miR-23c suppresses tumor growth of human hepatocellular carcinoma by attenuating ERBB2IP

Biomedicine & Pharmacotherapy 107 (2018) 424–432

Contents lists available at ScienceDirect

Biomedicine & Pharmacotherapy journal homepage: www.elsevier.com/locate/biopha

miR-23c suppresses tumor growth of human hepatocellular carcinoma by attenuating ERBB2IP

T

⁎

Lei Zhanga, Yufeng Wangb, Liang Wangb, Guozhi Yinb, Weimin Lic, Yao Xianc, Wei Yangb, , ⁎ Qingguang Liub, a

Department of Geriatric Surgery, The First Affiliated Hospital of Xi’an Jiaotong University, 277 Yanta West Road, Xi’an, Shaanxi Province 710061, China Department of Hepatobiliary Surgery, The First Affiliated Hospital of Xi’an Jiaotong University, 277 Yanta West Road, Xi’an, Shaanxi Province 710061, China c Department of Nutrition, The First Affiliated Hospital of Xi’an Jiaotong University, 277 Yanta West Road, Xi’an, Shaanxi Province 710061, China b

A R T I C LE I N FO

A B S T R A C T

Keywords: microRNA-23c Hepatocellular carcinoma ERR2IP Tumor growth Apoptosis

MicroRNAs (miRNAs) regulate a variety of development and physiologic processes, and play prominent roles in the initiation and progression of human cancers including hepatocellular carcinoma (HCC). MiR-23c is recently emerging as a cancer-associated miRNA, while its expression status and functional role in HCC are unrevealed yet. Here, we found that miR-23c underexpression was associated with the tumorigenesis of HCC based on TCGA data. qRT-PCR analysis revealed that miR-23c expression was reduced in HCC tissues and cell lines. Clinical analysis indicated that low miR-23c expression was correlated with large tumor size, high tumor grade, advanced tumor stage and poor survival of HCC patients. Our in vitro experiments found that overexpression of miR-23c inhibited cell proliferation and induced apoptosis of HCC cells. While miR-23c knockdown led to HCC cell growth arrest and apoptosis. Additionally, miR-23c overexpression repressed tumor growth of HCC in vivo. Mechanistically, erbb2 interacting protein (ERBB2IP) was identified as a direct target of miR-23c in HCC cells. miR-23c suppressed ERBB2IP expression in HCC cells and inversely correlated with ERBB2IP mRNA expression in HCC tissues. Notably, ERBB2IP silencing restrained HCC cell proliferation and induced apoptosis. ERBB2IP restoration reversed the inhibitory effects of miR-23c on HCC cell growth. In conclusion, our observations suggested that miR-23c inhibited cell proliferation and accelerated apoptosis by attenuating ERBB2IP. Targeting miR-23c might open a new avenue for HCC treatment.

1. Introduction Hepatocellular carcinoma (HCC) treatment is a knot that still cannot be solved by the surgeons and physicians worldwide. Surgical resection is the only and best option for early stage HCC [1]. However, most of HCC patients are often diagnosed at advanced stage accompanied with intrahepatic and distant metastases [1]. Even though patients with advanced HCC were treated with surgical resection, transarterial chemoembolization (TACE) or sorafinib, the survival is still dismal due to postoperative recurrence and metastasis [2–4]. Thus, it is imperative to explore the molecular mechanism underlying the initiation and progression of HCC.

MicroRNAs (miRNAs) are an abundant class of small non-coding RNA molecules (20–24 nt)), which are expressed in various tissues and cell types, and they are involved in post-transcriptional modulation of gene expression by targeting mRNA for degradation and translational inhibition [5]. Increasing evidence support that miRNAs play critical roles in distinct and diverse biological procedures, including cell proliferation, differentiation, apoptosis, cell cycle progression and stem cell division [6]. Importantly, miRNAs have been shown to be aberrantly expressed in many human diseases including cancer [7]. The specific miRNAs that function as oncogenes or tumor suppressers, contribute to the initiation and progression of human cancer [8]. Our research team have reported several hepatocarcinogenesis-associated miRNAs,

Abbreviations: HCC, hepatocellular carcinoma; TACE, transarterial chemoembolization; miRNA, microRNAs; LIHC, liver hepatocellular carcinoma; ELAVL1, ELAV like RNA binding protein 1; GSK3β, glycogen synthase kinase 3 beta; ECM, extracellular matrix; CRC, colorectal cancer; ERBB2IP, Erbb2 interacting protein; TGF-β, transforming growth factor beta; ERα, estrogen receptor-α; QSER1, glutamine and serine rich 1; FNIP1, folliculin interacting protein 1; BTLA, B and T lymphocyte associated; TNFIP3, tumor necrosis factor alpha-induced protein 3; BRCA, breast invasive carcinoma; CHOL, cholangiocarcinoma; LUAD, lung adenocarcinoma; LUSC, lung squamous cell carcinoma; PRAD, prostate adenocarcinoma; STAD, stomach adenocarcinoma ⁎ Corresponding authors. E-mail addresses: [email protected] (W. Yang), [email protected] (Q. Liu). https://doi.org/10.1016/j.biopha.2018.07.155 Received 29 June 2018; Received in revised form 26 July 2018; Accepted 31 July 2018 0753-3322/ © 2018 Elsevier Masson SAS. All rights reserved.

