MiR‐139‐5p, miR‐940 and miR‐193a‐5p inhibit the growth of hepatocellular carcinoma by targeting SPOCK1

Abstract The study was aimed to screen out miRNAs with differential expression in hepatocellular carcinoma (HCC), and to explore the influence of the expressions of these miRNAs and their target gene on HCC cell proliferation, invasion and apoptosis. MiRNAs with differential expression in HCC were screened out by microarray analysis. The common target gene of these miRNAs (miR‐139‐5p, miR‐940 and miR‐193a‐5p) was screened out by analysing the target genes profile (acquired from Targetscan) of the three miRNAs. Expression levels of miRNAs and SPOCK1 were determined by quantitative real time polymerase chain reaction (qRT‐PCR). The target relationships were verified by dual luciferase reporter gene assay and RNA pull‐down assay. Through 3‐(4,5‐dimethyl‐2‐thiazolyl)‐2,5‐diphenyl‐2‐H‐tetrazolium bromide,thiazolyl blue tetrazolium bromide (MTT) and transwell assays and flow cytometry, HCC cell viability, invasion and apoptosis were determined. In vivo experiment was conducted in nude mice to investigate the influence of three miRNAs on tumour growth. Down‐regulation of miR‐139‐5p, miR‐940 and miR‐193a‐5p was found in HCC. Overexpression of these miRNAs suppressed HCC cell viability and invasion, promoted apoptosis and inhibited tumour growth. SPOCK1, the common target gene of miR‐139‐5p, miR‐940 and miR‐193a‐5p, was overexpressed in HCC. SPOCK1 overexpression promoted proliferation and invasion, and restrained apoptosis of HCC cells. MiR‐139‐5p, miR‐940 and miR‐193a‐5p inhibited HCC development through targeting SPOCK1.

epithelial-mesenchymal transition (EMT) and metastasis, and its down-regulation is observed in HCC cells. 8 Li et al revealed the positive correlation between miR-139 down-regulation and poor HCC prognosis. 9 Meanwhile, miR-940 and miR-193a-5p have also been demonstrated to be remarkably down-regulated in HCC tissues 10,11 .
MiR-940 suppresses tumour cell invasion and migration through regulating chemokine CXCR2, 12 while miR-193a decreases proliferation and increased apoptosis of HCC cells as well as improved the effect of sorafenib therapy. 13 MiR-193b increased cisplatin sensitivity of HCC cells and induced apoptosis. 14 These previous studies suggest that miR-139, miR-940 and miR-193 play negative roles in HCC development.
Cysteine-rich acidic secreted protein/osteonectin, cwcv, and kazal-like domains proteoglycan 1 (SPOCK1), the gene that encodes the protein core of a seminal plasma proteoglycan, is an oncogene that involves in cell-cycle control, apoptosis, DNA repair and metastasis. 15 According to Li et al's study, SPOCK1 is able to block apoptosis and promote metastasis in HCC. 16 Its promotive effect was also found in various human malignancies, such as breast cancer, 17 colorectal cancer, 18 prostate cancer 18 and ovarian cancer. 19 Yan et al investigated the relationship between miR-129-5p and SPOCK1 and pointed out that SPOCK1 can be regulated by miR-129-5p in gastric cancer, and the suppression of SPOCK1 inhibits cancer deterioration. 20 Therefore, overexpression of SPOCK1 is adverse to cancer treatment. Since there are few researches at present to investigate the functions of SPOCK1 in HCC, further studies about SPOCK1 and its upstream regulators are essential.
In this study, the expression levels of miR-139-5p, miR-940 and miR-193a-5p in HCC were investigated and their biological functions were explored. The target relationships between these miRNAs and SPOCK1 were also investigated to uncover the mechanisms that underlie miRNAs' influence on HCC development. The results could provide novel insights into potential molecular targets for HCC treatment.  Table 1.

