Lncrna GAS5 Inhibits Cellular Proliferation by Targeting P27kip1

Published OnlineFirst April 10, 2017; DOI: 10.1158/1541-7786.MCR-16-0331

Cell Cycle and Senescence Molecular Cancer Research LncRNA GAS5 Inhibits Cellular Proliferation by Targeting P27Kip1 Gang Luo1,2, Dong Liu1, Chao Huang1, Miao Wang1, Xingyuan Xiao1, Fuqing Zeng1, Liang Wang1, and Guosong Jiang1

Abstract

Recent studies have demonstrated that long noncoding RNAs regulator of cell cycle, was positively regulated by GAS5 and (lncRNA) have important roles in cancer biology, and that the upregulation of GAS5 increased its promoter activity. E2F1, an downregulation of lncRNA growth arrest–specifictranscript5 important transcription factor, was shown to bind directly to (GAS5) has been reported in a variety of human cancers. and activate the P27Kip1 promoter. In addition, GAS5 interacted However, its role in prostate cancer is largely unknown. This with E2F1 and enhanced the binding of E2F1 to the P27Kip1 study aims to investigate the biological role and underlying promoter. Collectively, these findings determine that GAS5 mechanism of GAS5 on proliferation in prostate cancer. The functions as a tumor suppressor in prostate cancer develop- results demonstrate that GAS5 expression is significantly ment and progression via targeting P27Kip1. decreased in prostate cancer cells compared with prostate epithelial cells. Ectopic expression of GAS5 inhibited cell pro- Implications: This study reveals a molecular pathway involving Kip1 liferation and induced a cell-cycle arrest in G0–G1 phase, lncRNA GAS5/E2F1/P27 which regulates cell proliferation whereas GAS5 knockdown promoted the G1–S phase transi- and could be a potential therapeutic target in prostate cancer. tion. Subsequent analysis demonstrated that P27Kip1,aknown Mol Cancer Res; 15(7); 789–99. 2017 AACR.

Introduction identifying potential tumor-suppressor genes enriched during growth arrest (13). Previous studies have revealed that the expres- Prostate cancer is the second most frequently diagnosed cancer sion level of GAS5 is decreased in many kinds of cancers (14–16). in men worldwide, accounting for 15% of all male cancers (1). In Furthermore, overexpression of GAS5 could lead to cell-cycle 2015, there were 220,800 estimated new cases of prostate cancer arrest or cell apoptosis in several human cancers including lung, and 27,540 deaths by prostate cancer, making this disease the bladder, kidney, and other cancers (17–20). A recent study second leading cause of cancer-related death for North American showed that GAS5 gene expression was reduced in prostate cell men (2). Androgen deprivation therapy is an effective initial lines derived from metastases compared with those derived from treatment for prostate cancer, but almost all these patients even- normal prostate or primary prostate cancer (14). In addition, tually develop into castration-resistant prostate cancer, which is a further research revealed that high level of GAS5 promoted basal major cause of death (3–6). In consequence, better understanding apoptosis and enhanced the action of apoptotic stimuli in pros- of the tumorigenesis is essential for the development of diagnostic tate cancer cells (21). Despite these findings, the effects of GAS5 markers and novel effective therapies for prostate cancer patients. on prostate cell proliferation and the underlying mechanism Long noncoding RNAs (LncRNA), defined as transcripts con- remain largely unknown. taining >200 nucleotides without evident protein coding func- In the present study, we assessed the expression level of GAS5 in tion, were once considered to be transcriptional "noise" (7, 8). prostate cancer cells, investigated its effect on cell proliferation in Growing evidence indicates that lncRNAs participate in diverse vitro and in vivo, and explored the underlying mechanism. Our cellular processes, including cell differentiation, proliferation, results demonstrate that GAS5 plays an important role in the and apoptosis (9–12). The growth arrest–specific transcript 5 genesis and development of prostate cancer and may be a poten- (GAS5) was originally isolated from a subtraction cDNA library tial therapeutic target.

1Department of Urology, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China. 2Department of Urology, Materials and Methods The Central Hospital of Wuhan, Tongji Medical College, Huazhong University of Plasmid and RNA interference Science and Technology, Wuhan, China. The GAS5 sequence was synthesized according to the full- G. Luo, D. Liu, and C. Huang contributed equally to this article and should be length GAS5 sequence lacking a poly A tail (based on the GAS5 considered coﬁrst authors. sequence, NR_002578, in NCBI) and then subcloned into GV144 Corresponding Authors: Guosong Jiang, Union Hospital, Tongji Medical College, vector (GV144-GAS5; Genechem Co. Ltd.). Empty vector was Huazhong University of Science and Technology, NO. 1277, Jiefang Road, Wuhan used as a control. The shRNA targeting GAS5 (sh-GAS5) was 430022, Hubei Province, China. Phone: 86-27-8535-1624; Fax: 86-27-8365-1606; designed and synthesized by Genechem. The sequence was: sense, E-mail: [email protected]; and Liang Wang, 50-ccggGGACCAGCTTAATGGTTCTttcaagagaAGAACCATTAAGC- [email protected] TGGTCCtttttg-30; anti-sense, 50-aattcaaaaaGGACCAGCTTAATG- doi: 10.1158/1541-7786.MCR-16-0331 GTTCTtctcttgaaAGAACCATTAAGCTGGTCC-30. It was scrambled 2017 American Association for Cancer Research. to generate a negative control. GV122 vector expressing the shRNA

www.aacrjournals.org 789

Luo et al.

