MicroRNA-1 promotes apoptosis of hepatocarcinoma cells by targeting apoptosis inhibitor-5 (API-5)

FEBS Letters 589 (2015) 68–76

journal homepage: www.FEBSLetters.org

MicroRNA-1 promotes apoptosis of hepatocarcinoma cells by targeting apoptosis inhibitor-5 (API-5) Dong Li a,1, Yu Liu a,1, Hua Li a,1, Jing-Jing Peng a, Yan Tan b, Qiang Zou c, Xiao-Feng Song d, Min Du a, Zheng-Hui Yang e, Yong Tan a, Jin-Jun Zhou a, Tao Xu a, Zeng-Qiang Fu a, Jian-Qiong Feng a, Peng Cheng a, Tao chen a, Dong Wei a, Xiao-Mei Su a, Huan-Yi Liu a, Zhong-Chun Qi a, Li-Jun Tang f, Tao Wang f, Xin Guo g, Yong-He Hu a,⇑, Tao Zhang a,⇑ a

Department of Oncology, Chengdu Military General Hospital, Chengdu 610083, Sichuan Province, China Department of Scientific Research and Training, Chengdu Military General Hospital, Chengdu 610083, Sichuan Province, China c Institute of Aging and Immunity, Chengdu Medical College, Chengdu 610083, Sichuan Province, China d Department of Health Care, Chengdu Military General Hospital, Chengdu 610083, Sichuan Province, China e Department of Neurology, Chengdu Military General Hospital, Chengdu 610083, Sichuan Province, China f Department of General Surgery, Chengdu Military General Hospital, Chengdu 610083, Sichuan Province, China g Department of Medical Lab, Chengdu Military General Hospital, Chengdu 610083, Sichuan Province, China b

a r t i c l e

i n f o

Article history: Received 6 October 2014 Revised 16 November 2014 Accepted 17 November 2014 Available online 27 November 2014 Edited by Tamas Dalmay Keywords: microRNA-1 Apoptosis inhibitor 5 Apoptosis Hepatocellular carcinoma Hepatoma cell

a b s t r a c t Although microRNA-1 (miR-1) is a known liver cancer suppressor, the role of miR-1 in apoptosis of hepatoma cells has remained largely unknown. Our study shows that ectopic miR-1 overexpression induced apoptosis of liver hepatocellular carcinoma (HepG2) cells. Apoptosis inhibitor 5 (API-5) was found to be a potential regulator of miR-1 induced apoptosis, using a bioinformatics approach. Furthermore, an inverse relationship between miR-1 and API-5 expression was observed in human liver cancer tissues and adjacent normal liver tissues. Negative regulation of API-5 expression by miR-1 was demonstrated to promote apoptosis of HepG2 cells. Our study provides a novel regulatory mechanism of miR-1 in the apoptosis of hepatoma cells. Ó 2014 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved.

1. Introduction Hepatocellular carcinoma (HCC) is the fifth most common cancer and the third leading cause of cancer death worldwide [1]. Although the mechanism of cytological pathogenesis of HCC remains unclear, emerging evidence has revealed that microRNAs (miRNAs) play important roles in the initiation, progression, Abbreviations: HCC, hepatocellular carcinoma; API-5, apoptosis inhibitor 5; MET, MET proto-oncogene; HDAC4, histone deacetylase 4; FOXP1, forkhead box P1; CCND2, G1/S-specific cyclin-D2; CXCR4, C-X-C chemokine receptor type 4; SDF-1, stromal cell-derived factor 1 ⇑ Corresponding authors at: Chengdu Military General Hospital, No. 270, Rongdu Avenue, Jinniu District, Chengdu, Sichuan Province 610083, China. Fax: +86 02886570091 (Y.-H. Hu). Department of Oncology, Chengdu Military General Hospital, No. 270, Rongdu Avenue, Jinniu District, Chengdu, Sichuan Province 610083, China. Fax: +86 02886570406 (T. Zhang). E-mail addresses: [email protected] (Y.-H. Hu), [email protected] (T. Zhang). 1 Dong Li, Yu Liu and Hua Li equally contributed to this paper.

recurrence, and metastasis of HCC as well as its resistance to chemotherapy and radiotherapy [2–4]. microRNA-1 (miR-1) has been demonstrated to be a tumor-suppressive miRNA in various cancers including lung cancer, colorectal cancer, prostate cancer, thyroid carcinoma, rhabdomyosarcoma, and head and neck squamous cell carcinoma [5–11]. Some oncogenes, such as MET proto-oncogene (MET), histone deacetylase 4 (HDAC4), forkhead box P1 (FOXP1), G1/S-specific cyclin-D2 (CCND2), C-X-C chemokine receptor type 4 (CXCR4) and stromal cell-derived factor 1 (SDF-1), have been confirmed to be targets of miR-1 that mediate its tumor-suppressive effects [5,8,12]. Recently, it was reported that methylationmediated silencing of the miR-1 gene significantly induced the proliferation of hepatoma cells, whereas ectopic miR-1 overexpression in these cells inhibited cell proliferation, migration and invasion, indicating that miR-1 also exhibits tumor suppressor functionality in HCC [13,14]. In addition, our previous study further indicated that miR-1 inhibited the proliferation of hepatoma cells by targeting a potent mitogen: endothelin-1 (ET-1) [15,16]. Disruption of

http://dx.doi.org/10.1016/j.febslet.2014.11.025 0014-5793/Ó 2014 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved.

