PCK1 attenuates tumor stemness via activating the Hippo signaling pathway in hepatocellular carcinoma

Abstract

Liver cancer stem cells were found to rely on glycolysis as the preferred metabolic program. Phosphoenolpyruvate carboxylase 1 (PCK1), a gluconeogenic metabolic enzyme, is down-regulated in hepatocellular carcinoma and is closely related to poor prognosis. The oncogenesis and progression of tumors are closely related to cancer stem cells. It is not completely clear whether the PCK1 deficiency increases the stemness of hepatoma cells and promotes the oncogenesis of hepatocellular carcinoma. Herein, the results showed that PCK1 inhibited the self-renewal property of hepatoma cells, reduced the mRNA level of cancer stem cell markers, and inhibited tumorigenesis. Moreover, PCK1 increased the sensitivity of hepatocellular carcinoma cells to sorafenib. Furthermore, we found that PCK1 activated the Hippo pathway by enhancing the phosphorylation of YAP and inhibiting its nuclear translocation. Verteporfin reduced the stemness of hepatoma cells and promoted the pro-apoptotic effect of sorafenib. Thus, combined treatment with verteporfin and sorafenib may be a potential anti-tumor strategy in hepatocellular carcinoma.

Introduction

The incidence and mortality of primary liver cancer are among the top 10.¹ The high recurrence rate and chemotherapy resistance of hepatocellular carcinoma (HCC) are the main reasons for its poor prognosis. Increasing evidence supported cancer stem cells (CSCs) as a major driver of chemoresistance.² CSCs are a subpopulation of tumor cells with self-renewal and differentiation abilities, which contribute to poor prognosis and high mortality in patients with HCC. Several liver CSC markers have been found, including CD90, CD13, CD133, EpCAM, CD24, and CD44.³ Metabolic remodeling is one of the hallmarks of cancer. Tumor cells generally prefer to obtain cellular energy and biomacromolecules through glycolysis.⁴ Unlike tumor cells, CSCs utilize oxidative phosphorylation (OXPHOS) or glycolysis as their primary metabolic source, depending on the various cancer types.⁵ Liver CSCs are glycolysis-dependent and are characterized as an up-regulated expression of glycolysis-related genes including hexokinase 2 (HK2), phosphoglycerate mutant enzymes 1 (PGAM1), glucose transporter 1 (GLUT1), and pyruvate dehydrogenase kinase 4 (PDK4).⁶ Gluconeogenesis is the reverse reaction of glycolysis. As a key enzyme of gluconeogenesis, phosphoenolpyruvate carboxykinase 1 (PCK1) is down-regulated in HCC and is associated with poor prognosis. PCK1 plays a suppressive role in HCC, including inhibiting aerobic glycolysis, promoting oxidative stress and apoptosis, blocking cell cycle progression, inhibiting cell proliferation and metastasis, and restraining tumor angiogenesis.7, 8, 9, 10 A previous study has reported that PCK1 is up-regulated in melanoma tumor-repopulating cells and promotes the tumorigenicity of tumor-repopulating cells.¹¹ However, whether PCK1 is involved in the regulation of stemness of hepatoma cells is currently unclear.

The Hippo pathway has vital biological functions in organ growth and tumorigenesis. In the Hippo pathway, mammalian STE20-like protein kinase 1 and 2 and large tumor suppressors 1 and 2 (LATS1/2), as tumor suppressors restrain the oncogenic effects of YAP/TAZ-TEAD.¹² In addition to participating in cell proliferation, organ growth, and cancer development, recent evidence indicates that the Hippo pathway has important effects on the biological functions of CSCs, including drug resistance, self-renewal, and epithelial-to-mesenchymal transition.¹³

In this study, we found that PCK1 attenuated the stemness of hepatoma cells and inhibited its tumorigenicity. Furthermore, PCK1 increased the sensitivity of hepatoma cells to sorafenib. Further, we found that PCK1 deletion contributed to the stemness of hepatoma cells by up-regulating YAP. Verteporfin, a YAP/TAZ-selective inhibitor, sensitized hepatoma cells to sorafenib-induced apoptosis. Therefore, verteporfin combined with sorafenib could be a potential anti-tumor strategy for HCC.

Materials and methods

Full details are available in Supplementary Materials.

