Intrahepatic cholangiocarcinoma is a malignant tumor originating from intrahepatic bile ducts, which exerts potential harms to liver and other organs and endangers sufferers' life. The proportion of male patients with cholangiocarcinoma was higher than that of female patients. At present, the etiology of the disease is uncertain, and it is generally believed that cholangiocarcinoma may be related to bile duct stones, clonorchiasis sinensis infection, and primary sclerosing cholangitis, etc. The common clinical manifestations in patients with cholangiocarcinoma are persistent jaundice, abnormal color of stool and urine, biliary tract infection and biliary tract hemorrhage. When definitely diagnosed, it tends to be at the advanced stage, so the treatment of the disease was difficult. Currently, main clinical treatment is the comprehensive application of surgery, radiotherapy and chemotherapy, but after therapy, the recurrence and metastasis of tumors often occur. It is important that the recurrent and metastatic cholangiocarcinoma often have a higher degree of malignancy, which makes the survival of patients shorter and the overall prognosis worse.
MicroRNAs (miRNAs) are a kind of small molecules discovered in recent decades, which do not directly participate in protein translation but affect cellular processes and participate in the occurrence and development of diseases. Abnormal expression of miRNAs is involved in the occurrence and development of cholangiocarcinoma. It was reported that high expression level of miR‐378 is detected in cholangiocarcinoma tissues, and miR‐378 has an ability to influence the growth, migration, and invasion of cholangiocarcinoma cells. Besides, inhibition of miR‐191 expression has been shown to induce apoptosis of cholangiocarcinoma cells by acting on secreted frizzled‐related protein‐1. Hence, these suggested that miRNAs act key roles in cholangiocarcinoma development. miR‐373 suppresses TXNIP via targeting the 3'UTR of TXNIP to induce epithelial‐to‐mesenchymal transition (EMT) and metastasis of cancer cell, leading to promotion of breast cancer. miR‐373 was found to directly target mRNA of RelA and PIK3CA and shows tumor‐suppressive effects in lung cancer cells. Importantly, Wang et al. found that long non‐coding RNA (lncRNAs) H19 and highly upregulated in liver cancer (HULC) enhance migration and invasion of cholangiocarcinoma via a competing endogenous (ceRNA)/miR‐373 pathway. Such a study indicated a regulative effect of miR‐373 on the biological functions of cholangiocarcinoma cells. Nevertheless, functions of miR‐373 on the occurrence and development of cholangiocarcinoma and its specific mechanisms have not been fully elucidated.
Autophagy, is a vital phenomenon in cellular processes, in which its own cytoplasmic proteins or organelles are engulfed through autophagic vesicles. Especially, autophagy functions on degrading and recovering excess components, removing harmful subcellular substances, and meeting the metabolic needs of cells themselves and the renewal of some organelles. Growing researches indicated excessive enhanced and inhibited autophagy have adverse effects on the survival of cholangiocarcinoma cells. ULK1 participates in autophagosome formation, an initial event in autophagy, which functions in a complex with protein partners. ULK1 has been widely reported as the autophagic initiator that may decide subsequent tumor cell fate. For examples, knockdown of ULK1 suppresses cell survival and proliferation of AGS gastric cancer. A potent and selective ULK1 inhibitor, named as ULK‐101, inhibits autophagy and sensitizes cancer cells to nutrient stress. However, whether miR‐373 could bind to ULK1 to modulate the autophagy of cholangiocarcinoma cells has not been demonstrated.
Taken together, we speculated that miR‐373 is a key molecule to target ULK1, leading to inhibition of autophagy and subsequent promotion of apoptosis in cholangiocarcinoma cells. Our findings provided a new knowledge of miR‐373 for the treatment of cholangiocarcinoma.
Human cholangiocarcinoma tissues and paracancerous tissues were collected from 32 patients with cholangiocarcinoma through surgical operation in the Hunan Provincial People's Hospital, which were preserved in −80°C for real‐time quantitative PCR (qPCR) detection. The tissues use was authorized by the cholangiocarcinoma patients in present research. In addition, this study has been approved by the Hospital's Ethics Committee. Inform consent was obtained from all participants. Clinical characteristics of cholangiocarcinoma patients were listed in Table .