D. Li et al. / FEBS Letters 589 (2015) 68–76

the homeostasis between cell proliferation and apoptosis is a key cause of carcinogenesis, and reduced apoptosis leads to the occurrence of HCC and tumor growth. It has been reported that miR-1 can activate apoptosis by targeting prothymosin alpha in nasopharyngeal carcinoma cells [17]. The promotion of apoptosis was also observed following miR-1 transfection into bladder cancer and malignant mesothelioma cells [18,19]. In HCC, Datta et al. found that ectopic expression of miR-1 induced HepG2 cell apoptosis in vitro [13]. However, the mechanisms underlying this phenomenon are still largely unknown. In accordance with previous research, we observed that ectopic miR-1 overexpression induced the apoptosis of hepatoma cells in this study. Moreover, we found that miR-1 suppressed the malignant properties of HCC not only in vitro but also in vivo. Bioinformatics was used to elucidate the mechanism underlying these processes, and the anti-apoptotic factor apoptosis inhibitor-5 (API-5) was found to be a potential target of miR-1 and an important player in miR-1-induced apoptosis. Up-regulation of API-5 and down-regulation of miR-1 were detected in human liver cancer samples compared with adjacent non-tumor liver tissues. Negative regulation of API-5 mRNA and protein levels by miR-1 in hepatoma cells was further demonstrated, and dual-luciferase reporter assays confirmed that miR-1 directly bound to the 30 UTR of API-5 mRNA to silence API-5 expression. Furthermore, ectopic rescue of API-5 partially reversed the effect of miR-1 on HepG2 cells, indicating that miR-1 promoted apoptosis of HepG2 cells by directly targeting API-5. Due to the down-regulation of miR-1 and up-regulation of API-5 in human liver cancer tissues, our findings imply that miR1 plays a role in apoptosis mediated by API-5 in liver cancer cells, reaffirming the important role of miR-1 in the dysregulation of cellular apoptosis that may ultimately lead to tumorigenesis. 2. Materials and methods 2.1. Cell culture and clinical HCC specimen collection The human HCC cell lines HepG2 and Hep3B were obtained from the American Type Culture Collection (ATCC), and the Huh7 cell line was obtained from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China). Genotype validation of the cell lines was performed once a year to ensure that the genotypes remained stable. The HepG2 and Hep3B cells were maintained in Eagle’s minimum essential medium (MEM) (Gibco, Rockville, MD, USA) with 10% fetal bovine serum, penicillin (50 units/ml) and streptomycin (50 lg/ml) at 37 °C with 5% CO2. The Huh7 cells were incubated under the same conditions but cultured in Dulbecco’s minimal essential medium (DMEM) (Gibco, Rockville, MD, USA) containing 10% fetal bovine serum, penicillin (50 unit/ml) and streptomycin (50 lg/ml). Primary tumor and adjacent non-tumor tissues were obtained from 24 HCC patients during surgery at the Chengdu Military General Hospital (Chengdu, China) prior to any therapeutic intervention, such as chemotherapy, radiotherapy or interposition embolism. All the samples were subsequently verified by histology. Informed consent was obtained from all patients, and the study protocol was approved by the Clinical Research Ethics Committee of Chengdu Military General Hospital (Chengdu, China). The specimens were stored at 80 °C until RNA and whole-cell extracts were collected for real-time PCR and western blot analysis, respectively. 2.2. Plasmid construction and RNA synthesis The miRNA expression vector pSuper was obtained from OligoEngine. Sub-cloning of hsa-pre-miR-1 fragment into pSuper to form the expression vector pSuper-miR-1 has been described in previous studies [15,16].

69

The luciferase reporter vector pGL3 (Promega, Madison, WI, USA) was used to generate the luciferase constructs. A 2035-nt region of the wild-type 30 -UTR of the API-5 gene was amplified using the following primers: 50 -TATATCTAGAGGTT GGGGCACACGAGGAAA-30 (forward) and 50 -TATATCTAGAGCATAAGGGGA CATTTACATTTAA-30 (reverse). Then, the PCR products were digested with XbaI, and the fragment was cloned downstream of the open reading frame (ORF) of the luciferase reporter gene. The mutant pGL3-API-5-30 UTR was generated by PCR mutagenesis using the following primers: 50 -AGTGGTGGCAATAATCTCTAAAGA TTAGAAAA GACCATGAGCTGAACCTA-30 (forward) and 50 -TAGGTTCAGCTC ATGGTCTTTTCTAATCTTTAGAGATTATTGCCACCACT-30 (reverse). The eukaryotic expression plasmid pEX-1 (GenePharma, Shanghai, China) was used to construct an API-5 expression plasmid. A fragment of the API-5 gene coding sequence was amplified from cDNA obtained from HepG2 cells using the following primers: 50 -GATATGGCTAGCGCCACCATGCCGACAGT-30 (forward) and 50 -CT CTAGGAATTCTCAGTAGAGTCTTCCCCGACTACGATTTCCTC-30 (reverse). Then, the PCR product was inserted between the NheI and EcoRI sites in the multiple cloning site (MCS) of pEX-1. 20 -O-methyl-modified antisense RNA oligonucleotides for miR-1 were synthesized by Ambion (Ambion, Austin, Texas, USA) and transfected into hepatoma cells using Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA). The sequence of the oligonucleotide was 50 -AUACAUACUUCUUUACAUUCCA-30 . Scrambled oligonucleotide was used as a negative control, and its sequence was 50 -CAGUACUU UUGUGUAGUACAA-30 . 2.3. RNA extraction, real-time PCR and RT-PCR Total RNA was extracted using Trizol reagent according to the manufacturer’s instructions (Invitrogen, Carlsbad, CA, USA). Realtime PCR for miR-1 detection was performed as previously described [15,16]. Real-time PCR and semi-quantitative reverse transcription PCR (RT-PCR) for API-5 mRNA detection were performed using the same strategy. For semi-quantitative RT-PCR, the PCR reaction was terminated at 15 cycles, and the products were electrophoresed on a 2% agarose gel. The primers used for API-5 detection were as follows: 50 -TAGTGGGTTTGGAGAA GTTC30 (forward) and 50 -TCACTTGATAGGCATCTTTATG-30 (reverse). The mRNA levels were normalized to the housekeeping gene b-actin using the following primers: 50 -GTGGCATCCACGAAACTACC-30 (forward) and 50 -AGGGCAGTGAT CTCCT TCTG-30 (reverse). 2.4. Western blot analysis Proteins were extracted from total cell lysates using sodium dodecyl sulfate (SDS) buffer. The proteins were resolved by SDS-PAGE using a 12% polyacrylamide gel and transferred onto polyvinylidene fluoride membranes. The membranes were probed with monoclonal antibodies against API-5 (Santa Cruz Biotechnology, Santa Cruz, CA, USA) and b-actin (Santa Cruz Biotechnology, Santa Cruz, CA, USA). SuperSignal chemiluminescence reagent (Pierce, Rockford, IL, USA) was used to detect API-5 protein. Quantitative analysis was performed using Quantity One software (Bio-Rad, Hercules, CA, USA). 2.5. Bioinformatics analysis and luciferase assay A search for potential miR-1 targets was performed using the following websites for miRNA bioinformatics analysis: TargetScan (http://www.targetscan.org/) Pictar (http://pictar.mdc-berlin.de/) and BiBiserve (http://bibiserv.techfak.uni-bielefeld.de/rnahybrid/). For luciferase assays, HEK 293 cells were plated in 24-well plates and grown to 80–90% confluency. The cells were co-transfected with different amounts of the miR-1 expression vector pSuper-miR-1

