MiR‐552 promotes the proliferation, migration and EMT of hepatocellular carcinoma cells by inhibiting AJAP1 expression

Abstract Our goal was to explore the function of miR‐552 and its potential target AJAP1 in hepatocellular carcinoma (HCC) oncogenesis and progression. In this study, bioinformatics analysis was performed to detect abnormally expressed miRNAs. The relationship between miR‐552 and AJAP1 was validated using luciferase reporter assays. RT‐qPCR and Western blot assays were applied to explore the expression level of miR‐552, AJAP1 and epithelial‐mesenchymal transition (EMT) markers. HCC cell proliferation was examined using CCK8 assays, while migration and invasion were investigated using Transwell assays. Nude mouse tumourigenesis models were established to facilitate observation of HCC progression in vivo. Finally, prognostic analysis was performed to discover how the prognosis of HCC patients correlated with miR‐552 and AJAP1 expression. MiR‐552 overexpression in HCC cells promoted HCC cell migration, invasion and EMT by targeting/suppressing AJAP1. Poorer prognosis appeared in HCC patients with higher miR‐552 expression or lower AJAP1 levels. Our findings suggested that miR‐552 promotes HCC oncogenesis and progression by inhibiting AJAP1 expression.


| INTRODUCTION
Hepatocellular carcinoma (HCC) ranks as the fifth most prevalent malignancy worldwide. 1 As it is the most frequent primary liver cancer, HCC is the second most deadly cancer, with a 5-year survival rate of <20%. 2,3 Patients with chronic liver disease are at high risk for HCC oncogenesis. 4 However, because early-stage HCC is often not symptomatic, it is not easily identified; this feature, in combination with its common recurrence and metastasis after absolute resection, leads to poor recovery. 4,5 It timely to investigate the underlying pathogenic mechanisms of HCC to develop possible novel curative treatment methods. 1 MicroRNAs (miRNAs) are sequences of small noncoding RNAs of 18-22 nt in length. 6 They manipulate the expression levels of more than 60% of human genes at the posttranscriptional level by binding to the 3′ untranslated regions (3′UTR) of their target messenger RNAs (mRNAs). 7 This repression enables them to participate in cellular events, including proliferation, apoptosis and movement. 8 The first attempt to elucidate how miRNAs participate in human cancers was made by Chen et al 9  conclusion that miR-552 is possibly involved in the molecular mechanism of another cancer within the human digestive system. Cao et al 12 concluded that upregulation of miR-552 promoted colorectal cancer (CRC) cell growth by targeting DACH1. Wang et al. also found that ADAM28 is a target of miR-552 in CRC. 6 In addition, miR-552 was identified by Leivonen et al 7 as a negative regulator of HER2 in breast cancer. However, much remains unknown regarding the roles of miR-552 in the regulation of HCC oncogenesis and progression.
Adherens junctions-associated protein-1 (AJAP1) is located on 1p36, a chromosome in which nonrandom deletion is often detected in various human malignancies. 13 AJAP1 has been widely acknowledged as a biomarker for glioblastoma (GBM), for example, by Yang et al 14 Zeng et al 13 argued for its positive correlation with poorer GBM survival. In addition, AJAP1 could suppress cell adhesion and migration in oligodendrogliomas. 15 In HCC cell lines and tissues, AJAP1 loss was observed by Ezaka et al 16 , who highlighted not only its HCC-suppressive role but also its intermediate role in the epithelial-mesenchymal transition (EMT) process. As cell migration and invasion are the primary components of EMT, we therefore adopted EMT as a notable criterion. 14 There is less documentation on how AJAP1 is involved in HCC compared with other human cancers. Our study pioneered the exploration of the function of AJAP1 in HCC development.
In the present study, bioinformatics analysis was conducted to identify differentially expressed miRNAs in HCC. The miR-552 and AJAP1 expression levels in HCC cells and tissues were determined.
The expression levels of EMT markers were measured to confirm the influence of miR-552/AJAP1 on EMT. CCK8 and Transwell assays were used to study the regulatory effects of the AJAP1 and miR-552 interaction on HCC cell proliferation, migration and invasion. In addition to the prognostic analysis, an experiment using nude mice investigated this effect in vivo. Our research may open up a new path towards HCC treatment.