was constructed by Genechem. The siRNA targeting E2F1 (siE2F1) iodide (2 mg/mL, Keygen biotech) for 30 minutes at room and P27Kip1 (siP27) were also obtained from Genechem. si-E2F1 temperature, followed by flow cytometry analysis (FACScan, sequence: 50-AACUCCUCGCAGAUCGUCAUC-30. siP27 sequence: Becton Dickinson). 50-GGAGCAATGCGCAGGAAUATT-30. Tumor xenograft models Cell culture and transfection Male BALB/c-nude mice (4 weeks old) were purchased from The human prostate cancer cell lines PC3 and DU145 as well as Hunan SJA Laboratory Animal Co., Ltd. A total of 5 106 stably benign prostate epithelial cells PNT2C2 were purchased from transfected PC3 cells in 200 mL of sterile PBS were injected American Type Culture Collection. Cells were cultured in RPMI- subcutaneously into the right flanks of the mice. The tumors were 1640 (Gibco) supplemented with 10% FBS (HyClone), 100 units/ measured every 3 days. Three weeks after inoculation, the mice mL penicillin (Gibco), and 100 mg/mL streptomycin (Gibco) at were sacrificed by cervical dislocation, and the tumors were 37 C in a humid atmosphere with 5% CO2. GV144-GAS5, resected, measured, and weighed. The tumor volume was calcu- GV144, sh-GAS5, or sh-NC were transfected into PC3 and DU145 lated according to the following formula: volume ¼ 0.5 W2 L cells using Lipofectamine2000 Transfection Reagent (Invitrogen) (W, width; L, length). Animal care and protocols were approved according to the manufacturer's instructions. Stable cell lines were by the Institutional Animal Care and Use Committee of Huaz- screened by the treatment with G418 (Invitrogen). hong University of Science and Technology.

RNA extraction and RT-PCR analysis Immunohistochemistry analysis Total RNA was isolated from cultured cells with TRIzol reagent The primary antibody used to detect P27Kip1 was purchased according to the manufacturer's protocol (Invitrogen). The iso- from Santa Cruz Biotechnology. Tissue sections were deparaf- lated RNA was reverse transcribed into cDNA using a reverse finized in xylene and rehydrated with ethanol. Then, tissue transcription kit (Takara Bio Inc.). Quantitative real-time PCR sections were incubated with 10% normal goat serum followed was performed using a standard protocol from the SYBR Green with incubation with primary antibody overnight at 4C. After PCR Kit (Toyobo) on the StepOnePlus Real-Time PCR System being washed with PBS, the sections were incubated with (Applied Biosystems). The PCR primers were as follows: biotinylated secondary antibody, followed by further incubation with streptavidin–horseradish peroxidase (HRP) at 37C 50-GAAGGTGAAGGTCGGAGTC-30(GAPDH-forward), for 30 minutes. Color was developed using 3,3-diaminobenzi- 50-GAAGATGGTGATGGGATTTC-30(GAPDH-reverse), dine for 10 minutes, and then the sections were counterstained 50-ACAGGCATTAGACAGAAAGC-30(GAS5-forward), with hematoxylin. 50-TACCCAAGCAAGTCATCCA-30(GAS5-reverse), 0 0 Kip1 5 -AGACGGGGTTAGCGGAGCAA-3 (P27 -forward), Western blot assay 0 0 Kip1 5 -TCTTGGGCGTCTGCTCCACA-3 (P27 - reverse). Transfected cells were resuspended in RIPA lysis buffer fi m The PCR reaction was conducted at 95C for 30 seconds (Thermo Scienti c). Note that 20 g of total protein was followed by 40 cycles of 95C for 5 seconds and 60C for 30 separated by electrophoresis, transferred onto PVDF membranes, and incubated with primary antibodies for GAPDH, seconds. Each sample was analyzed in triplicate, and the relative DD Kip1 quantification of GAS5 expression was calculated using the 2 Ct P27 , P21, cyclin E, cyclin D1, and E2F1 overnight at 4 C. The fi method relative to GAPDH. speci c antibodies were purchased from Cell Signaling Tech- nology. Subsequent to being washed with Tris-buffered saline MTT assay containing Tween-20, the membranes were incubated with the After transfection, prostate cancer cells were seeded at a density specific HRP-conjugated secondary antibodies (Santa Cruz Bio- of 2,500 cells per well in 96-well plates. At the indicated time technology) at room temperature for 1 hour. Specific signals points, 100 mL MTT (0.5 mg/mL, Sigma) was added to each well were detected using the ECL Kit (Beyotime). Data analysis was and cells were incubated for 4 hours at 37C in a humidified performed using ImageJ Software to evaluate the expression chamber, followed by removal of the culture medium and addi- levels of proteins. ImageJ software was used to perform the gray tion of 200 mL DMSO (Sigma). Finally, plates were shaken for 15 analysis of Western blot. minutes for complete dissolution and the absorbance at 570 nm was measured. RNA-immunoprecipitation The RNA-immunoprecipitation (RIP) test was performed with EdU assay the Magna RIP RNA-Binding Protein Immunoprecipitation Kit Transfected PC3 and DU145 cells were cultured in 96-well (Millipore) according to the manufacturer's instructions. GV144- plates and incubated with 50 mmol/L EdU (RiboBio) for 2 hours. GAS5 or empty vector–transfected PC3 cells were lysed in com- Then, cells were fixed with 4% formaldehyde for 30 minutes at plete RIP lysis buffer, and the extract was incubated with magnetic room temperature. After permeabilization with 0.5% Triton-X, beads conjugated with E2F1 antibody or IgG for 6 hours at 4 C. the cells were incubated with Apollo staining solution for 30 Then, the beads were washed and incubated with Proteinase K to minutes. Subsequently, the DNA contents of the cells were stained remove proteins. Finally, purified RNA was subjected to PCR with Hoechst 33342 for 30 minutes. Photographs of the cells were analysis. captured with a fluorescent microscope. Luciferase reporter assay Flow cytometry analysis of cell cycle P27Kip1 promoter was cloned into pGL3-BASIC luciferase Transfected cells were harvested and fixed in 70% ethanol at reporter vector, which was a gift from Dr. Ceshi Chen (22). The 20 C overnight. Fixed cells were washed with PBS and incubated reporter promoters were transiently transfected along with a in 400 mL PBS, 50 mL RNase (1 mg/mL), and 10 mL propidium renilla control plasmid, concomitantly either with GV144,

