Hepatocellular carcinoma (HCC) ranks the sixth in incidence and the third in mortality among malignant tumors globally.[1–3] Surgery is still the first choice for the treatment of HCC. The postoperative recurrence rate in HCC patients is still high, and most recurrent HCC is not suitable for radical resection. Tumor metastasis is the most important factor in recurrence. Recent studies have shown that epithelial to mesenchymal transition (EMT) plays an important role in tumor invasion and metastasis. Cells which undergo EMT are more likely to be separated from the original tissue for distant metastasis. EMT is a key process in determining distant metastasis and intra-hepatic dissemination of HCC. So, identifying molecules that can inhibit EMT could help reduce the tendency for metastasis of HCC. Nowadays, although great efforts have been made to understand the biological mechanism of liver cancer, morbidity and mortality are still frustrating. Therefore, it is urgent to find better treatment strategies to supplement the present approach. The best way is to study the molecular mechanism of the occurrence and development of liver cancer.
Follistatin Like 5 (FSTL5) is an extracellular matrix secretory protein with a molecular structure similar to that of follistatin (FST). Follistatin Like (FSTL) family proteins is involved in cell migration, proliferation, differentiation, and organ development. FST can bind to the activin protein of transforming growth factor β (TGF-β) family, which can promote apoptosis and inhibit the proliferation of hepatocytes. In addition, FST can bind directly to TGF-β3 and inhibit EMT in mammary epithelial cells. Furthermore, FST could inhibit the expression of the Snail, Zeb, and twist1 to suppress EMT in bovine ovules. Therefore, the members of the FSTL family are considered to be an attractive therapeutic targets and prognostic indicators in cancers. As a derivative of FST, FSTL5 may play a similar role in HCC cells. FSTL5 may be one of the important molecules in the regulation of hepatocarcinogenesis and development, especially in the process of EMT. However, the expression and function of FSTL5 in HCC remain unclear. This study aimed to investigate the expression and function of FSTL5 in HCC and its role in the regulation of EMT.
The study was approved by the Ethics Committee of Bengbu Medical College (No. 2011035) and informed consent was obtained from all of the patients.
Patients and tissue specimens
A total of 321 HCC patients treated in the Department of Hepatological Surgery from January 2012 to December 2018 were included and tissue specimens were collected after radical resection. All patients underwent radical surgery, and no other comprehensive treatments such as chemotherapy and targeted therapy were performed before surgery. All specimens included cancer tissue and paracancerous tissue, and patient information including age, sex, lymph node metastasis, tumor size, tumor number, tumor stage, postoperative survival time, overall survival, hepatitis carrier, liver function as assessed by Child-Pugh classification, and pre-operative alpha-fetoprotein were collected. The follow-up period ranged from 6 to 90 months.
Paraffin sections were routinely dewaxed and hydrated, incubated with 3% H2O2 at 37°C for 10 min to inhibit endogenous peroxidase activity, and then incubated in citrate buffer solution at 95°C for 20 min for antigen retrieval. The sections were then incubated with rabbit anti-human monoclonal antibodies for FSTL5 (1:1000; Proteintech, Rosemont, IL, USA), E-cadherin (1:5000; Cell Signaling Technology, MA, USA) and vimentin (1:5000; Cell Signaling Technology), and streptavidin-peroxidase immunohistochemical staining kit and 3,3′-diaminobenzidine chromogenic reagent (Maixin Co., Fuzhou, China). The sections were counterstained with hematoxylin for microscopic examination. Positive staining for E-cadherin, vimentin, and FSTL5 was indicated by yellowish to tan granules in the cytoplasm or non-cytoplasm. A total of ten high-density visual fields were randomly selected, and 100 cells were counted in each field. Based on the proportion of stained cells, the number of positive cells ≤10% was scored 0; 10% to 50% was scored 1 point; >50% was scored 2 points. Based on staining intensity, the score was 0 for unstained, 1 for light yellow, 2 for yellow, and 3 for tan. The results of the two scores were multiplied: 0 to 3 was denoted negative or low expression, and ≥4 was denoted positive or high expression.
