MicroRNA-1915-3p Promotes Cell Metastasis and Progression by Targeting Bcl2L11 in Hepatocellular Carcinoma

Background: Increasing evidence suggests that miR-1915-3p plays vital regulatory roles in metastasis and progression of several types of cancer. However, the roles and underlying mechanism of miR-1915-3p in hepatocellular carcinoma (HCC) remains largely unclear. Methods: We carried out a bioinformatic meta-analysis to investigate a possible role of miR-1915-3p as prognostic biomarkers. In vitro cellular models of HCC were used for functional studies exploring the role of miR-1915-3p in HCC development and progression. Finally, in vivo studies were performed to demonstrate that miR-1915-3p is a viable therapeutic target. Results: This study showed that miR-1915-3p was signicantly increased in HCC tissue samples and cell lines, and high miR-1915-3p expression was associated with a poor overall survival (OS) and disease-free survival (DFS) time of HCC patients. Overexpression or ablation of miR-1915-3p expression resulted in accelerated or inhibited cell proliferation, migration, and invasion respectively in HCC cells. In addition, miR-1915-3p induced downregulation of proapoptotic factors, including caspase3, caspase8, BAD, Bcl2L11, and P53. It also induced upregulation of antiapoptotic Bcl-2, protecting HCC cells from apoptosis. A biological analysis indicated that miR-1915-3p could be directly targeted to Bcl2L11 to regulate the proliferation, invasion, and migration of HCC cells. Furthermore, in vivo studies conrmed that treatment with miR-1915-3p retarded the growth of tumor in nude mice. Conclusion: our study provided the evidence for the regulatory role of miR-1915-3p in HCC, which was causally linked to targeting of Bcl2L11. Medications that abrogate excessively expressed miR-1915-3p may offer novel targets for the management of HCC.


Introduction
Hepatocellular carcinoma (HCC) is one of the most common cancers of the liver and is ranked as the third leading cause of cancer death and the sixth most common neoplasm (1). Unfortunately, HCC is often diagnosed at an advanced stage, which is no longer amenable to curative therapies due to a de ciency of biomarkers for early diagnosis. Thus, radiotherapy and chemotherapy remaining the only options (2), which have shown antitumor activity (3), but no de nitive proof of survival bene t has been found (4). Presently, guring out the molecular mechanisms regarding the progression and metastasis in HCCs and looking for the biomarkers and therapeutic targets for HCC are urgent necessities.
MicroRNAs (miRNAs) are a class of highly conserved small RNA molecules that act as critical regulators of gene expression in multicellular eukaryotes (5). Since their initial discovery, a growing body of evidence has documented that miRNAs play an important role in cellular homeostasis, and dysfunction of these molecules are linked to several human diseases (6). Now, there are lots of examples linking dysregulated expression of miRNAs to cancer, and miRNAs are increasingly viewed as potential therapeutic, diagnostic, and monitoring targets (7,8). In HCC, miRNAs, including miR-101 (9), miR-221 (10), miR-21 (11) and miR-222 (12), can promote tumor development, whereas miR-203, miR-124 (13), miR-233 (14) and miR-122 (15) suppressed tumor growth and metastasis. These studies represent an important step toward the potential value of microRNA-based therapy for the treatment of HCC. miR-1915-3p was another dysregulated miRNA in a variety of cancers, including gastric cancer (16), breast cancer (17), diffuse glioma (18) and lung cancer (19). Although upregulated expression of miR-1915-3p have been reported to be linked to intracellular oxidative stress status in HCC (20), its role for tumor metastasis and progression is still largely unclear. Therefore, we examined the expression of miR-1915-3p in HCC, matching adjacent liver tissues and several HCC cancer cell lines with different metastatic potentials. We found that miR-1915-3p promoted proliferation, invasion, and migration of HCC cells in vitro. By using bioinformatics analysis, we identi ed that Bcl2L11 was the direct target of miR-1915-3p in regulating the proliferation, invasion, and migration of HCC cells. Moreover, we con rmed our nding by the xenograft tumor model. Therefore, the increased miR-1915-3p suppressed Bcl2L11 translation for apoptosis inhibition in promoting the growth of HCC cells might prove to be a novel therapeutic candidate for HCC.

