L-methionine promotes CD8⁺ T cells killing hepatocellular carcinoma by inhibiting NR1I2/PCSK9 signaling

Abstract

Background

Liver cancer has consistently high incidence and mortality rates among malignant tumors. PCSK9, a target for hypercholesterolemia therapy, has recently been identified as an inhibitor of anti-tumor immunity, and targeting PCSK9 effectively inhibits tumor progression. However, small molecule inhibitors are lacking due to its flat protein structure.

Methods

PCSK9 transcription inhibitor screening was conducted using a PCSK9 promoter-driven td-Tomato plasmid. Quantitative real-time PCR and immunoblotting were employed to assess the effect of L-methionine on PCSK9 expression in HCC cell lines. Co-culture experiments were performed to evaluate the impact of L-methionine on CD8⁺ T cell-mediated killing of liver cancer cells. RNA sequencing, CUT&Tag, gene editing, and luciferase reporter assays were utilized to identify the transcription factor regulating PCSK9. Additionally, liver cancer xenograft and spontaneous liver cancer mouse models were used to evaluate the anti-cancer efficacy of L-methionine.

Results

Our study identified L-methionine, an essential amino acid, as a transcriptional inhibitor of PCSK9. The optimal dose of L-methionine to inhibit PCSK9 expression and enhance CD8⁺ T cell-mediated killing of liver cancer cells in vitro is 50 μM. Furthermore, intraperitoneal injection of 5 mg/kg/day of L-methionine significantly inhibited tumor growth in both liver cancer xenograft and spontaneous liver cancer mouse models. Mechanistically, we identified NR1I2 as a key transcription factor for PCSK9 and their crucial binding site was TGCACCCTGACAC. L-methionine inhibits PCSK9 transcription by downregulating NR1I2.

Conclusions

This work demonstrates that L-methionine promotes CD8⁺ T cell-mediated killing of hepatocellular carcinoma by inhibiting NR1I2/PCSK9 signaling. Our study introduces a novel and convenient approach to inhibit PCSK9 and provides a theoretical basis for the rational supplementation of L-methionine in liver cancer patients.

Introduction

Liver cancer ranks as the sixth most common malignancy and the third leading cause of cancer-related death worldwide [1,2]. Metabolic reprogramming has emerged as a hallmark of cancer, enabling tumor cells to adapt to the hypoxic and nutrient-deprived microenvironment [3,4]. Increasing evidence demonstrates that lipids play a pivotal role in cancer progression. Low density lipoprotein receptors (LDLR) regulate plasma cholesterol levels by mediating the uptake of cholesterol via low-density lipoprotein (LDL) [5,6]. Recent studies have reported that LDLR levels are associated with the progression of various cancers, including liver cancer, glioblastoma, and glioma [7]. Given the critical role of lipids in cancer progression, targeting LDLR and related lipid metabolism pathways presents a promising therapeutic strategy [8].

Proprotein Convertase Subtilisin/Kexin Type 9 (PCSK9) is a serine protease primarily secreted by hepatocytes [9]. It binds to LDLR and promotes their lysosomal degradation, reducing LDLR levels on the cell membrane. This reduction prevents the endocytosis of LDL via binding with LDLR, leading to elevated plasma LDL levels [10,11]. Beyond its role in binding LDLR, PCSK9 also interacts with major histocompatibility complex class I (MHC I) through its Cys-His-rich domain (CHRD) M2 domain, promoting the degradation of MHC I in tumor cells. This process inhibits CD8⁺T cell responses and facilitates tumor immune escape. Blocking PCSK9 has been shown to enhance the anti-tumor activity of CD8⁺ T cells [[12], [13], [14]]. However, the flat structure of the PCSK9-LDLR interface poses challenges for the development of its small-molecule inhibitors [15]. Currently, small interfering ribonucleic acid (siRNA, inclisiran) and three monoclonal antibodies, alirocumab, evolocumab and tafolecimab, targeting PCSK9, are used clinically to treat hypercholesterolemia and atherosclerosis-related cardiovascular diseases [[16], [17], [18]]. However, their high cost and inconvenient administration have limited widespread use [19]. Therefore, developing new PCSK9 inhibitors is a promising approach for the clinical treatment of both cardiovascular diseases and cancer.

In this study, we constructed td-Tomato expression plasmid driven by PCSK9 promoter, and performed the PCSK9 transcription inhibitor screening in FDA approved drug library. We identified that L-methionine (L-Met) is a transcriptional inhibitor of PCSK9. Methionine, an essential amino acid, is mainly metabolized in the liver into S-adenosylmethionine (SAM), which is the main biological methyl donor and plays critical role in metabolism and epigenetic regulation in mammalian cells [20,21]. Additionally, methionine supports T cell differentiation and proliferation by facilitating DNA and RNA methylation [22]. However, tumor cells compete with T cells for methionine via Solute Carrier Family 43 (Amino Acid System L Transporter), Member 2 (SLC43A2), a key methionine transporter, and methionine deficiency has been shown to impair T cell function [23]. Notably, transcription factors such as Sterol Regulatory Element Binding Transcription Factor 2 (SREBP-2) and Hepatocyte Nuclear Factor 1-Alpha (HNF1α) have been reported to regulate the expression of PCSK9 [24,25]. And we found that nuclear receptor subfamily 1 group I member 2 (NR1I2), also known as pregnane X receptor (PXR), is another key transcription factor of L-methionine-reduced PCSK9 transcription. A moderate concentration of L-methionine inhibits NR1I2 expression, diminishing its binding to the PCSK9 promoter and ultimately leading to reduced PCSK9 transcription. Additionally, the moderate concentration of L-methionine effectively inhibits liver cancer progression by activating CD8⁺T cells in vivo.