Cell lines and clinical specimens

Human hepatoma cell line PLC/PRF/5 was purchased from the American Type Culture Collection (ATCC, Manassas, VA, USA), and Huh7 and MHCC-97H were from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China). Cells were maintained in DMEM supplemented with 10% FBS, 100 mg/mL penicillin, and 100 U/mL streptomycin (all were purchased from Gibco, Grand Island, NY, USA) in 5% CO₂ at 37 °C. HCC tumor specimens were obtained from the Second Affiliated Hospital of Chongqing Medical University with the informed consent of the patients according to the Institutional Review Board approval (reference number: 2022052).

Cells were cultured in medium supplemented with Sorafenib (HY-10201; MedChemExpress, New Jersey, USA) or Verteporfin (YAPi; HY-B0146; MedChemExpress), OSMI-1 (T16409; TargetMol, MA, USA), Thiamet G (TG; T6056; TargetMol, MA, USA) and then collected for analysis.

Western blot

Total protein was extracted by cell lysis buffer (Beyotime Biotechnology, Jiangsu, China) with 1 mM phenylmethanesulfonyl fluoride (Beyotime). The extraction and isolation of nuclear and cytoplasmic protein were performed using the Nuclear and Cytoplasmic Protein Extraction Kit (Beyotime) according to the manufacturer's instructions. Protein samples were separated by SDS/PAGE and electro-transferred to PVDF membranes (Millipore, Billerica, MA, USA). The immunoblots were probed with the indicated antibodies against PCK1 (1:1000; BS6870; Bioworld, Atlanta, GA, USA), YAP (1:1000; 14074; Cell Signaling Technology, Danvers, MA, USA), p-YAP (1:10,000; ab76252; Abcam, Cambridge, UK), O-GlcNAc (1:1000; ab2739; Abcam), Survivin (1:1000; 10508-1-AP; Proteintech Group Inc., Rosemont, IL, USA), H3 (1:5000; H0164; Millipore), β-Tubulin (1:50,000; 66240-1-Ig; Proteintech Group Inc.), and β-actin (1:4000, TA-09; ZSGB-BIO, Bejing, China). Protein bands were visualized with Clarity^TM Western ECL Substrate (Bio-Rad).

Immunofluorescence staining

Cell slides were fixed with 4% paraformaldehyde for 20 min, and penetrated with 0.5% Triton X-100 (Sigma-Aldrich, St Louis, MO, USA) for 15 min. Then, the cells were blocked with normal goat serum and incubated with anti-YAP (1:100; Cell Signaling Technology) at 4 °C overnight, and specific signals were visualized with Alexa Fluor 552 secondary antibody (ZF0316; ZSGB-BIO). The cells were counterstained with 1 μg/mL DAPI (10236276001, Roche Diagnostics GmbH, Mannheim, Germany) and visualized by a laser-scanning confocal microscope (Leica TCS SP8, Solms, Germany).

Sphere formation assay

Tumor cells were cultured in DMEM/F12 (DMEM/F12 basic, Gibco) supplemented with 1X B27 (B-27 Supplement (50X), serum-free, Gibco), 20 ng/mL human recombinant epidermal growth factor (Animal-Free Recombinant Human EGF, PeproTech, Cranbury, NJ, USA) and 20 ng/mL human recombinant basic fibroblast growth factor (Recombinant Human FGF-basic, PeproTech). Cells were cultured in 6-well ultra-low adhesion plates (Corning, Tewksbury, MA, USA) for 10–14 days.

Flow cytometry

Cells were collected and resuspended in 100 μL PBS containing PE anti-CD44 antibody (103007, BioLegend, San Diego, CA, USA) or APC anti-CD133 antibody (566596, BD Biosciences, San Jose, CA, USA). After 30-min incubation at 4 °C in the dark, cells were washed with PBS and resuspended with 300 μL PBS. Samples were detected by FACSCanto^TM II flow cytometer (BD Biosciences) and the data were analyzed by FlowJo.