Patients and specimens
Tissue sections in immunohistochemistry were enrolled at A liated Liutie Central Hospital of Guangxi Medical University and provided informed consent for tissue to be used for research studies. All tissues were snap-frozen and stored in liquid nitrogen immediately after surgical resection. Tissue microarray slides were purchased from Shanghai Outdo Biotech Co., Shanghai, China. The clinicopathologic features of the patients are described in Table 1. Each patient was diagnosed based on World Health Organization criteria, and tumor differentiation grade was classi ed according to Edmondson and Steiner (21). Liver function was evaluated using the Child-Pugh scoring system. Tumor stages were classi ed based on the International Union against Cancer TNM classi cation system. OS was de ned as the time from surgery to death or surgery to last observation time. The OS data were obtained from the last follow-up for surviving HCC patients. Time to tumor recurrence was calculated from the surgical resection to the date of any relapse, including both extrahepatic metastasis and intrahepatic recurrence (22).
The sample images were captured with 200X magni cation after incubation with 4-nitroblue-tetrazolium and nuclear fast red. ISH signals for miR-1915-3p were scored by combining the proportion of positive cells and staining intensity. The intensity of staining cells was scored as: strong (3) well as its negative control, were also purchased from Genomeditech (Shanghai, China). The transfected clones were selected with puromycin and G418 (Sigma-Aldrich, USA), according to the manufacturer's protocol, and the transfected e ciency was veri ed by quantitative PCR (qPCR) and western blot assays.
wt or mt 3'-UTR were sequenced from Bcl2L11 and were cloned into the pGL3-promoter vector (Promega, Madison, WI). All transfections were performed as previously described (24).

RNA extraction and quantitative real-time PCR (q RT-PCR)
Total RNA was extracted from cell lines and fresh-frozen HCC tissues with Trizol reagent (Thermo Fisher Scienti c). A synthesis of complementary DNA (cDNA) was performed by PrimeScript RT Reagent Kit (Takara, Shiga, Japan) and detected at the ABI7500 Sequence Detection System (Applied Biosystem, Foster City, CA, USA), according the manufacturer's guidelines. Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was used as an internal control for mRNA and U6 was used for microRNA. The gene expression relative to the endogenous control was analyzed by ∆C t and 2 -∆∆Ct . The following primers were used for qRT-PCR detection:
Wound-healing, cell proliferation and transwell invasion assay The wound healing assay was conducted with a modi ed method as previously described (25). Brie y, hepatocarcinoma cells were grown to 90% con uence and then starved for 12 hours. A scraped line was created with a p200 pipette tip and washed twice with PBS. The cells were subsequently incubated for 48 hours with a starvation medium (DMEM without FBS). Cellular migration toward the scratch was observed and recorded after 48 hours, and the percentage of open area was assessed. The CCK8 assay was used to measure the proliferation of HCC cells. Brie y, cells (104/well) were inoculated into 96-well assay plates. Then, the cells were incubated with 10 μl CCK-8 solution (Dojindo, Tokyo, Japan) for two hours. Absorbance was acquired by using an automatic microplate reader at 450 nm (Thermo Fisher Scienti c, Inc, Waltham, MA, USA). For the transwell invasion assays, the cells were plated in Transwell upper chambers (Corning, New York, USA) at a density of 2 × 104 in 200 μl of serum-free medium, while 800 μl of normal-containing culture medium was placed in the lower chambers. After 48 hours of incubation at 37℃, the cells that invaded the lower surface were xed with 4% paraformaldehyde for 15 minutes and stained with0.1% crystal violet dye (Ameresco, LIC, USA) for six minutes. They were photographed in an inverted microscope (×200 magni cation) and counted by using Image J software.

Luciferase reporter assay
For the luciferase assay, cells were seeded in 96-well plates and cultured for 24 hours. In all, 10 ng of Animal experiments Approximately 1×10 7 HCCSMMC-7721 cells, which stably transfected with either miR-1915-3p or the NC vector, were suspended in 100 μl Dulbecco's modi ed Eagle's medium and then cutaneously injected into the left ank of each BALB/c nude mouse. After injections for four weeks, the subcutaneous tumors were cut equally into small pieces and transplanted into the livers of nude mice. Transplanted mice were sacri ced six weeks later and tumor volume (V) was calculated with the following formula: (length × width 2 )/2 (26). The para n-xed livers were serial sectional and tunel assay was performed using a DeadEndTM Fluorometric TUNEL System (Promega), according to the manufacturer's instructions.

mRNA-seq
Total RNA was extracted from cells by Trizol reagent (Invitrogen, USA) and RNA samples were puri ed by magnetic oligo (dT). mRNA samples were then reverse transcribed into rst-strand cDNA, followed by second strand synthesis. Fragmented DNA samples were blunt ended and adapters were ligated to construct a library. DNA was quanti ed with Qubit (Invitrogen, CA, USA). Sequencing was carried out using an Illumina HiSeq 3000 SBS instrument according to the manufacturer's recommendations. We quanti ed transcript expression levels with a fragment per kilobase of transcripts per million fragments mapped (FPKM), and analyzed the gene transcripts between different samples.