Our study introduces a novel and convenient approach to inhibit PCSK9 and offers a new strategy for clinical tumor immunotherapy. Furthermore, our findings underscore the importance of optimal L-methionine intake for enhancing CD8⁺T cell-mediated anti-tumor immunity and provide a theoretical basis for the rational supplementation of L-methionine in clinical hepatocellular carcinoma patients.

Materials and methods

Cell lines

The human embryonic kidney cell line 293FT, human hepatoma cell line HepG2 and Huh7, mouse hepatoma cell line H22 and Hepa1-6 were purchased from ATCC. 293FT, HepG2, Hepa1-6 and Huh7 were cultured in DMEM (L110KJ, BasalMedia) supplemented with 10 % FBS (S711-001S, Lonsera) and 1 % penicillin/streptomycin. H22 were cultured in RPMI 1640 medium (L210KJ, BasalMedia) supplemented with 10 % FBS and 1 % penicillin/streptomycin. The L-methionine deficient DMEM and 1640 culture mediums (X087Z1 and X102J1) were customized from BasalMedia. L-methionine (HY-N0326) was obtained from MCE.

Bioinformatics analysis

PCSK9 differential expression between tumor and adjacent normal tissues across all The Cancer Genome Atlas (TCGA) tumors was obtained from TIMER2.0 (comp-genomics.org) Gene_DE module. The cancer types and their corresponding abbreviations in the TCGA database are listed in Supplementary Table 1. PCSK9 expression between liver cancer and normal tissues was extracted from the GSE84005 dataset on the GEO database. The relationship between PCSK9 and LIHC pathological stage was obtained from GEPIA 2 (cancer-pku.cn). Survival curves for liver cancer patients with high or low PCSK9 expression were generated using Kaplan-Meier plotter (kmplot.com).

Construction of PCSK9 promoter td-tomato plasmid

The PCSK9 promoter sequence was downloaded from the NCBI database and synthesized by BGI Genomics. The backbone plasmid (72486, Addgene) was digested with Agel (R3552S, NEB) and Smal (R0141V, NEB) enzymes, and the PCSK9 promoter sequence was inserted using the homologous recombination enzyme (C115, Vazyme).

Extraction and cultivation of mouse CD8⁺T cells

The 6-well plate was coated with CD3 (2 μg/mL, 16-0031-85, Invitrogen) and CD28 (1 μg/mL, 16-0281-85, Invitrogen) antibodies overnight at 4 °C. The spleens of C57BL/6 mice were grinded and filtered with a 70 μm cell strainer to obtain single-cell suspension. After lysis of red blood cells, the mouse CD8⁺ T cells were purified through EasySep™ Mouse CD8⁺ T Cell Isolation Kit (19853, Stemcell). CD8⁺T cells were stimulated with anti-CD3/CD28 for 2 days, and cultured in RPMI 1640 medium supplemented with 10 % FBS, 1 % Penicillin/Streptomycin/Gentamicin Solution (C0223, Beyotime), 25 mM HEPES solution (15630080, Gibco), 50 μM 2-Mercaptoethanol (21985023, Gibco), 1 mM sodium pyruvate (11360070, Gibco), and 1×MEM Non-essential Amino Acids (11140050, Gibco). The medium was supplemented with 50 U/mL mouse IL2 (212-12, PeproTech) every 2 days, and the T cells were re-stimulated with anti-CD3/CD28 every 7 days.

Real-time quantitative PCR (qRT-PCR)

Total RNA was extracted from cells with RNAiso Plus (9109, Takara). HiScript II One Step qRT-PCR SYBR Green Kits (Q221-01, Vazyme) were used to reverse transcribed the equal amounts of RNA for qRT-PCR. The reaction mixture performed reverse transcription for 3 mins at 50 °C, pre-denatured for 30 s at 95 °C, then denatured for 10 s at 95 °C, annealed and extended for 30 s at 60 °C, for a total 40 cycles. Dissolution curve was obtained in 15 s at 95 °C, 60 s at 60 °C and 15 s at 95 °C. The housekeeping gene Actin was used for normalization. The primers specific for the indicated genes used in qRT-PCR are listed in Supplementary Table 2.

Western blotting

The cells were collected and lysed by lysis buffer (P0013, Beyotime) with 1 mM PMSF (ST507, Beyotime) in ice bath for 30 mins, and then centrifuged at 13,000 rpm for 10 mins at 4 °C. The supernatant was collected and the concentration of proteins were measured with the Enhanced BCA Protein Assay Kit (P0009, Beyotime). After denaturation, equal amounts of proteins were separated by SDS-PAGE gels and transferred onto PVDF membranes (1620177, Biorad). After blocking with 5 % skimmed milk for 2 h at room temperature, membranes were incubated with primary antibodies at 4 °C overnight. Antibodies PCSK9 (bs-6060R, Bioss), NR1I2 (67912-1, Proteintech) and Actin (A1978, Sigma) were used. After washing three times with 1×TBST, membranes were then incubated with the appropriate HRP-linked secondary antibody for 2 h at room temperature. After washing three times with 1×TBST, membranes were incubated with ECL (1705061, biorad) and visualized under the Touch Imager Pro.

Cell viability (CCK8) assay

Ten thousand cells per well were added into a 96-well plate and cultured for 24 or 48 h. Then 10 μL of Enhanced Cell Counting Kit-8 (C0043; Beyotime) reagent was added and incubated at 37 °C for 30 mins, and the absorbance was measured at 450 nm in a microplate reader. The cell viabilities were normalized to the control group.

