary antibodies against β-catenin (#9587, Cell Signaling Technology, Danvers, MA, USA), AKT2 (#2964, Cell Signaling Technology), carbamoyl-phosphate synthetase 2, aspartate transcarbamylase, and dihydroorotase (CAD, #93925, Cell Signaling Technology), phospho CAD (p-CAD Ser¹⁸⁵⁹, #7030, Cell Signaling Technology), p-CAD Ser¹⁴⁰⁶ (PTM BioLab, Hangzhou, Zhejiang, China), cyclin D1 (#A19038, Abclonal), or Glyceraldehyde-3-phosphate dehydrogenase (GAPDH, #AC002, Abclonal) overnight at 4°C and then incubated with secondary antibodies (#68071 or #32210, LI-COR Biosciences, Lincoln, NE, USA) for 2 h at the room temperature. Finally, protein bands were detected by SuperSignal enhanced chemiluminescence (LI-COR Biosciences).

Histology and immunohistochemistry (IHC)

Hematoxylin and eosin (H&E) staining and IHC of Ki-67 (15580, Abcam, Cambridge, UK, 1:500) and cleaved-caspase 3 (#9664, Cell Signaling Technology, 1:200) were performed by Servicebio (Wuhan, Hubei, China). The sections were analyzed by using a Pannoramic DESK microscope equipped with Caseviewer (version C.V 2.3, 3DHISTECH, Budapest, Hungary).

Real-time quantitative polymerase chain reaction (RT-qPCR)

Total RNA extraction and RT-qPCR protocol were performed as described previously.^[20] RT-qPCR was performed using SYBR high-sensitivity qPCR SuperMix (#abs60086, Absin Biosciences, Shanghai, China). β-actin transcripts were used to normalize the expression of each gene.The primer sequences used in the study were listed in Supplementary Table 2, http://links.lww.com/CM9/B677.

Cell viability assay

A total of 3000-5000 cells were seeded per well in 96-well plates and then treated with dimethyl sulfoxide (DMSO)/PBS or different concentrations of compounds for indicated times. Cell viability was determined through the Cell counting kit 8 (CCK8) assay (Selleck Biotechnology, Shanghai, China) using microplate reader (Agilent, Santa Clara, CA, USA).

Cell proliferation and apoptosis assay

β-catenin^Δ(ex3)/+ MEFs or SNU398 cells were treated with PBS or MTX (100 nmol/L) for 48 h, and their proliferation was assayed using the 5-ethynyl-2'-deoxyuridine (EdU) Click Proliferation Kit (Beyotime, Shanghai, China) following the manufacturer's protocol. Flow cytometry was used to determine total cell apoptosis by the Annexin V-fluorescein isothiocyanate (FITC) Apoptosis Detection Kit (Beyotime).

Animal experiments

HBV; β-catenin^lox(ex3)/+ mice were previously reported.^[10] 5-month-old HBV; β-catenin^lox(ex3)/+ mice were intraperitoneally injected with MTX (15 mg/kg) three times per week for three months.

For subcutaneous tumor assay, MEFs (1 × 10⁶ cells per mouse), Hepa1-6 cells (1 × 10⁶ cells per mouse), SNU398 cells (5 × 10⁶ cells per mouse), or Huh7 cells (2 × 10⁶ cells per mouse) were subcutaneously injected into the right flanks of immune-deficient nude mice (HFK Bio-Technology, Beijing, China, 4–6 weeks old female mice). Tumor volumes were calculated according to formula: tumor volume = 1/2 (length × width²). Once tumor volume reached around 100 mm³, mice were randomly assigned for intraperitoneal treatment of vehicle, or MTX (15 mg/kg) for three times per week. When the maximum tumor volume reached 1000 mm³, tumors were harvested for analysis.

Statistical analysis

Data were presented as mean ± standard deviation. GraphPad Prism 9 (GraphPad software, Boston, MA, USA) was used for statistical analysis. The analysis was performed using Student's t-test or one-way analysis of variance (ANOVA). A P <0.05 was considered statistically significant.