70

D. Li et al. / FEBS Letters 589 (2015) 68–76

Fig. 1. miR-1 suppressed the malignant properties of HCC in vitro and in vivo. (A) MTT assay to evaluate HepG2 cell proliferation at 0, 24, 48, 72 and 96 h after pSuper-miR-1 (denoted as miR-1) or pSuper (denoted as NC) transfection. The growth curve of miR-1-transfected cells is represented by black squares, and NC-transfected cells are represented by white triangles. (B) Scratch wound assay to examine cell motility of HepG2 cells transfected with miR-1 and NC for 48 h. (C) Annexin V analysis of apoptotic cells after miR-1 and NC transfection. The lower right quadrant of the flow cytometry figure represents early apoptotic cells. The percentage of early apoptotic HepG2 cells is shown in the histogram on the right. (D) Clonogenic survival assay of HepG2 cells transfected with miR-1 and NC. (E) Tumorigenicity in vivo. The growth of subcutaneous tumors in BALB/c nude mice implanted with miR-1-overexpressing HepG2 cells or NC HepG2 cells after 6 weeks is shown on the left, and the volume and relative weight ratio (tumor weight/total mouse weight) of excised tumors are shown on the right. Data from at least three independent experiments are shown. ‘‘⁄’’ Represents P < 0.05; ‘‘⁄⁄’’ represents P < 0.01.

(0.5 lg, 1.0 lg and 2.0 lg) or equivalent amounts of the control vector pSuper, 100 ng of a luciferase reporter vector containing the wild-type API-5 30 UTR (pGL3-API-5-30 UTR-WT) or mutant API-5 30 UTR (pGL3-API-5-30 UTR-Mut), and 10 ng of the control renilla luciferase vector pRL-TK (Promega, Madison, Wisconsin, USA) using Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA). Luciferase activity was measured with a dual luciferase assay kit (Promega, Madison, Wisconsin, USA) using the M200 microplate fluorescence reader (TecanSafire, Männedorf, Switzerland) 48 h after transfection according to the manufacturer’s instructions. The pRL-TK vector, which constitutively expresses renilla luciferase, was used as an internal control to correct for differences in both transfection and harvest efficiencies. The transfections were

performed in duplicate and repeated at least three times in independent experiments. 2.6. Cell proliferation assay HepG2 cells were plated in 96-well plates at 4  103 cells/well and grown to 80–90% confluency before transfection. A total of 150 ng of pSuper-miR-1 or pSuper was used to transfect the cells in each well. At 24 h after transfection, cell proliferation was assessed using the CellTiter 96Ò AQueous One Solution Cell Proliferation Assay (MTS) (Promega, Madison, WI, USA) according to the manufacturer’s instructions; subsequently, cell proliferation was documented every 24 h for 4 days. Proliferation was assessed by

D. Li et al. / FEBS Letters 589 (2015) 68–76

measuring the absorbance at 450 nm using a microtiter plate reader (Molecular Devices, Sunnyvale, CA, USA). 2.7. Cell motility assay HepG2 cells were plated in 6-well plates and grown to 80–90% confluency before being transfected with 4 lg of pSuper-miR-1 or pSuper and placed in serum-free medium for 24 h. Wounds were generated in confluent HepG2 cells using a pipette tip, and the cells were rinsed with phosphate-buffered saline prior to the addition of fresh culture medium. Wounded areas were marked and photographed at 0 and 48 h using a phase-contrast microscope.

71

2 lg of pSuper-miR-1 and 2 lg of the API-5-expressing vector pEX-1-API-5. All the cells were collected 48 h after transfection, and apoptosis was assessed with the annexin V-APC apoptosis detection kit with propidium iodide (PI) (Biolegend, San Diego, CA, USA) following the manufacturer’s protocol. All cells were washed twice with cell staining buffer and suspended in binding buffer at a density of 3  106 cells/ml, and 5 ll of annexin V-APC and 10 ll of PI solution were added successively to the cell suspension. Then, the reaction was mixed gently and incubated in the dark at 25 °C for 15 min. Subsequently, flow cytometry was performed to examine the percentage of apoptotic cells. The early apoptotic cells were labeled with high annexin V-APC fluorescence and low PI fluorescence.