In vivo therapeutic efficacy

Animal experiments were performed in accordance with protocols approved by the Institutional Animal Care and Use Committee of Army Medical University. The subcutaneous xenograft tumor models were used to investigate in vivo antitumor efficacy of L-methionine. Briefly, BALB/c mice were subcutaneously inoculated with H22 cells (2 × 10⁵ cells/mouse), and C57BL/6 J mice were subcutaneously inoculated with 1:1 mixture of Hepa1-6 cells (2.5 × 10⁶ cells/mouse) and matrix gel (827215, ABW). After the tumor volume reached 50-100 mm³, mice were randomly divided and intraperitoneal injected with different doses of L-methionine every day. The body weight and tumor size were recorded every other day, and tumor size was calculated by the formula (length×width²)/2. Mice were sacrificed at the indicated time points or when the tumor size reached 2000 mm³ (Ethical endpoint). The blood, major organs and tumor tissues were collected for further examination.

Construction of spontaneous liver cancer model

Alb-iCre^+/− (T003814) and MYC^+/− (T009879) C57BL/6JGpt mice were purchased from Gempharmatech. Their respective offspring are used partly for conservation and partly for crossbreeding, overexpressing the proto-oncogene MYC specifically in the liver to form spontaneous hepatocellular carcinomas. The genotype we used was Alb-iCre^+/− MYC^+/−. The primers for genotyping are detailed in Supplementary Table 2.

Flow Cytometry

Tumor tissues were removed and digested with 1 mg/mL Collagenase IV (CC3801G, Coolaber) in RPMI 1640 medium for 30 mins at 37 °C. After filtering through the 70 μm cell strainers, single cell suspensions were subjected to a gradient centrifugation with 40 % and 70 % percoll (17089109, Cytiva) to obtain lymphocyte-enriched compartments. After lysing of red blood cells, lymphocytes were incubated with surface antibodies at 4 °C in the dark for 30 mins, including CD45 (557659, BDbiosciences), CD3 (100234, Biolegend), CD4 (100406, Biolegend), CD8a (45-0081-82, Invitrogen), NK1.1 (108736, Biolegend), PD1 (135206, Biolegend), TIGIT (11-9501-82, Ebioscience), CD69 (562920, BDbiosciences). Dead cells were excluded by the Zombie Aqua^TM Fixable Viability Kit (423101, Biolegend). For cytokine and CD107a detection, cells were stimulated with monensin (00-4505-51, Invitrogen), Brefeldin A (00-4506-51, Invitrogen) and/or CD107a antibody for 4 h and then fixed and permeabilized with Fixation and Permeabilization Solution Kit (554722, BDbiosciences), incubated with antibodies for 20 mins at 4 °C. The antibodies for cytokine detection were IFN-γ (163504, Biolegend), TNF-α (506333, Biolegend), perforin (154304, Biolegend) and GZMB (372214, Biolegend). Samples were analyzed on BD LSRFortessa, and flow cytometry data were analyzed using FlowJo.

Enzyme-linked immunosorbent assay (ELISA)

The 100 mg tumor tissues were homogenized and lysed by a cell lysis buffer with 1 mM PMSF. Then, cell suspensions were centrifuged at 12,000 g for 5 mins. The supernatant was collected to detect the PCSK9 concentration in tumor tissues by using PCSK9 ELISA detection kit (KE10050, Proteintech) according to the manufacturer's instruction. As for the blood sample, centrifuge at 5000 rpm for 5 mins, collect plasma, and then use the same kit for testing.

Hematoxylin-eosin staining (H&E) and Immunohistochemistry (IHC)

The tissues of mice were fixed with paraformaldehyde (PFA) and then sliced, embedded in paraffin and stained with H&E. Immunohistochemical staining was performed by the VECTASTAIN® ABC—HRP Kit (PK-4000, Vectorlabs). After blocked with non-specific proteins, sections were incubated with corresponding antibodies (PCSK9, NR1I2, and H2-Kd (sc-53852, Santa)) at 4 °C overnight. Then, sections were then incubated with a biotinylated IgG antibody at room temperature for 1 h. Finally, sections were visualized with diaminobenzidine (DAB) through a microscope.

Immunofluorescence

After treated with different concentrations of L-methionine, cells were fixed with paraformaldehyde and permeabilized with 0.5 % Triton X-100. Later, cells were stained with primary antibodies at 4 °C overnight. On the next day, secondary antibodies coupled with Alexa 488 were employed and incubated in the dark for 1 h at 37 °C. DAPI (C1006, Beyotime) were used to stain the nuclei. For the tissue sections, CD8a, IFN-γ, TNF-α and GZMB antibodies were used for staining. Immunofluorescence was examined by fluorescence microscopy (Olympus, Japan).

RNA sequencing and analysis

HepG2 cells were treated with 12.5 or 50 μM L-methionine for 24 h. Total RNA was isolated and extracted from cells using the RNAiso Plus. Then the samples were subjected to mRNA library construction and sequencing by BGI (BGISEQ-500 platform, Shenzhen, China). The log₂|(fold change)|≥ 1 in HepG2-L-Met50 vs HepG2-L-Met12.5 cells were defined as cut off for DEGs. The DEGs within the groups were further analyzed with Kyoto Encyclopedia of Genes and Genomes (KEGG) and Gene Set Enrichment Analysis (GSEA) for signaling pathway enrichment. ALGGEN (https://alggen.lsi.upc.es/) and JASPAR (https://jaspar.genereg.net/) was used to predict the transcription factors of PCSK9.