Animal models

BALB/c male nude mice were randomly grouped (n = 7/group). Amount of 50 μL of MHCC-97H cell suspension was mixed with 50 μL of the Matrigel (356234, Corning, NY, USA). The cell mixtures were subcutaneously injected into the mice. Tumor incidence was monitored 4 weeks after tumor cells inoculation. Tumor-initiating frequency was calculated by limiting dilution according to the protocol provided on the web (http://bioinf.wehi.edu.au/software/elda/).¹⁴ In addition, BALB/c nude mice were randomly divided into five groups (n = 6/group). Parental and PCK1-KO PLC/PRF/5 cells were collected and implanted into the livers of nude mice (1 × 10⁵ cells per injection). The mice implanted with PCK1-KO cells were randomly divided into four groups including (i) vehicle, (ii) sorafenib (30 mg/kg) intraperitoneally every other day, (iii) verteporfin (100 mg/kg) intraperitoneally every other day, and (iv) the combination of sorafenib and verteporfin for 2 weeks, respectively. Those mice were sacrificed after implantation for 5 weeks. All animal procedures were approved by the Research Ethics Committee of Chongqing Medical University (reference number: 2022052).

Data mining

Gene expression data and corresponding clinical data for patients with HCC were obtained from The Cancer Genome Atlas Liver Hepatocellular Carcinoma (TCGA-LIHC) dataset.^15,16 The Kaplan-Meier survival curves were generated by the “survminer” package of R (Version 3.6.3). Raw data from the GSE14897 and GSE25417 datasets were downloaded from the Gene Expression Omnibus (GEO) database (https://www.ncbi.nlm.nih.gov/geo/).

Statistical analysis

Data were shown as mean ± standard deviation (SD). Statistical analyses were performed using Student's t-test for two-group comparisons, and one-way ANOVA for multi-group comparisons. Pearson correlation coefficient (r) was used to test the linear correlation. P-values <0.05 were considered statistically significant (^∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001), and “ns” indicates no significance. Statistical analyses were conducted using GraphPad Prism 6.0 software (La Jolla, CA, USA).

Results

PCK1 attenuates the stemness of hepatoma cells and suppresses tumorigenicity in vitro and in vivo

CSCs are a crucial tumor cell population with high tumorigenicity in HCC.¹⁷ To investigate the effect of PCK1 on the stemness of hepatoma cells, we detected the mRNA expression of CSC markers. The results indicated that PCK1 deficiency significantly increases the mRNA expression of well-characterized stemness markers including CD133, EPCAM, CD44, and NANOG, as compared with parental cells (Fig. 1A). Conversely, PCK1 overexpression obtained the opposite effects (Fig. 1B). Moreover, in vitro sphere formation assays showed that PCK1 deletion enhanced the self-renewal ability of PLC/PRF/5 cells (Fig. 1C), but PCK1 overexpression inhibited the self-renewal ability of Huh7 and MHCC-97H cells (Fig. 1D). It has been reported that CD133⁺ CD44⁺ hepatoma cells showed CSCs' properties, including proliferation and self-renewal.¹⁸ In our study, flow cytometry analysis showed that the proportion of CD133⁺ or CD44⁺ HCC cells was increased in PLC/PRF/5 cells with PCK1 deficiency (Fig. 1E, F, upper panel) and was decreased in Huh7 and MHCC-97H cells overexpressing PCK1 (Fig. 1E, F, middle and lower panels). These results indicated that PCK1 reduces mRNA expression of CSC markers and the self-renewal property of hepatoma cells.

Figure 1.

PCK1 attenuates the stemness properties of hepatoma cells. (A–F) The effect of PCK1-KO (parental and PCK1-KO PLC/PRF/5 cells) and PCK1-OE (Huh7 or MHCC-97H cells infected with AdGFP, AdPCK1) on the stemness of HCC cells showed by mRNA level of CSC markers (A, B), abilities of sphere formation (C, D), and CD133⁺ cell population (E) and CD44⁺ cell population (F). Statistical analysis was shown as mean ± SD (^∗P < 0.05, ^∗∗P < 0.001, ^∗∗∗P < 0.0001; ns, no significance). Three samples one-way ANOVA was performed for (A, C) and two sample t-test for (B, D). Magnification: 40 × .

To further clarify the above findings, the tumorigenicity of HCC cells on PCK1 overexpression in vivo was examined in a nude mouse model, with subcutaneous inoculation and a limiting dilution approach with 1 × 10³, 1 × 10⁴, and 1 × 10⁵ cells. Consistent with the in vitro results, PCK1 attenuates the tumorigenicity of hepatoma cells in vivo (from 7/7 to 4/7 with 1 × 10⁵ cells; from 4/7 to 2/7 with 1 × 10⁴ cells; from 3/7 to 0/7 with 1 × 10³ cells). In addition, tumor size and mass were smaller in the AdPCK1 group when compared with the AdGFP group. Estimated tumor-initiating frequency was reduced from 1/7290 to 1/88308 after overexpression of PCK1 (Fig. 2A). These findings indicated that PCK1 attenuates the stemness properties of hepatoma cells and tumorigenicity in vitro and in vivo.