Results

β-catenin-activated cells are susceptible to MTX

To identify candidate chemotherapeutic agents that preferentially obliterate β-catenin-activated tumors, we first established constitutive β-catenin-activated MEFs by deleting exon 3 (β-catenin^Δ(ex3)/+), the most altered site in CTNNB1 gene [Figure 1A]. We then conducted a screening of 12 widely used chemotherapy drugs for the ones that selectively inhibited β-catenin-activated cells but not wild-type (WT) MEFs. β-catenin^Δ(ex3)/+ cells were more sensitive to MTX [Supplementary Figure 1, http://links.lww.com/CM9/B677 and Figure 1B,C]. Furthermore, MTX inhibited tumor formation of β-catenin^Δ(ex3)/+ MEFs, decreased tumor volumes and tumor weights with minimal effects on body weights in nude mice [Figure 1D–G]. Moreover, MTX inhibited proliferation and induced apoptosis of β-catenin^Δ(ex3)/+ MEFs [Supplementary Figure 2A–C, http://links.lww.com/CM9/B677 and Figure 1H]. Therefore, activated β-catenin cells were susceptible to MTX.

Figure 1.

β-catenin-activated MEFs are susceptible to MTX. (A)Western blotting of WT and β-catenin^Δ(ex3)/+ MEFs. (B, C) Validation of β-catenin^Δ(ex3)/+ MEFs' susceptibility to MTX. Cells were treated with MTX at different concentrations for 48 h (B). WT and β-catenin^Δ(ex3)/+ MEFs were treated with MTX (100 nmol/L) for the indicated time points (C). Cell proliferation ability was assessed by CCK8 assay. n = 3. (D-H) MTX suppresses tumor formation of β-catenin-activated MEFs. Nude mice subcutaneously inoculated with β-catenin^Δ(ex3)/+ MEFs were treated with PBS (n = 8) or MTX (n = 9, 15 mg/kg) three times per week. (D) Tumor weights were analyzed at the end of treatment. (E) Tumor growth was calculated as the mean value in tumor volumes. (F) Body weights of mice in each group were recorded. (G) Tumor images were presented at the end of treatment. (H) IHC staining of tumor lesions. Scale bar: 200 μm. Data were presented as mean ± SD and analysis was performed using t-test. ^*P <0.001. CCK8: Cell counting kit 8; H&E: Hematoxylin and eosin; IHC: Immunohistochemistry; MEFs: Mouse embryonic fibroblasts; MTX: Methotrexate; PBS: Phosphate buffered saline; SD: Standard deviation; WT: Wild-type.

MTX disrupts oncogenic β-catenin-stimulated nucleotide synthesis

Of the 12 tested chemotherapeutic drugs, MTX is the only one that targets nucleotide synthesis by inhibiting dihydrofolate reductase (DHFR) to block purine nucleotide and thymidylate synthesis. We previously reported that oncogenic β-catenin stimulated pyrimidine synthesis to support hepatic tumorigenesis.^[10] Therefore, we hypothesized that β-catenin-activated nucleotide synthesis made cells susceptible to MTX. Through untargeted metabolomics, we found that β-catenin^Δ(ex3)/+ MEFs were enriched with metabolites such as carbamoyl-aspartate (carbamoyl-asp), dihydroorotate, adenosine monophosphate (AMP) and guanosine monophosphate (GMP) in purine, and pyrimidine biosynthesis pathways in comparison with WT MEFs [Figure 2A,B]. The abundance of carbamoyl-asp, dihydroorotate, AMP and GMP in β-catenin^Δ(ex3)/+ MEFs were all reduced by MTX [Figure 2C]. Finally, the supplementation of calcium folinate (CF), widely used for offsetting the toxicity of MTX, reversed MTX-repressed proliferation of β-catenin^Δ(ex3)/+ MEFs [Figure 2D]. Taken together, MTX preferentially inhibited β-catenin-activated MEFs by repressing oncogenic β-catenin-boosted nucleotide synthesis.

Figure 2.

MTX disrupts oncogenic β-catenin-stimulated nucleotide synthesis. (A, B) β-catenin activation stimulated nucleotide synthesis. (A) Steady-state metabolite heatmaps of WT and β-catenin^Δ(ex3)/+ MEFs. (B) The steady state levels of nucleotide biosynthesis metabolites of MEFs were analyzed. (C) MTX disrupted oncogenic β-catenin-stimulated nucleotide synthesis. WT and β-catenin^Δ(ex3)/+ MEFs were treated with PBS or MTX (100 nmol/L) for 48 h, and nucleotide biosynthesis metabolites were then evaluated. (D) MTX-suppressed cell viability was reversed by CF. MEFs were treated with indicated agents (MTX 100 nmol/L, CF 5 μmol/L) for 48 h. Cell viability was determined by CCK8. Data were presented as mean ± SD and analysis was performed using t-test.^*P <0.001,^†P <0.01,^‡P <0.05. AMP: Adenosine monophosphate; CF: Calcium folinate; Carbamoyl-asp: carbamoyl-aspartate; CCK8: Cell counting kit 8; GMP: Guanosine monophosphate; MEFs: Mouse embryonic fibroblasts; MTX: Methotrexate; PBS: Phosphate buffered saline; SD: Standard deviation; WT: Wild-type.