Clinical pathological characteristics of patients with cholangiocarcinoma| Characteristic | Cases |
| Age (years) | |
| <50 | 13 |
| ≥50 | 19 |
| Gender | |
| Male | 15 |
| Female | 17 |
| Tumor size (cm) | |
| <5 | 20 |
| ≥5 | 12 |
| Tumor stage | |
| I/II stage | 18 |
| III/IV stage | 14 |
| Lymph node metastasis | |
| Negative | 21 |
| Positive | 11 |
The cholangiocarcinoma cell lines RBE, QBC939, HCCC9810, HUCCT‐1, and HIBEpiC were cultured in RPMI‐1640 (GIBCO, China) containing 10% of fetal bovine serum, as well as 1% of penicillin‐streptomycin solution (Sigma, China) in conventional atmosphere at 37°C with 5% CO2. All the cell lines were commercially obtained from American Type Culture Collection (ATCC) (Manassas, USA).
Cell transfection was performed according to the kit's instructions. In short, negative control (NC) mimic and NC inhibitor (duplex with random sequence), miR‐373 mimic, and miR‐373 inhibitor were all prepared by GenePharma (Shanghai, China). HCCC9810 and RBE cells were transfected with mixture which contained Lipofectamine 2000 (Invitrogen) and indicated mimics and inhibitors for 24 hours.
CCK‐8 assay was taken to detect cell proliferation. Briefly, after indicated treatment, the culture medium was discarded and the culture medium containing 10 μL of CCK‐8 solution (Dojindo, Japan) was added onto the 96‐well plate. After 4 hours incubation, the absorbance at 490 nm was assessed with a spectrophotometer.
HCCC9810 and RBE cells were transfected with indicated mimics and inhibitors, and then cultured in 6‐well plates for 24 hours. Cells were harvested using trypsin (Beyotime, China), washed twice with PBS, and then subjected to an apoptosis assay with an Annexin V‐FITC/PI apoptosis detection kit (Beyotime, China) referring to the product's instructions. The cell apoptosis rate was analyzed via BD LSRII Flow Cytometry System with FACSD (BD Bioscience).
Total RNA was extracted using RNA extraction kit (TaKaRa, China). Then the Prime Script RT Master Mix (TaKaRa, China) was used to reversely transcribe RNA into cDNA. Subsequently, qPCR was performed according to the direction of the SYBR Green master mixes (TaKaRa, China). The relative expression levels were calculated using the 2−ΔΔCt method. For mRNA, GAPDH was used as the internal control, and for miRNA, U6 was used as the internal control. Primer sequences were listed in Table .
Primer sequences for qPCR| Gene | Forward sequence | Reverse sequence |
| hsa‐miR‐373‐3p | 5'‐CGCGAAGTGCTTCGATTTTG‐3′ | 5'‐GTGCAGGGTCCGAGGT‐3′ |
| LC3 | 5'‐GAAGTTCAGCCACCTGCCAC‐3′ | 5'‐TCTGAGGTGGAGGGTCAGTC‐3′ |
| Beclin‐1 | 5'‐GCTGAAGACAGAGCGATGGT‐3' | 5'‐CCCCGATGCTCTTCACCTC‐3' |
| P62 | 5'‐GTACCAGGACAGCGAGAGGAA‐3' | 5'‐CCCATGTTGCACGCCAAAC‐3' |
| ULK1 | 5'‐GGAAGATGTCTCTGGGTGGA‐3' | 5'‐ACGACGTGCAAGTCAGACAG‐3' |
| U6 | 5'‐CTCGCTTCGGCAGCACA‐3' | 5'‐AACGCTTCACGAATTTGCGT‐3' |
| GAPDH | 5'‐CCAGGTGGTCTCCTCTGA‐3' | 5'‐GCTGTAGCCAAATCGTTGT‐3' |
After indicated transfection, HCCC9810 and RBE cells were washed with phosphate buffered saline and lysed using RIPA buffer. After protein concentration detection, the supernatants of cell lysates were subjected to electrophoresis in 10% SDS‐PAGE gel, which further transferred onto polyvinylidene fluoride membranes (Millipore). Furtherly, the membranes were blocked and then incubated with the primary antibodies Bcl‐2, Bax, Caspase‐3, Caspase‐9, ULK1, LC3‐I, LC3‐II, Beclin‐1, P62, and β‐actin (Cell Signaling Technology) overnight. Then the membranes were subjected to incubation with the secondary antibody (Cell Signaling Technology) were used as secondary antibodies for 2 hours. The bands of proteins were analyzed by an ECL detection kit (Beyotime, China).