790 Mol Cancer Res; 15(7) July 2017 Molecular Cancer Research

GAS5 Suppresses Prostate Cancer

Figure 1. The expression levels of GAS5 in prostate cancer cell lines. A, The relative expression levels of GAS5 in PC3 and DU145 cells were determined by qRT-PCR. B and C, GAS5 expression levels were analyzed in PC3 and DU145 cells transfected with GV144, GV144-GAS5, sh-NC, or sh-GAS5. The number of living PC3 (D and E) and DU145 (F and G) cells were determined by MTT assays. The experiments were repeated 3 times, respectively, and results were expressed as mean SD (, P < 0.05).

GV144-GAS5, sh-NC, or sh-GAS5. The cells were collected 40 Results hours after transfection, and luciferase activities were measured GAS5 suppresses cell proliferation in vitro using a dual luciferase reporter assay (Promega). In order to explore the effects of GAS5 in prostate cancer in vitro, ﬁ Chromatin immunoprecipitation we rst measured its expression in prostate cancer cell lines (PC3 The chromatin immunoprecipitation (ChIP) assay was per- and DU145) and prostate epithelial cell line (PNT2C2). Consis- formed according to the manufacturer's protocol using the EZ- tent with previous study (14), our results showed that the expres- ChIP Kit (Upstate Biotechnology). Chromatin DNA was immu- sion of GAS5 was reduced in PC3 and DU145 cells compared with noprecipitated with E2F1 antibody or normal IgG, washed, and that in PNT2C2 cells (Fig. 1A). Next, GV144, GV144-GAS5, sh- the DNA-protein cross-links were subsequently reversed. qRT- NC, or sh-GAS5 was transfected into PC3 and DU145 cells. qRT- PCR was performed by using primers for the p27Kip1 promoter (F: PCR analyses revealed that GAS5 expression was increased 11.5- 50-GGCCTCCCCCGCAGACCAC-30,R:50-GTTCCGCCACCTCC- and 8.7-fold by GV144-GAS5 (Fig. 1B), and reduced by 62% and CCTC GTTCC-30). 60% by sh-GAS5 (Fig. 1C) in PC3 and DU145 cells, respectively, compared with the control. Then, MTT assays were performed and Statistical analysis results showed that the number of living cells was decreased in Data were analyzed using SPSS 17.0 software (SPSS Inc.). All GV144-GAS5 group compared with GV144 group, while it was the experiments were carried out 3 times. Statistically signiﬁcant increased after knockdown of GAS5 (Fig. 1D–G). In addition, EdU differences (P < 0.05) were determined by the Student t test or assays showed that upregulation of GAS5 inhibited proliferation ANOVA, presented as mean SD. of PC3 and DU145 cells, whereas downregulation of GAS5

www.aacrjournals.org Mol Cancer Res; 15(7) July 2017 791

Luo et al.

Figure 2. The effects of GAS5 on prostate cancer cell proliferation. EdU assays were performed to detect the proliferation of PC3 (A) and DU145 (B) cells. Statistical analyses of EdU-positive PC3 (C) and DU145 (D) cells were shown. The experiments were repeated 3 times, respectively, and results were expressed as mean SD (, P < 0.05).

Kip1 promoted cell proliferation (Fig. 2). To further study the inhib- involving in G1–S phase transition: P27 , P21, cyclin D1, and itory effect of GAS5 on cell proliferation, flow cytometry analyses cyclin E (23–26). As shown in Fig. 5A and B, overexpression of of cell cycle were performed and results revealed that overexpres- GAS5 markedly increased the expression of P27Kip1 protein. In sion of GAS5 markedly increased the proportion of cells in G0–G1 contrast, transfection of sh-GAS5 induced downregulation of phase and decreased the proportion in S phase. After knockdown P27Kip1 protein. However, the variation in GAS5 expression did of GAS5, the G0–G1 phase cell ratio declined and the S phase cell not observably affect the expression levels of cyclin E, cyclin D1, or ratio ascended (Fig. 3). P21 protein. Next, qRT-PCR analyses were performed to detect the expression of P27Kip1 mRNA, and results showed that it was GAS5 suppresses cell proliferation in vivo positively regulated by GAS5 (Fig. 5C). Similar results were seen To investigate the effects of GAS5 in vivo,thegrowthratesof in immunohistochemistry analysis, which revealed that the PC3 cells stably transfected with GV144, GV144-GAS5, sh-NC, expression levels of P27Kip1 protein in xenograft tumor tissues or sh-GAS5 were examined after subcutaneous implantation were positively regulated by GAS5 (Fig. 5D and E). To further into BALB/c mice. Consistent with the results in vitro,the confirm this, protein and RNA were extracted from xenograft overexpression of GAS5 resulted in a remarkable retardation tumor tissues, and Western blot assays and qRT-PCR were per- of tumor growth in vivo. Conversely, the tumors formed in the formed. As shown in Fig. 5F–H, the expression of P27Kip1 protein sh-GAS5 group were obviously larger than those derived from and mRNA were positively regulated by GAS5. sh-NC group (Fig. 4). Taken together, these results showed that GAS5 negatively modulated cell growth in vivo. Downregulation of P27Kip1 abolishes the cell-cycle arrest induced by GAS5 GAS5 induces the expression of P27Kip1 In order to verify that GAS5 induces cell-cycle arrest via increas- In order to understand the underlying mechanism, potential ing P27Kip1 expression, siRNA targeting P27Kip1 was structured targets of GAS5 were determined in PC3 and DU145 cells. As and cotransfected with GV144-GAS5 into PC3 and DU145 cells. fl described above, GAS5 inhibited proliferation by inducing G0–G1 Then, ow cytometry analyses of cell cycle were performed and phase arrest, so we detected the expression levels of key regulators results revealed that upregualtion of GAS5 induced G0–G1 phase