Statistical analysis
All quanti ed data are presented as mean ± the standard deviation (SD). To compare quantitative data between groups, students' t-tests were used. For categorical parameters, the chi-square test was used to analyze the data. Kaplan-Meier survival analysis was used to estimate OS. The correlation between groups was determined using Pearson's correlation tests. All statistical analyses in this study were performed using SPSS-24 (IBM, NY, USA) and Prism 6 (GraphPad). P < 0.05 was considered signi cant.
Results miR-1915-3p is upregulated in HCC tissue samples and cell lines, and is associated with a poor prognosis for patients with HCC.
To characterize miR-1915-3p expression in human hepatic cancer, a tissue microarray containing 45 cases of normal, adjacent, and cancer tissue samples was immunohistochemically stained with a miR-1915-3p antibody (Fig. 1A). miR-1915-3p levels rose with a closing tumor, with the cancer tissue having signi cantly higher levels than normal or adjacent tissues. In addition, the level of miR-1915-3p was increased in most of the HCC cells lines (QGY-7701, QGY-7703, MHC-97H, HuH7, HepG2), whereas miR-1915-3p was decreased in SMMc-7721 cells in comparison with the normal hepatic cell line LO2 (Fig. 1B). Next, the expression of miR-1915-3p was examined in HCC specimens (n = 45). We found that miR-1915-3p was associated with signi cantly increased cancer tissue compared to the adjacent cells (Fig. 1C).
The correlation between the clinicopathologic features of 45 HCC patients and the expression of miR-1915-3p is shown in Table 1. The results indicated that the increase of the miR-1915-3p level was signi cantly correlated with the parameters of tumor differentiation (p = 006) and TNM stage (p = 0.02). However, the miR-1915-3p level was not correlated with other clinical characteristics, including sex, hepatitis B surface antigen, cirrhosis, tumor number, preoperative alpha-fetoprotein, alanine, and gamma glutamyl transferase (Table 1). In addition, a Kaplan-Meier analysis showed the one-year, three-year, and ve-year overall survival (OS) rates for HCC patients with high miR-1915-3p levels were signi cantly shorter than in patients with low miR-1915-3p levels. Moreover, HCC patients with high miR-1915-3p levels showed shorter one-year, three-year, and ve-year disease-free survival rates than those with low miR-1915-3p levels (Fig. 1D). These results indicated that upregulation of miR-1915-3p might play a potential role in promoting the malignant progression of HCCs.

