3- Mercaptopyruvate sulfurtransferase represses tumour progression and predicts prognosis in hepatocellular carcinoma

Background and Aims: The prognosis of hepatocellular carcinoma (HCC) remains dismal, and its molecular pathogenesis has not been completely defined. The enzyme 3- mercaptopyruvate sulfurtransferase (MPST) regulates endogenous hydrogen sulfide (H 2 S) biosynthesis. However, the role of MPST in HCC has never been inten-sively investigated. Methods: MPST protein expression was analysed in HCC tumour tissues and matched adjacent tissues. The effect of MPST on HCC progression was studied in vitro and in vivo. Results: The mRNA and protein expression of MPST was significantly downregulated in HCC samples compared with The enzyme 3- mercaptopyruvate sulfurtransferase regulates endogenous hydrogen sulfide biosynthesis. Our findings suggest that MPST may suppress cell proliferation associated with hydrogen sulfide production and inhibition of retinoblastoma protein signalling pathway in the development of hepatocellular carcinoma.


L ay Summar y
The enzyme 3-mercaptopyruvate sulfurtransferase regulates endogenous hydrogen sulfide biosynthesis. Our findings suggest that MPST may suppress cell proliferation associated with hydrogen sulfide production and inhibition of retinoblastoma protein signalling pathway in the development of hepatocellular carcinoma.

| INTRODUC TI ON
Hepatocellular carcinoma (HCC) is the third leading cause of cancerrelated death worldwide, and the second leading cause in China. 1 Despite major technical improvements in curative treatments such as medicine, surgical resection and transplantation, the clinical course of HCC is variable and the prognosis of HCC patients remains poor. The molecular pathogenesis of HCC is extremely complex and heterogeneous. 2 A better understanding of the molecular mechanism of this disease may help identify novel therapeutic targets to improve the outcome of patients with HCC.
Evolving information suggests that nonalcoholic fatty liver disease (NAFLD) may be an important cause of HCC in addition to viral hepatitis and alcohol-induced liver disease. 3 Recently, we demonstrated a role for 3-mercaptopyruvate sulfurtransferase (MPST), a key enzyme that regulates endogenous hydrogen sulfide (H 2 S) biosynthesis, in regulating lipid metabolism involved in the pathogenesis and development of NAFLD. 4 However, the information currently available on the functional role of MPST in HCC remains very limited.
Recent studies have indicated that H 2 S, acknowledged to be an important gasotransmitter, is involved in cancer biological processes. 5,6 H 2 S is produced in mammalian cells by three major enzymes including cystathionine β-synthase (CBS), cystathionine γ-lyase (CSE) and MPST. 7 Numerous studies have shown altered expression of these enzymes in the progression of various types of cancer cells. 8 Notably, accumulating evidence indicates that H 2 S-generating enzymes play important roles in HCC. CBS mRNA content was markedly decreased in HCC tissues. 9 Another study demonstrated that reduced CBS expression was associated with poor prognosis in HCC patients. 10 CSE protein is strongly expressed in tumour cell lines, and CSE/H 2 S promotes cell proliferation via cell cycle progression regulation in HCC. 11 Several studies have investigated the role of MPST in the context of cancer biology. One of the early studies focusing on the expression of MPST in cancer was conducted in 2006 in which MPST expression was markedly reduced in colon cancer and the decrease in expression was correlated with the depth of infiltration, suggesting that MPST is a tumour marker for colon cancer. 12 The expression and activity of MPST were also investigated in the human neoplastic cell lines, including astrocytoma U373, neuroblastoma SH-SY5Y and melanoma cell lines. 13 However, to date, studies that have shown the consequences of MPST modulation in cancer are limited, and the exact role of MPST in the development of HCC remains largely unknown.
In this study, we investigated the MPST expression pattern and determined its contribution to HCC progression. We also dissected the molecular mechanisms by which MPST promoted tumorigenesis.