792 Mol Cancer Res; 15(7) July 2017 Molecular Cancer Research

GAS5 Suppresses Prostate Cancer

A GV144 GV144-GAS5 sh-NC sh-GAS5 200 400 600 100 200 300 400 500 0 100 200 300 400 500 600 0 100 200 300 400 500 600 0 0 0 20 40 60 80 100 120 0 20 40 60 80 100 120 0 20 40 60 80 100 120 0 20 40 60 80 100 120 PC3

B 80 80 sh-NC GV44 sh-GAS5 GV44-GAS5 60 60

40 40

20 20 % Cell-cycle distribution % Cell-cycle % Cell-cycle distribution % Cell-cycle 0 0 G –G S G –M G0–G1 S G2–M 0 1 2 PC3

C GV144 GV144-GAS5 sh-NC sh-GAS5 800 600 600 400 400 600 800 400 200 2000 400 600 800 200 200 0 0 0 0 20 40 60 80 100 120 0 20 40 60 80 100 120 0 20 40 60 80 100 120 0 20 40 60 80 100 120 DU145

D 80 80 GV144 sh-NC GV144-GAS5 sh-GAS5 60 60

40 40

20 20 % Cell-cycle distribution % Cell-cycle % Cell-cycle distribution % Cell-cycle 0 0 G0–G1 S G2–M G0–G1 S G2–M DU145

Figure 3. The effects of GAS5 on cell cycle. Flow cytometry analyses were performed, and cell-cycle distributions of PC3 (A) and DU145 (C) cells were detected. The histograms were analyzed, and the proportions of PC3 (B) and DU145 (D) cells in G0–G1 phase, S phase, and G2–M phase were shown, respectively. The experiments were repeated 3 times, respectively, and results were expressed as mean SD (, P < 0.05).

www.aacrjournals.org Mol Cancer Res; 15(7) July 2017 793

Luo et al.

Figure 4. The effects of GAS5 in vivo. PC3 cells were subcutaneously injected into nude mice (n ¼ 5 for each group). Tumor growth curve (A and B), volume (C and D), and weight (E and F) were measured. Results were expressed as mean SD for three technical replicates (, P < 0.05).

arrest, but the cell-cycle arrest was abolished by transfection of normalized to that of firefly were significantly increased in the siP27 (Fig. 6). These observations revealed that GAS5 suppressed GV144-GAS5 group compared with the GV144 group, whereas cell proliferation through upregulation of P27Kip1. those were decreased in the sh-GAS5 group compared with the sh- NC group (Fig. 7A). Next, ChIP assays were performed to inves- GAS5 enhances the binding of E2F1 to P27Kip1 promoter and tigate whether GAS5 could enhance the binding of E2F1 to activates P27Kip1 promoter P27Kip1 promoter. As shown in Fig. 7B, the binding of E2F1 to Previous studies reported that GAS5 exerted its functions by P27Kip1 promoter was increased in cells transfected with GV144- regulating E2F1 protein expression (17, 20). In addition, as an GAS5 compared with cells transfected with GV144. Finally, West- important transcription factor, E2F1 could bind directly to and ern blot assays were performed to examine whether GAS5 could activate P27Kip1 promoter (27–29). So we hypothesized that directly regulate the expression of E2F1 protein. Interestingly, GAS5 affected P27Kip1 expression by regulating E2F1 expression results showed that up- or downregulation of GAS5 did and further activating P27Kip1 promoter. To confirm this, lucif- not notably change the expression of E2F1 mRNA or protein erase reporter assays were used to determine the promoter activity (Fig. 7C–E), indicating that GAS5 activated P27Kip1 promoter in of P27Kip1. Results showed that the renilla luciferase activities other ways.

794 Mol Cancer Res; 15(7) July 2017 Molecular Cancer Research

GAS5 Suppresses Prostate Cancer

ABGV144 GV144 GV144 GV144-GAS5sh-NC sh-GAS5 GV144 GV144-GAS5sh-NC sh-GAS5 GV144-GAS5 GV144-GAS5 sh-NC sh-NC Kip1 Kip1 P27 P27 sh-GAS5 1.5 sh-GAS5 1.0 Cyclin E Cyclin E 1.0 Cyclin D1 Cyclin D1 0.5 0.5 P21 P21

GAPDH GAPDH 0.0 0.0 Kip1 P27Kip1 Cyclin E Cyclin D P21 P27 Cyclin E Cyclin D P21 PC3 DU145 PC3 DU145 CD expression/GAPDH Relative protein expression/GAPDH Relative protein

3 3

2 2 mRNA expression mRNA expression 1 1 Kip1 Kip1

0 0

GV144 sh-NC GV144 sh-NC GV144 GV144-GAS5 sh-NC sh-GAS5 Relative P27 sh-GAS5 Relative P27 sh-GAS5 GV144-GAS5 GV144-GAS5 PC3 DU145