SK-Hep1 and MHCC-LM3 HCC cell lines were purchased from Shanghai Academy of Life Sciences of the Chinese Academy of Sciences, Shanghai, China. The cells were cultured in Dulbecco's modification of Eagle's medium containing 5% fetal bovine serum at 37°C in an incubator with 5% CO2 and saturated humidity.
A total of 50 × 104 SK-Hep1 and MHCC-LM3 cells were seeded into each well of six-well plates. When the cell density reached 60%, the cells were infected by lentiviral vector over-expressing FSTL5 (Genecopoeia, Guangzhou, China), and the cells were screened using puromycin (2 μg/mL). HCC cell line stably expressing FSTL5 was obtained 2 weeks later. Puromycin (2 μg/mL) was routinely added to stabilize the expression of FSTL5.
Western blot assay
Cells were lysed in radio-immunoprecipitation assay buffer containing 1 mmol/L phenylmethanesulfonyl fluoride on ice for 10 min and the supernatant was collected following centrifugation at 4°C and 1000×g for 5 min. The protein concentration was calibrated by bicinchoninic acid method and 20 μg protein per well were separated by electrophoresis in 12% sodium dodecyl sulfate polyacrylamide gel electrophoresis, and then transferred to polyvinylidene fluoride membrane. The membrane was incubated in phosphate buffer solution tween containing 5% skimmed milk powder for 1 h, and then incubated overnight with antibodies against E-cadherin, vimentin, FSTL5, and glyceraldehydes phosphatedehydrogenase (GAPDH) (1:5000; Cell Signaling Technology) at 4°C. Next the membranes were incubated with goat anti-rabbit antibody (1:5000 diluted) at room temperature for 2 h, then washed and exposed to a chemiluminescence instrument.
Quantitative real-time polymerase chain reaction
Trizol reagent (Invitrogen, USA) was used to extract total ribonucleic acid (RNA) from the cells. RNA was reverse transcribed into complementary deoxyribonucleic acid following the instructions of the PrimeScript reverse transcription reagent kit (Takara Bio Inc, Ohtu, Japan). Real-time polymerase chain reaction (PCR) was performed in accordance with the instructions of Synergy BrandsSynergy Brands Premix Ex Taq TM II kit (Takara Bio Inc). The sequences of primers (synthesized by Shanghai Bioengineering Co, Ltd, Shanghai, China) were as follows: E-cadherin: upstream, 5′-ATTCTGATTCTGCTGCTT G-3′, downstream, 5′-AGTAGTCATAGTCCTGGTCTT-3′; vimentin: upstream, 5′-GAAGAA CTTG CCGTT GAAG-3′, downstream, 5′-GAAGGTGACGAGCCATTTC-3′; GAPDH, upstream, 5′-GGAGCGAGATCCC TCCAAAAT-3′, downstream, 5′-GGCTGGTCATACTT CTCATGG-3′; FSTL5: upstream, 5′-TGAAGTGCACAGAGCTGCTT-3′, downstream, 5′-AGCATATTTTCT TGCTGTATTC-3′. The expression of target genes was calculated using the 2−ΔΔct method.
Cell counting kit-8 assay
A total of 5000 cells in the logarithmic growth phase were seeded into 96-well plate and cultured for 1, 2, 3, 4, and 5 days, respectively. Next, 10 μL of cell counting kit-8 reagent was added to each well and the absorbance at 450 nm was measured.
Matrigel (Thermo Fisher Scientific, Darmstadt, Germany) was coated onto the membrane of a transwell chamber. The cells were seeded into the upper chamber and medium containing 10% fetal bovine serum was added to the lower chamber. After 40 h, the cells on the upper membrane surface were lightly removed. Membrane was fixed with methanol and then stained with crystal violet and visualized under microscopy. A total of ten different visual fields were photographed using a microscope.
