A total of 20 human liver cancer tissues and 20 adjacent non-tumor liver tissue specimens and their sera were used (Table I), we also recruited 20 healthy control subjects (Table II) from physical examination and the donating blood were collected for experiments. Patients with liver cancer were treated, and frozen tissue samples were obtained from The Affiliated Hospital of Hebei University, same as the healthy control subjects. The tumor type was confirmed by a pathologist. The study was approved by the Ethics Committee of The Affiliated Hospital of Hebei University, and written informed consent was obtained from all participants. All human materials were used in accordance with the Declaration of Helsinki Principles and relevant policies and regulations of China and the policies of the Institutional Review Board of the Hospital of Hebei University.

Human cell lines HepG2 and SUN387 were purchased from the Type Culture Collection of the Chinese Academy of Sciences (Beijing, China). The cell lines were maintained in RPMI-1640 medium (cat no. 11875; Gibco; Thermo Fisher Scientific, Inc., Waltham, MA, USA) and supplemented with 10% fetal bovine serum (cat no. 10437028; Gibco; Thermo Fisher Scientific, Inc.). Cells were incubated at 37°C in a humidified incubator, supplemented with 5% CO₂.

Total RNA was extracted from liver cancer cells using guanidinium thiocyanate (cat no. 5596026; Invitrogen; Thermo Fisher Scientific, Inc.). cDNA was generated by RT using oligo (dT) primers as outlined in the manufacturer's protocol (cat no. R233-01; Vazyme, Piscataway, NJ, USA). For the thermocyling conditions, 50°C 15 min, 85°C 5 sec. Target genes and controls were treated under the same conditions and analyzed by qPCR using SYBR Premix Ex Taq™ (cat. no. RR420L; Takara Biotechnology Co., Ltd., Dalian, China), according to the manufacturer's protocol. The primers used were as follows: MASTL forward, 5′-ATCCTGTATGCCACATCAGAC-3′, and reverse, 5′-TTTCTTCACCTTCTGGCCAAG-3′; GAPDH forward, 5′-GCACCGTCAAGGCTGAGAAC-3′, and reverse, 5′-TGGTGAAGACGCCAGTGGA-3′. These results were normalized by the Pfaffi method (19) using CFX96 C1000 Touch Thermal Cycler (Bio-Rad Laboratories, Inc., Hercules, CA, USA).

RNAi experiments were performed using 30 nM small interfering RNA (siRNA) oligos (Shanghai GenePharma Co., Ltd, Shanghai, China) by targeting human MASTL. Liver cancer cells were seeded at 2×10⁵ cells per well in a 6-well plate. Control siRNAs consisting of a scrambled sequence were used as a negative control. Transfection was performed with Lipofectamine^® 2000 reagent (cat no. 11668019; Invitrogen; Thermo Fisher Scientific, Inc.), according to the manufacturer's protocol. The efficiency of RNAi was confirmed by western blot analysis 48 h post-transfection, and the phenotype was detected following confirmation of MASTL knockdown by siRNA. The MASTL siRNA sequence was 5′-GGACAAGTGTTATCGCTTA-3′ (8).

Protein extracts were prepared with radioimmunoprecipitation assay buffer according to the manufacturer's protocol (cat no. 89900; Pierce; Thermo Fisher Scientific, Inc.). The protein concentration was detected by BCA assay (cat. no. CW0014S; CWBIO; Beijing, China). In brief, 40 µg of whole cell lysate was separated by SDS-PAGE at 12% polyacrylamide and then transferred on to a PVDF membrane. Then, the membrane was blocked with 5% BSA (cat. no. A8010; Beijing Solarbio Science & Technology Co., Ltd., Beijing, China) in PBST for 1h at room temperature and incubated with the primary antibodies overnight at 4°C. Western blot analysis was performed using antibodies against MASTL (1:1,000) (cat no. ab86387; Abcam; Cambridge, MA, USA) and β-actin (1:5,000) (cat. no. 4967; Cell Signaling Technology, Inc., Danvers, MA, USA). The secondary antibodies conjugated with HRP (1:10,000) for 90 min at room temperature. The blots were developed using enhanced chemiluminescence reagent (cat. no. GERPN2109; GE Healthcare, Chicago, IL, USA) by GeneGnome XRQ (Syngene, Syngene Division of Synoptics Ltd; UK).

HepG2 and SUN3r7 cells were plated at 1×10⁴ cells per well in 96-well plates and 10 µl (cat no. C0009; Beyotime Institute of Biotechnology, Haimen, China) was added to the cells. MTT was removed 4 h post incubation, followed by pipetting of dimethyl sulfoxide to solubilize the formazan products. Absorbance was measured with a microplate reader (Gen5; BioTek Instruments, Inc., Winooski, VT, USA) at 570 nm.