| Cell culture
Eighty-one pairs of human HCC tissues and the corresponding adjacent tissues were obtained from Qingdao No. 6 People's Hospital. The 81 patients had undergone neither chemotherapy nor radiotherapy before the absolute resection. The detailed clinicopathological characteristics of these patients are provided in   The reverse transcription reaction was carried out by incubating the extracted RNA in a water bath for 1 hour at 37°C to synthesize cDNA, which was then used as a template in PCR reactions for amplification. Agarose gel electrophoresis was used to analyse the PCR products. The primer sequences used in the RT-qPCR (Invitrogen) are listed in Table 3. GAPDH was used as an internal control for calculating the PRM1 mRNA content. Meanwhile, PRM1 standards with concentrations of 100, 10, 1, 0.1 and 0.01 ng, along with a negative contrast group were prepared. The expression levels of PRM1 mRNA were calculated after the plotting of a PCR reaction standard curve. The results were analysed using an Applied Biosystems 7300 Fast Real-Time PCR System.

| Western blotting
Normal hepatic or HCC tissues were ground in LN and mixed with  Proteintech Group, Chicago, IL, USA) was used as a loading control.
Band intensities were quantified using ImageJ software. The obtained images were converted into 8-bit format to perform uncalibrated optical densitometry analysis. After conversion, the films were compared with the densitometry quantification setting the same ball radius value (50.0 pixels) for background subtraction. Each band was individually selected and circumscribed with the rectangular ROI selection tool and "Gels" function followed by quantification of the peak area in the obtained histograms. The data were acquired as arbitrary area values.

| Cell transfection
The miRNA mimics were synthesized via chemical synthesis to enhance the function of the endogenous miRNAs. The miRNA inhibitors are chemically modified suppressors that target specific miRNAs in the cells. The negative controls (NC) were scrambled oligonucleotides. Cell transfection was performed as described previously. 17

| Luciferase reporter assay
The target gene AJAP1 was identified via TargetScan before investigating the correlation between AJAP1 and miR-552 using luciferase

| Survival and statistical analysis
The

| MiR-552 expression level was correlated with clinicopathological characteristics
When the mean value of miR-552 expression was regarded as a criterion, 35 of the 81 patients belonged to the low-expression group and 46 belonged to the high-expression group. A chi-squared test was used to analyse the correlation between miR-552 expression level and clinicopathological characteristics of HCC patients. As shown in Table 1, high miR-552 expression tended to be observed in the higher-grade category (P = 0.001) and advanced HCC category (P = 0.043). To summarize, miR-552 level was significantly correlated with tumour histological grade and TNM stage, but was not correlated with the patient's gender, age, AFT level, tumour size and tumour number (all P > 0.05).

| MiR-552 promoted HCC cell proliferation and mobility
We thus suggested that miR-552 could affect HCC progression. We then conducted in vitro proliferation and mobility experiments. In The cells in the inhibitor control group showed no significant difference compared with the NC group. **, P < 0.01 compared to the NC group. (C) CCK8 assay results indicated that miR-552 mimics led to a higher absorbance at OD 450 nm while the miR-552 inhibitor led to a reduced absorbance in both Hep3B and HepG2 cells. *, P < 0.05, **, P < 0.01 compared to the NC group. (D, E) Transwell assay results showed that miR-552 mimics promoted both migration and invasion in selected HCC cell lines, whereas the miR-552 inhibitor acted as a suppressor. **, P < 0.01 compared to the control group both the Hep3B and HepG2 cell lines, the increase in miR-552′s relative RNA level due to transfection with mimics and its decrease due to transfection with inhibitors were marked as significant in comparison with the NC group (Figure 2A,B, P < 0.05). The CCK8 assay results demonstrated that miR-552 upregulation promoted Hep3B and HepG2 cell proliferation, whereas inhibition of miR-552 expression suppressed proliferation. As illustrated in Figure 2C, the miR-552 inhibitor group resulted in a lower absorbance than the NC group, while the mimics group had a higher absorbance (P < 0.05).
In the Transwell assays, more cells from the miR-552 mimics group migrated and invaded. Meanwhile, downregulation of miR-552 limited the two processes ( Figure 2D,E, P < 0.05). Thus, miR-552 promotes the proliferation, migration and invasion processes of HCC cells. The RT-qPCR results from the 81 pairs of HCC and adjacent tissues showed that AJAP1 had significantly higher expression in tumour-adjacent tissues than in HCC tissues ( Figure 3E, P < 0.0001).

| MiR-552 targeted AJAP1 and inhibited its expression
In addition, there was a significant inverse correlation between the relative expression levels of miR-552 and AJAP1 in HCC tissues (Figure 3F, P < 0.0001).
In conclusion, miR-552 targets AJAP1 in HCC tissues and cell lines. The expression levels of the two factors are very likely negatively correlated.