All the experiments were repeated three times. The measured data were expressed as mean ± standard deviation, and the count data were expressed using the percentile system. The Chi-squared test was used to analyze the relationship between FSTL5 and clinical parameters. The Spearman rank correlation test was used to detect the correlation between the expression of FSTL5 and E-cadherin or vimentin. The t test was used to analyze the difference between groups. P < 0.05 indicated significant difference. SPSS 19.0 statistical software (SPSS Inc, Chicago, IL, USA) was used to process data.
Relationship between FSTL5 and HCC patient survival
Immunohistochemistry (IHC) showed that 56 HCC cancer specimens were positive and 265 were negative for FSTL5 expression. In the corresponding paracancerous tissues, 289 were positive and 32 were negative for FSTL5 expression. The expression of FSTL5 was significantly lower in HCC cancer specimens than in paracancerous tissues (9.97% vs. 82.55%, χ2 = 340.15, P
< 0.001). In addition, survival analysis showed that patients with high FSTL5 expression had significantly longer survival than those with low FSTL5 expression (χ2 = 8.22, P = 0.004) [Figure 1].
Relationship between FSTL5 and HCC indexes
The early tumor stage (χ2 = 11.29, P = 0.001), small tumor diameter (χ2 = 11.29, P = 0.001), a reduced number of tumors (χ2 = 45.52, P
< 0.001), lower alpha-fetoprotein value (χ2 = 24.36, P < 0.001), more probability of hepatitis carrying (χ2 = 40.9, P < 0.001), and less lymph node metastasis (χ2 = 5.58, P = 0.02) were correlated with high expression of FSTL5 (P
< 0.05) [Table 1].
Relationship between FSLT5 and EMT in HCC tissues
To analyze the role of FSTL5 in EMT of HCC, we assessed the correlation between FSTL5 expression and EMT markers E-cadherin and vimentin in HCC tissues. IHC showed that high expression of FSTL5 in HCC was accompanied by high expression of E-cadherin and low expression of vimentin (r
= 0.38, P
< 0.001), while low expression of FSTL5 was accompanied by low expression of E-cadherin and high expression of vimentin (r
= −0.385, P
< 0.001) [Figure 1 and Table 2]. These data suggest that FSTL5 is implicated in the regulation of EMT in HCC.
FSLT5 over-expression inhibits EMT of HCC cells
To confirm that FSTL5 regulates HCC EMT, we detected changes in the expression of EMT markers at the protein and mRNA levels in HCC cells with the over-expression of FSTL5 by Western blotting and quantitative real-time PCR, respectively. We found that over-expression of FSTL5 up-regulated the expression of E-cadherin and down-regulated the expression of vimentin in SK-Hep1 and MHCC-LM3 cells, both at protein and mRNA (HCC-LM3, negative control [NC] vs. FSTL5-interfering group [Lv-FSTL5]: FSTL5 [t = 31.36, P < 0.001], E-cadherin [t = 50, P < 0.001], vimentin [t = 72.75, P < 0.001]; SK-Hep1, NC vs. Lv-FSTL5: FSTL5 [t = 60.85, P < 0.001], E-cadherin [t = 45.03, P < 0.001], vimentin [t = 67, P < 0.001]) levels [Figure 2]. These results indicate that FSTL5 can inhibit EMT in HCC.