| Immunohistochemistry staining
Immunohistochemistry (IHC) staining was conducted as described previously. 4 Briefly, tissue sections were dewaxed and rehydrated before performing antigen retrieval. The slides were incubated with anti-MPST (GeneTex), anti-p-Rb (CST) or anti-Ki67 (CST) overnight at 4°C, and incubated with an horseradish peroxidase-conjugated secondary antibody for 1 h at 37°C in an incubator. Immunoreactivity was detected using a DAB kit (ZSGB-BIO) and visualized as brown staining. Slides were counterstained with haematoxylin. The strength of positivity was semiquantified by taking into account the staining intensity and the percentage of positive cells measured by Image-Pro Plus 6.0 software. Paraffin sections were scored semiquantitatively as follows: Grade 0: 0% immunoreactive cells; Grade 1: ≦5% immunoreactive cells; Grade 2: >5%-50% immunoreactive cells; Grade 3: ≧50 immunoreactive cells. For statistical purposes, cases with Grade 0 and 1 were considered to have low expression, and those with Grade 2 and 3 were considered to have high expression.

| Cell lines and cell culture
Human HCC cell lines (HepG2, MHCC-LM3, Huh7 and Hep3B) were purchased from the Cell Bank of Type Culture Collection of Chinese Academy of Sciences, Shanghai Institute of Cell Biology, Chinese Academy of Sciences and were cultivated as described by the suppliers.

K E Y W O R D S
cell proliferation, hydrogen sulfide, liver cancer, retinoblastoma protein

| H 2 S donor treatment, H 2 S and reactive oxygen species measurement
The cells were exposed to different concentrations of NaHS (an H 2 S donor; Sigma-Aldrich) for 24 h. The phosphate-buffered saline (PBS) group served as a control. The measurement of H 2 S production was performed as previously described. 4 The assay of reactive oxygen species (ROS) was described in Appendix S1.

| Plasmid transfection and cell treatment
An overexpression plasmid containing full-length MPST DNA was transfected into HCC cells using Lipofectamine 3000 (Invitrogen) according to the manufacturer's instructions. Two days after transfection, 0.9 mg/ ml G418 solution (Sigma-Aldrich) was added to cells for selection of stable clones. The sequences of MPST siRNA used in this study and details regarding siRNA transfection were described in Appendix S1.

| Cell colony formation assay
Cells were seeded at 40 cells/cm 2 in 6-well plates and incubated for 2 weeks. Cells were fixed with methanol and stained with 0.1% crystal violet. Visible clones containing over 50 cells were counted.

| Cell apoptosis assessment
Cell apoptosis was assessed with a PI and Annexin V-APC apoptosis detection kit (KeyGen BioTECH). Briefly, the cells were trypsinized, resuspended in AnnexinV-binding buffer and incubated with AnnexinV-APC/PI in the dark for 20 min. The samples were detected using a FACS Calibur flow cytometer (BD Biosciences) and analysed by FloMax software. The measurement of mitochondrial membrane potential (MMP), associated with early apoptotic process, was described in detail in Appendix S1.

| RNA isolation and quantitative real-time PCR (qRT-PCR)
Total RNA isolation and qRT-PCR analysis were carried out as previously described. 4 Levels of relative expression were calculated and quantified with the 2 −ΔΔCt method after normalization with reference to the expression of GAPDH.

| Western blot analysis
Western blot analysis was carried out as previously described 4 SPSS 22.0 software was used for statistical analysis. The experimental data are expressed as the mean ± SD and were assessed by a two-tailed Student's t test or Mann-Whitney U tests. The OS rate was calculated with the Kaplan-Meier method, and the significant difference between survival curves was determined with the logrank test. Statistical significance was accepted if p < .05.

| Downregulation and clinical significance of 3-mercaptopyruvate sulfurtransferase expression in hepatocellular carcinoma
Based on the data from gene expression omnibus (GEO) database (GSE62232), compared to normal liver tissues, MPST was identified to be downregulated in HCC ( Figure 1A) caused by various aetiologies ( Figure S1A). We then performed q-PCR to verify the results and found that the mRNA expression of MPST was significantly downregulated in 30 primary HCC tissues compared with their adjacent nontumour tissues ( Figure 1B). Besides, the relative MPST mRNA level was negatively correlated with the tumour size ( Figure 1C). Next, we found protein level of MPST expression is markedly decreased in HCC caused by common aetiologies, among which hepatitis B virus (HBV)-related HCC is the most notable ( Figure S1B).  Figure 1D).