Biomedicine & Pharmacotherapy 107 (2018) 424–432

L. Zhang et al.

including miR-1468 [9], miR-876-5p [10], miR-542-3p [11], miR-367 [12], miR-1296 [13], miR-519a [14] and miR-187-3p [15]. Analysis of TCGA data from ONCOMIR (http://www.oncomir.org) reveals that there are 398 miRNAs associated with tumorigenesis of liver hepatocellular carcinoma (LIHC). Among these miRNAs, miR-23c is poorly characterized in HCC. miR-23c is found to repress renal tubular epithelial pyroptosis by targeting ELAV like RNA binding protein 1 (ELAVL1) in diabetic nephropathy [16]. Moreover, miR-23c directly targets glycogen synthase kinase 3 beta (GSK3β) and affects nucleus pulposus cell proliferation and extracellular matrix (ECM) synthesis via Wnt signaling in intervertebral disc degeneration [17]. The first evidence for the involvement of miR-23c in human cancer originates from the higher expression of miR-23c in colorectal cancer (CRC) tissues with recurrence [18]. Furthermore, steviol suppresses gastrointestinal cancer cell proliferation partially by regulating miR-23c [19]. However, the expression and functional role of miR-23c are rarely known in HCC. Erbb2 interacting protein (ERBB2IP), also known as Erbin, is a leucine-rich repeat and PDZ domain protein [20]. Increasing studies indicate that ERBB2IP regulates cancer cell apoptosis, proliferation and inflammatory response [21–24]. The function of ERBB2IP depends on cell types and contexts. ERBB2IP acts as a tumor suppressor via inhibiting oncogenic pathways including ERK [25] and transforming growth factor beta (TGF-β) [26]. In contrast, ERBB2IP contributes to the tumorigenesis of breast cancer [27], CRC [23] and HCC [28]. ERBB2IP expression is elevated and functions as an oncogene by destabilizing estrogen receptor-α (ERα) in HCC [28]. However, the mechanism underlying aberrant overexpression of ERBB2IP in HCC is still unclear. At present, we confirmed the prognostic significance of miR-23c in HCC. Then we investigated the role of miR-23c in HCC progression through functional experiments in vitro and xenograft tumor model in vivo. miR-23c regulation of ERBB2IP was demonstrated by bioinformatics and luciferase reporter assays.

Table 1 Correlation between the clinicopathologic characteristics and miR-23c expression in hepatocellular carcinoma. Characteristics

n = 80

miR-23c

P

Low expression (n = 40)

High expression (n = 40)

Age (y)

< 60 ≥60

45 35

20 20

25 15

0.260

Sex

Male Female

63 17

29 11

34 6

0.172

HBsAg

Negative Positive

13 67

5 35

8 32

0.363

Serum AFP level (ng/mL)

< 20 ≥20

36 44

14 26

22 18

0.072

Tumor size (cm)

<5 ≥5

47 33

18 22

29 11

0.012*

Liver cirrhosis

Absent Present

22 58

9 31

13 27

0.317

Microvascular invasion

Absent Present

35 45

14 26

21 19

0.115

Edmondson-Steiner grade

I + II III + IV

46 34

18 22

28 12

0.024*

TNM tumor stage

I + II III + IV

64 16

28 12

36 4

0.025*

HBV, hepatitis B virus; AFP, alpha-fetoprotein; TNM, tumor-node-metastasis. * Statistically significant.

ERBB2IP and non-targeting (NT) siRNA were purchased from GenePharma (Shanghai, China). These oligonucleotides and vectors were transfected into HCC cells using Lipofectamine 2000 (Thermo Fisher Scientific, Waltham, MA, USA).

2. Materials and methods 2.3. Quantitative real-time polymerase chain reaction (qRT-PCR) 2.1. Clinical samples and cell lines Total RNAs were extracted from tissues and cultured cells with Trizol Reagent (Thermo Fisher Scientific). Total RNA was reverse transcribed into cDNA using a RevertAid First Strand cDNA Synthesis Kit (Thermo-Fisher Scientific). qRT-PCR analyses were performed using SYBR® Premix Ex Taq™ II (Takara, Dalian, China) and an All-in-One miRNA qPCR Detection kit (GeneCopoeia, Rockville, MD, USA) on an ABI PRISM 7300 Sequence Detection system (Applied Biosystems, Foster City, CA, USA). The primer sequences were listed as follows: miR-23c forward, 5′-CCA GAA GGA CGT AGA AG-3′and reverse, 5′-CTT CAC TGT GAT GGG CTC-3′; U6 forward, 5′-GCT TCG GCA GCA CAT ATA CTA AAA T-3′ and reverse, 5′-CGC TTC ACG AAT TTG CGT GTC AT-3′; ERBB2IP forward, 5′- CTC TGT GGG GAC TTC AAC G-3′ and reverse, 5′-TGG GTG TCA GCT TGG TGT T-3′; GAPDH forward, 5′-CCA TGT TCG TCA TGG GTG TG-3′ and reverse, 5′-GGT GCT AAG CAG TTG GTG GTG-3′. The 2−ΔΔCt method was used to calculate the relative gene expression normalized by GAPDH and U6.

All 80 pairs of HCC and tumor-adjacent tissues were obtained from the First Affiliated Hospital of Xi’an Jiaotong University (Xi’an, China). All patients signed written informed consent before enrolling in this study. The collected specimens were pathologically confirmed and initially snap frozen in liquid nitrogen, then stored at −80 °C until use. All patients did not receive immunotherapy, radiotherapy or chemotherapy prior to operation. This study was reviewed and approved by the Ethic Committee of 1st Affiliated Hospital of Xi’an Jiaotong University in accordance with the guidelines outlined in the Declaration of Helsinki. The clinicopathological features of HCC patients were summarized in Table 1. Human immortalized normal hepatocyte cell line (LO2) and HCC cell lines (Huh7, SMMC-7721, Bel-7402, MHCC97H) were previously maintained in our lab. HCC cells were grown in Dulbecco’s modified Eagle medium (DMEM, Gibco, Carlsbad, CA, USA) supplemented with 10% fetal bovine serum (FBS, Gibco) and 1% penicillin-streptomycin (Sigma-Aldrich, Inc., St-Louis, MO, USA) in the incubator at 37 °C at 5% CO2.