FSTL5 inhibits the proliferation of HCC cells
To investigate the inhibitory effects of FSTL5 on HCC cells, we assessed changes in the proliferation ability of HCC cells with the over-expression of FSTL5. Cell counting kit-8 assay showed that FSTL5 over-expression inhibited SK-Hep1 (NC vs. Lv-FSTL5: 1 day [t = 0.65, P = 0.581], 2 days [t = 0.56, P = 0.632], 3 days [t = 7.324, P = 0.018], 4 days [t = 6.23, P = 0.021], 5 days [t = 10.21, P = 0.003]) and MHCC-LM3 (NC vs. Lv-FSTL5: 1 day [t = 0.46, P = 0.911], 2 days [t = 0.53, P = 0.712], 3 days [t = 4.32, P = 0.037], 4 days [t = 7.49, P = 0.012], 5 days [t = 9.3661, P = 0.009]) cell proliferation [Figure 3].
FSTL5 inhibits the invasion ability of HCC cells
Furthermore, we investigated the invasion of HCC cells with the over-expression of FSTL5. Transwell assay showed that HCC cells with the over-expression of FSTL5 had significantly lower cell invasion (SK-Hep1, NC vs. Lv-FSTL5: t = 21.57, P < 0.001; MHCC-LM3, NC vs. Lv-FSTL5: t = 18.04, P < 0.001) [Figure 4].
EMT is a process of dedifferentiation and transformation of epithelial cells into mesenchymal cells. During EMT, the expression of E-cadherin and β-catenin is decreased, while the expression of vimentin, fibronectin, N-cadherin, and matrix metalloproteinases is up-regulated. In addition, the expression of EMT related transcription factors such as snail, slug, twist, and Zeb increases. Cells which undergo EMT are more likely to be separated from the original tissue for distant metastasis, and develop chemotherapeutic drug resistance and even tumor stem cell-like characteristics.[11–13] EMT is a key process in determining distant metastasis and intra-hepatic dissemination of HCC. Therefore, identifying molecules that can inhibit EMT could help reduce the tendency for metastasis of HCC.
Several members of FSTL family such as FSTL1 and FSTL3 have been investigated for their role in HCC. FSTL3 demonstrated low expression in HCC and inhibited TGF-β signaling, while FSTL1 was highly expressed in Hep3B HCC cell line and related to EMT in HCC.[15,16] It was reported that higher expression of FSTL5 was associated with worse prognosis of Wnt/sonic hedgehog (SHH)-independent neuroblastomas. However, the role of FSTL5 in HCC remains largely unclear. In this study, we found that the expression of FSTL5 in HCC tissues was lower than that in paracancerous tissues, and the expression level of FSTL5 was negatively correlated with the prognosis of HCC patients. Patients with high expression of FSTL5 had a longer survival than patients with low expression. However, we only included 321 patients who were followed up, and detected the corresponding HCC tissue samples of these patients. Our number of cases is not very large, and we will continue to increase the number of detection cases in the future to study the value of FSTL5 in evaluating the prognosis of patients with HCC.
To analyze the relationship between FSTL5 and EMT, we detected the expression of FSTL5 and EMT markers in HCC tissues. IHC analysis showed that the expression of FSTL5 was positively correlated with the expression of E-cadherin and negatively correlated with the expression of vimentin. Moreover, using HCC cell lines we demonstrated that the over-expression of FSTL5 down-regulated the expression of vimentin and up-regulated the expression of E-cadherin, confirming that FSTL5 could inhibit EMT. Furthermore, we found that the over-expression of FSTL5 inhibited the proliferation and invasion of HCC cells, consistent with the inhibitory effects of FSTL5 on EMT. Different from other researchers, we found low expression of FSTL5 in HCC, while some researchers found high expression of FSTL1 and low expression of FSTL3 in HCC. Similarly, we all found that FSTL family molecules are closely related to EMT in HCC. EMT is an important process of invasion and metastasis of HCC. So we think that FSTL family plays an important role in the process of EMT of HCC.