2.8. Colony-forming assay 2.10. Animals and in vivo xenograft tumorigenicity assay HepG2 cells seeded in 6-well plates were transfected with 4 lg of pSuper-miR-1 or pSuper using Lipofectamine 2000. At 48 h posttransfection, the cells were collected and seeded (3  103/well) in 6-well plates for 14 days. Colonies were then fixed with 70% ethanol, stained with crystal violet solution, and counted. 2.9. FACS assay for apoptosis

Athymic nude mice (BALB/C-nu/nu, 4–6 weeks old, male) were obtained from the Animal Center of Sichuan University (Chengdu, China), housed under specific pathogen-free conditions and cared for in accordance with the guidelines of the laboratory animal center of Chengdu Medical College (Chengdu, China). All studies involving animals were approved by the Research Animal Care and Use Committee of Chengdu Medical College.

HepG2 cells seeded in 24-well plates were transfected with either 2 lg of pSuper-miR-1 or pSuper using Lipofectamine 2000. For the API-5 rescue assay, HepG2 cells were co-transfected with

Fig. 2. Apoptosis inhibitor-5 (API-5) is a putative target of miR-1. (A) Base-pairing of miR-1 and its predicted binding site within the 30 UTR of API-5 mRNA. An eightnucleotide ‘‘seed sequence’’, which is the main determining factor for the interaction between the miRNA and its target gene, was located between nucleotides 1550 and 1557 of the API-5 30 UTR. (B) Secondary structure of the target site duplex consisting of the API-5 30 UTR and miR-1 with the free energy of hybrids (Mef). (C) The 8-bp ‘‘seed sequence’’ for miR-1 in the 30 UTR of API-5 is conserved across species; mRNA pairing is outlined in red, and nucleotides that differ across species are highlighted in blue. Hsa: human, Ptr: chimpanzee, Mmu: mouse, Rno: rat, Cpo: guinea pig, Cfa: dog, Fca: cat, Eca: horse, Bta: cow.

Fig. 3. Analysis of API-5 protein and miR-1 expression in human primary liver cancer and matching healthy liver tissues. (A) Western blot analysis of API-5 protein expression in 24 clinical primary liver cancer tissue samples and adjacent paired non-cancerous liver tissue samples. (B) Quantification and statistical analysis of Western blots was performed using Quantity One software (Bio-Rad). (C) Realtime PCR analysis of miR-1 levels in the 24 human primary liver cancer tissue samples and matching liver tissue samples. T denotes tumor tissue from HCC patients, whereas N represents the paired non-tumorous liver tissues. Data from at least three independent experiments are shown. ‘‘⁄’’ Represents P < 0.05.

72

D. Li et al. / FEBS Letters 589 (2015) 68–76

For the construction of miR-1-expressing lentiviruses, the hsa-pre-miR-1 fragment was cloned into the pWPXLd plasmid (Addgene, Cambridge, MA, USA) between the EcoRI and SpeI sites located downstream of the ORF of the EGFP gene. The pWPXLdmiR-1 plasmid was co-transfected into 293FT cells with the lentiviral packaging plasmids psPAX2 and pLP/VSVG (Invitrogen, Carlsbad, CA, USA), and 48 h later, the lentivirus in the media was collected, filtered, and used to infect HepG2 cells. For the establishment of stable miR-1-overexpressing and control cell lines, HepG2 cells were infected with either the miR-1-expressing lentivirus (Lenti-miR-1) or the control lentivirus (Lenti-NC). Next, flow cytometry sorting was performed to isolate the infected cells (GFP-positive). The cells were maintained in culture medium for 5–7 days before undergoing GFP selection again, and stable clones were obtained after 2 rounds of GFP selection. Overexpression of miR-1 in the stable cell lines was confirmed by real-time PCR. BALB/c nude mice were assigned to 2 groups of 6 mice each. Either HepG2 cells stably overexpressing miR-1 or control HepG2 cells were implanted into the left flanks of the mice at a density of 1.0  106 cells/100 ll in H-DMEM without FBS. The tumor-bearing mice were sacrificed 6 weeks after inoculation, and the tumors were excised and weighed. Tumor volume was calculated using

the formula L  S2/2, where L is the longest tumor diameter and S is the shortest tumor diameter. The weight of tumor was normalized to the total weight of the mouse. 2.11. Statistical analysis Quantitative data were expressed as the mean ± S.D. Differences between groups were determined using a one-way ANOVA followed by a Student–Newman–Keuls test. A P value <0.05 was considered to denote statistical significance. 3. Results 3.1. miR-1 suppressed the malignant properties of HCC in vitro and in vivo To address the anti-tumorigenic function of miR-1 in liver cancer cells, the effects of miR-1 on the malignant properties of HCC in vitro were examined. We transiently transfected HepG2 cells with pSuper-miR-1 to overexpress mature miR-1. Control cells were transfected with a comparable amount of empty vector. Proliferation assays revealed that the cell growth of miR-1-overexpressing

Fig. 4. Ectopic expression of miR-1 suppressed API-5 mRNA and protein expression in hepatoma cells. (A) Real-time PCR analysis of miR-1 levels in HepG2 cells transfected with miR-1 (denoted as miR-1) and control vector (denoted as NC). (B) Real-time PCR (left panel) and semi-quantitative RT-PCR (right panel) analysis of API-5 mRNA levels in HepG2 cells transfected with miR-1 or NC. (C) Western blot analysis of API-5 protein levels in HepG2 cells transfected with miR-1 or NC. (D) Western blot analysis of API-5 protein levels in HepG2 cells transfected with miR-1 antisense oligonucleotide (AS-miR-1) or scrambled oligonucleotide (AS-NC). The housekeeping gene b-actin was used for normalization. Data from at least three independent experiments are shown. ‘‘⁄’’ Represents P < 0.05; ‘‘⁄⁄’’ Represents P < 0.01.