Generation of overexpressing cells

NR1I2 (NM_022002) cDNA (G103867, YouBio) and mPcsk9 (NM_153565) cDNA (G126040, YouBio) were cloned into the pCDH-EF1-copGFP-T2A-Puro vector (72263, Addgene). 293FT cells were co-transfected with lentiviral system VSVG and pSPAX2 together with the overexpression plasmids mentioned above using Lipofectamine 8000 (C0533, Beyotime) for lentivirus package. Two days after transfection, the supernatant was filtered by 0.45 μm and centrifuged with 10 % sucrose solution at 80,000 g to concentrate virus. HepG2 cells were infected with virus for 24 h and then selected with 2 μg/mL puromycin (ST551, Beyotime). Overexpressing efficiency was validated by immunoblotting and fluorescence screening.

Cleavage Under Targets and Tagmentation (CUT & Tag)

CUT & Tag assay was performed with the CUT & Tag kit (HD101, Vazyme). Briefly, HepG2 cells treated with 12.5 (control) or 50 μM L-methionine were bound using Concanavalin A-coated Magnetic Beads Pro, and cell membranes were permeabilized using Digitonin. Through immunoprecipitation using NR1I2 and Protein G antibodies, the MNase nuclease fused to Protein G can accurately target and cut the DNA sequence near NR1I2. Finally, the putative binding site of NR1I2 on PCSK9 promoter were identified by qPCR. The primers specific for this test are listed in Supplementary Table 2.

Dual-luciferase reporter assay

Human PCSK9 promoter and its deletant were cloned into the pGL6-TA basic vector (D2105, Beyotime). Based on the predicted sites by JASPAR and literature reports, we predicted that the key binding site between NR1I2 and PCSK9 promoter was TGCACCCTGACAC. Therefore, we deleted it and inserted the deletant into the pGL6-TA vector. HepG2 cells were transfected with the PCSK9 promoter-pGL6-TA or its deletant and pRL-SV40-N Renilla luciferase reporter plasmids (D2762, Beyotime) for 24 h. Then, the cell lysates were harvested and subjected to luciferase assay with the Dual Luciferase Reporter Gene Assay Kit (RG027, Beyotime).

Statistical analysis

Statistical analysis was performed using GraphPad Prism statistical software. Data are presented as mean ± Standard Deviation (SD). Unpaired Student's t-test was used to compare two groups. One-way ANOVA was used to compare >2 groups. The results were considered significant at *p < 0.05, **p < 0.01, ***p < 0.001.

Results

L-methionine is a transcriptional inhibitor of PCSK9

PCSK9 is overexpressed in multiple cancer types compared to normal tissues and is positively correlated with pathological stage and poor survival in liver cancer patients (Supplementary Fig. 1). To develop small-molecule inhibitors of PCSK9, we constructed a reporter plasmid expressing td-Tomato driven by PCSK9 promoter sequence (Supplementary Fig. 2). Human hepatocellular carcinoma HepG2 cells were stably transfected with this reporter plasmid and used for screening an FDA-approved drug library (Supplementary Table 3). Cells were treated with drugs at a concentration of 100 μM for 24 h, and four drugs that reduced td-Tomato fluorescence intensity were selected for further western blot validation (Fig. 1A-B). Through initial screening, we identified three active pharmaceutical ingredients that reduced PCSK9 transcription, with L-methionine showing the most significant inhibitory effect (Fig. 1C). We then cultured human HepG2 and mouse liver cancer H22 cells in L-methionine-deficient medium, supplemented with varying concentrations of L-methionine, and found that L-methionine significantly inhibited the mRNA and protein expression of PCSK9 in a dose-dependent manner (Fig. 1D-F). These data collectively demonstrate that L-methionine is a transcriptional inhibitor of PCSK9.

Fig. 1.

L-methionine reduces the mRNA and protein expression of PCSK9 in liver cancer cell lines. (A) Hunam PCSK9 (hPCSK9) promoter-tdTomoto plasmids were constructed and transferred into HepG2 cells for FDA-approved drug library screening. (B) Schematic diagram of fluorescence changes during drug screening. (C) The protein level of hPCSK9 in HepG2 cells treated with four candidate drugs at 100 μM for 24 h. (D) The mRNA level of PCSK9 in HepG2 and H22 cells treated with 0, 6.25, 12.5, 25, 50, 100 μM L-methionine, detected by qRT-PCR (n = 3). (E-F) The protein level of PCSK9 in HepG2 and H22 cells treated with 0, 12.5, 25, 50, 100 μM L-methionine for 24 h, validated by western blotting.