Figure 2.

PCK1 suppresses in vivo tumorigenicity. MHCC-97H cells infected with AdGFP or AdPCK1 were subcutaneously inoculated with 1 × 10³, 10⁴, or 10⁵ cells in nude mice. Tumor growth, tumor incidence, and latency period were evaluated. Tumor latency data were shown as mean ± SD. Two sample t-test; n = 7; ^∗P < 0.05, ^∗∗P < 0.001.

PCK1 enhances the sensitivity of hepatoma cells to sorafenib treatment

Liver CSCs contribute to initiating tumor development, inducing tumor progression, and correlating with chemotherapy resistance.¹⁹ Sorafenib-resistant hepatoma cells exhibited significant stemness features.²⁰ To explore the role of PCK1 in the sensitivity of hepatoma cells to sorafenib treatment, we examined the cell proliferation ability in PCK1 knockout (PCK1-KO) and PCK1 overexpression (PCK1-OE) cells treated with sorafenib. The results of cell growth curves and colony formation assays indicated that PCK1-OE significantly restricted cell growth and sensitized hepatoma cells to sorafenib (Fig. 3A–D; Fig. S1A, B). To observe the effects of sorafenib on apoptosis, we examined the apoptosis rate of cells after sorafenib treatment. Flow-cytometric analysis showed that sorafenib induced cell apoptosis. More importantly, PCK1-KO attenuated the proapoptotic effect of sorafenib on hepatoma cells (Fig. 3E). Conversely, PCK1-OE obtained the opposite effects (Fig. 3F; Fig. S1C). Together, these results implied that PCK1 enhances the sensitivity of hepatoma cells to sorafenib via promoting apoptosis.

Figure 3.

PCK1 enhances the sensitivity of hepatoma cells to sorafenib. (A–D) The effect of PCK1-KO or PCK1-OE (Huh7 cells) on cell growth curves and colony formation assays. (E, F) Flow-cytometric analysis of apoptosis. PLC/PRF/5 cells were treated with 5 μM sorafenib, and Huh7 cells were treated with 3 μM sorafenib. Statistical data were shown as mean ± SD. Two sample t-test; ^∗P < 0.05, ^∗∗P < 0.001, ^∗∗∗P < 0.0001).

PCK1 inhibits YAP nuclear translocation

The Hippo pathway has important effects on CSCs' self-renewal and drug resistance.¹³ YAP/TAZ-TEAD is a tumor promoter in the Hippo pathway.²¹ Thus, we examined the effect of PCK1 on YAP activation. Immunoblotting revealed YAP and downstream Survivin were significantly increased and p-YAP decreased in PCK1-KO cells (Fig. 4A). Conversely, PCK1 overexpression decreased protein levels of YAP and Survivin but increased the expression of p-YAP (Fig. 4B; Fig. S2A). Previous studies have found that YAP is modified by O-GlcNAc and its phosphorylation is inhibited.22, 23, 24, 25 Our previous study found that PCK1 deletion up-regulated global cellular O-GlcNAcylation.⁹ Therefore, we hypothesized that PCK1 deletion could up-regulate YAP O-GlcNAcylation and inhibit its phosphorylation. To verify this hypothesis, we examined whether PCK1 regulates YAP O-GlcNAcylation in PCK1-KO or PCK1-OE cells. The results showed that PCK1-KO strengthened YAP O-GlcNAcylation and decreased p-YAP, while OSMI-1 (the OGT inhibitor) partially alleviated YAP O-GlcNAcylation and increased p-YAP (Fig. 4C, D). In contrast, PCK1-OE decreased YAP O-GlcNAcylation and increased p-YAP, while TG (the OGA inhibitor) partially reversed YAP O-GlcNAcylation and weakened p-YAP (Fig. S2B–E), suggesting PCK1 deletion up-regulates YAP O-GlcNAcylation, inhibits its phosphorylation, and inhibits the activation of the Hippo/YAP signaling pathway. Next, we detected YAP expression in cytoplasm and nucleus extraction in hepatoma cells. Surprisingly, endogenous YAP presents a strong nuclear localization in PCK1-KO cells. Whereas, PCK1 overexpression caused a significant down-regulation of nuclear YAP in Huh7 and MHCC-97H cells (Fig. 4E; Fig. S2F, G). The results were also confirmed by immunofluorescence (Fig. 4F–H). In addition, immunohistochemical staining showed that YAP and Survivin were highly expressed in human HCC tissues, while p-YAP was lowly expressed (Fig. 4I; Fig. S2H). Together, these results suggested that PCK1 down-regulates YAP O-GlcNAcylation, promotes YAP phosphorylation, and inhibits YAP nuclear translocation.

