HCC specimens used in the present study were obtained from HCC patients who underwent liver resection and gave their informed consent. The following liver tumor-derived cell lines SSMC-7721, QGY-7703, BEL-7404, BEL-7405, YY-8103 and Huh-7, were employed in the present study. These cell lines were grown in Dulbecco’s modified Eagle’s medium (DMEM), supplemented with heat-inactivated 10% fetal bovine serum (Gibco, Carlsbad, CA, USA) at 37ºC in a 5% CO₂ humidified incubator.

Antibodies against FAM9C, GAPDH, Lamin A and β-actin were from Santa Cruz Biotechnology Inc. (Santa Cruz, CA, USA). Antibodies against phospho-Akt (pAkt and Ser473) and Akt were from Cell Signaling Technology (Beverly, MA, USA). Propidium iodide (PI), 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT), 5-Aza-2′-deoxycytidine (5-Aza-dC), 4′-6-diamidino-2-phenylindole (DAPI) and PI3K inhibitor LY294002 were purchased from Sigma Chemical Co. (St. Louis, MO, USA).

SSMC-7721 and BEL-7405 cells were seeded at a density of 5×10⁵/100-mm dishes, cultured for 48 h, and treated with 2 mM 5-Aza-dC (Sigma Chemical), a demethylating agent. Forty-eight hours after treatment, cells are washed with PBS and fresh medium was added. Cells are further incubated for another 48 h before isolation of total cellular RNA.

RNA from HCC specimens and cell lines were extracted using TRIzol^® Reagent (Invitrogen Life Technologies, Carlsbad, CA, USA) according to the manufacturer’s instructions. Total RNA was treated with DNase I to avoid DNA contamination, the quality and quantity of extracted RNA was confirmed by NanoDrop 2000 (Thermo Fisher Scientific, Waltham, MA, USA). RNA reverse transcription reaction was performed using M-MLV reverse transcriptase (Promega), according to the manufacturer’s instructions. The primers sequences used in this study were as follows: upstream: 5′-TAG GAGCTGGAGGCTGAGAG-3′ and downstream: 5′-GGTTG AAGAGACAATGAAGTGCC-3′ for FAM9A gene amplification (NM_174951.2); upstream: 5′-GCAGTGCAGTGGTG TGATCT-3′ and downstream: 5′-CTTTCCTGCATGCTTCT TCC-3′ for FAM9B gene amplification (NM_205849.1); upstream: 5′-CTGAGGGAGCAACATCAGGG-3′ and downstream: 5′-TTGGTATCAACCCCCGTGTG-3′ for FAM9C gene amplification (NM_174901.4); and β-actin, a typical housekeeping gene used as the internal control, upstream: 5′-AATCGTGCGTGACATTAAGGAG-3′ and downstream: 5′-ACTGTGTTGGCGTACAGGTCTT-3′. All reactions were performed in triplicate in an Applied Biosystems 7300 Real-time PCR system. The relative mRNA level of the target gene was calculated using the comparative threshold cycle method (2^−ΔΔCt) normalized by β-actin expression (13).

Formalin-fixed HCC samples were paraffin-embedded and cut into 4-μm sections. The sections were deparaffinized and dehydrated, and then treated with methanol containing 0.3% H₂O₂ to inhibit endogenous peroxidase. The slides were incubated with anti-FAM9C goat polyclonal antibody (Santa Cruz Biotechnology) at 37ºC for 2 h and then at 4ºC overnight, followed by incubation with a horseradish peroxidase-conjugated anti-goat antibody (Dako Japan Ltd., Kyoto, Japan) at 37ºC for 1 h. The signals were detected using Diaminobenzidine Substrate kit (Vector Laboratories, Burlingame, CA, USA). Counterstaining was performed with hematoxylin. In addition, the slides with HCC specimens and the corresponding adjacent non-HCC livers were simultaneously used in the immunohistochemistry staining, and were then assessed by visual inspection and the estimation of the percentage of immunopositive cells. Immunofluorescence assay was performed to detect endogenous FAM9C in BEL-7404 cells. These cells were plated on polylysine-treated slides and then incubated at 37ºC for 1 h. The fixed cells were blocked with phosphate-buffered saline buffer containing 5% bovine serum albumin and then stained with anti-FAM9C antibody (Santa Cruz Biotechnology) at 4ºC overnight, followed by incubation with fluorescein isothiocyanate-conjugated goat anti-IgG antibody (Gibco-BRL, Grand Island, NY, USA) at 4ºC for 2 h. After rinsing, the slides were analyzed using immunofluorescence microscopy.