| MiR-552 promoted HCC cell proliferation, migration and invasion by inhibiting AJAP1
The results from RT-qPCR and Western blot experiments showed that the relative mRNA levels and expression levels of AJAP1 were significantly higher in the AJAP1 groups, indicating successful transfection ( Figure 4A,B, P < 0.001). These cells were then used in the CCK8 assay to investigate their proliferative ability. The results showed that in comparison with the control group, overexpression of only AJAP1 decreased HCC cell proliferation, but that the proliferation rate was not affected by cotransfection with miR-552 mimics and AJAP1 plasmids ( Figure 4C, showed that in both cell lines, AJAP1 overexpression suppressed HCC cell proliferation, whereas miR-552 + AJAP1 cotransfection had no significant influence on cell proliferation. (D) In the Transwell assay to assess cell migration, fewer cells migrated when AJAP1 was overexpressed compared with the control group; however, the mimics+AJAP1 group had almost no difference compared to the control group. (E) In the Transwell assay to assess cell invasion, there was less invasion by AJAP1-overexpressing cells compared to the control group; however, the mimics+AJAP1 group had almost no difference. *, P < 0.05 compared to the control group miR-552 in HCC cells occurred through inhibition of AJAP1 expression.

| MiR-552 promoted EMT and oncogenesis of HCC by inhibiting AJAP1
Hep3B cell lines were used for the upcoming assays. Western blot outcome indicated that in contrast to the NC group, miR-552 over-  Figure 5C,D, P < 0.05). Based on the above results, we concluded that miR-552 prompted HCC growth and EMT by regulating AJAP1.

| Consideration of MiR-552 and AJAP1 together provided a better prognosis index for HCC patients
The survival outcome of HCC patients (data downloaded from TCGA database) was analysed using KM survival analyses. The results indicated that after resection, HCC patients with a high AJAP1 expression level and a low miR-552 expression level had higher survival rates than did those who had a low AJAP1 expression level and high miR-552 expression level ( Figure 6A,B, P < 0.05). These results indicated that miR-552 and AJAP1 levels were closely correlated with HCC prognosis.

| DISCUSSION
In our results, it is obvious that miR-552 was overexpressed in both the HCC cell lines and in the tissues and was significantly correlated  Figure 7. We demonstrated that miR-552 is a potential HCC biomarker and therapeutic target. that AJAP1 is associated with the cytoskeleton in endothelial cells. 25 Tanaka et al 26  and Vimentin, which also support the effects of miR-552 and AJAP1 on EMT in HCC. Therefore, we concluded that miR-552 could facilitate HCC by promoting the EMT pathway.
There are still limitations in the current paper. For instance, a previous study proved that transfection of miRNA mimics at high concentrations altered the gene expression in a nonspecific manner and can cause cell death, 28 suggesting that the results may be affected by artefacts. The transfections used in our research were all transient transfection. As it is efficient in vitro, lentiviral transfection is more valid in vivo. These methods still need to be improved. On the other hand, metastasis assays could also be used in further studies to explore whether miR-552 could influence HCC metastasis. In vivo experiments on the inhibition of distant metastases will be performed in future studies. Because HCC is highly metastatic, further study of HCC metastasis would increase the clinical significance of using miR-552 as a therapy target for its treatment. Furthermore, we only proved a part of our hypothesis, and further study will be performed in the future.
MiR-552 was expressed at a high level in HCC tissues and cell lines. Upregulation of miR-552 was positively correlated with HCC cell proliferation and migration, along with poor clinicopathological characteristic of postoperative patients. MiR-552 played an oncogenic role by restricting AJAP1 expression and manipulating EMTrelated protein levels. Higher miR-552 and lower AJAP1 levels also corresponded to poorer HCC prognoses. In conclusion, miR-552 induced HCC progression by downregulating AJAP1 and may be significant for future HCC treatment.

ETHICS APPROVAL
The study was approved by the Ethics Boards of Qingdao No. 6 People's Hospital.

CONSENT FOR PUBLICATION
Written informed consent was collected from all patients.

CONFLI CT OF INTERESTS
The authors declare that they have no competing interests.

AUTHORS' CONTRIBUTI ONS
Contributing to the conception and design: Keli Su, Weiqing Qu; Analysing and interpreting data: Xinyuan Wen; Drafting the article: Weiqing Qu and Xinyuan Wen; Revising it critically for important intellectual content: Wei Gou; Approving the final version to be published: All authors.