E FG 0.8 0.5 0.6 0.4 GV144 GV144-GAS5sh-NC sh-GAS5 0.6 0.4 0.3 Kip1

0.4 P27 expression protein 0.2

Kip1 0.2

Mean density 0.2 Mean density 0.1

0.0 0.0 GAPDH 0.0 GV144 GV144-GAS5 sh-NC sh-GV144

Relative P27 GV144 sh-NC sh-GAS5 H GV144-GAS5

2.5

2.0

1.5 mRNA expression 1.0 Kip1 0.5

0.0

Relative P27 GV144 sh-NC sh-GAS5 GV144-GAS5

Figure 5. GAS5 induces the expression of P27Kip1. A, The protein expression levels of P27Kip1, cyclin E, cyclin D1, and P21 were detected by Western bolt assays. B, Gray analyses of the proteins were shown. C, P27Kip1 mRNA expression levels were measured by qRT-PCR. D, P27Kip1 proteins in xenograft tumor tissues were assessed by IHC assay. E, The mean density of the section was detected by IPP 6.0 software. F and G, The expression levels of P27Kip1 protein in xenograft tumor tissues were determined. H, The expression levels of P27Kip1 mRNA in xenograft tumor tissues were detected. The experiments were repeated 3 times, respectively, and results were expressed as mean SD (, P < 0.05).

GAS5 binds directly to E2F1 and knockdown of E2F1 transfected along with GV144-GAS5 and Western blot assays were diminishes the upregulation of P27Kip1 protein performed to assess the expression of P27Kip1. As shown in Fig. 7H inducing by GAS5 and I, upregulation of GAS5 increased P27Kip1 protein expression Given that binding to proteins including transcription factors level, whereas this effect was reversed by knockdown of E2F1. was an important way for lncRNAs to perform their functions (30, Taken together, these results demonstrated that GAS5 induced 31), it was reasonable to hypothesize that GAS5 enhanced the P27Kip1expression by binding to E2F1. combination of E2F1 and P27Kip1promoter by binding to E2F1, without changing its protein or mRNA expression levels. Then, RIP assays were performed and results showed that GAS5 was Discussion enriched with E2F1 compared with IgG control (Fig. 7F and G). To Prostate cancer is a common malignancy of the male urinary further conﬁrm the role of E2F1 in the process, siE2F1 was system (32), so ﬁnding new molecular targets for its diagnosis and

www.aacrjournals.org Mol Cancer Res; 15(7) July 2017 795

Luo et al.

A GV144 GV144-GAS5 GV144-GAS5+siNC GV144-GAS5+siP27 200 400 600 0 200 400 600 0 200 400 600 0 0 200 400 600 0 20 40 60 80 100 120 0 20 40 60 80 100 120 0 20 40 60 80 100 120 0 20 40 60 80 100 120 PC3

B GV144 GV144-GAS5 GV144-GAS5+siNC GV144-GAS5+siP27 0 200 400 600 0 200 400 600 0 200 400 600 0 200 400 600 0 20 40 60 80 100 120 0 20 40 60 80 100 120 0 20 40 60 80 100 120 0 20 40 60 80 100 120 DU145

CD GV144 GV144 GV144-GAS5 GV144-GAS5 100 100 GV144-GAS5+siNC GV144-GAS5+siNC 80 GV144-GAS5+siP27 80 GV144-GAS5+siP27

60 60

40 40

20 20

% Cell-cycle distribution % Cell-cycle 0 distribution % Cell-cycle 0 1 S 1 S –M –M –G 2 –G 2 G 0 G G 0 G PC3 DU145

Figure 6. Knockdown of P27Kip1 abolishes the cell-cycle arrest induced by GAS5. Flow cytometry analyses were performed, and cell-cycle distributions of PC3 (A) and DU145

(B) cells were detected. The histograms were analyzed and the proportions of PC3 (C) and DU145 (D) cells in G0–G1 phase, S phase, and G2–M phase were shown, respectively. The experiments were repeated 3 times, respectively, and results were expressed as mean SD (, P < 0.05).

treatment has the potential to improve the clinical strategies and and upregulation of MEG3 inhibits intrinsic cell survival pathway outcomes of this disease. A mass of evidence has revealed that and induces cell-cycle arrest in G0–G1 phase (35). Prostate cancer lncRNAs are functional in carcinogenesis through regulatory antigen 3 (PCA3) is highly expressed in prostatic tumors and mechanisms such as promoter looping, alternative splicing, anti- described as a novel biomarker of prostate cancer (36, 37). sense gene silencing, transcriptional regulation, and DNA repair, The association between lncRNA GAS5 and the progression of thus potentially serving as tumor markers (33). To date, many various cancers has been demonstrated in previous studies. lncRNAs have been identiﬁed, and their involvement in prostate Mourtada-Maarabouni and colleagues found that GAS5 played cancer has been reported. Differentiation Antagonizing Non- an essential role in normal growth arrest in both T-cell lines and Protein Coding RNA (DANCR) is highly expressed in prostate nontransformed lymphocytes and controlled apoptosis in breast cancer tissues and promotes cancer invasion and metastasis cancer (14, 38). Cao and colleagues demonstrated that GAS5 through repressing the expression of TIMP2/3 (34). Maternally regulated cell-cycle progression and inhibited bladder cancer expressed gene 3 (MEG3) is decreased in prostate cancer tissues, cell proliferation, partially via regulation of CCL1 expression

796 Mol Cancer Res; 15(7) July 2017 Molecular Cancer Research

GAS5 Suppresses Prostate Cancer

A BC 3 4 30 40 1.5 1.5 3 2 30 20 1.0 1.0 2 20 1 10 1 10 0.5 0.5 Relative to input (%) Relative 0 0 to input (%) Relative 0 0 0.0 0.0 IgG GV144IgG GV144-GAS5 IgG GV144IgG GV144-GAS5 Luciferase activity (relative level) activity (relative Luciferase level) activity (relative Luciferase