2.4. Western blot Total proteins from cultured cells were lysed using RIPA buffer (Beyotime, Shanghai, China) and were quantified with a BCA protein assay kit (BIO-RAD, Hercules, CA, USA). Proteins were fully electrophoresed on 10% SDS polyacrylamide gels (SDS-PAGE) and transferred to polyvinylidene difluoride (PVDF) membranes. The membranes were incubated with primary antibodies at 4 °C overnight. The rabbit-antihuman ERBB2IP primary antibody was purchased from Bethyl Laboratories (A303-763A, Montgomery, TX, USA). The mouse-antihuman GAPDH primary antibody was obtained from Santa Cruz Biotechnology (sc-47724, Santa Cruz, Dallas, TX, USA). Then, the

2.2. Cell transfection Lentiviral vector-mediated miR-23c (miR-23c, HLMIR1272) and negative control (NCLMIR001) were purchased from Sigma. miR-23c inhibitors (anti-miR-23c, HmiR-AN1916-AM03) and negative control (CmiR-AN0001-AM03) were obtained from GeneCopoeia, Inc (Guangzhou, China). Lentivirus infection of HCC cells was performed in the presence of Polybrene (8 ng/ml). ERBB2IP expression plasmid (pcDNA3.1-ERBB2IP), small interfering RNA (siRNA) targeting 425

Biomedicine & Pharmacotherapy 107 (2018) 424–432

L. Zhang et al.

Table 2 hsa-miR-23c expression is associated with the tumorigenesis of human cancers. miRNA name

Cancer abbreviation

T-test p-value

T-test FDR

Unregulated in:

Tumor log2 Mean expression

Normal log2 Mean expression

hsa-miR-23c hsa-miR-23c hsa-miR-23c hsa-miR-23c hsa-miR-23c hsa-miR-23c hsa-miR-23c

BRCA CHOL LIHC LUAD LUSC PRAD STAD

3.74e-02 8.50e-07 2.17e-05 1.48e-04 1.88e-02 4.41e-10 5.35e-03

6.28e-02 4.36e-05 9.41e-05 7.36e-04 3.88e-02 5.24e-09 1.19e-02

Tumor Normal Normal Tumor Tumor Normal Normal

0.14 0.17 0.27 0.39 0.22 1.00 0.11

0.04 2.20 0.88 0.06 0.05 2.70 0.44

BRCA: breast invasive carcinoma; CHOL: cholangiocarcinoma; LIHC: liver hepatocellular carcinoma; LUAD: lung adenocarcinoma; LUSC: lung squamous cell carcinoma; PRAD: prostate adenocarcinoma; STAD: stomach adenocarcinoma. Fig. 1. The expression and prognostic significance of miR-23c in HCC. (A) The expression of miR-23c was reduced in 80 HCC tissues compared to adjacent noncancerous tissues as determined by qRT-PCR. P < 0.0001 by t-test (B) The expression differences of miR-23c between HCC cell lines (Huh7, MHCC97H, SMMC-7721 and Bel-7402) and human hepatocyte cell line (LO2). n = three independent experiments, *P < 0.05 by ANOVA. (C) The HCC patients were divided into miR-23c low-expressed (n = 40) and miR23c high-expressed group (n = 40) with the median value of miR-23c expression as a cut off value. The overall survival of HCC patients in the miR-23c low-expressed group was poorer than cases in high-expressed group. P < 0.05 by Log-rank test. (D) TCGA data from OncoLnc (http://www.oncolnc.org/) confirmed that low miR-23c expression indicated poor survival of HCC patients. P < 0.05 by Log-rank test.

Table 3 hsa-miR-23c expression is associated with clinical parameters in LIHC based on TCGA data. miRNA Name

Clinical Parameter

ANOVA P-value

ANOVA FDR

Multivariate Log Rank P-value

Multivariate Log Rank FDR

hsa-miR-23c hsa-miR-23c hsa-miR-23c hsa-miR-23c hsa-miR-23c hsa-miR-23c

Histologic Grade Pathologic M Status Pathologic N Status Pathologic Stage Pathologic T Status Sex

6.93e-06 1.07e-04 1.06e-08 3.49e-04 1.54e-03 1.52e-02

5.63e-04 1.24e-02 1.43e-06 2.02e-02 4.03e-02 1.72e-01

4.11e-05 1.53e-04 2.17e-04 6.76e-03 2.42e-03 1.11e-04

1.39e-03 2.96e-03 4.19e-03 5.95e-02 3.20e-02 2.90e-03

LIHC: liver hepatocellular carcinoma.

2.6. CCK-8 assay

membranes were incubated with horseradish peroxidase (HRP)-conjugated secondary antibodies (NXA931-1ML and NA934-1ML, GE Healthcare Life Sciences, Beijing, China) and visualized with ECL reagents (Millipore, Plano, TX, USA). Blots were semi-quantified by ImageJ software (1.46; National Institutes of Health, Bethesda, MD, USA).

The cell proliferation was measured by cell counting kit-8 assay (meilunbio, Dalian, China). HCC cells were cultured in a 96-well plate and transfected with corresponding plasmids, and then collected from 0, 24, 48 and 72 h after transfection. 10 μL CCK-8 reagent was added into the cells and incubated for another 1 h at 37 °C. Next, the resulting product was measured the absorbance at 450 nm using a microplate reader (Thermo Scientific).

2.5. Colony formation assay HCC cells (500 cells per well) were seeded in a 6-well plate and cultured for around two weeks. Then, cell colonies were stained with hematoxylin solution after washing three times with PBS. The numbers of visible colonies were calculated under a microscope.