D. Li et al. / FEBS Letters 589 (2015) 68–76

HepG2 cells (measured by absorbance at 450 nm) was significantly reduced by 21% after 48 h (1.134 ± 0.078 vs. 1.436 ± 0.135, P < 0.01) and 17% after 72 h (1.348 ± 0.093 vs. 1.627 ± 0.143, P < 0.05) relative to the control cell growth (Fig. 1A). Flow cytometry assays revealed that miR-1 transfection significantly increased the percentage of apoptotic HepG2 cells (20.80 ± 3.60% vs. 1.93 ± 0.35%, P < 0.01) compared with the control group (Fig. 1C). A scratch wound assay was performed to test cell motility and revealed that control HepG2 cells nearly closed the wound within 48 h, while miR-1-transfected cells did not (Fig. 1B). In colony-forming assays, we found that the clonogenic survival of HepG2 cells was reduced by nearly 60% upon ectopic expression of miR-1 (79 ± 16.7 vs. 188.67 ± 12.5, P < 0.01) (Fig. 1D). To validate the tumor-suppressive effect of miR-1 on HCC in vivo, a HepG2 cell line stably overexpressing miR-1 was established by miR-1 lentivirus infection of HepG2 cells followed by GFP selection (Supplementary Fig. 1A). miR-1 overexpression in this stable cell line was confirmed by real-time PCR (Supplementary Fig. 1B). Stable HepG2 cell lines infected with lentivirus expressing miR-1 (Lenti-miR-1) or control lentivirus (Lenti-NC) were subcutaneously injected into the left flanks of BALB/c nude mice. We observed that xenograft tumors were significantly smaller in the lenti-miR-1 group compared with the control group at 6 weeks post-implantation (Fig. 1E, left). Additionally, Lenti-miR1-infected HepG2 tumors had a significantly reduced volume (47.32 ± 6.64 vs. 172.66 ± 21.03, P < 0.01) and weight (7.31 ± 1.33 vs. 24.89 ± 4.23, P < 0.01) compared with Lenti-NC-infected tumors (Fig. 1E, middle and right), indicating that miR-1 markedly diminished HCC tumorigenicity and displayed a significant antitumor effect in vivo.

73

3.4. miR-1 negatively regulated both the mRNA and protein expression of API-5 in hepatoma cells To confirm that API-5 is indeed a target of miR-1 in HCC cells, we examined endogenous API-5 expression in hepatoma cells expressing ectopic miR-1. Real-time PCR analysis showed that the miR-1 level was increased over 300-fold in HepG2 cells after miR-1 transfection compared with control cells (Fig. 4A). As a result, the mRNA level of API-5 was dramatically reduced by approximately 7-fold in miR-1-overexpressing HepG2 cells compared with control cells. (Fig. 4B, left). Semi-quantitative RT-PCR was used to visualize alterations in the API-5 mRNA level after miR-1 transfection (Fig. 4B, right). Furthermore, western blot analysis also revealed that API5 protein was reduced in a dose-dependent manner with increasing expression of ectopic miR-1 (Fig. 4C). Similarly, reductions in the levels of API-5 protein and mRNA after miR-1 transfection were observed in two additional hepatoma cell lines: Hep3B and Huh7 (Supplementary Fig. 2A and B). In addition, miR-1-overexpressing HepG2 cells that were infected with Lenti-miR-1 also exhibited lower API-5 protein expression compared with control cells that were infected with Lenti-NC. (Supplementary Fig. 1C). On the other hand, when the miR-1 expression was down-regulated by transfection of HepG2 cells with 40 nM or 80 nM antisense miR-1 oligonucleotide (As-miR-1), the level of API-5 protein was increased by

3.2. API-5 is a potential target gene of miR-1 To investigate the underlying mechanism of apoptosis induction by miR-1 in HCC, TargetScan and Pictar software programs were used to identify putative miR-1 targets with anti-apoptotic properties. Among these predicted targets, we were particularly interested in the anti-apoptotic factor API-5. There was only one putative miR-1 binding site in the API-5 30 UTR, which was identified by the TargetScan and Pictar algorithms. The putative basepairing and secondary structure of miR-1 bound to the target site in the API-5 30 UTR are shown in Fig. 2A and B. This miR-1 target site was located between nucleotides 1535 and 1557 of the API-5 30 UTR. An eight-nucleotide sequence (nucleotides 1550–1557) in the binding site was designated as the ‘‘seed sequence’’, and its continuous complementarity to miR-1 was recognized to be crucial for their interaction. The free energy of the hybrid (Mef) was 18.5 kcal/mol. The sequence of the ‘‘seed sequence’’ was highly conserved across species, indicating the importance of miRNA regulation during evolution (Fig. 2C). 3.3. miR-1 and API-5 expression is inversely correlated in human primary HCC tissues We further analyzed the abundance of API-5 protein in 24 human primary HCC tissue samples and paired adjacent normal tissue samples. Western blot assays showed significant up-regulation of API-5 in 19 HCC samples compared with the paired adjacent normal liver tissue samples (Fig. 3A). Statistical analysis confirmed an approximate 3-fold up-regulation of API-5 protein levels in HCC samples compared with matching normal liver tissue samples (Fig. 3B). In contrast, miR-1 expression, as detected by real-time PCR, was significantly reduced (>50%) in liver tumors relative to their matched controls (Fig. 3C). These results indicated that the abundance of API-5 protein and level of miR-1 were inversely related in both HCC tissues and normal liver tissues.

Fig. 5. Luciferase reporter and mutation assay. (A) Schematic representation of the predicted sequence of the miR-1 binding site in the 30 UTR of API-5 (upper panel). The nucleotides in red are nucleotides that have been mutated to abolish miR-1 binding to the API-5 30 UTR (lower panel). (B) Luciferase assay to assess the effect of miR-1 on the expression of a pGL3 luciferase reporter construct carrying the wild-type or mutant 30 UTR of API-5. Wild-type (pGL3-API-5-30 UTR-WT) or mutant (pGL3-API-530 UTR-Mut) reporter constructs were co-transfected with pSuper-miR-1 (miR-1) or pSuper (NC) as shown. Data from at least three independent experiments are shown. ‘‘⁄’’ Represents P < 0.05; ‘‘⁄⁄’’ represents P < 0.01.