Small interference RNAs (siRNAs) targeting FAM9C, Akt and negative control siRNA (si-NC) were chemically synthesized (Invitrogen-Life Technologies). Two siRNAs against FAM9C with high efficiency were used in the functional experiment assays, and their sequences were as follows: FAM9C-si1, sense, 5′-GGACCAGUUGGAGGUU CAAdTdT-3′, antisense, 5′-UUGAACCUCCAACUGGUCC dTdT-3′; FAM9C-si2, sense, 5′-CCAAGGAGCUAGAAGAU AUdTdT-3′, antisense, 5′-AUAUCUUCUAGCUCCUUGT dT-3′; si-NC, sense, 5′-UUCUCCGAACGUGUCACGUdT dT-3′, antisense, 5′-ACGUGACACGUUCGGAGAAdTdT-3′. As well, the sequences of a siRNA against Akt were as follows: sense, 5′-GCACCUUCAUUGGCUACAAdTdT-3′, antisense, 5′-UUGUAGCCAAUGAAGGUGCdTdT-3′.

The full-length FAM9C ORF (501 bp, GenBank accession number NM_174901.4) was amplified by RT-PCR from BEL-7404 cells cDNA. The primers for FAM9C ORF were as follows: forward, 5′-TCAGAATTCATGGCTGCCAAGGACCAGT TGGAGG-3′; reverse, 5′-TCAGGTACCTCAATCTCTTTCA GTTCCTTTCCCA-3′. The PCR product was then inserted into pcDNATM3.1B (Invitrogen) through EcoRI- and KpnI- adapter. In addition, synthesized DNA nucleotide fragments of short hairpin RNA (shRNA) for knocking down endogenous FAM9C, was inserted into pSUPER (Oligoengine; Seattle, WA, USA). The sequences of these synthesized oligonucleotides for FAM9C silencing were as follows: forward, 5′-GAT CCCCGGACCAGTTGGAGGTTCAATTCAAGAGATTGA ACCTCCAACTGGTCCTTTTTGGAAA-3′, reverse, 5′-AGCT TTTCCAAAAAGGACCAGTTGGAGGTTCAATCTCTTGA ATTGAACCTCCAACTGGTCCGGG-3′ (FAM9C-sh1); forward, 5′-GATCCCCCCAAGGAGCTAGAAGATATTTC AAGAGAATATCTTCTAGCTCCTTGGTTTTTGGAAA-3′, reverse, 5′-AGCTTTTCCAAAAACCAAGGAGCTAGAAG ATATTCTCTTGAAATATCTTCTAGCTCCTTGGGGG-3′ (FAM9C-sh2); pSUPER-shNC with irrelevant nucleotide sequence serves as a negative control, and sequences of shNC were as follows: forward, 5′-GATCCCCTTCTCCGAACGT GTCACGTTTCAAGAGAACGTGACACGTTCGGAGAATT TTTGGAAA-3′, reverse, 5′-AGCTTTTCCAAAAATTCTCCG AACGTGTCACGTTCTCTTGAAACGTGACACGTTCGGA GAAGGG-3′.

Cell transfection was performed by Lipofectamine 2000 (Invitrogen) according to the manufacturer’s instructions.

HCC cells with FAM9C overexpression or knockdown were seeded in triplicate into 96-well plates (3,000 cells/well) respectively and then incubated with 5 mg/ml (10 μl) MTT for 4 h at 37ºC, then the medium was replaced with 100 μl dimethylsulfoxide (DMSO) for 10 min every 24-h intervals for 5–6 days. Cell growth curves were generated by calculating the mean value of the optical density measurements at 492 nm using a microplate reader. All experiments were independently repeated at least 3 times.