HepG2 cells transfected with siRNA were harvested, washed with PBS, and fixed in 70% ethanol at 4°C overnight. The fixed cells were stained with propidium iodide (PI; cat no. C0080; Beijing Solarbio Science & Technology Co., Ltd.) at 37°C for 30 min prior to analysis via fluorescence-activated cell sorting (FACS) (Becton, Dickinson and Company, Franklin Lakes, NJ, USA). The cell cycle profiles were interpreted by BD CellQuest Pro software version 5.1 (BD Biosciences; Becton, Dickinson and Company).

Briefly, to create the MASTL kinase-dead mutation, the nucleotide residues 718–816 of MASTL were deleted. The DNA coding sequences of MASTL and the kinase-dead mutant of MASTL were cloned into the pcDNA3.1 eukaryotic expression vectors (Invitrogen; Thermo Fisher Scientific, Inc., USA) and were confirmed by sequencing.

Cells were cultured overnight on a glass coverslip and fixed using 4% paraformaldehyde for 30 min at room temperature. The cells were incubated with primary anti-MASTL antibody and secondary fluorescein isothiocyanate (FITC)-goat anti-rabbit IgG antibody (cat no. ab6717; Abcam; USA). DAPI (cat. no. C1002; Beyotime Institute of Biotechnology) was employed for nuclear staining. Paraffin sections were compared with non-tumor liver samples fixed using 4% paraformaldehyde overnight at room temperature (tissue sections 8 mm), 0.5% Triton X-100 in PBS was used for reducing non-specific interactions at room temperature for 1 h, and subsequently stained with anti-MASTL antibody (1:100) and FITC-goat anti-rabbit IgG antibodies (1:50). Images were acquired using a fluorescence microscope (magnification, ×100) (E600; Nikon Corporation, Tokyo, Japan) and the results were evaluated by fluorescence signal intensity using Zeiss 510 META software version 3.5.

IL-6 and TNF-α quantification in healthy and liver cancer patient sera by ELISA (cat. no. EK106/2&EK1822; MULTI SCIENCES, Beijing; China) according to the manufacturer's protocol. The concentrations of IL-6 and TNF-α were calculated based on standard curves provided with the kits, and the results of IL-6 and TNF-α were expressed in pg/ml.

Cells were processed according to the protocol described in the ChIP Assay kit (cat no. P2078; Beyotime Institute of Biotechnology, Haimen, China) with anti-H3K4Me3 antibody (1:100) (cat no. 07-473; EMD Millipore, Billerica, MA, USA). Magnetic protein G beads (cat no. 88847; Pierce; Thermo Fisher Scientific, Inc.) were used to pull-down the antibody-chromatin complexes. The ChIP primers used were as follows: MASTL pro (0–0.2 K) forward, 5′-TTAAGACTTTCTACAGCTT-3′, and reverse, 5′-TGAGCAGGAAGCGAGTCC-3′. The NF-κB inhibitor pyrrolidine dithiocarbamate (PDTC; cat. no. P8765) was purchased from Sigma-Aldrich (Merck KGaA, Darmstadt, Germany). The STAT3 inhibitor NSC74859 (cat. no. SD4794) was purchased from Beyotime (Beyotime Institute of Biotechnology).

One-way ANOVA followed by Newman-Keuls post-test was used for multi-group comparisons. The two-tailed Student's t-test of SPSS 18.0 (SPSS, Inc., Chicago, IL, USA) was used for two-group comparisons. All the experiments were repeated at least three times. Results are displayed as the mean ± standard error of the mean. P<0.05 was considered to indicate a statistically significant difference.

Cancer of the liver is one of the most common causes of cancer-associated mortality worldwide (11); in China, the majority of liver cancer cases occur due to sustained, chronic infection with HBV or HCV (12,13,20). Numerous studies have confirmed that inflammatory cytokines, including IL-6 and TNF-α, secreted by Kupffer cells and hepatocytes, contribute to the development of hepatocacinogenesis (21–23). Additionally, these inflammatory cytokines may activate signal transducer and activator of transcription 3 (STAT3) and NF-κB to induce hepatocarcinogenesis (24,25), though the detailed mechanisms are yet to be clarified.

To investigate whether IL-6 or TNF-α may affect the expression of MASTL in the human liver cancer cell lines HepG2 and SUN387, these cells were stimulated with 100 pg/ml TNF-α or IL-6 for 1–5 days. RT-qPCR and western blotting were performed to determine the mRNA and protein expression levels of MASTL. MASTL mRNA expression was induced in HepG2 and SUN387 cells by stimulation with IL-6 or TNF-α (Fig. 1A and B). Similarly, MASTL protein expression was increased by the introduction of IL-6 and TNF-α (Fig. 1C and D).

A number of studies have demonstrated that Gwl kinases serve a critical role in regulating mitosis in Drosophila and Xenopus. In addition, Gwl kinases inhibit the activation of PP2A by interacting with Ensa/Arpp19 (26,27). Previous studies of MASTL confirmed that it had a similar function in regulating mitotic entry and cytokinesis in human cell lines, including HeLa and U2OS (8).