74

D. Li et al. / FEBS Letters 589 (2015) 68–76

approximately 50% compared with that in the control cells transfected with the scrambled oligonucleotide (Fig. 4D). Induction of API-5 protein expression by miR-1 knockdown was also observed in Hep3B and Huh7 cell lines (Supplementary Fig. 2C). These results demonstrated that miR-1 could negatively regulate the expression of API-5 mRNA and protein in hepatoma cells. 3.5. The predicted miR-1 binding site in the API-5 30 UTR mediated API5 down- regulation To further investigate the mechanism by which miR-1 regulates API-5 expression, a region (1–2192) of the wild-type API-5 30 UTR containing the putative miR-1 binding site (1535–1557) was cloned into the firefly luciferase reporter vector pGL3 to generate

pGL3-API-5-30 UTR-WT (Fig. 5A, upper). The mutant construct pGL3-API-5-30 UTR-Mut was also obtained by substituting the cytosines at the second, third and seventh nucleotide positions (from the 30 end) of the predicted ‘‘seed sequence’’ with guanine, adenine and guanine, respectively, to abolish the continuous complementarity to miR-1 (Fig. 5A, lower). Luciferase reporter assays showed that ectopic miR-1 transfection significantly reduced the luciferase activity of pGL3- API-5-30 UTR-WT in a dose-dependent manner. In contrast, miR-1 transfection had no significant influence on the luciferase activity of pGL3-API-5-30 UTR-Mut (Fig. 5B, lower). Taken together, these results suggested that miR-1 could inhibit the expression of a chimeric protein harboring the sequence of the miR-1 recognition site in the 30 UTR of API-5 at the post-transcriptional level.

Fig. 6. Rescue of API-5 expression partially reversed miR-1-induced apoptosis in hepatoma cells. (A) Western blot analysis of API-5 protein expression in non-transfected HepG2 cells (Mock), pSuper-transfected HepG2 cells (NC), pSuper-miR-1-transfected HepG2 cells (miR-1) and HepG2 cells cotransfected with pSuper-miR-1 and the API-5expressing vector pEX-1-API-5 (miR-1 + API-5). (B) Percentage of early apoptotic HepG2 cells after the transfections. (C) Flow cytometry analyses of early apoptotic cells dually stained with annexin V and propidium iodide (PI) after the transfections. Data from at least three independent experiments are shown. ‘‘⁄’’ Represents P < 0.05; ‘‘⁄⁄’’ represents P < 0.01.

D. Li et al. / FEBS Letters 589 (2015) 68–76

3.6. API-5 overexpression partially reversed the ability of miR-1 to promote apoptosis in hepatoma cells To elucidate whether miR-1 promotes HepG2 cell apoptosis by targeting API-5, we further performed a rescue assay by co-overexpressing miR-1 and API-5 in HepG2 cells. A eukaryotic plasmid, pEX-1, was used to construct the API-5 expression plasmid pEX-1API-5. Consistent with previous results, ectopic miR-1 transfection led to a significant decrease in API-5 protein expression (Fig. 6A) and a marked increase in the percentage of early apoptotic HepG2 cells compared with the mock-transfected group (20.83 ± 3.60% vs. 2.10 ± 0.26%, P < 0.01) and NC-transfected group (20.83 ± 3.60% vs. 1.93 ± 0.35%, P < 0.01) (Fig. 6B). However, when HepG2 cells were co-transfected with pEX-1-API-5 and pSuper-miR-1, the reduction of API-5 protein level by miR-1 transfection was dramatically restored (Fig. 6A). As a result, the percentage of early apoptotic HepG2 cells was decreased by approximately 50% when both miR-1 and API-5 were overexpressed compared with when miR-1 alone was overexpressed (11.01 ± 2.17% vs. 20.83 ± 3.60%, P < 0.05, Fig. 6B and C). Therefore, the results indicated that miR-1 promoted hepatoma cell apoptosis by directly targeting API-5. 4. Discussion Previous studies have focused attention on the tumor suppressive effect of miR-1 in HCC. It has been found that ectopic expression of miR-1 reduces the growth, replication potential, and clonogenic survival of HepG2 cells while inducing apoptosis and cell cycle arrest of these cells in vitro [13]. Additionally, miR-1 has also been reported to suppress the invasion and migration of HCC [14]. However, the tumor suppressive effects of miR-1 in vivo and the underlying mechanism of the pro-apoptotic effect of miR-1 in HCC have remained largely unknown. In our study, we demonstrated for the first time that miR-1 suppressed tumor growth in vivo. We also found that API-5 is one of the target genes of miR-1 that contributes to the induction of apoptosis in HCC. API-5, also named fibroblast growth factor-2-interacting factor (FIF) and anti-apoptosis clone 11 (AAC-11), is a 55 kD nuclear protein with biological properties that are relevant to cancer [20]. Many studies have demonstrated marked API-5 expression in human cancer cells, including leukemia cells and nasopharyngeal, renal, cervical and colorectal cancer cells [21]. It has also been observed that patients with API-5-expressing non-small cell lung cancer have a poorer prognosis than patients with non-API-5expressing cancers, suggesting that this protein is an independent parameter that can be used to predict outcomes for tumor patients [22]. Functionally, API-5 was found to support cell viability after withdrawal of growth factors in BALB/c3T3 fibroblasts [20]. It also been reported that API-5 overexpression could protect cervical cancer cells from apoptosis [23]. Moreover, API-5 depletion markedly increased cellular sensitivity to anticancer drugs, whereas its expression interfered with drug-induced cell death, revealing that API-5 is an important anti-apoptotic factor for cancer cell survival [24]. In HCC, a recent study demonstrated that API-5 can be phosphorylated and activated by PIM2 kinase and inhibit apoptosis of hepatoma cells through nuclear factor-jB (NF-jB) signaling, indicating that API-5 is also a potent anti-apoptotic factor in HCC [25]. Notably, miRNA has also been found to participate in the regulation of apoptosis by modulating API-5 expression. For example, miR-224 has been reported to negatively regulate API-5 expression, promoting the apoptosis of hepatocarcinoma cells [26]. Our study revealed that miR-1 also inhibits the expression of API-5 to induce apoptosis of hepatoma cells, indicating that a regulatory network of miRNAs may coordinately control API-5 expression in HCC.