| Kaplan-Meier survival analysis
The mRNA/miRNA normalized gene expression data and clinical information of 377 liver hepatocellular carcinoma (LIHC) samples were obtained from the Cancer Genome Atlas (TCGA) database (https://por tal.gdc.cancer.gov/projects/TCGA-LIHC). These patients were divided into two groups according to the median value of expression level of three selected miRNAs and SPOCK1. All the data were applied to perform Kaplan-Meier survival analysis and log-rank statistics. Hazard ratio (HR) and 95% CI of the association between SPOCK1/miRNAs and overall survival were calculated using Cox proportional hazards analysis with R "Survival", "Survplot", and "Kmplot" packages.   The microplate reader was used to measure absorbance at 490 nm. The number of invaded cells was counted by microscope.

| Cell apoptosis
Cells harvested after 48 hours transfection were stained by PI Annexin V Apoptosis Detection kit (BD Bioscience) and then detected by FACS Calibur flow cytometry (BD Bioscience) to determine apoptosis. Data analysis was performed by FACS Diva software.

| Statistics
All statistical analyses were performed using GraphPad Prism 6.0.
Data were expressed as mean ± SD. Student's t test and one-way ANOVA were used to compare the significance between two groups and among multiple groups respectively. Two-sided P values were calculated, and P < 0.05 was chosen for statistical significance. All experiments were repeated for at least three times.

| MiR-139-5p, miR-940 and miR-193a-5p were down-regulated in HCC tissues and HepG2 cell
MiRNAs with differential expression (fold change > 2, P < 0.05) in HCC tissues were screened out ( Figure 1A). MiR-139-5p, miR-940 and miR-193a-5p were found to target multiple common genes in human cancers, which could affect cancer development, thus were chosen as the miRNAs of interest in this study. Compared with NL tissues, their expression levels were aberrantly lower in HCC tissues ( Figure 1B). The relative expressions in 46 pairs of HCC and matched adjacent tissues confirmed the down-regulation of miR-139-5p, miR-940 and miR-193a-5p in tumour tissues ( Figure 1C).
According to Table 1

139-5p, miR-940 and miR-193a-5p
The possible target genes of miR-139-5p, miR-940 and miR-193-5p were predicted by Targetscan database. Among them, 11 common target genes were screened out through the analysis of the potential target genes profile ( Figure 4A). Then, we used another two databases, starBase v2.0 and microRNA.org, to verify our prediction. By literature review, we have confirmed four predicted common target genes, SPOCK1, GDE1, TRAF3 and TAOK1, that were frequently found associated with tumourigenesis. We then conducted RNA pull-down assay to determine the binding relationship between the three miRNAs and the four genes. Although all of them were predicted to be able to bind to miR-139-5, miR-940 and miR-193a-5p ( Figure 4B), only SPOCK1 had strong binding capacity with all three selected miRNAs ( Figure 4C). In 46 pairs HCC tissues, SPOCK1 mRNA and protein expressions were also distinctively elevated (Figure 4D,E). In addition, the mRNA and protein expression levels of SPOCK1 were also higher in HepG2 and Hep3b cells ( Figure 4F,G).
After miR-139-5p/miR-940/miR-193a-5p was overexpressed, the level of SPOCK1 was restrained than NC group both in mRNA and protein expression. However, in cotransfection group, the expression of SPOCK1 mRNA and protein was lower than those in single miRNA mimics group, but the difference was not significant ( Figure 5D). In brief, SPOCK1 could closely bind to miR-139-5p, miR-940 and miR-193-5p as their common target gene.

| SPOCK1 promoted growth and invasion of HCC cells
Overexpression Meanwhile, cell apoptosis was also suppressed by SPOCK1 overexpression and remained stable in SPOCK1+miRNAs groups, si-SPOCK1 promoted the cell apoptosis ( Figures 6F & 7D). In short, SPOCK1 could promote growth and invasion of HCC cells, and its promotive effect could be attenuated by miR-139-5p, miR-940 and miR-193a-5p.