Figure 4.

PCK1 inhibits YAP nuclear translocation and promotes its phosphorylation. (A, B) The protein level of YAP, p-YAP, and Survivin in PCK1-KO cells (A) and PCK1-OE cells (Huh7) (B). (C, D) PCK1-KO cells were treated with 50 μM OSMI-1 for 12 h, followed by sWGA pull-down assay (C) or Western blot (D). The presented input YAP was adjusted to a similar level (C). (E–H) YAP expression in cytoplasm and nucleus extraction in hepatoma cells. Cell samples were detected by Western blotting (E) or immunofluorescence (F–H). Scale bar, 20 μm. Nuclei were counterstained with DAPI. (I) PCK1 and p-YAP expression in non-tumor tissues (N) and tumor tissues (T) from 14 patients with HCC and semi-quantitative analyses of immunohistochemistry (IHC) data. Scale bars, 100 μm. Statistical data were shown as mean ± SD. Two sample t-test; ^∗∗∗P < 0.0001.

Verteporfin reduces hepatoma cell stemness and increases sensitivity to sorafenib

Previous studies indicate that Hippo pathway plays important roles in self-renewal and cell apoptosis.^26,27 We hypothesized that loss of PCK1 could promote stemness of CSCs via Hippo pathway. To verify this hypothesis, we examined the mRNA expression of CD133, EPCAM, CD44, and NANOG, and the self-renewal ability of PCK1-KO cells treated with verteporfin or of PCK1 and YAP combined knockout cells. The results showed that both YAP inhibitor and YAP knock-out reduced the mRNA expression level of CSC markers and sphere formation (Fig. 5A, B; Fig. S3A, B). Furthermore, PCK1-KO cells treated with verteporfin in combination with sorafenib showed significantly higher apoptosis (Fig. 5C). Verteporfin combined with sorafenib induced a synergistic effect on inhibiting proliferation (Fig. 5D, E). Similarly, YAP deficiency retarded the proliferation ability of PCK1-KO cells (Fig. S3C, D). To confirm the synergistic effect of verteporfin and sorafenib in inhibiting tumor growth in vivo, PCK1-KO cells were implanted into the nude mice livers. The synergistic inhibitory effect of verteporfin and sorafenib was confirmed in the orthotopic implantation model, consistent with the results of cell experiments in vitro (Fig. 5F–H). Collectively, these data indicated targeting YAP with verteporfin could attenuate stemness properties and increase sensitivity to sorafenib in vitro and in vivo.

Figure 5.

Verteporfin weakens CSCs' properties and potentiates the anti-tumor effect of sorafenib. (A, B) The mRNA expression of stemness-related genes (A) and abilities of sphere formation (B) in PCK1-KO cells treated with 20 μM YAP inhibitor (verteporfin). (C–E) Flow-cytometric analysis of apoptosis (C), cell growth curves (D), and colony formation assays (E) were performed after treatments with sorafenib and verteporfin as indicated. PCK1-KO cells were treated with 5 μM sorafenib and 20 μM verteporfin. YAPi, YAP inhibitor, verteporfin. (F) Representative images of orthotopic liver implantation of PCK1-KO cells in nude mice. (G) Liver tumor weight was assessed (n = 6 for each group). (H) Ki67 staining in liver tumor tissues. Scale bar, 100 μm. Statistical data were shown as mean ± SD. Multi samples one-way ANOVA; ^∗P < 0.05, ^∗∗P < 0.001, ^∗∗∗P < 0.0001. Magnification: 40 × .