GV144 sh-NC GV144 sh-NC sh-GAS5 sh-GAS5 PC3 DU145 E2F1 mRNA expression Relative GV144 sh-NC E2F1 mRNA expression Relative GV144 sh-NC sh-GAS5 sh-GAS5 GV144-GAS5PC3 GV144-GAS5DU145 GV144-GAS5 GV144-GAS5 D E PC3 DU145 1.5 1.5 F GV144 GV144-GAS5sh-NC sh-GAS5 GV144 GV144-GAS5sh-NC sh-GAS5

1.0 1.0 Input IgG E2F1 Input IgG E2F1 E2F1 E2F1

0.5 0.5 GAS5 GAS5 GAPDH GAPDH 0.0 0.0 PC3 DU145 PC3 DU145 sh-NC GV144 sh-NC GV144 Relative protein expression/GAPDH Relative sh-GAS5 protein expression/GAPDH Relative sh-GAS5 GV144-GAS5PC3 GV144-GAS5DU145 GH I

80 80 1.0 1.0

60 60 GV144 GV144-GAS5GV144-GAS5GV144-GAS5 + siNC + siE2F1 GV144 GV144-GAS5 GV144-GAS5GV144-GAS5 + siNC + siE2F1 40 expression/GAPDH 0.5 expression/GAPDH 0.5 40 Kip1

Kip1 Kip1 Kip1 P27 P27 20 20 0.0 0.0 Relative to input (%) Relative Relative to input (%) Relative 0 0 GAPDH GAPDH IgG E2F1 IgG E2F1 Relative P27 Relative GV144 P27 Relative GV144 PC3 DU145 PC3 DU145 GV144-GAS5GV144+siNC GV144-GAS5GV144+siNC GV144+siE2F1 GV144+siE2F1 PC3 DU145

Figure 7. GAS5 activates P27Kip1 promoter via E2F1. A, The P27Kip1 promoter activities were detected by luciferase reporter assay. B, ChIP assays were performed to investigate the binding of E2F1 to P27Kip1 promoter. C, The expression levels of E2F1 mRNA were detected by qRT-PCR. D and E, The expression levels of E2F1 protein were detected and analyzed. F and G, RIP assays were performed to investigate the combination of GAS5 and E2F1. H and I, The P27Kip1 protein levels were examined after knockdown of E2F1. The experiments were repeated 3 times, respectively, and results were expressed as mean SD (, P < 0.05).

(39). Sun and colleagues showed that ectopic expression of estimated that lncRNA NONHSAG023333 could interact with GAS5 decreased gastric cancer cell proliferation and induced E2F1 to regulate its transcription (46). Unlike earlier reports apoptosis partly via regulating E2F1 and P21 expression (20). In which showed that GAS5 exerted its effects via regulating E2F1 addition, recent research verified that GAS5 could inhibit autop- expression (17, 20), our study revealed that GAS5 enhanced the hagy in NSCLC cells (40). Despite these findings, the role of GAS5 activity of E2F1 by binding directly to it, without changing its in prostate cancer remains largely unknown. In this study, we mRNA or protein expression. It is suggested that P53 is involved in found that GAS5 induced G0–G1 arrest and inhibited cell prolif- downregulation of E2F1 expression by GAS5 (17). However, P53 eration in prostate cancer cells. However, whether GAS5 could is deleted in PC3 cells and mutated in DU145 cells (47). Thus, loss regulate other biological behavior of prostate cancer cells, such as of P53 function is probably the reason that GAS5 is unable to migration and invasion, is still unknown and more research is regulate E2F1 expression in PC3 and DU145 prostate cancer cells. needed to confirm it. E2F1 belongs to E2F family, whose members show structural Up to now, studies have found that GAS5 exerted its effects homology with each other and can be divided into two categories: through several mechanisms. GAS5 could compete with DNA transcriptional activator (E2F1–3) and repressors (E2F4–8). E2F glucocorticoid response element (GRE) for binding to the DNA- subunits collectively play crucial roles in cellular proliferation, binding domain of the glucocorticoid receptor (GR), blocked differentiation and metabolism, and so on (48). This brings us glucocorticoid–GR–GRE interaction and then induced targeted some new thinking: whether GAS5 can interact with other E2F gene transcription, such as cIAP2 and SGK1 (41). GAS5 directly members and whether GAS5 can regulate other biological beha- targeted miRNA, like miR-21 and miR-222, resulting in the viors of prostate cancer cells, like differentiation and metabolism? change of the downstream signaling molecules (42–44). In addi- We will get into those aspects in our future work. tion, GAS5 was found to interact with transcriptional activator Y- In conclusions, we demonstrate that GAS5 inhibits prostate box binding protein 1 (YBX1). GAS5 knockdown accelerated cancer cell proliferation and acts as a tumor suppressor. Our study YBX1 protein turnover without affecting YBX1 transcription, reveals a novel regulatory pathway that GAS5 can bind directly to which decreased YBX1-transactivated p21 expression and abol- E2F1 and activate P27Kip1 promoter. Our findings enriched our ished G1 phase cell-cycle arrest (16). In our study, we proved that knowledge on the versatile function and mechanism of GAS5. GAS5 could bind directly to transcription factor E2F1, enhance the binding of E2F1 to P27Kip1 promoter, and then activate P27Kip1 Disclosure of Potential Conflicts of Interest promoter. Those results improve our understanding of lncRNA No potential conflicts of interest were disclosed. GAS5 in prostate cancer. Kip1 P27 is a member of the KIP/CIP family of cyclin/cyclin- Authors' Contributions dependent kinase (Cdk) inhibitors and regulates cell-cycle pro- fi Conception and design: G. Luo, D. Liu, M. Wang, L. Wang, G. Jiang gression at the G1 to S phase transition (45). It is veri ed that Development of methodology: G. Luo, D. Liu, C. Huang, M. Wang, G. Jiang Kip1 transcription factor E2F1 could induce P27 expression Acquisition of data (provided animals, acquired and managed patients, through activating its promoter (28, 29). Moreover, it has been provided facilities, etc.): G. Luo, M. Wang, X. Xiao

www.aacrjournals.org Mol Cancer Res; 15(7) July 2017 797

Luo et al.