| MiR-139-5p, miR-940 AND miR-193a-5p inhibited tumour growth in vivo
Transfection of miR-139-5p mimics, miR-940 mimics and miR-193a-5p mimics in HepG2 cells reduced the volume and weight of liver tumours in vivo ( Figure 8A-C). The SPOCK1 expression in these tumour tissues showed that overexpression of miRNAs significantly decreased the expression of SPOCK1 ( Figure 8D). The protein expression of SPOCK1 in tumour tissues confirmed the down-regulation of SPOCK1 resulted from miR-139-5p, miR-940 and miR-193a-5p ( Figure 8E). Consistently, miR-139-5p/miR-940/miR-193a-5p mimics reduced the PCNA and Ki67 expression which were highly expressed in proliferating cells ( Figure 8F). The ectopic overexpression of miR-139-5p/miR-940/miR-193a-5p promoted the protein levels of E-cadherin, while reducing N-cadherin and Vimentin in the tumour tissues ( Figure 8G). Moreover, Western blot was F I G U R E 4 SPOCK1 was up-regulated in hepatocellular carcinoma (HCC) tissues and HepG2 cell. A, Eleven common target genes of miR-139-5p, miR-940 and miR-193a-5p were selected by Venn diagram. (B) SPOCK1, GDE1, TRAF3 and TAOK1 were common target genes of miR-139-5p, miR-940 and miR-193a-5p. C, According to RNA pull-down assay, the three selected miRNAs expressions of Bio-SPOCK1 probes were all increased. *P < 0.05 compared with Bio-NC-probe. D, The relative mRNA expression of SPOCK1 detected by qRT-PCR was higher in HCC tissues. **P < 0.01 compared with normal tissues. E, The relative protein expression of SPOCK1 detected by western blot was higher in HCC tissues. **P < 0.01 compared with normal tissues. F, The relative mRNA expression of SPOCK1 detected by qRT-PCR was higher in HepG2 and Hep3b cell lines. **P < 0.01 compared with normal HL-1107 cell line. G, The relative protein expression of SPOCK1 detected by western blot was higher in HepG2 and Hep3b cell lines. **P < 0.01 compared with normal HL-1107 cell line. H, The relative mRNA expression of SPOCK1 detected by qRT-PCR was reduced after overexpression of miR-139-5p/miR-940/miR-193a-5p.*P < 0.05, **P < 0.01 compared with NC group. I, The relative protein expression of SPOCK1 detected by western blot was reduced after overexpression of miR-139-5p/miR-940/ miR-193a-5p. *P < 0.05, **P < 0.01 compared with NC group. J, Kaplan-Meier curve analysis of overall patient survival between those with high SPOCK1 expression and low SPOCK1 expression in the TCGA-LIHC database.
used to examined the key cell apoptosis regulators and found that miR-139-5p/miR-940/miR-193a-5p mimic enhanced the protein expression of the active forms of caspase 3, caspase 7 and caspase 8 in the tumour tissues ( Figure 8H). Therefore, the results indicated that miR-139-5p, miR-940 and miR-193a-5p inhibited tumour growth in vivo through down-regulating the expression of SPOCK1.