Previous studies have shown that interference with FSTL5 expression in lung cancer can activate the expression of Yes-associated protein (YAP), and increase the sensitivity of Kirsten Rat Sarcoma (KRAS) mutated lung cancer cells to Exportin 1 (XPO1) inhibitors. YAP is closely related to invasion, metastasis, and EMT in many kinds of tumors. In colon cancer cells, YAP can promote the expression of stroma markers such as vimentin and inhibit the expression of epithelial markers such as E-cadherin and occludin. In addition, YAP directly binds to the promoters of slug and vimentin to regulate EMT.[19,20] In addition, interference with YAP can inhibit proliferation, invasion, and EMT of HCC cells. YAP can directly bind to the promoter of p53 and increase chemotherapeutic sensitivity of HCC cells. The expression of YAP was higher in HCC tissues and lower in paracancerous tissues. Our previous results showed that over-expression of FSTL5 can regulate the proliferation and apoptosis of HCC via Wnt/β-catenin pathway, which may be regulated by YAP. Combined with the results of this study, we speculate that FSTL5 can regulate EMT in HCC through Wnt/β-catenin/YAP pathway, but further investigations are needed to reveal the molecular mechanism. For example, the rescue experiment on the regulation of EMT process of HCC YAP has not been studied. This is what we will study later.
This study speculates that FSTL5 is closely related to prognosis in HCC, and plays an important role in regulating invasion, metastasis, and EMT in HCC. Targeting FSTL5 may represent a new direction for the treatment of liver cancer.
This study was supported by the grants from the Anhui Province Science Foundation for Youths (No. 1808085QH288), the University Top-notch Talent Cultivation Project (No. gxbjZD28), the Bengbu Medical College Science Foundation (No. BYKY1734ZD), and the Science Foundation of the First Affiliated Hospital of Bengbu Medical College (No. Byyfykj201727).
Conflicts of interest
1. Chen W, Zheng R, Baade PD, Zhang S, Zeng H, Bray F, et al. Cancer statistics in China, 2015. Ca Cancer J Clin
2016; 66:115–132. doi: 10.3322/caac.21338.
2. Heimbach JK, Kulik LM, Finn RS, Sirlin CB, Abecassis MM, Roberts LR, et al. AASLD guidelines for the treatment of hepatocellular carcinoma
2018; 67:358–380. doi: 10.1002/hep.29086.
3. Wang ZL, Zhang LZ, Wang YL, Zheng WP, Zheng H. Clinical features and clonal origin of diffuse hepatocellular carcinoma
. Chin Med J
2018; 131:1128–1131. doi: 10.4103/0366-6999.230725.
4. Zhang HM, Shi YX, Sun LY, Zhu ZJ. Hepatocellular carcinoma
recurrence in living and deceased donor liver transplantation: a systematic review and meta-analysis. Chin Med J
2019; 132:1599–1609. doi: 10.1097/CM9.0000000000000287.
5. Campbell K, Casanova J. A common framework for EMT and collective cell migration. Development
2016; 143:4291–4300. doi: 10.1242/dev.139071.
6. Kreidl E, Oztürk D, Metzner T, Berger W, Grusch M. Activins and follistatins: emerging roles in liver physiology and cancer. World J Hepatol
2009; 31:17–27. doi: 10.4254/wjh.v1.i1.17.
7. Nogai H, Rosowski M, Grün J, Rietz A, Debus N, Schmidt G, et al. Follistatin antagonizes transforming growth factor-beta3 -induced epithelial-mesenchymal transition in vitro: implications for murine palatal development supported by microarray analysis. Differentiation
2008; 76:404–416. doi: 10.1111/j.1432-0436.2007.00223.x.
8. Kusama K, Bai R, Ideta A, Aoyagi Y, Okuda K, Imakawa K. Regulation of epithelial to mesenchymal transition
in bovine conceptuses through the interaction between follistatin and activin A. Mol Cell Endocrinol
2016; 434:81–92. doi: 10.1016/j.mce.2016.06.017.