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, The costs of publication of this article were defrayed in part by the computational analysis): D. Liu, C. Huang, M. Wang payment of page charges. This article must therefore be hereby marked Writing, review, and/or revision of the manuscript: G. Luo, M. Wang, F. Zeng, advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate G. Jiang this fact. Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): G. Luo, D. Liu, C. Huang, M. Wang Received September 29, 2016; revised November 1, 2016; accepted April 5, Study supervision: D. Liu, F. Zeng, G. Jiang 2017; published OnlineFirst April 10, 2017.

References 1. Torre LA, Bray F, Siegel RL, Ferlay J, Lortet-Tieulent J, Jemal A. Global cancer 23. Chu IM, Hengst L, Slingerland JM. The Cdk inhibitor p27 in human cancer: statistics, 2012. CA A Cancer J Clin 2015;65:87–108. Prognostic potential and relevance to anticancer therapy. Nat Rev Cancer 2. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2015. CA Cancer J Clin 2008;8:253–67. 2015;65:5–29. 24. Bertoli C, Skotheim JM, de Bruin RA. Control of cell cycle tran- 3. Mostaghel EA, Lin DW. Practical guide to the use of abiraterone in scription during G1 and S phases. Nat Rev Mol Cell Biol 2013; castration resistant prostate cancer. Can J Urol 2014;21:57–63. 14:518–28. 4. Sakamoto S, Ichikawa T. [Current status of castration resistant prostate 25. Pestell RG. New roles of cyclin D1. Am J Pathol 2013;183:3–9. cancer basic research]. Nihon Rinsho Japanese Journal of Clinical Medicine 26. Siu KT, Rosner MR, Minella AC. An integrated view of cyclin E function and 2014;72:2097–102. regulation. Cell Cycle 2012;11:57–64. 5. Attar RM, Takimoto CH, Gottardis MM. Castration-resistant prostate 27. Gopinath S, Alapati K, Malla RR, Gondi CS, Mohanam S, Dinh DH, et al. cancer: locking up the molecular escape routes. Clin Cancer Res 2009; Mechanism of p27 upregulation induced by downregulation of cathepsin 15:3251–5. B and uPAR in glioma. Mol Oncol 2011;5:426–37. 6.ChenCD,WelsbieDS,TranC,BaekSH,ChenR,VessellaR,etal. 28. Wang C, Hou X, Mohapatra S, Ma Y, Cress WD, Pledger WJ, et al. Activation Molecular determinants of resistance to antiandrogen therapy. Nat Med of p27Kip1 expression by E2F1. A negative feedback mechanism. J Biol 2004;10:33–9. Chem 2005;280:12339–43. 7. Struhl K. Transcriptional noise and the fidelity of initiation by RNA 29. Hodul PJ, Dong Y, Husain K, Pimiento JM, Chen J, Zhang A, et al. Vitamin E polymerase II. Nat Struct Mol Biol 2007;14:103–5. delta-tocotrienol induces p27(Kip1)-dependent cell-cycle arrest in pancre- 8. Wang KC, Chang HY. Molecular mechanisms of long noncoding RNAs. atic cancer cells via an E2F-1-dependent mechanism. PLoS One 2013;8: Mol Cell 2011;43:904–14. e52526. 9. Gupta RA, Shah N, Wang KC, Kim J, Horlings HM, Wong DJ, et al. Long 30. Bartonicek N, Maag JL, Dinger ME. Long noncoding RNAs in cancer: non-coding RNA HOTAIR reprograms chromatin state to promote cancer Mechanisms of action and technological advancements. Mol Cancer 2016; metastasis. Nature 2010;464:1071–6. 15:43. 10. Gutschner T, Diederichs S. The hallmarks of cancer: A long non-coding 31. Fatica A, Bozzoni I. Long non-coding RNAs: New players in cell differen- RNA point of view. RNA Biol 2012;9:703–19. tiation and development. Nat Rev Genet 2014;15:7–21. 11. Guttman M, Donaghey J, Carey BW, Garber M, Grenier JK, Munson G, et al. 32. Hsing AW, Tsao L, Devesa SS. International trends and patterns of prostate lincRNAs act in the circuitry controlling pluripotency and differentiation. cancer incidence and mortality. Int J Cancer 2000;85:60–7. Nature 2011;477:295–300. 33. Salameh A, Lee AK, Cardo-Vila M, Nunes DN, Efstathiou E, Staquicini FI, 12. Ernst C, Morton CC. Identification and function of long non-coding RNA. et al. PRUNE2 is a human prostate cancer suppressor regulated by the Front Cell Neurosci 2013;7:168. intronic long noncoding RNA PCA3. Proc Natl Acad Sci U S A 2015; 13. Schneider C, King RM, Philipson L. Genes specifically expressed at growth 112:8403–8. arrest of mammalian cells. Cell 1988;54:787–93. 34. Jia J, Li F, Tang XS, Xu S, Gao Y, Shi Q, et al. Long noncoding RNA DANCR 14. Mourtada-Maarabouni M, Pickard MR, Hedge VL, Farzaneh F, promotes invasion of prostate cancer through epigenetically silencing Williams GT. GAS5, a non-protein-coding RNA, controls apo- expression of TIMP2/3. Oncotarget 2016;7:37868–81. ptosis and is downregulated in breast cancer. Oncogene 2009; 35. Luo G, Wang M, Wu X, Tao D, Xiao X, Wang L, et al. Long non-coding RNA 28:195–208. MEG3 inhibits cell proliferation and induces apoptosis in prostate cancer. 15. Lu X, Fang Y, Wang Z, Xie J, Zhan Q, Deng X, et al. Downregulation of gas5 Cell Physiol Biochem 2015;37:2209–20. increases pancreatic cancer cell proliferation by regulating CDK6. Cell 36. Feibus AH, Sartor O, Moparty K, Chagin K, Kattan MW, Ledet E, et al. Tissue Res 2013;354:891–6. Clinical utility of PCA3 and TMPRSS2:ERG urinary biomarkers in 16. Liu Y, Zhao J, Zhang W, Gan J, Hu C, Huang G, et al. lncRNA GAS5 enhances African American men undergoing prostate biopsy. J Urol 2016; G1 cell cycle arrest via binding to YBX1 to regulate p21 expression in 196:1053–60. stomach cancer. Sci Rep 2015;5:10159. 37. Cui Y, Cao W, Li Q, Shen H, Liu C, Deng J, et al. Evaluation of prostate 17. Shi X, Sun M, Liu H, Yao Y, Kong R, Chen F, et al. A critical role for the long cancer antigen 3 for detecting prostate cancer: A systematic review and non-coding RNA GAS5 in proliferation and apoptosis in non-small-cell meta-analysis. Sci Rep 2016;6:25776. lung cancer. Mol Carcinog 2015;54Suppl 1:E1–E12. 38. Mourtada-Maarabouni M, Hedge VL, Kirkham L, Farzaneh F, Williams 18. Liu Z, Wang W, Jiang J, Bao E, Xu D, Zeng Y, et al. Downregulation of GAS5 GT. Growth arrest in human T-cells is controlled by the non-coding promotes bladder cancer cell proliferation, partly by regulating CDK6. RNA growth-arrest-specific transcript 5 (GAS5). J Cell Sci 2008;121: PLoS One 2013;8:e73991. 939–46. 19. Qiao HP, Gao WS, Huo JX, Yang ZS. Long non-coding RNA GAS5 functions 39. CaoQ,WangN,QiJ,GuZ,ShenH.LongnoncodingRNAGAS5actsasa as a tumor suppressor in renal cell carcinoma. Asian Pac J Cancer Prev tumor suppressor in bladder transitional cell carcinoma via regulation 2013;14:1077–82. of chemokine (CC motif) ligand 1 expression. Mol Med Rep 2016;13: 20. Sun M, Jin FY, Xia R, Kong R, Li JH, Xu TP, et al. Decreased expression of 27–34. long noncoding RNA GAS5 indicates a poor prognosis and promotes cell 40. Zhang N, Yang GQ, Shao XM, Wei L. GAS5 modulated autophagy is a proliferation in gastric cancer. BMC Cancer 2014;14:319. mechanism modulating cisplatin sensitivity in NSCLC cells. Eur Rev Med 21. Pickard MR, Mourtada-Maarabouni M, Williams GT. Long non-coding Pharmacol Sci 2016;20:2271–7. RNA GAS5 regulates apoptosis in prostate cancer cell lines. Biochim 41. Kino T, Hurt DE, Ichijo T, Nader N, Chrousos GP. Noncoding RNA gas5 is a Biophys Acta 2013;1832:1613–23. growth arrest- and starvation-associated repressor of the glucocorticoid 22. Wang C, Nie Z, Zhou Z, Zhang H, Liu R, Wu J, et al. The receptor. Sci Signal 2010;3:ra8. interplay between TEAD4 and KLF5 promotes breast cancer par- 42. Zhang Z, Zhu Z, Watabe K, Zhang X, Bai C, Xu M, et al. Negative tially through inhibiting the transcription of p27Kip1. Oncotarget regulation of lncRNA GAS5 by miR-21. Cell Death Differ 2013; 2015;6:17685–97. 20:1558–68.