The immunofluorescence was conducted to evaluate the expression of LC3 in HCCC9810 cells. Briefly, sterile slides were put into plates, prior to seeding HCCC9810 into each well. Furtherly, after transfection, HCCC9810 cells were fixed with 4% paraformaldehyde for 30 minutes. Then Triton‐XTM 100 was added for increasing membrane permeability. After washing and blocking, HCCC9810 cells were incubated with LC3 antibody (Cell Signaling Technology) overnight at 4°C. After washing again, HCCC9810 cells were incubated with secondary antibody for 1 hour. Then cells were counterstained with DAPI and sealed by cover glass, and finally observed with a confocal fluorescence microscope (LSM880, Zeiss, Germany).
Tissues were fixed in 4% paraformaldehyde for 24 hours and dehydrated with a gradient of ethanol (100%, 95%, 80%, and 70%). Tissues were embedded in paraffin and sectioned in 3 μm thickness. For IHC staining, the thickness was soaked in 3% H2O2 for 10 minutes and blocked with nonimmune goat serum at room temperature for 10 minutes, followed by incubating with anti‐ULK1 (Abcam, China) at 4°C for about 12 hours. Then, the thickness was washed with PBS buffer for three times, and incubated with secondary antibodies at 25°C for 30 minutes. Subsequently, the thicknesses were stained with DAB reagent (Sigma) for 5 minutes, and then stained with haematoxylin for 2 minutes. Observing and photographing the histomorphological changes under the microscope (LEICADMLB2, Germany).
The 3′‐UTR sequence of ULK1 was amplified, which further loaded to the luciferase reporter vector. Subsequently, HCCC9810 cells were seeded and cultured in cell plates, which were transfected with wild‐type ULK1 3′‐UTR vectors as well as mutant ULK1 3′‐UTR vectors and miR‐373 mimic or NC mimic using Lipofectamine 2000 (Invitrogen). Then, the luciferase activity was detected through the dual‐luciferase reporter assay system (Promega).
Data were showed as mean ± SD. All experiments were conducted at least three times. Data were analyzed by one‐way ANOVA or Student's t‐test. P value less than .05 was regarded as significant differences.
To elucidate the correlation between miR‐373 expression and cholangiocarcinoma, qPCR was performed to detect relative expression level of miR‐373. The results showed a lower level of miR‐373 in cholangiocarcinoma tissues compared with paracancerous tissues (P < .05; Figure A). Furthermore, miR‐373 level was significantly lower in patients with advanced stages of cholangiocarcinoma (III/IV phase) than that in patients with early stages of cholangiocarcinoma (I/II phase) (P < .01; Figure B). On the other hand, miR‐373 expression had no effects on lymph node metastasis (Figure C). And the results showed a lower level of miR‐373 in cholangiocarcinoma cell lines RBE, QBC939, HCCC9810, HUCCT‐1 compared with human intrahepatic biliary epithelial cells HIBEpiC (P < .05; Figure D). Our results suggested that miR‐373 may play an important role in regulating the development of cholangiocarcinoma.
Enlarge this image.miR‐373 is downregulated in tissues and cell lines of cholangiocarcinoma. A, qPCR was used to detect relative expression levels of miR‐373 in cancerous tissues and paracancerous tissues from 32 patients with cholangiocarcinoma. B, qPCR was applied to analyze the expression levels of miR‐373 in the tissues of patients with different clinical stages of cholangiocarcinoma. C, qPCR was used to detect the relationship between the expression level of miR‐373 and lymph node metastasis. D, qPCR was applied to analyze relative expression levels of miR‐373 in cholangiocarcinoma cell lines RBE, QBC939, HCCC9810, HUCCT‐1, and human intrahepatic biliary epithelial cells HIBEpiC. Data were shown as mean ± SD. N = 3. *P < .05; **P < .01
To further investigate the regulative effects of miR‐373 on biofunctions of cholangiocarcinoma cells, knockdown and overexpression of miR‐373 were conducted via transfection, and the proliferation and apoptosis were determined through flow cytometry. we found an obviously decreased proliferation rate in miR‐373 mimic group, while a significantly increased proliferation rate in miR‐373 inhibitor group, compared with control group (Figure A). As shown in Figure B, miR‐373 overexpression evidently elevated apoptosis rate of HCCC9810 and RBE, but knockdown of miR‐373 extremely inhibited HCCC9810 and RBE apoptosis. Meanwhile, western blot analysis showed that miR‐373 overexpression significantly reduced relative expression level of Bcl‐2, and increased those of Bax, Caspase‐3 and Caspase‐9, but miR‐373 downregulation exerted absolutely opposite effects on expression levels (Figure C). Taken together, these results revealed that upregulated miR‐373 inhibits proliferation and promotes apoptosis of cholangiocarcinoma cells HCCC9810 and RBE.