75

An important observation that prompted us to investigate the ability of miR-1 to regulate API-5 expression in hepatocarcinoma was that API-5 expression was inversely correlated with miR-1 expression in HCC patients. Although the expression of API-5 was demonstrated to be up-regulated in many cancers such as leukemia and nasopharyngeal, renal, cervical and colorectal cancers [21], there have been few studies exploring its expression in HCC. In this study, we first examined API-5 protein expression in HCC tissues and found that API-5 expression was increased in human liver cancer tissues compared with adjacent non-cancerous liver tissue. This result indicated that increased expression of API-5 could lead to the development of hepatocarcinoma, and API-5 also could be investigated as a new parameter to predict prognosis in HCC patients. A recent study reported that transfection of hepatoma cells with miR-1 resulted in a differentiated phenotype similar to that of hepatocytes; thus, miR-1 may also be involved in the regulation of cellular differentiation in liver cancer cells or liver cancer stem cells [27,28]. Moreover, because API-5 was reported to strengthen the chemoresistance of cancer cells [24], it can be speculated that miR-1 might attenuate the drug resistance of HCC by targeting API5. Further investigation into the role and underlying mechanism of miR-1 in other aspects of hepatocarcinoma such as differentiation and drug resistance will be beneficial for understanding the tumor-suppressive effects of miR-1 in HCC. In conclusion, our study demonstrated that miR-1 promoted the apoptosis of hepatoma cells by directly targeting API-5. This finding adds further insight into the functional roles of miR-1 in HCC. Conflict of interest statement The authors report no conflict of interest. Acknowledgments We acknowledge doctors Tao Chen, Shu Zou, Tie-jun Zhao and Dong-xuan Li of the General Surgery Department of Chengdu Military General Hospital for providing and preparing surgically resected clinical liver cancer tissue samples. This work was supported by the National Natural Science Foundation of China (Grant 81101634), the Scientific Research Foundation of the Science and Technology Department of Sichuan Province (Grants 2012SZ0058 and 2012JY0031), the Scientific Research Project of the Health Department of Sichuan Province (Nos. 110472 and 110478), the Key Project of Medical Science and Technology of Chengdu Military Region during the 12th FiveYear Plan (Nos. B12018 and B14005) and the Scientific Research Project of the Administration of Traditional Chinese Medicine of Sichuan Province (No. 2014K057). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.febslet.2014.11. 025. References [1] Farazi, P.A. and DePinho, R.A. (2006) Hepatocellular carcinoma pathogenesis: from genes to environment. Nat. Rev. Cancer 6, 674–687. [2] Huang, S. and He, X. (2011) The role of microRNAs in liver cancer progression. Br. J. Cancer 104, 235–240. [3] Aravalli, R.N., Steer, C.J. and Cressman, E.N. (2008) Molecular mechanisms of hepatocellular carcinoma. Hepatology 48, 2047–2063. [4] Giordano, S. and Columbano, A. (2013) MicroRNAs: new tools for diagnosis, prognosis, and therapy in hepatocellular carcinoma? Hepatology 57, 840–847.

76

D. Li et al. / FEBS Letters 589 (2015) 68–76

[5] Nohata, N., Hanazawa, T., Enokida, H. and Seki, N. (2012) MicroRNA-1/133a and microRNA-206/133b clusters: dysregulation and functional roles in human cancers. Oncotarget 3, 9–21. [6] Liu, Y.N., Yin, J.J., Abou-Kheir, W., Hynes, P.G., Casey, O.M., Fang, L., Yi, M., Stephens, R.M., Seng, V., Sheppard-Tillman, H., Martin, P. and Kelly, K. (2013) MiR-1 and miR-200 inhibit EMT via Slug-dependent and tumorigenesis via Slug-independent mechanisms. Oncogene 32, 296–306. [7] Reid, J.F., Sokolova, V., Zoni, E., Lampis, A., Pizzamiglio, S., Bertan, C., Zanutto, S., Perrone, F., Camerini, T., Gallino, G., Verderio, P., Leo, E., Pilotti, S., Gariboldi, M. and Pierotti, M.A. (2012) MiRNA profiling in colorectal cancer highlights miR-1 involvement in MET-dependent proliferation. Mol. Cancer Res. 10, 504–515. [8] Yan, D., Dong Xda, E., Chen, X., Wang, L., Lu, C., Wang, J., Qu, J. and Tu, L. (2009) MicroRNA-1/206 targets c-Met and inhibits rhabdomyosarcoma development. J. Biol. Chem. 284, 29596–29604. [9] Nasser, M.W., Datta, J., Nuovo, G., Kutay, H., Motiwala, T., Majumder, S., Wang, B., Suster, S., Jacob, S.T. and Ghoshal, K. (2008) Down-regulation of micro-RNA1 (miR-1) in lung cancer. Suppression of tumorigenic property of lung cancer cells and their sensitization to doxorubicin-induced apoptosis by miR-1. J. Biol. Chem. 283, 33394–33405. [10] Liu, R., Zhang, C., Hu, Z., Li, G., Wang, C., Yang, C., Huang, D., Chen, X., Zhang, H., Zhuang, R., Deng, T., Liu, H., Yin, J., Wang, S., Zen, K., Ba, Y. and Zhang, C.Y. (2011) A five-microRNA signature identified from genome-wide serum microRNA expression profiling serves as a fingerprint for gastric cancer diagnosis. Eur. J. Cancer 47, 784–791. [11] Wang, F., Song, G., Liu, M., Li, X. and Tang, H. (2011) MiRNA-1 targets fibronectin1 and suppresses the migration and invasion of the HEp2 laryngeal squamous carcinoma cell line. FEBS Lett. 585, 3263–3269. [12] Leone, V., D’Angelo, D., Rubio, I., de Freitas, P.M., Federico, A., Colamaio, M., Pallante, P., Medeiros-Neto, G. and Fusco, A. (2011) MiR-1 is a tumor suppressor in thyroid carcinogenesis targeting CCND2, CXCR4, and SDF1alpha. J. Clin. Endocrinol. Metab. 96. E1388-98. [13] Datta, J., Kutay, H., Nasser, M.W., Nuovo, G.J., Wang, B., Majumder, S., Liu, C.G., Volinia, S., Croce, C.M., Schmittgen, T.D., Ghoshal, K. and Jacob, S.T. (2008) Methylation mediated silencing of microRNA-1 gene and its role in hepatocellular carcinogenesis. Cancer Res. 68, 5049–5058. [14] Wei, W., Hu, Z., Fu, H., Tie, Y., Zhang, H., Wu, Y. and Zheng, X. (2012) MicroRNA-1 and microRNA-499 downregulate the expression of the ets1 proto-oncogene in HepG2 cells. Oncol. Rep. 28, 701–706. [15] Li, D., Yang, P., Li, H., Cheng, P., Zhang, L., Wei, D., Su, X., Peng, J., Gao, H., Tan, Y., Zhao, Z., Li, Y., Qi, Z., Rui, Y. and Zhang, T. (2012) MicroRNA-1 inhibits proliferation of hepatocarcinoma cells by targeting endothelin-1. Life Sci. 91, 440–447.