2.7. EdU incorporation assay Ethynyl deoxyuridine (EdU) incorporation assay was performed 426

Biomedicine & Pharmacotherapy 107 (2018) 424–432

L. Zhang et al.

Fig. 2. miR-23c regulates HCC cell proliferation and apoptosis. (A) SMMC-7721 and Bel-7402 cells that were transfected with miR-23c or control vector were detected by qRT-PCR for miR-23c expression. n = three independent experiments, *P < 0.05 by t-test (B) CCK-8 assay revealed that miR-23c overexpression repressed the proliferation of HCC cells. n = three independent experiments, *P < 0.05 by ANOVA. (C) The colony forming number in miR-23c overexpressed HCC cells was obviously lower compared to control cells. n = three independent experiments, *P < 0.05 by t-test (D) The percentage of EdU-positive HCC cells was significantly reduced by miR-23c overexpression. n = five fields of three independent experiments, *P < 0.05 by t-test Scale bar: 20 μm. (E) miR-23c restoration led to increased apoptosis in both SMMC-7721 and Bel-7402 cells. n = three independent experiments, *P < 0.05 by t-test.

1000 rpm/min for 5 min. The supernate was discard. The residue was resuspended with 100 μL binding buffer. 4 μL Annexin V-FITC and 3 μL propidium (PI) were added in the situation. After incubating for 15 min at room temperature, 200 μL binding buffer was added and measured by FCM flow cytometry (BD, Bioscience, San Jose, CA, USA).

with the EdU kit (Roche, Indianapolis, IN, USA) in accordance with the manufacturer’s instruction. Results were acquired using the Zeiss fluorescence photomicroscope (Carl Zeiss, Oberkochen, Germany) and quantified via counting at least five random fields. 2.8. Cell apoptosis analysis The transfected cells were washed by cold PBS and centrifuged at 427

Biomedicine & Pharmacotherapy 107 (2018) 424–432

L. Zhang et al.

Fig. 3. miR-23c overexpression suppresses tumor growth of HCC in vivo. (A) SMMC-7721 cells that were stably transfected with miR-23c or control vector were implanted into nude mice via subcutaneous injection. The tumor volume in miR-23c overexpression group (n = 5) was obviously smaller that this in control group (n = 5). *P < 0.05 by ANOVA. Scale bar: 1cm. (B) The tumor weight in miR-23c overexpression group was significantly lighter that this in control group. n = 5, * P < 0.05 by t-test (C) The expression of miR-23c in miR-23c overexpression group was markedly higher that this in control group. n = 5, *P < 0.05 by t-test (D) The percentage of Ki-67 staining tumor cells in miR-23c overexpression group was prominently lower compared to control group. n = three fields of five tissue sections, *P < 0.05 by t-test Scale bar: 50 μm.

2.9. HCC xenograft model

2.11. Statistical analyses

Male BALB/c nude mice (four-week-old) purchased from Shanghai Laboratory Animal Co. Ltd. (Shanghai, China). Briefly, SMMC-7721 cells with stably overexpressed miR-23c and control cells were implanted into the flank of nude mice via subcutaneous injection. At the endpoint of 3 weeks, mice were sacrificed by cervical dislocation under anesthesia with ether and tumor tissues were harvested, photographed, measured and examined by immunohistochemical staining of Ki-67 [29]. Tumor sizes were evaluated by the formula: Volume (mm3) = [width2 (mm2) × length (mm)]/2. All animal experiments were approved by the Animal Care Committee of Xi’an Jiaotong University.

Data were analyzed using GraphPad Prism 6.0 Software (GraphPad Inc., San Diego, CA, USA). The Student’s t-test was used to analyze differences between two groups, and ANOVA was used when more than two groups were compared. The correlation between miR-23c and ERBB2IP mRNA expression was analyzed using the Spearman’s correlation test. Overall survival curves were protracted using the KaplanMeier method and estimated by the Log-rank test. Differences were defined as statistically significant if p < 0.05. 3. Results 3.1. miR-23c inversely correlates with HCC progression Analysis of TCGA data from ONCOMIR (http://www.oncomir.org) revealed that miR-23c underexpression was significantly associated with the tumorigenesis of LIHC (P < 0.0001, Table 2). Next, we found that miR-23c expression was obviously down-regulated in HCC tissues relative to adjacent normal liver tissues (P < 0.05, Fig. 1A). Besides, miR-23c level was promiently reduced in HCC cell lines (Huh7, MHCC97H, SMMC-7721, Bel-7402) as compared with LO2 cells (P < 0.05, Fig. 1B). Then, the patients were divided into miR-23c lowexpressed (n = 40) and miR-23c high-expressed group (n = 40) with the median value of miR-23c expression as a cut off value. As shown in Table 1, miR-23c expression was negatively associated with tumor size (P = 0.012), Edmondson-Steiner grade (P = 0.024) and TNM tumor stage (P = 0.025). Analysis of TCGA data from ONCOMIR (http://

2.10. Luciferase reporter assay Luciferase reporter vector containing 3′UTR of ERBB2IP was purchased from GeneCopoeia, Inc (HmiT059016-MT06, wt ERBB2IP3′UTR). The potential miR-23c binding sites were mutated by the Quick-change site-directed mutagenesis kit (Agilent Technologies, Santa Clara, CA, USA, mt ERBB2IP-3′UTR). wt ERBB2IP-3′UTR or mt ERBB2IP-3′UTR and miR-23c mimics or negative control were transfected into SMMC-7721 cells. Forty-eight hours after transfection, luciferase activities were obtained using the Dual-Luciferase Reporter Assay System (Promega, Madison, WI, USA).