Enlarge this image.Upregulated miR‐373 promotes apoptosis in cholangiocarcinoma cells. A, HCCC9810 and RBE were transfected with NC mimic, miR‐373 mimic, NC inhibitor, miR‐373 inhibitor, as well as without vector. Cell proliferation was detected by CCK8 method. B, After transfection, apoptosis was analyzed via flow cytometry. C, After indicated treatment, western blot was taken to evaluate relative expression levels of Bcl‐2, Bax, Caspase‐3, and Caspase‐9. Data were shown as mean ± SD. N = 3. *P < .05; **P < .01
In addition, to determine the regulative effect of upregulation and downregulation of miR‐373 on autophagy of HCCC9810, we subsequently examined the autophagy via immunofluorescence, qPCR and western blot. Immunofluorescence assay suggested that positive expression levels of LC3 in HCCC9810 was decreased by miR‐373 overexpression but was increased by miR‐373 downregulation (Figure A). Additionally, at both mRNA and protein levels, after miR‐373 overexpression, expression levels of LC3 and Beclin‐1 were reduced and that of P62 was increased; however, after miR‐373 downregulation, expression levels of LC3, Beclin‐1 and P62 was reversely changed (Figure B,C). Accordingly, we concluded that miR‐373 overexpression markedly inhibits autophagy of cholangiocarcinoma cells.
Enlarge this image.miR‐373 overexpression inhibits autophagy of cholangiocarcinoma cells. A, HCCC9810 was transfected with NC mimic, miR‐373 mimic, NC inhibitor, miR‐373 inhibitor, or without vector. Immunofluorescence analysis was conducted to determine LC3 expression. Blue area indicated nucleus and red area indicated positively expressed protein. B, The relative expression levels of LC3, Beclin‐1, and P62 on mRNA level was examined by qPCR. C, Western blot was performed to assess the protein expression levels of Beclin‐1 and P‐62, as well as value of LC3‐II/LC3‐I. Data were shown as mean ± SD. N = 3. *P < .05; **P < .01
Next, the potential target genes of miR‐373 were predicted in TargetScan database (
Enlarge this image.miR‐373 regulates ULK1 expression by directly targeting bind. A, The potential binding site between miR‐373 and ULK1 was predicted. B, Dual luciferase reporter gene assay was performed to verify the direct binding relationship between miR‐373 and ULK1. The miR‐373 mimic and luciferase reporter plasmids with wild‐type or mutant ULK1 3'‐UTR, were cotransfected into cells. C, HCCC9810 was transfected with NC mimic, miR‐373 mimic, NC inhibitor, miR‐373 inhibitor, and then qPCR was taken to evaluate relative expression levels of ULK1 and miR‐373. D, The expression level of ULK1 was analyzed by western blot. E, The expression level of ULK1 in cholangiocarcinoma tissues was analyzed by immunohistochemistry. F, The expression level of ULK1 in cholangiocarcinoma tissues was analyzed by qPCR. G, qPCR was used to detect the expression level of ULK1 in cholangiocarcinoma tissues of patients with different clinical stages. H, qPCR was used to detect the relationship between ULK1 expression level and lymph node metastasis. I, The correlation analysis between miR‐373 and ULK1. N = 3. *P < .05; **P < .01
Our previous study found that miR‐373 inhibits autophagy and promotes apoptosis in cholangiocarcinoma cells, and miR‐373 targets ULK1. To further validate the role of ULK1 in miR‐373‐mediated autophagy, we conducted a study of this part. As shown in Figure A, miR‐373 upregulation suppressed LC3 expression, but combined overexpression of miR‐373 as well as ULK1 promoted LC3 expression, in comparison to miR‐373 mimic group. As shown in Figure B, we found that overexpression of miR‐373 reduced expression levels of LC3, ULK1, and Beclin‐1, and raised P62 expression level in HCCC9810 and RBE cells. Importantly, in comparison to miR‐373 mimic group, overexpression of miR‐373 and ULK1 elevated expression levels of LC3, ULK1 and Beclin‐1, and abated P62 expression level (Figure B). In