| DISCUSSION
Down-regulation of miR-139-5p, miR-940 and miR-193a-5p was identified in HCC, and the inhibitory influence of these miRNAs imposed on HCC progression was revealed in this study. SPOCK1, the common target gene of miR-139-5p, miR940 and miR-193a-5p, enhanced HCC cell viability and invasion. Therefore, the regulatory F I G U R E 5 SPOCK1 was the common target gene of miR-139-5p, miR-940 and miR-193a-5p. A, Correlation analysis indicated that the expression of miR-139-5p was negatively correlated with SPOCK1 level. B, Correlation analysis indicated that the expression of miR-940 was negatively correlated with SPOCK1 level. C, Correlation analysis indicated that the expression of miR-193a-5p was negatively correlated with SPOCK1 level. D, According to dual luciferase reporter assay, the relative luciferase activities of HepG2 cells in SPOCK1-WT + miRNA mimics (miR-139-5p mimics, miR-940 mimics and miR-193a-5p mimics) groups were significantly decreased, while those in SPOCK1-MUT + miRNA mimics groups showed no significant variation. **P < 0.01 compared with NC + SPOCK1-WT group. MiR-193a-3p dictated the deterioration of HCC-clinical analysis 10 and reduced the 5-FU resistance of HCC cells by mediating SRSF2. 23 Similarly, miR-193b also facilitated sorafenib and cisplatin induced apoptosis, 14,24 indicating that miR-193 could improve chemosensitivity. In summary, the antitumour effect of miR-139-5p has been reported in many researches. Our study confirmed its inhibitory  16 Meanwhile, SPOCK1 also played promotive roles in the progression of other cancers, 25 so that it was considered as a novel metastasis biomarker and potential target. 26 Many studies showed that SPOCK1 could activate PI3K/AKT signalling pathway, leading to the promotion of gliomas, 27 colorectal cancer 18 and gall bladder cancer. 28 In gastric cancer, miR-129-5p could inhibit the expression of SPOCK1, leading to the suppression of cancer cell processes. 20 Nonetheless, the target relationship between SPOCK1 and three miRNAs has not been properly investigated before. We found that SPOCK1 could bind miR-139-5p, miR-940 and miR-193a-5p, and it was suppressed by these miRNAs directly and specifically in HCC cells. This finding might be one of the potential mechanisms in inhibiting HCC development. As a common target gene of these miRNAs, SPOCK1 played a vital role in stimulating HCC progression and had great potential to be a novel target for HCC treatment.
However, there still existed several limitations in our study. The sample size in our study is small and the relationships among these miRNAs need further exploration. In addition, we found that F I G U R E 7 SPOCK1 promoted growth and invasion of Hep3b cells. A, qRT-PCR was used to detect the expression of SPOCK1. B, Cell viability of Hep3b cells was detected by MTT assay. C, Transwell assay revealed the invasion ability of Hep3b cells. D, Flow cytometry was performed to access the apoptosis rate of Hep3b cells cotransfection of these miRNAs can further suppress the progression of HCC, while its effect on the inhibition of SPOCK1 has no significant difference compared with single transfection. We suppose that these miRNAs have other target mRNAs which are tumour promoter and overexpression of these miRNAs inhibits these mRNAs the expression of these mRNAs leading to a better antitumour effect.
However, its potential mechanism remains to be further investigated and confirmed. Since many previous studies report the involvement of SPOCK1 in PI3K/AKT signalling pathway, further researches are worth conducting to investigate the signalling pathways regulated by SPOCK1 in HCC and to identify their associations with miR-139-5p, miR-940 and miR-193a-5p.
In conclusion, miR-139-5p, miR-940 and miR-193a-5p were down-regulated in HCC and their overexpression suppressed cell proliferation and invasion, induced apoptosis and inhibited tumour growth in vivo. SPOCK1, the common target gene of these miRNAs, was overexpressed in HCC and promoted its development. Therefore, through down-regulating SPOCK1, miR-139-5p, miR-940 and miR-193a-5p successfully restrained HCC deterioration.

ETHICS APPROVAL AND CONSENT TO PARTICIPATE
This study was approved by the Human Research Ethics Committee of the First Affiliated Hospital of Guangzhou University of Chinese Medicine. Moreover, the experiments were undertaken with the understanding and written consent of each subject.

CONFLI CT OF INTEREST
The authors confirm that there are no conflicts of interest.

AUTHOR CONTRI BUTIONS
P.L. substantially contributed to the concept and design of the work; P.L. and Y.Z involved in the analysis and interpretation of the data.; F I G U R E 8 MiR-139-5p, miR-940 and miR-193a-5p inhibited tumour growth in vivo. (A,B) The volume and weight of liver tumours of nude mice were significantly decreased in miR-139-5p, miR-940 and miR-193a-5p groups. C, The tumours were taken out from killed nude mice after 4 wk. Tumours in NC group were obviously larger than those in miRNAs groups. (D,E) Relative SPOCK1 mRNA expression detected by qRT-PCR and protein expression detected by western blot were decreased in miR-139-5p, miR-940 and miR-193a-5p groups. *P < 0.05, compared with NC group. F, Relative proliferation markers including PCNA and Ki67 expression detected by western blot were reduced in miR-139-5p, miR-940 and miR-193a-5p groups. G, The relative EMT marker E-cadherin detected by western blot was increased and Ncadherin and Vimentin expressions were reduced in miR-139-5p, miR-940 and miR-193a-5p group. H, Relative cell apoptosis regulators including caspase 3, caspase 7 and caspase 8 detected by western blot were increased in miR-139-5p, miR-940 and miR-193a-5p group P.L., J.L. and Z.X. drafted the manuscript; L.L. critically revised the work for important intellectual content; all authors involved in the final approval of the work.