428

Biomedicine & Pharmacotherapy 107 (2018) 424–432

L. Zhang et al.

Fig. 4. ERBB2IP is a direct target of miR-23c. (A) The putative binding sequences between miR-23c and 3′-UTR of ERBB2IP mRNA was shown. Mutated sequences in 3′-UTR of ERBB2IP mRNA were underlined. (B) Overexpression of miR-23c reduced the luciferase activity of vectors containing wt 3′-UTR of ERBB2IP rather than mt 3′-UTR of ERBB2IP in SMMC-7721 cells. n = three independent experiments, *P < 0.05 by t-test (C) miR-23c overexpression reduced the levels of ERBB2IP mRNA and protein in SMMC-7721 cells. n = three independent experiments, *P < 0.05 by t-test (D) miR-23c knockdown increased ERBB2IP abundance in Huh7 cells. n=three independent experiments, *P < 0.05 by t-test (E) The expression of ERBB2IP mRNA was up-regulated in HCC tissues compared to normal liver tissues in both TCGA database and our cohort of 80 HCCs. P < 0.0001 by t-test (F) An inverse correlation between miR-23c and ERBB2IP mRNA expression was observed in 80 HCC samples. P < 0.0001 by Spearman’s correlation test.

7402 cells (P < 0.05, Fig. 2E), whereas Huh7 cells with miR-23c knockdown had lower percentage of apoptotic cells than the control cells (P < 0.05, Supplementary Fig. 1E). Collectively, miR-23c repressed cell proliferation and induced apoptosis of HCC cells.

www.oncomir.org) also indicated that miR-23c expression was correlated with histological grade, pathologic M status, pathologic N stage, pathologic stage and pathologic T status (P < 0.05, respectively, Table 3). Furthermore, the overall survival of HCC patients in the miR23c low-expressed group was poorer than cases in high-expressed group (P = 0.0047, Fig. 1C). TCGA data from OncoLnc (http://www.oncolnc. org/) confirmed that low miR-23c expression indicated poor survival of HCC patients (P = 0.0032, Fig. 1D). These findings suggested that miR23c could inhibit HCC progression and was a potential favorable prognostic marker.

3.3. miR-23c suppresses tumor growth in vivo Next, a xenograft tumor model was constructed to investigate the effects of miR-23c on tumor growth in vivo. Three weeks after implantation, the tumor volume in miR-23c overexpressed SMMC-7721 group was obviously smaller than the control group (P < 0.05, Fig. 3A). In accordance, the tumor weight in miR-23c overexpressed SMMC-7721 group was significantly lighter than the control group (P < 0.05, Fig. 3B). The xenograft tumors arising form miR-23c overexpressed SMMC-7721 group expressed higher level of miR-23c but less Ki-67 staining than the control group (P < 0.05, Fig. 3C and D). These results suggested that ectopic forced expression of miR-23c markedly inhibited tumor growth of HCC in mice.

3.2. MiR-23c inhibits HCC cell proliferation and induces apoptosis To disclose whether miR-23c displays a specific role in HCC progression, SMMC-7721 and Bel-7402 cells were constructed as miR-23c overexpressed cell lines (P < 0.05, Fig. 2A). Notably, CCK-8, colony formation and EdU incorporation assay indicated that the proliferative capability of miR-23c overexpressed HCC cells were decreased compared to control cells (P < 0.05, Fig. 2B–D). In contrast, miR-23c was knocked down by transfecting miR-23c inhibitors into Huh7 cells (P < 0.05, Supplementary Fig. 1A), and proliferation of Huh7 cells with miR-23c knockdown was enhanced relative to control cells (P < 0.05, Supplementary Fig. 1B–D). Moreover, miR-23c overexpression significantly induced apoptosis of both SMMC-7721 and Bel-

3.4. ERBB2IP is a direct target of miR-23c To disclose the molecular mechanism underlying tumor suppressive role of miR-23c in HCC, bioinformatics analysis was performed to search for the potential targets of miR-23c. Based on miRWalk (http:// 429

Biomedicine & Pharmacotherapy 107 (2018) 424–432

L. Zhang et al.

Fig. 5. Restoration of ERBB2IP attenuates the tumor suppressive effects of miR-23c. (A) SMMC-7721 cells that were co-transfected with corresponding vectors were subjected to immunoblotting for ERBB2IP expression. (B) CCK-8 assay revealed that ERBB2IP restoration promoted the proliferation of miR-23c overexpressed SMMC-7721 cells. (C) The colony forming number in SMMC-7721 cells co-transfected with miR-23c and ERBB2IP vector was obviously higher compared to control cells. (D) The percentage of EdU-positive SMMC-7721 cells with miR-23c overexpression was significantly increased by ERBB2IP restoration. Scale bar: 20 μm. (E) ERBB2IP restoration led to reduced apoptosis in miR-23c overexpressed SMMC-7721 cells. n = three independent experiments, *P < 0.05 versus Control+EV by ANOVA, #P < 0.05 versus miR-23c+EV by ANOVA.

data from UALCAN (http://ualcan.path.uab.edu/index.html) revealed that ERBB2IP expression was up-regulated in HCC compared to normal liver tissues (P < 0.05, Fig. 5E). Elevated expression of ERBB2IP also confirmed in HCC tissues based on our cohort (P < 0.05, Fig. 5E). And a reverse relationship between miR-23c and ERBB2IP levels was observed in HCC tissues (r = −0.312, P < 0.0001, Fig. 4F). Altogether, ERBB2IP was a direct target of miR-23c in HCC.