L-methionine promotes the cytotoxicity of CD8⁺T cells target liver cancer cells

L-methionine is an essential amino acid crucial for tumor cell metabolism and the cytotoxic function of tumor-infiltrating CD8⁺T cells. CD8⁺T cells are more dependent on L-methionine than tumor cells [23], while excessive L-methionine leads to T cell exhaustion in the tumor microenvironment [26]. To investigate the effects of varying L-methionine concentrations on cell viability of liver cancer cells and CD8⁺T cells, we found that 12.5 μM was the optimal concentration of L-methionine for the growth of liver cancer cells, and L-methionine showing no toxicity to both HepG2 and H22 cells even at 400 μM (Fig. 2A-B). Moreover, Ki67 immunofluorescence staining showed that L-methionine induced the highest Ki67 expression in HepG2 cells at 12.5 μM (Supplementary Fig. 3A). Clone formation experiment revealed that 50 μM L-methionine exhibits the strongest effect in promoting cancer HepG2 clone formation (Supplementary Fig. 3B), we guessed that was because culture medium didn't change for a long time in the cloning experiment resulting in the consumption of L-methionine. For CD8⁺T cells, L-methionine slightly enhanced viability at concentrations below 50 μM (Fig. 2C). More importantly, we co-cultured CD8⁺ T cells with liver cancer cells and found that L-methionine in the range of 25-50 μM significantly enhanced the cytotoxic activity of CD8⁺T cells against liver cancer cells (Fig. 2D). When whole spleen cells are cultured in vitro, L-methionine concentrations up to 100 μM promoted proliferation and increased secretion of IFN-γ, TNF-α, and Perforin of CD8⁺ T cells (Fig. 2E), the large amount of other cells contained in the spleen consume some methionine. These findings indicate that L-methionine enhances the cytotoxicity of CD8⁺T cells against liver cancer cells without causing toxicity at concentrations of 25 and 50 μM, and 50 μM L-methionine provides the greatest improvement in CD8⁺T cell viability, while higher doses above 50 μM increase toxicity. Therefore, we selected 12.5 μM as baseline control dose and 50 μM as the optimal administration dose of L-methionine for further in vitro studies.

Fig. 2.

L-methionine promotes the cytotoxicity of CD8⁺ T cells target liver cancer cells. (A-B) CCK-8 assay was performed to detect the cell viability of HepG2 and H22 cells after L-methionine treatment for 24 h and 48 h (n = 3). The data marked with significant statistical variations are evaluated against the 0 μM reference points. * Represented statistical significance in 24 h, while # in 48 h. (C) CCK-8 assay was performed to detect the cell viability of CD8⁺T cells after L-methionine treated for 24 and 48 h (n = 3). (D) Hepa1-6 and mouse CD8⁺T cells were co-cultured and treated with different concentrations of L-methionine for 24 h. The cell viability of Hepa1-6 was detected by CCK-8. (E) Flow cytometry detected the proportion of CD8⁺T cells and their secretion of IFN-γ, TNF-α and Perforin after treated with L-methionine at different concentrations for 24 h (n = 3). (F) Hepa1-6 cells and CD8⁺T cells were treated with 12.5 (Control) or 50 μM L-methionine for 24 h, respectively, then co-cultured for another 24 h. The mortality rate of Hepa1-6 was detected by Calcein-AM. Bar: 75 μm. (G) td-Tomato expressing H22 cells and CoGFP expressing H22 cells were treated with 12.5 (Control) and 50 μM L-methionine for 24 h respectively, then 1:1 ratio mixed and co-cultured with CD8⁺T cells for another 24 h. Bar: 100 μm. (H) The protein level of PCSK9 in wild type (WT) Hepa 1-6 and mPCSK9-OV Hepa1-6 cells. (I) mPcsk9-OV Hepa1-6 or WT Hepa1-6 cells were co-cultured with CD8⁺T cells at a 1:50 ratio with 12.5 (Control) or 50 μM L-methionine for 24 h. The cell viability of Hepa1-6 was calculated by CCK-8 assay. Bar: 75 μm.

To further investigate whether the regulation of CD8⁺T cell function by L-methionine is related to the reduction of PCSK9 expression in tumor cells, we treated Hepa1-6 with 50 μM L-methionine for 24 h firstly and then co-cultured the treated Hepa1-6 and control Hepa1-6 with CD8⁺T cells for another 24 h. Surprisingly, treated Hepa1-6 were easier killed than control (Fig. 2F). Following that, H22 cells labeled with red fluorescence were designated as the control group, while H22 cells labeled with green fluorescence were exposed to 50 μM L-methionine for a duration of 24 h. Subsequently, the two cell types were combined in equal proportions and co-cultured with T cells for an additional 24 h. It was obvious that the control red-fluorescent H22 survived better (Fig. 2G). These data suggest that L-methionine, in addition to directly affecting T cell proliferation and function, has a pathway that indirectly affects T cells through tumor cells. Then we overexpressed PCSK9 in Hepa1-6 cells (Fig. 2H) and found that PCSK9 overexpression significantly reduced the cytotoxicity effect of CD8⁺T cells, and further reversed the L-methionine-promoted cytotoxicity of CD8⁺T cells target liver cancer cells (Fig. 2I). These data collectively suggest that L-methionine promotes the CD8⁺ T cells-mediated killing effects targeting liver cancer cells by reducing PCSK9.

L-methionine reduce PCSK9 transcription by repressing NR1I2

To investigate the mechanism by which L-methionine inhibits PCSK9 transcription, we conducted RNA sequencing analysis on L-methionine-treated HepG2 cells. Differential expression gene (DEG) analysis revealed that L-methionine significantly increased the expression of 1,720 genes and decreased the expression of 329 genes (Fig. 3A). Gene Set Enrichment Analysis (GSEA) confirmed that L-methionine activates TNF-α signaling pathways in HepG2 cells (Fig. 3B). Kyoto Encyclopedia of Genes and Genomes (KEGG) and biological process (BP) pathway analysis further showed that DEGs were primarily enriched in antigen processing and presentation, toll-like receptor signaling, and lipid metabolism and atherosclerosis (Fig. 3C). To identify transcription factors regulating PCSK9, we predicted the PCSK9 transcription factors in ALGGEN and JASPAR database, and overlapped with our DEGs, identifying four overlapping genes as potential PCSK9 transcription regulators (Fig. 3D). Among these, NR1I2 showed the greatest reduction in L-methionine-treated HepG2 cells (Fig. 3E). Quantitative real-time PCR confirmed that L-methionine significantly downregulated NR1I2 expression (Fig. 3F). Overexpression of NR1I2 increased both mRNA and protein levels of PCSK9, reversing the transcriptional inhibition of PCSK9 mediated by L-methionine (Fig. 3G-H). These findings demonstrate that L-methionine reduces PCSK9 expression by repressing the transcription factor NR1I2.