This finding was further validated by western blot analysis of MPST
To further explore the prognostic value of MPST in HCC, IHC studies were performed in TMA containing paired tumour and adjacent tissue specimens from 90 HCC patients. The protein expression level of MPST in HCC tissues was weak in 80 cases compared with their adjacent tissue specimens, including one with negative staining ( Figure 1E). And lower MPST expression was significantly associated with larger tumour size (Table 1). In addition, Kaplan-Meier analysis revealed that HCC patients with low expression of MPST had worse OS ( Figure 1F).

| Overexpression of 3-mercaptopyruvate sulfurtransferase represses hepatocellular carcinoma cell proliferation
The association of lower MPST expression with larger tumour size and increased OS suggests its potential inhibitory role in HCC progression. To test this hypothesis, we assessed the effect of MPST on the proliferation and growth of HCC cells. As shown in Figure 2A

3-Mercaptopyruvate sulfurtransferase is an important enzyme that
regulates H 2 S synthesis, which is involved in the development and progression of many types of cancer. To examine the effect of H 2 S on HCC, we treated LM3 cells with NaHS, the H 2 S donor, for 24 h.
H 2 S significantly promoted cell proliferation in LM3 cells in a dosedependent manner ( Figure 4A). Therefore, we hypothesized that MPST regulated HCC development partly through the regulation of H 2 S production. As illustrated in Figure 4B

| Overexpression of 3-mercaptopyruvate sulfurtransferase induces G1-phase cell cycle arrest and regulates the phosphorylation of Rb in hepatocellular carcinoma
To further determine the mechanism associated with growth inhibi-

| 3-Mercaptopyruvate sulfurtransferase promotes apoptosis
In addition to cell cycle regulation, it is well established that Rb affects tumour progression by regulating apoptosis, which plays a critical role in tumorigenesis. 16 Flow-cytometric analysis revealed significantly higher percentage of apoptosis in the MPST-overexpressing HCC cells ( Figure 6A). Mitochondrial dysfunction was demonstrated to be an important early step in the regulation of the apoptotic process, in which a reduction in MMP is an early event. 17 We found that overexpression of MPST led to a significant decrease in MMP in HCC cells ( Figure 6B). Since apoptosis is often mediated by the activation of caspases that lead to PARP binding to fragmented DNA, western blot analysis was then used to detect caspase activation. The results showed that cleavages of caspase-3, and PARP were dramatically increased in the MPSToverexpressing cells compared with the control cells ( Figure 6C).

| The combination of 3-mercaptopyruvate sulfurtransferase and p-Rb levels has better prognostic value for hepatocellular carcinoma
The correlation between p-Rb expression and survival data of HCC was further evaluated. The results of Kaplan-Meier analysis showed that OS was significantly lower in HCC patients with high expression of p-Rb ( Figure 7A). We further analysed the association between the expression levels of MPST and p-Rb in 85 pairs of HCC samples by IHC. TMA analysis revealed a significantly negative correlation between MPST and p-Rb levels ( Figure 7B,C).
Patients whose tumours had low levels of MPST and high p-Rb had worse OS ( Figure 7D). In summary, evaluation of MPST and p-Rb expression is a powerful predictor of poor prognosis for HCC patients.