mirwalk.umm.uni-heidelberg.de/), glutamine and serine rich 1 (QSER1), folliculin interacting protein 1 (FNIP1), ERBB2IP, B and T lymphocyte associated (BTLA) and tumor necrosis factor, alpha-induced protein 3 (TNFIP3) were recognized as potential targets of miR23c in databases including TargetScan, miRDB and miRTarBase. Among these, only ERBB2IP was reported to be an oncogene in HCC [28]. Thus, ERBB2IP was selected as a candidate target of miR-23c. The luciferase reporter vector containing wild-type (wt) or mutated (mt) 3′UTR of ERBB2IP were used (Fig. 4A). The luciferase activities were obviously reduced in the reporter vector with wt 3′UTR of ERBB2IP, but not in vector with mt 3′UTR of ERBB2IP after miR-23c overexpression (P < 0.05, Fig. 4B). Moreover, the levels of ERBB2IP mRNA and protein were reduced in SMMC-7721 cells with miR-23c overexpression but increased in Huh7 cells with miR-23c knockdown (P < 0.05, Fig. 4C and D). However, miR-23c overexpression did not have significant impact on the levels of QSER1, FNIP1, BTLA, and TNFIP3 mRNA in SMMC-7721 cells (Supplementary Fig. 2). Analysis of TCGA

3.5. ERBB2IP is essential for the role of miR-23c in HCC Since ERBB2IP contributes to tumorigenesis in HCC [28]. Thus, we aimed to explore whether ERBB2IP mediated the biological behavior of miR-23c in HCC. ERBB2IP was knocked down by a specific siRNA in SMMC-7721 cells (P < 0.05, Supplementary Fig. 3A). We found that ERBB2IP silencing prominently retrained cell proliferation and induced apoptosis of SMMC-7721 cells (P < 0.05, Supplementary Fig. 3B–E). Furthermore, the expression of ERBB2IP was restored in miR-23c 430

Biomedicine & Pharmacotherapy 107 (2018) 424–432

L. Zhang et al.

effective therapeutic candidates for HCC.

overexpressed SMMC-7721 cells (P < 0.05, Fig. 5A). CCK-8, colony formation and EdU incorporation assays revealed that ERBB2IP restoration attenuated the suppressive role of miR-23c in the proliferation of SMMC-7721 cells (P < 0.05, Fig. 5B–D). Flow cytometry analysis indicated that ERBB2IP restoration reversed apoptosis of miR-23c overexpressed SMMC-7721 cells (P < 0.05, Fig. 5E). Taken together, these results suggested that the tumor suppressive role of miR-23c was at least partially attributed by attenuating ERBB2IP in HCC.

5. Conclusions To conclude, we have demonstrated that downregulation of miR23c is a frequent event in HCC. miR-23c underexpression is associated with poor prognostic features and indicates reduced survival of HCC patients. miR-23c inhibits HCC cell proliferation and induces apoptosis in vitro and in vivo. ERBB2IP is a crucial downstream target of miR-23c to mediate the antitumor effects. The identification of miR-23c/ ERBB2IP axis can provide a new insight into the understanding of molecular mechanism involved in HCC initiation and progression, and it can bestow diagnostic and therapeutic advantages to the development of novel therapy for HCC.

4. Discussion Increasing studies have confirmed that miRNAs are main drivers on the initiation and progression of HCC via post-transcritptionally regulating oncogenes or tumor suppressors [30,31]. Therefore, discovery of oncogenic or tumor suppressive miRNAs is important to disclose the molecular mechanisms underlying HCC progression. Only several studies focus on the role of miR-23c in human cancers [18,19]. Based on TCGA data, we found that miR-23c expression was associated with the tumorigenesis of breast invasive carcinoma (BRCA), cholangiocarcinoma (CHOL), LIHC, lung adenocarcinoma (LUAD), lung squamous cell carcinoma (LUSC), prostate adenocarcinoma (PRAD) and stomach adenocarcinoma (STAD). Reduced expression of miR-23c was observed in CHOL, LIHC, PRAD and STAD, whereas elevated expression of miR23c was found in BRCA, LUAD and LUSC, suggesting context-dependent expression of miR-23c in human cancers. Our data further confirmed the underexpression of miR-23c in HCC tissues. Further clinical association analysis indicated that miR-23c underexpression positively correlated with large tumor sized, high tumor grade and advanced tumor stage. Moreover, both our follow-up data and TCGA data revealed that low miR-23c level indicated poor prognosis of HCC patients. There data suggest that miR-23c may act as a promising prognostic marker for HCC patients. Previous study reports that miR-23c is implicated in CRC cell proliferation [19]. In the present study, through gain- and loss-of-function experiments, we verified that miR-23c functioned as a critical tumor suppressor to repress cell proliferation and induce apoptosis in different HCC cell lines by directly attenuating ERBB2IP, which suggested a novel molecular mechanism corresponding for HCC progression. Through systematic bioinformatics prediction and mechanistic experiments, we confirmed that ERBB2IP was a direct target of miR-23c in HCC cells. MiR-130b-3p regulation of ERBB2IP has been reported in early lupus nephritis [32]. ERBB2IP has been recognized as an antitumor factor in human cervical and breast cancer [33,34]. ERBB2IP loss facilitates cervical cancer cells resistance to anoikis and contributes to tumor growth and metastasis in vivo via negatively regulation of STAT3 [33]. ERBB2IP knockdown promotes Her-2 overexpressing breast cancer cell migration and invasion, and attenuates the therapeutic effect of trastuzumab by inactivating AKT [34]. Whereas, another study shows that ERBB2IP promotes proliferation of breast cancer cells and facilitates the tumorigenesis in transgenic mice by acting as a ERBB2 regulator [27]. Furthermore, ERBB2IP contributes to tumorigenesis and tumor growth in CRC through interacting with c-Cbl [23]. And ERBB2IP overexpression promotes the formation and growth of HCC via inversely modulating ERα signaling [28]. Here, both our data and TCGA data suggested a upregulation of ERBB2IP in HCC. We found that ERBB2IP knockdown resulted in cell growth arrest and apoptosis of HCC cells. Notably, we revealed that ERBB2IP restoration promoted the cell proliferation and reversed apoptosis of miR-23c-overexpressing HCC cells. These data may provide a new insight into miR-23c/ ERBB2IP axis in HCC. In summary, our study revealed that downregulation of miR-23c in HCC tissues correlated with malignant clinical features and poor prognosis of patients. miR-23c inhibited HCC cell proliferation and induced apoptosis by directly targeting ERBB2IP. Abnormal decrease of miR-23c might be an important driver for hepatocarcinogenesis and HCC progression. More importantly, miR-23c/ERBB2IP axis might be