To demonstrate the function of MASTL in HepG2 cells, MASTL was knocked down with siRNA (Fig. 2A). FACS analysis was performed following PI staining to examine whether MASTL silencing altered cell cycle progression. The results indicated that cells transfected with siRNA-MASTL were arrested at the G2/M phase compared with the control cells (Fig. 2B and C). Morphological differences were observed by immunofluorescence. Specifically, HepG2 cells treated with siRNA exhibited a greater proportion of multinuclear cells compared with control cells (Fig. 2D and E). This indicated that the cells failed to enter mitosis and complete cell division. These results were consistent with previous reports (8,28), which demenstrated that cells depleted of MASTL by siRNA remain in the G2 phase, and exhibit slow chromosomal condensation.

Based on the FACS data, it was hypothesized that knockdown of MASTL may inhibit liver cancer cell proliferation. To confirm this hypothesis, an MTT assay was performed to examine the proliferation of HepG2 and SUN387 cells transfected with MASTL siRNA or control siRNA, and western blotting was performed to test the efficiency of silencing in SUN387 cells (Fig. 2A, right). The results indicated that the proliferation of MASTL-silenced HepG2 and SUN387 cells was significantly inhibited compared with the control samples (Fig. 2F), suggesting that MASTL influences the proliferation of liver cancer cells.

The function of MASTL in cell cycle progression and cell proliferation was examined by overexpression in HepG2 and SUN387 cell lines. Exogenous expression of MASTL was confirmed using an anti-Flag antibody (Fig. 3A), and a subsequent MTT assay revealed that overexpression of MASTL in HepG2 and SUN387 cells resulted in more rapid proliferation, compared with the empty vector controls (Fig. 3B). These results are partially consistent with the MASTL silencing data in these cell lines (predominantly in HepG2), further confirming the role of MASTL in liver cancer proliferation.

Furthermore, MASTL wild type and kinase-dead vectors were transfected into HepG2 cells to assess alterations in morphology (Fig. 3C). The nuclei of HepG2 cells transfected with the wild type MASTL vector were much smaller, and aggregated together compared with those transfected with kinase-dead vector, indicating that MASTL promotes liver cancer cell proliferation. FACS analysis was performed with cells expressing mutant MASTL and the corresponding empty vector. The results indicated that the number of cells expressing mutant MASTL in the G1 phase was significantly decreased compared with the control (Fig. 3D and E). Taken together, this suggested the involvement of MASTL in liver cancer cell proliferation by regulating cell cycle progression. These results were consistent with previous reports (8,28) which illustrated that MASTL knockdown impaired cell proliferation.

To confirm whether the inflammatory cytokines IL-6 and TNF-α induced MASTL expression in liver cancer tissues, 20 patients with liver cancer and 20 healthy control subjects were recruited. The results indicated higher concentrations of IL-6 and TNF-α in the sera of diseased patients compared with those of the control subjects (Fig. 4A). IHF and IHC results revealed that 14 of the 20 diseased patients (70%) exhibited high expression of the MASTL protein, and all of these patients were chronically infected with HBV. Chronic HBV infection was determined by persistent HBV surface antigenemia lasting more than six months. By contrast, 4 of the 20 matched non-tumor liver tissues (20%) displayed increased MASTL protein expression (Fig. 4B). Lower expression of the MASTL protein was revealed in 16 of the 20 non-tumor liver tissues (Table III). These data indicated that the high expression of MASTL promoted liver cancer carcinogenesis.

The mechanisms involved in the IL-6 and TNF-α-induced MASTL expression in liver cancer cell lines were investigated. A ChIP assay was performed to analyze whether stimulation with cytokines effected the methylation of histone H3K4 at the MASTL promoter, which critically influences the regulation of transcription by increasing chromatin accessibility. The results revealed that by day 4, H3K4me3 at the MASTL promoter was significantly increased in HepG2 cells treated with IL-6 or TNF-α (Fig. 4C). This suggested that cytokine stimulation of liver cancer cells promoted chromatin accessibility at the MASTL promoter, inducing MASTL expression.

Numerous studies have demonstrated that IL-6 and TNF-α signaling promotes the proliferation of liver cancer cells by NF-κB activation (29–32); therefore, the role of NF-κB in IL-6 and TNF-α-induced expression of MASTL was investigated. HepG2 cells were pretreated with the NF-κB inhibitor PDTC, and the MASTL mRNA expression level was measured in response to IL-6 or TNF-α stimulation. The promotion of MASTL mRNA expression was inhibited by PDTC (Fig. 4D). In addition, a ChIP assay was performed, which revealed that on day 4, H3K4me3 at the MASTL promoter was significantly decreased in HepG2 cells treated with PDTC (Fig. 4E). Thus NF-κB activation was required to induce MASTL mRNA expression by IL-6 or TNF-α.

In summary, stimulation of liver cancer cells with IL-6 or TNF-α promoted trimethylation of histone H3 lysine 4 at the MASTL promoter to facilitate chromatin accessibility, and NF-κB was involved in cytokine-induced MASTL mRNA expression.