FAM9C was overexpressed in HCC cell lines Huh-7 and YY-8103 through the transfection with recombinant pcDNA3.1-FAM9C plasmid, additionally endogenous FAM9C was knocked down in QGY-7703, BEL-7404 cells via pSUPER-shFAM9C plasmids. Transfected cells were cultured on 100-mm dishes for colony formation, followed by the addition of G418 (Life Technologies) to the medium at a final concentration of 0.6–1 mg/ml. After 3–4 weeks, the remaining colonies were washed twice with PBS and counted on crystal violet-stained dishes. All experiments for observing colony formation were independently repeated at least three times.

To establish a stably FAM9C-expressing HCC cell line, HuH-7 cells were transfected with the recombinant plasmid pcDNA3.1B-FMA9C, then cultured by G418 (Life Technologies). An immunoblot assay was employed to examine the expression of FAM9C in nine individual colonies. One colony with steady and strong FAM9C expression was selected as study object, and another colony without ectopic FAM9C expression was used as mock object. In addition, BEL-7404 cell line with stable FAM9C knockdown was also established by transfection with plasmid pSUPER-FAM9C-sh2, with the methods mentioned above. A colony of BEL-7404 cells transfected with plasmid pSUPER-shNC containing irrelevant sequence was used as a negative control.

Cultured HCC cells were harvested with trypsin-EDTA and then centrifuged at 500 × g for 5 min. Cytoplasmic and nuclear proteins were extracted using NE-PER^® Nuclear and Cytoplasmic Extraction reagents (Thermo Scientific Pierce), according to the manufacturer’s instructions. Protein extracts were separated by 12% SDS-PAGE and transferred onto Hybond-C nitrocellulose membranes (Amersham Life Science, Buckinghamshire, UK). After blocking with PBS containing 5% non-fat milk, the membrane was incubated for immunoblot analysis with primary antibodies, followed by incubation with an IRDye 800DX-conjugated, affinity-purified secondary antibody. Signals were detected using the Odyssey Infrared Imaging system (LI-COR Biosciences).

Flow cytometry was used to analyze cell cycle and cell apoptosis. For DNA content analysis, cells were fixed in 70% ethanol, rehydrated in PBS, and treated for 30 min with RNase A (10 mg/ml) and for 5 min with propidium iodide (10 μg/ml). For UV-induced apoptosis analysis, 3J/cm², 5 min UV-irradiated cells (1×10⁶ cells) were harvested and washed twice with cold PBS. Then, Annexin V-FITC staining was performed according to the protocol provided by the manufacturer (BD Biosciences Pharmingen, San Diego, CA, USA). Briefly, cells were resuspended in binding buffer at a concentration of 1×10⁶ cells/ml, and 5 μl of Annexin-FITC and 100 μl cell suspension (1×10⁵ cells) were gently mixed and incubated for 15 min at room temperature in the dark. Cells were analyzed for apoptosis by flow cytometry on a FACSCalibur (Becton-Dickinson, Mountain View, CA, USA).

A total of 2×10⁶ offspring Huh-7 cells stably expressing ectopic FAM9C were subcutaneously injected into the flank of nude mice, and the same amount of progeny Huh-7 cells without ectopic FAM9C expression as mock control were injected into the opposite flank of the same mice. Cells (2×10⁶) derived from BEL-7404 with stable knockdown of endogenous FAM9C were injected subcutaneously into the flank of nude mice, and the same amount of BEL-7404 cells from the colony transfected with pSUPER-shNC were injected into the opposite flank of the same mice. Tumor growth kinetics was estimated by measuring tumor size and volume every 3–4 days. Tumor volume was determined using the following formula: 1/2 × width² × length.

Statistical analysis was performed by the Student’s t-test using SSPS software (SPSS, Inc., Chicago, IL, USA). P-value <0.05 was considered to indicate a statistically significant result.