9. Zhang DY, Ma X, Sun WL, Cui P, Lu Z. Down-regulated FSTL5 promotes cell proliferation and survival by affecting Wnt/β-catenin signaling in hepatocellular carcinoma
. Int J Clin Exp Pathol
10. Vaquero J, Guedj N, Clapéron A, Nguyen Ho-Bouldoires TH, Paradis V, Fouassier L. Epithelial-mesenchymal transition in cholangiocarcinoma: from clinical evidence to regulatory networks. J Hepatol
2017; 66:424–441. doi: 10.1016/j.jhep.2016.09.010.
11. Jayachandran A, Dhungel B, Steel JC. Epithelial-to-mesenchymal plasticity of cancer stem cells: therapeutic targets in hepatocellular carcinoma
. J Hematol Oncol
2016; 9:74–86. doi: 10.1186/s13045-016-0307-9.
12. Giannelli G, Koudelkova P, Dituri F, Mikulits W. Role of epithelial to mesenchymal transition
in hepatocellular carcinoma
. J Hepatol
2016; 65:798–808. doi: 10.1016/j.jhep.2016.05.007.
13. Diepenbruck M, Christofori G. Epithelial-mesenchymal transition (EMT) and metastasis: yes, no, maybe? Curr Opin Cell Biol
2016; 43:7–13. doi: 10.1016/j.ceb.2016.06.002.
14. Fan M, Shen J, Liu H, Wen Z, Yang J, Yang P, et al. Downregulation of PRRX1 via the p53-dependent signaling pathway predicts poor prognosis
in hepatocellular carcinoma
. Oncol Rep
2017; 38:1083–1090. doi: 10.3892/or.2017.5785.
15. Deli A, Kreidl E, Santifaller S, Trotter B, Seir K, Berger W, et al. Activins and activin antagonists in hepatocellular carcinoma
. World J Gastroenterol
2008; 14:1699–1709. doi: 10.3748/wjg.14.1699.
16. Slany A, Haudek VJ, Zwickl H, Gundacker NC, Grusch M, Weiss TS, et al. Cell characterization by proteome profiling applied to primary hepatocytes and hepatocyte cell lines Hep-G2 and Hep-3B. J Proteome Res
2009; 9:6–21. doi: 10.1021/pr900057t.
17. Remke M, Hielscher T, Korshunov A, Northcott PA, Bender S, Kool M, et al. FSTL5 is a marker of poor prognosis
in non-WNT/non-SHH medulloblastoma. J Clin Oncol
2011; 29:3852–3861. doi: 10.1200/JCO.2011.36.2798.
18. Kim J, McMillan E, Kim HS, Venkateswaran N, Makkar G, Rodriguez-Canales J, et al. XPO1-dependent nuclear export is a druggable vulnerability in KRAS-mutant lung cancer. Nature
2016; 538:114–117. doi: 10.1038/nature19771.
19. Shao DD, Xue W, Krall EB, Bhutkar A, Piccioni F, Wang X, et al. KRAS and YAP converge to regulate EMT and tumor survival. Cell
2014; 158:171–184. doi: 10.1016/j.cell.2014.06.004.
20. Lamar JM, Stern P, Liu H, Schindler JW, Jiang ZG, Hynes RO. The Hippo pathway target, YAP, promotes metastasis through its TEAD-interaction domain. Proc Natl Acad Sci U S A
2012; 109:E2441–E2450. doi: 10.1073/pnas.1212021109.
21. Bai N, Zhang C, Liang N, Zhang Z, Chang A, Yin J, et al. Yes-associated protein (YAP) increases chemosensitivity of hepatocellular carcinoma
cells by modulation of p53. Cancer Biol Ther
2013; 14:511–520. doi: 10.4161/cbt.24345.
22. Zhang X, Fan Q, Li Y, Yang Z, Yang L, Zong Z, et al. Transforming growth factor-beta1 suppresses hepatocellular carcinoma
proliferation via activation of Hippo signaling. Oncotarget
2017; 8:29785–29794. doi: 10.18632/oncotarget.14523.