PCK1 expression was negatively correlated with the stemness of hepatoma cells

Finally, we assessed the mRNA level of PCK1 in spheres derived from PLC/PRF/5 cells. The result showed PCK1 weakly expressed in spheres (Fig. 6A). In addition, PCK1 was also lowly expressed in embryonic stem cells by analysis of its expression profiles derived from GEO datasets (GSE14897²⁸ and GSE25417²⁹) (Fig. 6B, C). Furthermore, PCK1 was negatively correlated to EPCAM or CD133 mRNA level (Fig. 6D, E) and survival analysis showed that patients with lower PCK1 but higher CD44 or CD133 had poorer overall survival (Fig. 6F, G). In summary, these results supported that PCK1 suppresses the stemness properties in HCC.

Figure 6.

PCK1 is down-regulated in liver CSCs. (A) The mRNA level of PCK1 in spheres derived from PLC/PRF/5 cells. (B)PCK1 was analyzed in undifferentiated embryonic stem cells (ESCs) (GSE14897). Un-dif, undifferentiated; Hep-dif, hepatic-differentiation. Statistical data were shown as mean ± SD (^∗P < 0.05; ^∗∗P < 0.001). (C)PCK1 expression levels were analyzed on day 5 and day 15 during hepatic differentiation from ES cells (GSE25417). (D, E) Correlation analysis of PCK1 and EPCAM(D) or CD133(E) mRNA levels in HCC using data from the TCGA database (Pearson's correlation coefficient). (F, G) Prognostic significance of the combination of PCK1 with CD44(F) or CD133(G) with overall survival (OS). Two sample t-test was performed for (A–C) and Pearson correlation coefficient (r) for (D, E). (H) Schematic illustration of the role of PCK1 in liver CSCs. Loss of PCK1 increases the stemness of hepatoma cells by promoting YAP nuclear translocation. Inhibition of YAP can attenuate the stemness of hepatoma cells and promote cell apoptosis.

Discussion

CSCs are a population of tumor cells with a self-renewal property. CSCs have distinct metabolic manners compared with tumor cells and primarily rely on glycolysis or OXPHOS in a tumor type-dependent manner. The CSCs of lung cancer, pancreatic cancer, glioblastoma, ovarian cancer, and AML relied on OXPHOS.30, 31, 32, 33 Nevertheless, the CSCs of breast cancer, nasopharyngeal cancer, and liver cancer mainly utilized glycolysis.34, 35, 36 A previous study observed that the expression of glycolytic genes (HK2, PDK4, LUT1, PGAM1) was up-regulated and the expression of gluconeogenetic enzymes (G6Pase, PCK) was down-regulated in CD133⁺ liver CSCs.⁶ In this study, we found that PCK1 attenuates the stemness of hepatoma cells and tumorigenicity via Hippo pathway. Also, PCK1 enhances the sensitivity of hepatoma cells to sorafenib. Furthermore, verteporfin, a YAP/TAZ-selective inhibitor, could reduce the stemness of hepatoma cells and enhance the pro-apoptotic effect of sorafenib (Fig. 6H). This study uncovered the importance of the gluconeogenic rate-limiting enzyme PCK1 in the self-renewal and tumorigenic properties of liver CSCs.

Recent studies have shown that metabolic enzymes have essential roles in the maintenance and regulation of stemness in CSCs. Glutaminase 1 promoted CSCs' properties via ROS/Wnt/β-catenin pathway.³⁷ Fructose-1,6-bisphosphatase suppressed glycolysis and promoted apoptosis in cancer stem-like cells.³⁸ High levels of phosphoenolpyruvate carboxykinase 2 (PCK2) maintained tumor-initiating cells and promoted tumor aggressiveness in prostate cancer.³⁹ In another study on melanoma, down-regulation of PCK2 was essential for tumor-repopulating cell growth in vitro and tumorigenesis in vivo in melanoma.⁴⁰ Unexpectedly, up-regulation of PCK1 was necessary for the growth of tumorigenic melanoma tumor-repopulating cells.¹¹ Previous studies have shown that the role of PCK1 in cancers originating in different tissues may be inconsistent.^41,42 PCK1 promotes tumor cell proliferation and survival in cancers originating in non-gluconeogenic tissues, including melanoma, colon carcinoma, gastric cancer, and pancreatic cancer.^11,43, 44, 45 In contrast, PCK1 expression was decreased in cancers derived from gluconeogenic tissues, such as the liver and kidney. PCK1 inhibits tumor cell proliferation and survival in HCC and renal cell carcinoma.^7,46, 47, 48 It has been reported that acetylation-induced isoenzyme conversion resulted in PCK1 down-regulation and PCK2 up-regulation in HepG2-resistant sorafenib cells, and high expression of PCK2 reduced the sensitivity of HCC cells to sorafenib.⁴⁹ However, it did not completely deny the role of PCK1 in the sensitivity of hepatoma cells to sorafenib. The biological phenotype and gene expression of sorafenib-resistant HCC cells have been altered. In this study, PCK1 expression increased gradually with cell differentiation. PCK1 inhibits the expression of CSC markers, the stemness of hepatoma cells, and tumorigenicity. This study helps improve the understanding of the involvement of PCK1 in the tumorigenesis and development of different cancer types.