To evaluate the role of FAM9 members in HCC, we first investigated the transcriptional levels of three FAM9 family members, FAM9A, FAM9B and FAM9C, in 22 pairs of human HCC specimens by semi-quantitative and real-time RT-PCR. The data showed that only FAM9C mRNA was obviously elevated in more than a half of HCC specimens examined, as compared with that of corresponding non-cancerous livers, while FAM9A mRNA was not specifically amplified and FAM9B exhibited disordered expression in these specimens (Fig. 1A and B). To confirm this result, FAM9C was further evaluated in additional 46 paired human HCC specimens by semi-quantitative and real-time PCR. The result showed the relative mRNA level of FAM9C was significantly upregulated in HCC tumor specimens compared with corresponding adjacent non-tumor livers (P<0.001) (Fig. 1C), and the upregulation of FAM9C mRNA was obvious in 25 of the 46 HCC specimens as shown by semi-quantitative PCR (Fig. 1D). Furthermore, we also examined the expression and location of FAM9C protein in a pair of HCC specimens using immunohistochemical staining with a specific antibody against FAM9C. The resulting data showed that FAM9C protein was located in the nuclei, and presented more expression in tumor tissue than adjacent non-tumor sample (Fig. 1E). Collectively, these data demonstrated that FAM9C was frequently elevated in human HCCs and as a novel cancer testis gene.

The expression pattern of FAM9C was further evaluated in some HCC cell lines by RT-PCR. The result showed that FAM9C presented relatively high expression in QGY-7703 and BEL-7404 cells but low or no expression in other cell lines examined (Fig. 2A). To investigate whether DNA methylation regulates the FAM9C expression in HCC cells, we treated two HCC cell lines SSMC-7721 and BEL-7405 with 5-Aza-dC, a demethylating agent. RT-PCR analysis showed that FAM9C expression was upregulated in these treated cells as compared with untreated controls (Fig. 2B). In addition, we determined the expression and subcellular location of FAM9C protein in HCC cells by a western blot assay of cytoplasmic and nuclear extracts, as well as a subcellular immunofluorescence assay. In conformity with the mRNA expression levels of FAM9C, FAM9C protein exhibited also relatively high expression in nuclear extracts of QGY-7703 and BEL-7404 cells, but without expression in cytoplasmic extracts of these cell lines (Fig. 2C). FAM9C was located in the nucleus in BEL-7404 cells, as indicated by immunofluorescence assay (Fig. 2D).

To evaluate the role of FAM9C on HCC cells, recombinant pcDNA3.1-FAM9C was transiently transfected into Huh-7 and YY-8103 cells with little FAM9C expression. The result showed that overexpressed FAM9C, as demonstrated by western blot assay, significantly promoted cell growth of the two HCC cell lines compared with those transfected with empty vector (Fig. 3A and B). Furthermore, FAM9C also promoted the anchorage-dependent colony formation of Huh-7 and YY-8103 cells (Fig. 3C and D). These data suggest that FAM9C overexpression plays an important role in promoting HCC cell growth, colony formation in vitro.

To further investigate the effect of FAM9C on HCC cell proliferation and colony formation, we used chemically synthesized siRNAs and shRNAs derived from recombinant pSUPER to knock down endogenous FAM9C in BEL-7404 and QGY-7703 cells. Through evaluating the knockdown efficiency of four siRNAs against FAM9C (data not shown), the two efficient siRNAs (FAM9C-si1 and FAM9C-si2) were used to silence endogenous FAM9C in BEL-7404 and QGY-7703 cells (Fig. 4A). The growth curve results showed that FAM9C knockdown significantly inhibited the proliferation of BEL-7404 and QGY-7703 cells (Fig. 4B). Moreover, pSUPER vectors carrying shRNAs (FAM9C-sh1 and FAM9C-sh2) against FAM9C were employed to FAM9C knockdown, where endogenous FAM9C was silenced in the above two HCC cell lines respectively, as demonstrated by western blot assay (Fig. 4C). The colony formation assay demonstrated that FAM9C knockdown significantly suppressed the anchorage-dependent clonogenicity of BEL-7404 and QGY-7703 cells (Fig. 4D). These collective data implied that endogenous FAM9C could be essential for maintaining the proliferation and anchorage-dependent colony formation of HCC cells in vitro.