addition, cell apoptosis induced by combined overexpression of miR‐373 and ULK1, was slighter than that induced by overexpression of miR‐373 only in HCCC9810 and RBE cells (Figure C). Finally, western blot detection uncovered that overexpression of miR‐373 as well as ULK1 dramatically augmented Bcl‐2 expression level and lowered expression levels of Bax, Caspase‐3, and Caspase‐9, compared with only miR‐373 overexpression (Figure D). As shown in Figure E, flow cytometry results showed that the apoptosis rate was significantly increased after overexpression of miR‐373; after overexpression of miR‐373 as well as ULK1, the apoptosis rate was significantly reduced. However, the apoptosis rate was significantly increased after overexpression of miR‐373 as well as ULK1, and autophagy degradation inhibitor 3‐MA treatment (Figure E). Western bolt results showed that overexpression of miR‐373 obviously reduced the expression levels of LC3 and Bcl‐2, while overexpression of P62 and cleaved caspase‐3 was upregulated (Figure F). Overexpression of miR‐373 as well as ULK1 significantly up‐regulated the expression levels of LC3 and Bcl‐2, and the expression levels of P62 and cleaved caspase‐3 were downregulated (Figure F). However, the expression levels of LC3 and Bcl‐2 were significantly downregulated, and P62 and cleaved caspase‐3 expressions were upregulated after overexpression of miR‐373 as well as ULK1, and 3‐MA treatment (Figure F). Overall, we further demonstrated that miR‐373 suppresses autophagy and subsequently enhances apoptosis of cholangiocarcinoma cells by directly targeting ULK1.
Enlarge this image.miR‐373 inhibits autophagy and promotes apoptosis of cholangiocarcinoma cells by targeting ULK1. A, HCCC9810 was transfected with miR‐373 mimic only, or miR‐373 mimic combined with ULK1 overexpression vector. Immunofluorescence analysis was used to detect LC3 expression. Blue area indicated nucleus and red area indicated positively expressed protein. B, Protein expression levels of Beclin‐1, P‐62, and ULK1, as well as value of LC3‐II/LC3‐I were examined by western blot. C, Apoptosis assay of HCCC9810 and RBE was carried out by flow cytometry. D, Expression levels of Bcl‐2, Bax, Caspase‐3, and Caspase‐9 were analyzed by western blot. E, Apoptosis assay of HCCC9810 and RBE cells after transfected indicated plasmids was detected by flow cytometry. F, Expression levels of LC3, P62, Bcl‐2, and cleaved‐Caspase‐3 in HCCC9810 and RBE cells after transfected indicated plasmids were analyzed by western blot. N = 3. *P < .05; **P < .01
Upon participating in basically physiological and pathological processes, miRNAs could recognize related binding‐sites and adjust the downstream genes expression. Up to date, increasing evidences concluded involvement of miRNAs in modulation of cancer cells. Cao et al. confirmed that miR‐29b overexpression attenuates growth of cholangiocarcinoma cells line QBC939 and aggravates apoptosis of QBC939. The apoptosis of cholangiocarcinoma cells is enhanced via miR‐191 inhibition and further reactivation of secreted frizzled‐related protein‐1. Some miRNAs including miR‐21, miR‐26, miR‐122 as well as miR‐150, even are selected to be biomarkers for prognosis of cholangiocarcinoma. miR‐373 is a tumor‐related miRNA in which abnormal expression has been widely reported in various human cancers including malignant cholangiocytes. Moreover, previous reports demonstrated that miR‐373 plays an important role in tumourigenesis and development of cholangiocarcinoma through methylation pathway. Meanwhile, our results indicated a low level of miR‐373 in cholangiocarcinoma tissues and cell lines, which is consistent with the results of Chen et al. And we found that miR‐373 overexpression decreased proliferation and increased apoptosis in cholangiocarcinoma via elevating expressions of Bax, Caspase‐9 as well as Caspase‐3 and suppressing Bcl‐2 expression.