[16] Li, D., He, B., Zhang, H., Shan, S.F., Liang, Q., Yuan, W.J. and Ren, A.J. (2012) The inhibitory effect of miRNA-1 on ET-1 gene expression. FEBS Lett. 586, 1014– 1021. [17] Wu, C.D., Kuo, Y.S., Wu, H.C. and Lin, C.T. (2011) MicroRNA-1 induces apoptosis by targeting prothymosin alpha in nasopharyngeal carcinoma cells. J. Biomed. Sci. 18, 80. [18] Yoshino, H., Chiyomaru, T., Enokida, H., Kawakami, K., Tatarano, S., Nishiyama, K., Nohata, N., Seki, N. and Nakagawa, M. (2011) The tumour-suppressive function of miR-1 and miR-133a targeting TAGLN2 in bladder cancer. Br. J. Cancer 104, 808–818. [19] Xu, Y., Zheng, M., Merritt, R.E., Shrager, J.B., Wakelee, H., Kratzke, R.A. and Hoang, C.D. (2013) MiR-1 induces growth arrest and apoptosis in malignant mesothelioma. Chest 144, 1632–1643. [20] Tewari, M., Yu, M., Ross, B., Dean, C., Giordano, A. and Rubin, R. (1997) AAC-11, a novel cDNA that inhibits apoptosis after growth factor withdrawal. Cancer Res. 57, 4063–4069. [21] Koci, L., Chlebova, K., Hyzdalova, M., Hofmanova, J., Jira, M., Kysela, P., Kozubik, A., Kala, Z. and Krejci, P. (2012) Apoptosis inhibitor 5 (API-5; AAC-11; FIF) is upregulated in human carcinomas in vivo. Oncol. Lett. 3, 913–916. [22] Sasaki, H., Moriyama, S., Yukiue, H., Kobayashi, Y., Nakashima, Y., Kaji, M., Fukai, I., Kiriyama, M., Yamakawa, Y. and Fujii, Y. (2001) Expression of the antiapoptosis gene, AAC-11, as a prognosis marker in non-small cell lung cancer. Lung Cancer 34, 53–57. [23] Kim, J.W., Cho, H.S., Kim, J.H., Hur, S.Y., Kim, T.E., Lee, J.M., Kim, I.K. and Namkoong, S.E. (2000) AAC-11 overexpression induces invasion and protects cervical cancer cells from apoptosis. Lab. Invest. 80, 587–594. [24] Rigou, P., Piddubnyak, V., Faye, A., Rain, J.C., Michel, L., Calvo, F. and Poyet, J.L. (2009) The antiapoptotic protein AAC-11 interacts with and regulates Acinusmediated DNA fragmentation. EMBO J. 28, 1576–1588. [25] Ren, K., Zhang, W., Shi, Y. and Gong, J. (2010) Pim-2 activates API-5 to inhibit the apoptosis of hepatocellular carcinoma cells through NF-kappaB pathway. Pathol. Oncol. Res. 16, 229–237. [26] Wang, Y., Lee, A.T., Ma, J.Z., Wang, J., Ren, J., Yang, Y., Tantoso, E., Li, K.B., Ooi, L.L., Tan, P. and Lee, C.G. (2008) Profiling microRNA expression in hepatocellular carcinoma reveals microRNA-224 up-regulation and apoptosis inhibitor-5 as a microRNA-224-specific target. J. Biol. Chem. 283, 13205–13215. [27] Zhang, X., Zhang, E., Ma, Z., Pei, R., Jiang, M., Schlaak, J.F., Roggendorf, M. and Lu, M. (2011) Modulation of hepatitis B virus replication and hepatocyte differentiation by microRNA-1. Hepatology 53, 1476–1485. [28] Mishra, P.J. and Merlino, G. (2009) MicroRNA reexpression as differentiation therapy in cancer. J. Clin. Investig. 119, 2119–2123.