To determine whether the dysregulated FAM9C contributes to hepatocarcinogenesis in vivo, Huh-7 cells (2×10⁶) stably expressing FAM9C were subcutaneously injected into a flank of athymic nude mice, and the same amount of subcloned Huh-7 cells transfected with pcDNA3.1-FAM9C but without ectopic FAM9C expression as Mock control were subcutaneously injected into the other flank of the same nude mice. As expected, the stably FAM9C-expressing Huh-7 cells formed tumors faster than the mock cells without FAM9C expression under the observation for four weeks (P<0.05) (Fig. 5A). The removed xenograft tumors from nude mice demonstrated that FAM9C overexpression significantly promoted the tumor size and weight as compared with that of mock objects (P<0.05) (Fig. 5B and C). These data indicated that FAM9C overexpression significantly promoted tumorigenicity in vivo of Huh-7 cells. Moreover, a total of 2×10⁶ BEL-7404 cells, whose endogenous FAM9C was stably knocked down by plasmid pSUPER-FAM9C-sh2, were subcutaneously injected into a flank of athymic nude mice, and the same amount of cells transfected with control plasmid (pSUPER-shNC) as negative control was injected into the opposite flank of the same mice. Notably, FAM9C knockdown led to significant tumor growth retardation under observation for four weeks (P<0.01) (Fig. 5D), where BEL-7404 cells with FAM9C silence had low tumorigenicity in vivo in four of six mice tested, whereas cells used as controls formed tumors in all six mice (Fig. 5E). The removed xenograft tumors from nude mice showed that FAM9C silence significantly restrained the tumor size and weight as compared with that of controls (P<0.01) (Fig. 5F). The result indicated that silencing of FAM9C significantly inhibited tumorigenicity in vivo of BEL-7404 cells.

To explore the molecular mechanisms by which FAM9C contributes to HCC cell proliferation, clonogenicity in vitro and tumorigenicity in vivo, we first analyzed the cell cycle of HCC cells with FAM9C overexpression or knockdown. However, FAM9C overexpression and knockdown did not alter cell cycle distribution of Huh-7 and BEL-7404 cells respectively (Fig. 6A and B). As reported, FAM9C protein exhibited similarity to SYCP3, a component of the synaptonemal complex (12). More importantly, SYCP3 led to activation of Akt, resulting in the upregulation of anti-apoptotic proteins (14). To determine whether FAM9C as a homologous gene of SYCP3 activates Akt signaling pathway, we examined the activity of Akt pathway by western blot assay while FAM9C overexpression or knockdown in HCC cells. We found that the phosphorylation levels of Akt significantly increased in Huh-7 cells expressing FAM9C, whereas the phosphorylation levels of Akt obviously decreased in BEL-7404 cells with FAM9C knockdown (Fig. 6C). These data indicated that FAM9C regulated activation of Akt.

To further address the role of FAM9C-induced activation of Akt in HCC cells, we investigated cell apoptosis of UV-irradiated HCC cells with FAM9C overexpression or knockdown by flow cytometry method. The resulting data demonstrated that FAM9C overexpression significantly reduced the proportion of apoptotic Huh-7 cells (P<0.01), while FAM9C knockdown significantly increased the apoptotic percentage of BEL-7404 cells (P<0.01), indicating that FAM9C endowed HCC cells with resistance to apoptosis (Fig. 6D and E). In addition, to determine whether the activation of PI3K-Akt pathway is directly responsible for the anti-apoptotic role of FAM9C in HCC cells, we further performed cell apoptosis assays when Akt was silenced by a validated siRNA (data not shown) or blocked by LY294002, a known inhibitor of PI3K-Akt signaling pathway. Interestingly, the results demonstrated that both siRNA against Akt and LY294002 inhibitor significantly abolished or even prevented the anti-apoptotic effect of FAM9C in Huh-7 and BEL-7404 cells with UV exposure (P<0.01) (Fig. 6D and F). Thus, these collective data suggest that FAM9C plays the anti-apoptotic role through mediating PI3K-Akt signaling pathway in HCC cells.