Conflicts of interest All authors declare no conflicts of interest. Acknowledgments This study was supported by grants from the National Natural Science Foundation of China (81572847), Natural Science Basic Research Plan in Shaanxi Province of China (2017JM8002), International Science and Technology Cooperation Program in Shaanxi Province of China (2018KW-062). Appendix A. Supplementary data Supplementary material related to this article can be found, in the online version, at doi:https://doi.org/10.1016/j.biopha.2018.07.155. References [1] A. Diaz-Gonzalez, M. Reig, J. Bruix, Treatment of hepatocellular carcinoma, Dig. Dis. 34 (5) (2016) 597–602. [2] Z.Y. Tang, S.L. Ye, Y.K. Liu, L.X. Qin, H.C. Sun, Q.H. Ye, L. Wang, J. Zhou, S.J. Qiu, Y. Li, X.N. Ji, H. Liu, J.L. Xia, Z.Q. Wu, J. Fan, Z.C. Ma, X.D. Zhou, Z.Y. Lin, K.D. Liu, A decade’s studies on metastasis of hepatocellular carcinoma, J. Cancer Res. Clin. Oncol. 130 (4) (2004) 187–196. [3] J.R. Desai, S. Ochoa, P.A. Prins, A.R. He, Systemic therapy for advanced hepatocellular carcinoma: an update, J. Gastrointest. Oncol. 8 (2) (2017) 243–255. [4] R.S. Finn, A.X. Zhu, W. Farah, J. Almasri, F. Zaiem, L.J. Prokop, M.H. Murad, K. Mohammed, Therapies for advanced stage hepatocellular carcinoma with macrovascular invasion or metastatic disease: a systematic review and meta-analysis, Hepatology 67 (1) (2018) 422–435. [5] G.A. Calin, C.M. Croce, MicroRNA signatures in human cancers, Nat. Rev. Cancer 6 (11) (2006) 857–866. [6] C.M. Croce, G.A. Calin, miRNAs, cancer, and stem cell division, Cell 122 (1) (2005) 6–7. [7] R. Garzon, G.A. Calin, C.M. Croce, MicroRNAs in cancer, Annu. Rev. Med. 60 (2009) 167–179. [8] M.D. Jansson, A.H. Lund, MicroRNA and cancer, Mol. Oncol. 6 (6) (2012) 590–610. [9] Z. Liu, Y. Wang, C. Dou, L. Sun, Q. Li, L. Wang, Q. Xu, W. Yang, Q. Liu, K. Tu, MicroRNA-1468 promotes tumor progression by activating PPAR-gamma-mediated AKT signaling in human hepatocellular carcinoma, J. Exp. Clin. Cancer Res. 37 (1) (2018) 49. [10] Q. Xu, Q. Zhu, Z. Zhou, Y. Wang, X. Liu, G. Yin, X. Tong, K. Tu, MicroRNA-876-5p inhibits epithelial-mesenchymal transition and metastasis of hepatocellular carcinoma by targeting BCL6 corepressor like 1, Biomed. Pharmacother. 103 (2018) 645–652. [11] J. Tao, Z. Liu, Y. Wang, L. Wang, B. Yao, Q. Li, C. Wang, K. Tu, Q. Liu, MiR-542-3p inhibits metastasis and epithelial-mesenchymal transition of hepatocellular carcinoma by targeting UBE3C, Biomed. Pharmacother. 93 (2017) 420–428. [12] Y. Wang, Z. Liu, B. Yao, Q. Li, L. Wang, C. Wang, C. Dou, M. Xu, Q. Liu, K. Tu, Long non-coding RNA CASC2 suppresses epithelial-mesenchymal transition of hepatocellular carcinoma cells through CASC2/miR-367/FBXW7 axis, Mol. Cancer 16 (1) (2017) 123. [13] Q. Xu, X. Liu, Z. Liu, Z. Zhou, Y. Wang, J. Tu, L. Li, H. Bao, L. Yang, K. Tu, MicroRNA-1296 inhibits metastasis and epithelial-mesenchymal transition of hepatocellular carcinoma by targeting SRPK1-mediated PI3K/AKT pathway, Mol. Cancer 16 (1) (2017) 103. [14] K. Tu, Z. Liu, B. Yao, S. Han, W. Yang, MicroRNA-519a promotes tumor growth by targeting PTEN/PI3K/AKT signaling in hepatocellular carcinoma, Int. J. Oncol. 48 (3) (2016) 965–974.