The Hippo pathway is involved in cell proliferation, organ growth, tumorigenesis, and stemness of CSCs.⁵⁰ YAP/TAZ is a tumor promoter in the Hippo pathway. LATS1/2 phosphorylates YAP/TAZ and inhibits nuclear translocation of YAP/TAZ, thereby inhibiting downstream gene transcription.⁵¹ Our results showed that PCK1 promotes the activation of Hippo pathway, decreases YAP nuclear translocation, and inhibits the stemness of hepatoma cells. Previous studies have shown that YAP is modified by O-GlcNAc, and YAP O-GlcNAcylation disrupts its interaction with LATS1, preventing its phosphorylation and activating its activity.22, 23, 24, 25 We have found that PCK1 deficiency increases global O-GlcNAcylation in HCC.⁹ In this study, we found that PCK1 deficiency suppresses the phosphorylation of YAP by increasing its O-GlcNAcylation. These results suggest that gluconeogenesis dysregulation and abnormal energy metabolism play an important role in maintaining the stemness and tumorigenesis of hepatoma cells.

Sorafenib is approved as a first-line treatment for patients with HCC, but its clinical efficacy is poor. Drug resistance was identified as a major barrier to the failure of sorafenib therapy in HCC.⁵² Drug resistance was thought to be related to tumor heterogeneity. Tumors are considered a heterogeneous population of cells, and CSCs represent a small population of cancer cells.⁵³ In HCC treated with sorafenib, most cancer cells were apoptotic, but liver CSCs remained and showed resistance to sorafenib.⁵⁴ Stemness-related signaling pathways, including Hedgehog, Hippo, Notch, and Wnt pathways, play vital roles in the acquisition of CSC phenotypes. Previous studies have shown that blocking the Wnt, Hedgehog, and Notch pathways can sensitize sorafenib in HCC.55, 56, 57 In this study, we found that PCK1 attenuates the stemness of CSCs via activating Hippo pathway and increases the sensitivity of HCC cells to sorafenib. Importantly, the combination of sorafenib and verteporfin, a YAP inhibitor, significantly suppressed hepatoma cell proliferation and promoted apoptosis.

In summary, our study showed that PCK1 suppresses the stemness of hepatoma cells and enhances the sensitivity of hepatoma cells to sorafenib by promoting YAP phosphorylation. Inhibition of YAP can attenuate the stemness of hepatoma cells and inhibit cell proliferation. Therefore, the combination of verteporfin with sorafenib may be a potential anti-tumor strategy for HCC.

Author contributions

RL performed most experiments, analyzed the data, and wrote the original draft. YL, WLZ, and GJZ performed most experiments and analyzed the data. ZRZ assisted with animal experiments. LYH provided technical assistance. NT and KW contributed to supervision, writing – review & editing, and funding acquisition.

Conflict of interests

The authors declare that there is no conflict of interests.

Funding

This work was supported by the National Natural Science Foundation of China (No. 82073251, 82303854), the Natural Science Foundation Project of Chongqing, China (No. cstc2019jcyj-msxmX0587, cstc2019jscx-dxwtBX0019), the Science and Technology Research Program of Chongqing Municipal Education Commission of China (No. HZ2021006, KJZDM202000401), the Future Medical Youth Innovation Team of Chongqing Medical University (No. W0036, W0101), the Natural Science Foundation of Sichuan Province (No. 22NSFSC1284), and the Kuanren Talents Program of the Second Affiliated Hospital of Chongqing Medical University.

Appendix A. Supplementary data

The following is the Supplementary data to this article.