Fig. 3.

L-methionine inhibits PCSK9 transcription by repressing NR1I2. (A) Volcano plot of RNA-seq data from control HepG2 cells and 50 μM L-methionine-treated HepG2 cells. (B) GSEA of all transcriptomic genes in RNA-seq. (C) KEGG and BP pathway enrichment analyses of DEGs in RNA-seq. (D) Intersection of transcription factors predicted by ALGGEN and JASPAR for hPCSK9 and DEGs obtained from RNA-seq. (E) The Log₂(FC) expression changes of 4 potential PCSK9 transcription regulators in RNA-seq. (F) The mRNA level of hNR1I2 after 12.5 (Control) or 50 μM L-methionine treatment in HepG2 cells, detected by qRT-PCR (n = 3). (G) The mRNA level changes of HepG2 (WT) and NR1I2-overexpressing HepG2 (NR1I2-OV) cells treated with 12.5 (Control) or 50 μM L-methionine (n = 3). (H) The protein levels of NR1I2 and PCSK9 in WT and NR1I2-OV HepG2 cells. (I-J) The putative NR1I2-binding sites on PCSK9 promoter were predicted by the JASPAR. Cut & Tag assay was performed in HepG2 cells by using anti-NR1I2, and qRT-PCR was used to detect the fragments of putative NR1I2-binding sites in immunoprecipitates (n = 3). (K) WT and site deletion PCSK9 promoter-driven luciferase activity was detected in WT and NR1I2-OV HepG2 cells (n = 3).

Next, JASPAR predicted two putative NR1I2 binding sites on the PCSK9 promoter, which our CUT-Tag assay confirmed as authentic, evidenced by successful amplification of both site fragments in the anti-NR1I2 immunoprecipitates. Notably, treatment with 50 μM L-methionine significantly reduced NR1I2 binding at these two sites compared to the control group, with site 1 exhibiting a more pronounced alteration than site 2 (Fig. 3I-J). Subsequent analysis of potential binding sites, predicted by JASPAR, revealed no known NR1I2-associated repeat sequences typically linked to DNA binding at site 2 [27] (Supplementary figure 2). This finding highlights the critical role of site 1. Therefore, we then deleted the TGCACCCTGCACAC sequence from site 1 and constructed a fluorescent reporter plasmid. Deletion of this sequence markedly reduced the fluorescence signal, and even NR1I2 overexpression failed to restore it (Fig. 3K). These results indicate that TGCACCCTGCACAC is essential for NR1I2 binding and NR1I2-mediated PCSK9 transcription. Collectively, these data demonstrate that NR1I2 is a key transcription factor for PCSK9, and L-methionine inhibits PCSK9 transcription by reducing NR1I2 expression.

The moderate concentration of L-methionine effectively inhibits the growth of liver cancer in vivo

Our in vitro findings suggest that the appropriate concentration range of L-methionine is crucial for enhancing CD8⁺T cell-mediated antitumor immunity. To identify the optimal dosage of L-methionine for activating antitumor immunity in liver cancer mouse model, we inoculated Balb/c mice were with H22 liver cancer cells and randomly divided into four groups, receiving 2.5, 5, or 10 mg/kg/day L-methionine or normal saline, respectively. Tumor growth curves showed that 2.5 mg/kg/day L-methionine slightly inhibited tumor growth, while 5 mg/kg/day exhibited the strongest antitumor effects. Interestingly, a higher dose of 10 mg/kg/day did not inhibit tumor growth (Fig. 4A-B). Repeating the setup for both the 5mg/kg/day group and the control group yielded identical outcomes (Fig. 4C-D). This result was consistent with our in vitro findings, where 100 μM L-methionine induced toxicity in CD8⁺T cells. Moreover, PCSK9 levels in the tumor microenvironment of H22 tumors were significantly reduced in mice treated with 5 mg/kg/day L-methionine (Fig. 4E). We further confirmed that 5 mg/kg/day L-methionine effectively inhibited tumor growth in the Hepa1-6 transplanted tumor model (Fig. 4F-H). H&E staining of tumor tissues revealed that sections from mice treated with 5 mg/kg/day L-methionine had fewer densely packed cells and showed sparse areas of apoptotic/necrotic cells (Fig. 4I). Immunohistochemistry staining confirmed that 5 mg/kg/day L-methionine significantly reduced the expression of PCSK9 and NR1I2, while increasing MHC-I (H2-Kd) expression in tumor tissues (Fig. 4J).

Fig. 4.

The moderate concentration of L-methionine effectively inhibits the growth of liver cancer in vivo. (A) Schematic diagram of Balb/c mice inoculated with H22 transplanted tumors and treated with L-methionine. (B) H22 tumor growth curve of Balb/c mice treated with the dose of 2.5, 5, 10 mg/kg/day L-methionine (n = 5). (C-D) H22 tumor volume curve and image under 5 mg/kg/day treatments of L-methionine with control (n = 5). (E) PCSK9 level in H22 tumors from control and 5 mg/kg/day L-methionine treated mice, detected by ELISA (n = 3). (F) Schematic diagram of C57 mice inoculated with Hepa1-6 transplanted tumors and treated with L-methionine. (G-H) Hepa1-6 tumor growth curves and image of control and 5 mg/kg/day L-methionine treated mice (n = 5). (I) H&E staining of H22 tumor sections. Bar: 100 μm. (J) Immunohistochemistry staining of PCSK9, NR1I2 and H2-Kd in H22 tumor sections. Bar: 100 μm.