431

Biomedicine & Pharmacotherapy 107 (2018) 424–432

L. Zhang et al.

[24] N. Liu, J. Zhang, S. Liu, Y. Liu, D. Zheng, Erbin-regulated sensitivity of MCF-7 breast cancer cells to TRAIL via ErbB2/AKT/NF-kappaB pathway, J. Biochem. 143 (6) (2008) 793–801. [25] P. Dai, W.C. Xiong, L. Mei, Erbin inhibits RAF activation by disrupting the sur-8Ras-Raf complex, J. Biol. Chem. 281 (2) (2006) 927–933. [26] F. Dai, C. Chang, X. Lin, P. Dai, L. Mei, X.H. Feng, Erbin inhibits transforming growth factor beta signaling through a novel Smad-interacting domain, Mol. Cell. Biol. 27 (17) (2007) 6183–6194. [27] Y. Tao, C. Shen, S. Luo, W. Traore, S. Marchetto, M.J. Santoni, L. Xu, B. Wu, C. Shi, J. Mei, R. Bates, X. Liu, K. Zhao, W.C. Xiong, J.P. Borg, L. Mei, Role of Erbin in ErbB2-dependent breast tumor growth, Proc. Natl. Acad. Sci. U. S. A. 111 (42) (2014) E4429–38. [28] H. Wu, S. Yao, S. Zhang, J.R. Wang, P.D. Guo, X.M. Li, W.J. Gan, L. Mei, T.M. Gao, J.M. Li, Elevated expression of Erbin destabilizes ERalpha protein and promotes tumorigenesis in hepatocellular carcinoma, J. Hepatol. 66 (6) (2017) 1193–1204. [29] K. Tu, W. Yang, C. Li, X. Zheng, Z. Lu, C. Guo, Y. Yao, Q. Liu, Fbxw7 is an independent prognostic marker and induces apoptosis and growth arrest by regulating YAP abundance in hepatocellular carcinoma, Mol. Cancer 13 (2014) 110. [30] K.L. Xie, Y.G. Zhang, J. Liu, Y. Zeng, H. Wu, MicroRNAs associated with HBV infection and HBV-related HCC, Theranostics 4 (12) (2014) 1176–1192. [31] C.M. Wong, F.H. Tsang, I.O. Ng, Non-coding RNAs in hepatocellular carcinoma: molecular functions and pathological implications, Nat. Rev. Gastroenterol. Hepatol. 15 (3) (2018) 137–151. [32] W. Wang, S. Mou, L. Wang, M. Zhang, X. Shao, W. Fang, R. Lu, C. Qi, Z. Fan, Q. Cao, Q. Wang, Y. Fang, Z. Ni, Up-regulation of serum MiR-130b-3p level is associated with renal damage in early lupus nephritis, Sci. Rep. 5 (2015) 12644. [33] Y. Hu, H. Chen, C. Duan, D. Liu, L. Qian, Z. Yang, L. Guo, L. Song, M. Yu, M. Hu, M. Shi, N. Guo, Deficiency of Erbin induces resistance of cervical cancer cells to anoikis in a STAT3-dependent manner, Oncogenesis 2 (2013) e52. [34] D. Liu, M. Shi, C. Duan, H. Chen, Y. Hu, Z. Yang, H. Duan, N. Guo, Downregulation of Erbin in Her2-overexpressing breast cancer cells promotes cell migration and induces trastuzumab resistance, Mol. Immunol. 56 (1–2) (2013) 104–112.

[15] C. Dou, Z. Liu, M. Xu, Y. Jia, Y. Wang, Q. Li, W. Yang, X. Zheng, K. Tu, Q. Liu, miR187-3p inhibits the metastasis and epithelial-mesenchymal transition of hepatocellular carcinoma by targeting S100A4, Cancer Lett. 381 (2) (2016) 380–390. [16] X. Li, L. Zeng, C. Cao, C. Lu, W. Lian, J. Han, X. Zhang, J. Zhang, T. Tang, M. Li, Long noncoding RNA MALAT1 regulates renal tubular epithelial pyroptosis by modulated miR-23c targeting of ELAVL1 in diabetic nephropathy, Exp. Cell Res. 350 (2) (2017) 327–335. [17] S. Yang, L. Li, L. Zhu, C. Zhang, Z. Li, Y. Guo, Y. Nie, Z. Luo, Bu-Shen-Huo-Xue-Fang (BSHXF) modulates nucleus pulposus cell proliferation and extracellular matrix (ECM) remodeling in intervertebral disc degeneration through miR-483 regulation of Wnt pathway, J. Cell. Biochem. (2018), https://doi.org/10.1002/jcb.26760 [Epub ahead of print]. [18] N. Yamazaki, Y. Koga, H. Taniguchi, M. Kojima, Y. Kanemitsu, N. Saito, Y. Matsumura, High expression of miR-181c as a predictive marker of recurrence in stage II colorectal cancer, Oncotarget 8 (4) (2017) 6970–6983. [19] J. Chen, Y. Xia, X. Sui, Q. Peng, T. Zhang, J. Li, J. Zhang, Steviol, a natural product inhibits proliferation of the gastrointestinal cancer cells intensively, Oncotarget 9 (41) (2018) 26299–26308. [20] J.P. Borg, S. Marchetto, A. Le Bivic, V. Ollendorff, F. Jaulin-Bastard, H. Saito, E. Fournier, J. Adelaide, B. Margolis, D. Birnbaum, ERBIN: a basolateral PDZ protein that interacts with the mammalian ERBB2/HER2 receptor, Nat. Cell Biol. 2 (7) (2000) 407–414. [21] H. Huang, Y. Song, Y. Wu, N. Guo, Y. Ma, L. Qian, Erbin loss promotes cancer cell proliferation through feedback activation of Akt-Skp2-p27 signaling, Biochem. Biophys. Res. Commun. 463 (3) (2015) 370–376. [22] C.M. Xie, D. Wei, L. Zhao, S. Marchetto, L. Mei, J.P. Borg, Y. Sun, Erbin is a novel substrate of the Sag-betaTrCP E3 ligase that regulates KrasG12D-induced skin tumorigenesis, J. Cell Biol. 209 (5) (2015) 721–737. [23] S. Yao, P. Zheng, H. Wu, L.M. Song, X.F. Ying, C. Xing, Y. Li, Z.Q. Xiao, X.N. Zhou, T. Shen, L. Chen, Y.H. Liu, M.D. Lai, L. Mei, T.M. Gao, J.M. Li, Erbin interacts with c-Cbl and promotes tumourigenesis and tumour growth in colorectal cancer by preventing c-Cbl-mediated ubiquitination and down-regulation of EGFR, J. Pathol. 236 (1) (2015) 65–77.

432