| DISCUSS ION
The MPST is localized in both the cytosol and mitochondria. 19 High levels of MPST have been discovered in numerous cancer cell lines or tissues, including the astrocytoma U373, the neuroblastoma SH-SY5Y, the melanoma cell lines A375 and WM35, human lung adenocarcinoma cells and gliomas tissue. 8,20 In contrast, MPST was only slightly expressed in papillary thyroid cancer tissues, similar to that in adjacent normal tissues, 21 while in renal cell carcinoma resections, the MPST expression was highly variable. 22 Ramasamy et al. identified MPST as a potential tumour marker for colorectal cancer since its expression was markedly reduced in advanced colon cancer and correlated with disease progression. 12 Interestingly, MPST expression was found to be upregulated in HepG2 cells that exhibit stem-like properties when they recover from a cytotoxic stimulus. 23 Here, we observed that the expression of MPST was decreased in most of our HCC tissues; moreover, low MPST expression was associated with worse OS and larger tumour size in HCC patients. This discrepancy might be attributed to the fact that the MPST system in cancer cells appears to be highly complex and may well be highly cell-type and context-dependent. 24 However, to date, the modulatory effect of MPST on cancer progression remains largely unknown. A recent study showed that silencing of MPST reduced the proliferation rate of the human lung adenocarcinoma cell and attenuated the mitochondrial DNA repair rate. 20,24 In the murine hepatoma cell line Hepa1c1c7, siRNAmediated inhibition of MPST reduced bioenergetic parameters; however, the mechanism of how MPST regulates HCC development remains unclear. 25 In this study, we focused on the biological function Recent studies suggest that the MPST/H 2 S system plays a functional role in cancer progression through several mechanisms. 24  Our data also indicated that the molecular mechanism by which MPST inhibited HCC cell proliferation and tumorigenicity involves triggering G1-phase arrest and inducing apoptosis via suppression of the AKT/FOXO3a signalling pathway, hypophosphorylation of Rb and accumulation of p27. PI3K/AKT is an important signalling pathway involved in cell survival, inhibition of apoptosis, cell cycle progression and proliferation in human cancers. 27 Activated AKT is known to phosphorylate specific targets such as FOXO3a and leads to its cytoplasmic translocation, while hypophosphorylated FOXO3a is released from 14-3-3 protein and translocates into the nucleus, where it transactivates target genes that control cell death and thereby induce apoptosis. 28,29 Besides, the AKT/FOXO pathway has been shown to promote cell growth by inactivating the negative cell-cycle regulator p27. 30 Our study also suggested an important role of MPST in the regulation of apoptosis involved in HCC development. Mitochondria manage apoptotic signals that include changes in electron transport, loss of MMP and release of caspase activators. 31 A breakdown in the MMP is an invariant feature of early apoptosis. In the current study, we found that overexpression of MPST induced a significant decrease in MMP in HCC cells. In view of this, the apoptosis promoted by MPST is partly mediated by the mitochondrial pathway.
The Rb pathway acts as a master checkpoint in cell cycle progression, and regulates cell apoptosis. It is mediated by the interaction of Rb with other cellular proteins, especially E2F transcription factors. 32 Phosphorylation activated by CDK4-cyclin D1 complexes enables E2F release from Rb, resulting in the transcription of a number of genes that are necessary for DNA synthesis. 33 However, hypophosphorylated Rb actively inhibits cell cycle progression from the G1 to S phase and induces premature senescence. 34 Further studies will be necessary to elucidate the regulatory mechanism of MPST on Rb signalling pathways. Notably, the predictive value of MPST expression levels combined with the p-Rb signal was more sensitive than that of MPST alone for OS rate. This result identified the combined evaluation of MPST and p-Rb levels as a new prognostic marker in patients with HCC. This finding is important because these factors provide not only a new criterion for prognosis but also a potential therapeutic target.
In summary, our findings showed that MPST plays an important role in suppressing proliferation and tumorigenesis in HCC and is associated with H 2 S production and the AKT/FOXO3a/Rb signalling pathway. Therefore, we propose that MPST may function as a potential tumour suppressor and serve as a candidate predictor of outcome in HCC patients. Further studies are warranted to determine the molecular mechanism driving the loss of MPST during HCC progression and how H 2 S interacts with the various constituents of the cellular microenvironment so as to regulate cell signalling functions in HCC development. The restoration of MPST could be an effective therapeutic strategy for HCC treatment.