Further flow analysis of tumor tissues revealed a significantly increase in the proportion tumor-infiltrating CD8⁺ T cells, while decrease the proportion of tumor-infiltrating CD4⁺ T cells, in 5 mg/kg/day L-methionine-treated mice (Fig. 5A-B). We also observed significant increase in the percentage of tumor-infiltrating CD8⁺ T cells express activation markers CD69, and cytokines including IFN-γ, TNF-α, Perforin and GZMB in 5 mg/kg/day L-methionine-treated group (Fig. 5C-D). Immunofluorescence staining further confirmed that 5 mg/kg/day L-methionine treatment markedly increased the number of tumor-infiltrating CD8⁺ T cells, and increased the CD8⁺ T cells activation makers (IFN-γ, TNF-α and GZMB) expression in H22 tumor microenvironment (Fig. 5E). Furthermore, H&E staining and blood biochemistry tests revealed that 5 mg/kg/day L-methionine treatment had no significant toxicity to the mouse heart, liver, spleen, lung and kidney (Supplementary figure 4A). Notably, 5 mg/kg/d L-methionine caused significantly decrease in LDL-C level in blood of mice (Supplementary figure 4B), suggesting 5 mg/kg/day L-methionine treatment not only exhibits efficient anti-liver cancer activity, but also promotes LDL-C degradation by inhibiting PCSK9 in vivo.

Fig. 5.

L-methionine promotes the cytotoxicity of CD8⁺T cells in liver cancer mouse model. (A) Gating strategy of flow cytometry detecting the H22 tumor tissues from control and 5 mg/kg/day L-methionine treated mice. (B) The percentage of tumor-infiltrating CD4⁺and CD8⁺T cell in H22 tumors (n = 5). (C) Frequency of tumor-infiltrating CD8⁺T cells expressing CD69 in H22 tumors (n = 5). (D) Frequency of tumor-infiltrating CD8⁺T cells expressing IFN-γ, TNF-α, Perforin and GZMB in H22 tumors (n = 5). (E) Immunofluorescence staining of CD8, IFN-γ, TNF-α and GZMB in H22 tumor sections. Bar: 30 μm.

The moderate concentration of L-methionine effectively inhibits the development of spontaneous liver cancer

We then investigated the efficacy of L-methionine in activating CD8⁺T cell-mediated anti-tumor immunity in a spontaneous liver cancer mouse model (Fig. 6A). This model was generated by crossing MYC^±and Alb-Cre mice, with the Alb-iCre^±MYC^+/− offspring used for the study (Fig. 6B). Mice were randomly divided into two groups and treated with either 5 mg/kg/day L-methionine or normal saline. The results showed that 5 mg/kg/day L-methionine reversed body weight loss in mice with spontaneous liver cancer (Fig. 6C). Additionally, L-methionine treatment inhibited liver cancer progression and reduced the number of tumor nodules in the liver (Fig. 6D-E). Furthermore, 5 mg/kg/day L-methionine significantly decreased the level of PCSK9 in the blood and liver (Fig. 6F-G) and expression of NR1I2 in the liver (Fig. 6H). The results of flow cytometry revealed an increase in the secretion of IFN-γ and TNF-α by tumor-infiltrating CD8⁺T cells in these mice (Fig. 6I), while fluorescence staining further verified the increase in CD8⁺T cells and their secreted factors (Fig. 6J). These findings demonstrate that moderate L-methionine supplementation inhibits liver cancer development by reducing tumor cell PCSK9 secretion and enhancing CD8⁺T cell-mediated anti-tumor immunity, with the optimal dose being 5 mg/kg/day in mice.

Fig. 6.

L-methionine inhibits the development of spontaneous liver cancer. (A) Schematic diagram of spontaneous liver cancer mice treated with L-methionine. (B) Genotyping identification of spontaneous liver cancer mice. KI: knock-in. (C) Body weight changes of spontaneous liver cancer mice. (D) Representative images of the livers from WT mice, spontaneous liver cancer mice, and 5 mg/kg/day L-methionine-treated spontaneous liver cancer mice. (E) H&E staining of liver tissues. Bar: 100 μm. (F) PCSK9 level in blood from control and 5 mg/kg/day L-methionine treated spontaneous liver cancer mice, detected by ELISA (n = 3). (G) Immunohistochemistry staining of PCSK9 in liver tissues. Bar: 100 μm. (H) Immunohistochemistry staining of NR1I2 in liver tissues. Bar: 100 μm. (I) Frequency of tumor-infiltrating CD8⁺T cells expressing IFN-γ, TNF-α in tumors from the spontaneous liver cancer mice (n = 5). (J) Immunofluorescence staining of CD8, IFN-γ, TNF-α and GZMB in liver tissues. Bar: 30 μm.

Discussion

Methionine is essential for synthesizing various biologically active substances, lipid metabolism, and tumor progression [28]; however, its role remains controversial [29,30]. Studies indicate that a high-methionine diet increases fat/body weight ratios, elevates plasma triglyceride (TG) levels, decreases high-density lipoprotein cholesterol (HDL) [31]. Conversely, a methionine-deficient diet has been linked to fatty liver, slowed growth, weakness, edema, skin lesions, and extended lifespan in mice [32]. Both dietary methionine restriction and supplementation have been shown to either enhance or inhibit antitumor immunity and tumor progression, highlighting the complexity of methionine's function [[33], [34], [35]]. PCSK9 is a newly identified protease that contributes to lipid metabolism disorders and tumor progression. Notably, we discovered that L-methionine inhibits PCSK9 transcription in both mouse and human hepatocellular carcinoma cells. Our study shows that moderate concentrations of L-methionine can promote CD8⁺ T cell-mediated killing of liver cancer cells and inhibit the development of liver cancer cell xenograft tumors and spontaneous liver cancer.