miR-1915-3p promoted proliferation, invasion, and migration of HCC cells in vitro.
To further investigate the role of miR-1915-3p in HCCs, both loss and gain-of-function experiments were performed in HCCHepG2 and SMMC-7721 cell lines. As detected by qRT-PCR, we con rmed that the miR-
The Bcl-2 family is categorized into anti-apoptotic and proapoptotic molecules, and the cell life and death are modulated by the balance between these two subsets. As shown in Figs. 3A and B, the expression of the antiapoptotic molecule Bcl-2 was increased after miR-1915-3p upregulation in HCCSMMC-7721 cells, whereas the levels of proapoptotic molecules, including caspase3, caspase8, BAD, Bcl2L11, and P53, were all decreased. In contrast, inhibition of miR-1915-3p results in the downregulation of Bcl-2 and the upregulation of proapoptotic molecules in HepG2 cells. Therefore, these data demonstrate that miR-1915-3p plays a critical role in HCC cell survival by increasing Bcl-2 expression.
Bcl2L11 was found as one of the targets for miR-1915-3p.
In order to investigate the underlying mechanisms for how miR-1915-3p exerts its functional impacts on HCC cells, TargetScan (http://www.targetscan.org) was used to identify potential miR-1915-3p target genes (Fig. 4A). Among the identi ed potential targets, Bcl2L11 was rst due to its crucial role in the apoptosis in HCC cells (27). Bcl2L11 has been previously reported to be a target of several miRNAs in HCC (28) and other cancers (29,30). We found the complementary sequence of miR-1915-3p in the 3'-UTR of Bcl2L11 mRNA (Fig. 4B). To further con rm Bcl2L11 was a direct target of miR-1915-3p, we performed a luciferase reporter gene assay. The results demonstrated that miR-1915-3p overexpression resulted in a decrease luciferase activity in wild-type (wt) construct of Bcl2L11 3'UTR, but there was no signi cant change of luciferase activity observed in mutant (mt) (Fig. 4B). The relationship between miR-1915-3p and Bcl2L11 was analyzed in qRT-PCR. The results indicated that miR-1915-3p overexpression signi cantly decreased the Bcl2L11 level in SMMC-7721 cells, whereas a miR-1915-3p knockdown increased the level of Bcl2L11 in HepG2 (Fig. 4C). Then, to further elucidate whether miR-1915-3p executed its function by suppressing Bcl2L11, we transfected SMMC7721-miR-1915-3p or HepG2-anti-miR-1915-3p cells with Bcl2L11 overexpression or Bcl2L11 silence, respectively. The results showed that Bcl2L11 expression was signi cantly decreased in SMMC7721-miR-1915-3p cells, even though Bcl2L11 overexpression existed. Moreover, Bcl2L11 expression was signi cantly increased despite the Bcl2L11 silencing in HepG2-anti-miR-1915-3p cells (Fig. 4D). Therefore, these data suggested that Bcl2L11 could be a direct downstream target of miR-1915-3p in HCCs.
Bcl2L11 mediates the functional effects of miR-1915-3p on HCC cells.
Then, rescue experiments were performed to ascertain whether miR-1915-3p executed its functional effects by suppressing its target genes. A restoration of Bcl2L11 blocked the miR-1915-3p-mediated promotive effects on proliferation (Fig. 5A), migration (Fig. 5B), and invasion ( Fig. 5C) in SMMC-7721 cells. In addition, silencing of Bcl2L11 results in a reverse effect, which abrogates the inhibitory function of miR-1915-3p loss on HepG2 cells (Fig. 5A-C). Taken together, our results revealed that Bcl2L11 acted as a functional downstream target of miR-1915-3p in HCC cells.
The roles of miR-1915-3p-Bcl2L11 signaling on modulating tumor metastasis and tumor growth were examined in vivo. Restoration of Bcl2L11 expression (low-dose, medium-dose, and high-dose Bcl2L11) in SMMC7721-miR-1915-3p cells resulted in a signi cant decrease in tumor weight and volume compared to controls (Fig. 6A-C). In addition, different Bcl2L11 concentration treatments induced signi cant apoptosis in a dose-dependent manner compared with the group of mice treated with saline (Fig. 6D).