MTX preferentially inhibits β-catenin-activated liver cancer cells

The efficacy and selectivity of MTX on β-catenin WT and mutant liver cell lines were investigated. According to the mutation state of Wnt pathway-related genes including CTNNB1, adenomatous polyposis coli (APC), axis inhibition protein 1 and 2 (AXIN1 and AXIN2), four of the eight hepatic cell lines were classified as CTNNB1 WT lines and the remaining were classified as CTNNB1 mutant lines. The CTNNB1 mutant cells were more susceptible to MTX than the WT cells. MTX inhibited the proliferation and induced apoptosis of CTNNB1-mutated SNU398 cells [Supplementary Figure 2D–F,3 http://links.lww.com/CM9/B677]. β-catenin WT Huh7 cells were then transfected with plasmids harboring S33A, S37A, T41A, and S45A mutations (β-catenin^mut) [Figure 3A]. Compared to the Huh7 cells transfected with vector control, mutant β-catenin-overexpressed Huh7 cells were sensitive to MTX [Figure 3B]. Moreover, β-catenin-silenced CTNNB1-mutated HCCLM3 cells were resistant to MTX [Figure 3D,E]. Altogether, β-catenin-activated liver cancer cells were more susceptible to MTX.

Figure 3.

MTX preferentially inhibits β-catenin-activated liver cancer cells and does not affect β-catenin signaling. (A, B) Overexpressed mutated β-catenin rendered cells susceptible to MTX. Western blotting of Huh7 cells transfected with vector or β-catenin^mut plasmid (A). The inhibition rate was assessed by CCK8 assay (B). (D, E) Depletion of β-catenin abrogated CTNNB1 mutant cell's susceptibility to MTX. HCCLM3 cells were transfected with control or β-catenin siRNA and the relevant protein abundances were examined by western blotting (D). The sensitivity of cells to MTX was compared by CCK8 assay (E). (C, F–H) β-catenin^Δ(ex3)/+ MEFs or SNU398 cells were treated with PBS or MTX (100 nmol/L) for 48 h. qRT-PCR (C, F) or westernblotting (G, H) was performed. (I) 293T cells treated with PBS or MTX (100 nmol/L) were utilized for luciferase assay with TCF/LEF reporter. Data were presented as mean ± SD and analysis was performed using t-test. ^*P <0.001. AKT2: AKT serine/threonine kinase 2; CTNNB1: Catenin beta 1; CAD: Carbamoyl-phosphate synthetase 2, aspartate transcarbamylase, and dihydroorotase; CCK8: Cell counting kit 8; GS: Glutamine synthetase; LGR5: Leucine-rich repeat containing G protein-coupled receptor; MEFs: Mouse embryonic fibroblasts; MTX: Methotrexate; NC: Negative control; n.s: No significance; OE: Overexpression; p-CAD: Phospho-CAD; PBS: Phosphate buffered saline; SD: Standard deviation; TCF/LEF: T-cell factor/lymphoid enhancer factor.

MTX does not affect β-catenin signaling

To determine whether MTX influences β-catenin signaling, we examined the effects of MTX on the expression of glutamine synthetase (GS), cyclinD1, and leucine-rich repeat containing G protein-coupled receptor (Lgr5), which are well-recognized genes transcriptionally regulated by β-catenin. However, we did not find any changes in the mRNA levels of these genes [Figure 3C,F]. We also evaluated AKT2/p-CAD signaling cascade and found that MTX did not affect AKT2 expression but slightly decreased p-CAD levels [Figure 3G,H]. The result of luciferase assay with T-cell factor/lymphoid enhancer factor (TCF/LEF) reporter further suggested that MTX did not alter β-catenin signaling [Figure 3I]. Therefore, MTX did not affect β-catenin signaling cascade.