Autophagy, as an important process, maintained intracellular homeostasis and regulated cell differentiation. Two‐side effects of autophagy in occurrence and development of cancers regarding induction of drug resistance of cancer cells, as well as promotion of apoptotic death of some cancer cells, have been found. Besides, Hu et al. demonstrated that autophagy in cholangiocarcinoma cells induced by phenformin treatment could be a potential mechanism to antagonize cholangiocarcinoma. Inhibition of active autophagy induces apoptosis and increases chemosensitivity in cholangiocarcinoma. Also, Hou et al. found that autophagy inhibition facilitated apoptosis as well as chemosensitivity of cholangiocarcinoma for blocking development of cholangiocarcinoma. In our research, we came to a conclusion that miR‐373 overexpression restrained autophagy of cholangiocarcinoma cells HCCC9810 through the enhancement of P62 expression and the decrease of Beclin‐1 expression level and the value of LC3‐II/LC3‐I. And our conclusion is basically identical with the work from Liu and colleagues that miR‐34a inhibits autophagy in cancer cells.
Growing evidence supported the conclusion that ULK1 affects the biological characteristics of tumors by mediating autophagy. Knockdown of ULK1 inhibits proliferation of non‐small cell lung cancer cell and augments apoptosis via modulating autophagy. Furthermore, Dower et al. uncovered that ULK1 inhibition promotes apoptosis in neuroblastoma. Accordingly, both apoptosis and autophagy might be modulated by ULK1 expression in cancer cells. In our study, miR‐373 was proved to regulate the downstream gene ULK1 via direct binding. We found that the apoptosis rate was significantly increased after overexpression of miR‐373 as well as ULK1, and 3‐MA treatment, suggesting that miR‐373 overexpression suppressed autophagy and subsequently boosted cholangiocarcinoma cells apoptosis via ULK1 inhibition.
Although we initially reveal a novel downstream molecular mechanism of the apoptosis of cholangiocarcinoma cells caused by miR‐373 in vitro, there is still much to be done to deeply understand this complex mechanism. As the situation in the animal body is more complicated and has more uncertainty than cell models, whether this mechanism could be applied to an animal model is not certain. Therefore, in the future, this novel molecular mechanism needs further verification in mice model and prove its potential of therapeutic targets.
In summary, we demonstrated for the first time that miR‐373 is reduced in cholangiocarcinoma tissues as well as cell lines, and negatively regulates ULK1 expression further to restrain autophagy and enhance apoptosis in cholangiocarcinoma cells (Figure ). Therefore, our findings suggested that miR‐373 might be a great potential target for cholangiocarcinoma treatment.
Enlarge this image.miR‐373 targets ULK1 to inhibit autophagy and further promotes apoptosis of cholangiocarcinoma cells. miR‐373 is a key molecule to target ULK1, leading to inhibition of autophagy and subsequent promotion of apoptosis in cholangiocarcinoma cells
The authors declare no potential conflict of interest.
All data generated or analyzed during this study are included in this published article.
© 2020. This work is published under http://creativecommons.org/licenses/by-nc-nd/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.
Details
1 Department of Hepatobiliary Surgery, Hunan Provincial People's Hospital, The First Affiliated Hospital of Hunan Normal University, Changsha, Hunan Province, People's Republic of China; Research Laboratory of Biliary Diseases, Hunan Provincial People's Hospital, The First Affiliated Hospital of Hunan Normal University, Changsha, Hunan Province, People's Republic of China
2 Department of Hepatobiliary Surgery, Hunan Provincial People's Hospital, The First Affiliated Hospital of Hunan Normal University, Changsha, Hunan Province, People's Republic of China
3 Institute of Clinical Medicine, Hunan Provincial People's Hospital, The First Affiliated Hospital of Hunan Normal University, Changsha, Hunan Province, People's Republic of China
4 Department of Hepatobiliary Surgery, Hunan Provincial People's Hospital, The First Affiliated Hospital of Hunan Normal University, Changsha, Hunan Province, People's Republic of China; Laboratory of Hepatobiliary Molecular Oncology, Hunan Provincial People's Hospital, The First Affiliated Hospital of Hunan Normal University, Changsha, Hunan Province, People's Republic of China