Discussion
Recurrence after resection or ablation and progression after effective chemoembolization are major drawbacks in the management of hepatocellular carcinomas. Thus, to understand the underlying mechanisms of HCC recurrence and to target the progression in order to provide more effective treatments of HCCs are urgent necessities. Since the initial application of miRNA in HCC, signi cant effort has been devoted to the development of therapeutics that utilize this pathway (31). Numerous microRNAs, including microRNA-195 (32), microRNA-494 (33) and microRNA-657(34), signi cantly suppressed or promoted HCC metastasis and progression, indicating the modulation of miRNA activity represented an attractive strategy for recurrence and progression of HCCs. Unfortunately, the underlying molecular mechanisms by these miRNAs in HCCs have not been fully revealed. In our study, the expression of miR-1915-3p, including HCC cell lines and tissue samples from humans, was a signi cant upregulation, which was consistent with previous reports (20). However, previous reports revealed only miR-1915-3p as an oxidative stress-responsive microRNA, but revealed its biological effect in HCC. This prompted us to speculate that miR-1915-3p may be important, but frequently neglected when correlated to metastasis or recurrence of HCC.
Similar with other ndings, the serum level of miR-1915-3p was upregulated in breast cancer (17) and diffuse glioma patients (18) when compared with healthy volunteers. Our study has, for the rst time to our knowledge, shown that the level of miR-1915-3p was higher in human liver cancer samples versus adjacent and normal samples. In addition, we also found miR-1915-3p was upregulated in HCC cell lines and associated with a poor prognosis for patients with HCC. In light of the expression level of miR-1915-3p that was decreased in the gastric cancer (16) and breast cancer tissues (35) and their cell lines, we postulated that miR-1915-3p may play different roles in various cancers, although further investigation is needed.
The aim of treatment of HCC is to increase survival while maintaining the highest quality of life. Survival bene ts have been proven by surgical resection, ablation (36), lenvatinib (37), and regorafenib (38) treatments in HCC. However, tumor recurrence, including true recurrence due to de novo tumors with the oncogenic liver, complicates 70% of case at ve years out (39). Our results showed that upregulation of miR-1915-3p promoted HCC progression by accelerating proliferation, migration, and invasion in vitro, whereas miR-1915-3p knockdown inhibited these activities. A recent study demonstrated that miR-1915-3p promotes the migration and proliferation of certain human breast cancer cells (17), which supports the oncogene role of miR-1915-3p in HCC. It should be noted that high miR-1915-3p expression levels also resulted in the reduction of proliferation and migration in breast cancer cell MCF-7 (35), and showed longer overall survival time than those with low miR-1915-3p expression levels via activation of Bcl-2 in gastric cancer patients (16). In our study, however, high miR-1915-3p expression predicted a reduced OS and DFS in HCC patients. The ndings that both high and low miR-1915-3p expressions could augment the biological behavior of cancer cells seems contradictory. Nevertheless, previous studies have demonstrated that ERK1/2 could be activated by high miR-1915-3p expression (17). Moreover, ERK1/2 could act as a positive-regulator of Bcl-2 (40,41). Thus, we have postulated that the biological behavior of cancer cells could be promoted through an indirect activation of Bcl-2 via ERK1/2 signaling at a high miR-1915-3p expression in breast cancer, and through direct activation of Bcl-2 via miR-1915-3p-binding mechanism at a low miR-1915-3p expression in gastric cancer.
The scienti c interests in exploring the mechanisms responsible for cell apoptosis in HCC have blossomed for decades. Previous studies have demonstrated that the caspase and Bcl-2 family play an important role during the initiation of apoptosis in HCC invasion and metastasis (42)(43)(44). In our in vitro study, miR-1915-3p increased antiapoptotic molecule expression (Bcl-2) and decreased the expression of proapoptotic molecules (caspase3, caspase8, BAD, Bcl2L11, and P53) in HCC cells. The evidence indicated that miR-1915-3p possibly contributes to the development and progression of HCC by regulating the caspase and Bcl-2 family. Our ndings may provide possible research directions for pathological mechanisms of HCC recurrence or progression, which better early detection and isolation.
Bcl2L11, a member of the Bcl-2 family, reportedly conferred dual regulatory cell growth and apoptosis in gastric cancer (29), non-small cell lung cancer (45) and ovarian cancer (46).

Conclusion
Collectively, to our knowledge, we presented the rst evidence that miR-1915-3P was highly expressed in metastasis HCC tissue samples and cell lines. In addition, its upregulation was related to the poor prognosis of HCC patients, as well as the idea that miR-1915-3p promoted proliferation, invasion, and metastasis of HCC cells by suppression of Bcl2L11. While there clearly remains signi cant work to be done in HCC therapies, our ndings highlight the roles of miR-1915-3p involved in HCC metastasis and progression, which will enable us to use it as a potential therapeutic target.  independent experiments, *P < 0.05, **P < 0,01). C. Wound healing as assessed by scratch assay and the quanti cation of the percentage of wound area were shown (n = 3 independent experiments, Scal bar: 50 μm, **P < 0,01). D. Cell invasion was examined using transwell invasion assays after knockdown or overexpression of miR-1915-3p in HCC cells (n = 3 independent experiments, Scal bar: 50 μm, **P < 0,01).

Declarations
Data are presented as mean ± standard deviation.
Page 20/23   assay showed the proliferation of SMMC-7721-miR1915-3p that were transfected with Bcl2L11 vector and HepG2-miR1915-3p cells that were transfected with Bcl2L11 vector and their corresponding control (n = 3 independent experiments, **P < 0,01). The efects of Bcl2L11 on the migration and invasion of HCC cells was determined by B wound healing assay and C transwell assay (n = 3 independent experiments, scal bar: 50 μm, **P < 0,01). Data are presented as mean ± standard deviation.