MTX suppresses tumor formation of β-catenin-activated liver cancer cells

To determine the therapeutic efficacy of MTX against liver cancer, we first established a subcutaneous tumor model with Ctnnb1 exon 3-deleted mouse Hepa1-6 cells. During the 19 days treatment, MTX suppressed tumor growth without affecting the mouse body weights [Figure 4A–E]. In addition, xenografts of human HCC were generated with Ctnnb1 WT Huh7 cells and Ctnnb1 mutant SNU398 cells. MTX blocked the tumor growth of SNU398 cells but not Huh7 cells [Figure 4F–J and Supplementary Figure 4, http://links.lww.com/CM9/B677]. Therefore, MTX inhibited β-catenin-activated liver tumor formation.

Figure 4.

MTX suppresses tumor formation of β-catenin-activated liver cancer cells. (A–E) After 9 days of subcutaneous inoculation of Hepa 1–6 cells, nude mice (n = 6) were treated with PBS or MTX (15 mg/kg) thrice a week. Tumor formation was evaluated. (A) Tumor weights were plotted and analyzed. (B) Tumor growth was calculated as the mean value in tumor volumes. (C) Body weights of mice after treatment were recorded. (D) Tumor images were captured at the end of treatment. (E) IHC staining of tumor lesions. Scale bar: 200 μm. (F–J) After 10 days of subcutaneous inoculation of SNU398 cells, nude mice (n = 6) were treated with PBS or MTX (15 mg/kg) thrice a week. (F) Tumor weights were plotted and analyzed. (G) Tumor growth was calculated as the mean value in tumor volumes. (H) Body weights of mice after treatment were recorded. (I) Tumor images were captured at the end of treatment. (J) Immunohistochemistry staining of tumor lesions. Data were presented as mean ± SD and analysis was performed using t-test. ^*P <0.01, ^†P <0.001. H&E: Hematoxylin and eosin; IHC: Immunohistochemistry; MTX: Methotrexate; PBS: Phosphate buffered saline; SD: Standard deviation.

MTX abrogates oncogenic β-catenin-mediated hepatocarcinogenesis

HBV-induced chronic hepatitis is the major risk factor of HCC. Overall, 13.4–19.0% of the HBV-infected liver cancer patients harbor CTNNB1 mutations in their tumor lesions.^[8,9] By mimicking clinical settings with transgenic HBV and Ctnnb1 active mutation in liver, HBV; β-catenin^lox(ex3)/+ mice experienced multiple-step pathological processes of human hepatocarcinogenesis.^[10] Therefore, HBV; β-catenin^lox(ex3)/+ mouse is an ideal pre-clinical liver cancer model for the drug evaluation. The HBV; β-catenin^lox(ex3)/+ mice that had not yet formed hepatic tumors were intraperitoneally injected with MTX. After 3 months of treatment, MTX-injected mice exhibited lower liver weight to body weight and tumor incidence than PBS-treated mice [Figure 5A–F]. Moreover, MTX decreased tumor numbers and maximum tumor volumes without significantly affecting body weights [Figure 5B–E]. Furthermore, MTX inhibited the proliferation of liver tumors and induced their apoptosis in HBV; β-catenin^lox(ex3)/+ mice [Figure 5G]. In conclusion, MTX blunted oncogenic β-catenin-mediated hepatocarcinogenesis.

Figure 5.

MTX abrogates oncogenic β-catenin-mediated hepatocarcinogenesis. HBV; β-catenin^lox(ex3)/+ mice were injected via tail-vein with Cre-adenovirus by 7 weeks, treated with vehicle (PBS) or MTX (15 mg/kg) intraperitonially thrice a week by 5 months and then sacrificed by 8 months after birth (A–C, E, F). Ratio of liver weight to body weight (A), tumor incidence (B), tumor number (C), and the maximum liver tumor size (D) of 8-month mice were analyzed, representative liver pictures (F); (E) body weights of mice after treatment were analyzed. (G) IHC staining of mouse liver tissues. Scale bar: 200 μm. Data are presented as mean ± SD. n = 6. ^*P <0.05. IHC: Immunohistochemistry; H&E: Hematoxylin and eosin; LW/BW: Lung weight / body weight; MTX: Methotrexate; PBS: Phosphate buffered saline; SD: Standard deviation.