Given the essential role of methionine in tumor progression and the greater uptake capacity of tumor cells compared to immune cells in the tumor microenvironment, it is crucial to determine an appropriate concentration of methionine intake for cancer patients. Reports indicate that plasma methionine levels in healthy individuals average around 30 μM, whereas levels in tumor patients are often lower [23,36]. One study recommends a safe dosage range of 0.3 % to 0.9 % (236-705 mg/kg/day) for methionine supplementation, while emphasizing that a dietary restriction to 0.17 % has the potential to extend lifespan in rodents [37]. Our results suggested that liver cancer growth can be inhibited when cultured with L-methionine at a dose of 50 μM in vitro and supplemented with an additional 5 mg/kg/day of L-methionine under normal dietary conditions in mice. This dose did not adversely affect hepatic or renal function and lowered plasma LDL-C (Supplementary Fig. 4). Based on our findings and dose conversion formulas [38], we recommend hepatocellular carcinoma patients intake L-methionine at a dose of 0.375 mg/kg/day which ensured enter the bloodstream. However, all current studies lack clinical evidence to support these findings. Indeed, methionine is irreplaceable for humans in maintaining health as an essential amino acid. So, it is urgent and important to explore the appropriate range of intake doses.

Methionine serves as a precursor for the universal methyl group donor and regulates gene transcription and translation through epigenetic modifications, including DNA and histone methylation [39,40]. Previous studies have indicated that methionine oxidation activates transcription factors in response to oxidative stress [41]. HNF1α and SREBP-2 have been shown to regulate PCSK9 expression in HepG2 cells and the liver [42]. In this study, we identified NR1I2 as a PCSK9 transcription factor, which is a member of the nuclear receptor family, activated by various exogenous ligands and is highly expressed in the liver, small intestine, and colon [43]. Activation of NR1I2 in both humans and mice induces PCSK9 expression and raises plasma LDL levels, contributing to atherosclerosis [24,44]. Furthermore, NR1I2 activation may reduce the efficacy of the antitumor drug irinotecan in colon cancer patients [45]. Notably, NR1I2-deficient mice exhibit upregulation of the Toll-like receptor signaling pathway and increased inflammatory factors such as TNF-α [46]. Our study found that L-methionine decreases the expression of both PCSK9 and NR1I2, while overexpression of NR1I2 reverses the inhibition of PCSK9 by L-methionine. We further clarified that L-methionine represses PCSK9 transcription by affecting the binding of NR1I2 to the TGCACCCTGCACAC region of the PCSK9 gene. Our findings are consistent with existing studies indicating that NR1I2 functions as an upstream regulator of PCSK9. Furthermore, our study provides novel mechanistic insights by demonstrating that NR1I2 directly modulates PCSK9 expression at the transcriptional level.

In summary, our findings offer new insights into the inhibitory effects of L-methionine on hepatocellular carcinoma at a dose of 50 μM in vitro and 5 mg/kg/day in mice. Furthermore, L-methionine enhances CD8⁺T cell-mediated killing of hepatocellular carcinoma by inhibiting PCSK9 transcription through the reduction of NR1I2 (Fig. 7). These results provide a theoretical basis for the rational supplementation of methionine in liver cancer patients and highlight NR1I2 and PCSK9 as potential therapeutic targets.

Fig. 7.

Diagram showing the molecular mechanisms of L-methionine on anti-tumor activity.

List of Abbreviation

L-Met, L-methionine

HCC, hepatocellular carcinoma

LDL, low-density lipoprotein

LDL-R, low-density lipoprotein receptor

PCSK9, proprotein convertase subtilisin/Kexin Type 9

MHC I, major histocompatibility complex class I

NR1I2, nuclear receptor subfamily 1 group I member 2

PXR, pregnane X receptor

ATCC, American Type Culture Collection

DMEM, Dulbecco's modified Eagle's medium

RPMI 1640, Roswell Park Memorial Institute 1640 medium

FBS, fetal bovine serum

PMSF, phenylmethanesulfonyl fluoride

SDS-PAGE, sodium dodecyl-sulfate polyacrylamide gel electrophoresis

PVDF, polyvinylidene fluoride

ECL, enhanced chemiluminescence

CCK-8, Cell Counting Kit-8

IFN-γ, interferon gamma

TNF-α, tumor necrosis factor α

GZMB, granzyme B

DEGs, differentially expressed genes

TF, transcription factor

APIs, active pharmaceutical ingredients

TCGA, The Cancer Genome Atlas

SLC43A2, Solute Carrier Family 43 (Amino Acid System L Transporter), Member 2

HNF1α, Hepatocyte Nuclear Factor 1-Alpha

SREBP-2, Sterol Regulatory Element Binding Transcription Factor 2

CRediT authorship contribution statement

Chengsha Yuan: Writing – original draft, Methodology, Data curation, Conceptualization. Changpeng Hu: Software, Funding acquisition, Formal analysis, Conceptualization. Huyue Zhou: Visualization, Validation, Investigation, Formal analysis. Wuyi Liu: Visualization, Validation, Data curation. Wenjing Lai: Visualization, Methodology. Yafeng Liu: Visualization. Yue Yin: Software. Guobing Li: Writing – review & editing, Supervision, Project administration, Conceptualization. Rong Zhang: Writing – review & editing, Supervision, Funding acquisition, Conceptualization.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.