Discussion

CTNNB1 is the most commonly proto-oncogene in liver cancer.^[4] Targeted therapy for β-catenin-activated tumors is unavailable. In this study, we identified MTX as a selective inhibitor for β-catenin-activated cells among 12 widely used chemotherapeutic drugs. Mechanistically, oncogenic β-catenin-stimulated nucleotide synthesis made CTNNB1 mutant cells susceptible to MTX. Moreover, MTX abrogated oncogenic β-catenin-mediated subcutaneous tumor growth and hepatocarcinogenesis in mice.

Limited response restricts the use of chemotherapy in liver cancer treatment.^[21] Influences of genetic alterations on the efficacy of chemotherapy drugs should be considered. CTNNB1 is the most frequently activated proto-oncogene in liver cancer and its hepatocarcinogenic significance has been demonstrated.^[10,11] However, there is no approved targeted therapy for β-catenin-activated tumors. In pursuing potential chemotherapy for β-catenin-activated liver cancer, we screened 12 widely used chemotherapeutic agents and identified MTX as a preferential drug against β-catenin-activated cells. MTX, a folic acid antagonist, is an inhibitor of nucleotide metabolism. It blocks the conversion of dihydrofolate to tetrahydrofolate, thereby hindering nucleotide synthesis in proliferating cancer cells.^[22]

Alterations of proto-oncogenes and tumor suppressor genes reprogram cell metabolism, including DNA biosynthesis, for tumor development. For example, Phosphatase and tensin homolog (PTEN) knockout stimulates glutamine flux through de novo pyrimidine synthesis pathway to support rapid cell proliferation.^[23] Deficient tumor protein P53 (P53) upregulates methylenetetrahydrofolate dehydrogenase 2 (MTHFD2) to increase serine utilization for folate metabolism and purine synthesis.^[24] Carbamoyl phosphate synthetase 1-maintained pyrimidine metabolism and DNA synthesis are essential for Kirsten rat sarcoma viral oncogene homologue (KRAS)/Liver kinase B1 (LKB1)-mutant lung cancer cell proliferation.^[25] These metabolic alterations are targetable vulnerabilities in the cancer treatment. In this study, we carried out untargeted metabolomic profiling and found that oncogenic β-catenin stimulated metabolite generation in nucleotide biosynthesis, such as carbamoyl-asp, dihydroorotate, AMP, and GMP levels, which were largely reversed by MTX. MTX did not alter the expression of well-recognized genes transcriptionally regulated by β-catenin, indicating that MTX did not affect the transcriptional activity of β-catenin. The nucleotide metabolism alteration is a critical consequence of β-catenin activation and is downstream of β-catenin signaling cascade. For instance, we previously found that pyrimidine synthesis was regulated by β-catenin/AKT2 cascade.^[10] Although MTX does not directly affect β-catenin signaling, β-catenin-activated nucleotide synthesis sensitizes β-catenin mutant cells to MTX treatment.

With the advantages of lower risk, shorter timelines, and less investment, repurposing known drug strategy expedites drug discovery and development. MTX is an old drug that has been used worldwide for the treatment of various diseases such as autoimmune disease, rheumatoid arthritis, and cancer.^[26,27] However, little is known about its efficacy in liver cancer. In this study, we established the preclinical therapeutic efficacy of MTX against mouse liver cancer, which mimicked a subtype of human HCC with concurrent CTNNB1-activated mutation and HBV infection.

In conclusion, β-catenin-boosted nucleoside biosynthesis is essential for the tumor cell proliferation and tumor growth. The enhanced nucleotide synthesis sensitized β-catenin-activated cells to MTX intervention. MTX blocks mouse hepatocarcinogenesis, which recapitulated human HCC development driven by CTNNB1-activated mutation and HBV infection. Repurposing MTX might become an affordable and readily translatable therapeutic strategy against oncogenic β-catenin-associated liver cancer.

Funding

This work was supported by grants from the Chinese Academy of Medical Sciences Innovation Fund for Medical Sciences (No. 2021-1-I2M-018), the Haihe Laboratory of Cell Ecosystem Innovation Fund (No. 22HHXBSS00012), and the National Natural Science Foundation of China (Nos. 81730078 and 81872287).

Conflicts of interest

None.

Supplementary Material

cm9-137-181-s001.pdf ^{(610.8KB, pdf)}