798 Mol Cancer Res; 15(7) July 2017 Molecular Cancer Research

GAS5 Suppresses Prostate Cancer

43. Yu F, Zheng J, Mao Y, Dong P, Lu Z, Li G, et al. Long non-coding RNA 46.HeDX,ZhangGY,GuXT,MaoAQ,LuCX,JinJ,etal.Genome- growth arrest-specific transcript 5 (GAS5) inhibits liver fibrogenesis wide profiling of long non-coding RNA expression patterns in through a mechanism of competing endogenous RNA. J Biol Chem anthracycline-resistant breast cancer cells. Int J Oncol 2016;49: 2015;290:28286–98. 1695–703. 44. Zhao X, Wang P, Liu J, Zheng J, Liu Y, Chen J, et al. Gas5 exerts tumor- 47. Isaacs WB, Carter BS, Ewing CM. Wild-type p53 suppresses growth of suppressive functions in human glioma cells by targeting miR-222. Mol human prostate cancer cells containing mutant p53 alleles. Cancer Res Ther 2015;23:1899–911. 1991;51:4716–20. 45. Marinoni I, Pellegata NS. p27kip1: A new multiple endocrine neoplasia 48. Poppy Roworth A, Ghari F, La Thangue NB. To live or let die - complexity gene? Neuroendocrinology 2011;93:19–28. within the E2F1 pathway. Mol Cell Oncol 2015;2:e970480.

www.aacrjournals.org Mol Cancer Res; 15(7) July 2017 799

LncRNA GAS5 Inhibits Cellular Proliferation by Targeting P27Kip1

Gang Luo, Dong Liu, Chao Huang, et al.

Mol Cancer Res 2017;15:789-799. Published OnlineFirst April 10, 2017.

Updated version Access the most recent version of this article at: doi:10.1158/1541-7786.MCR-16-0331

Cited articles This article cites 48 articles, 7 of which you can access for free at: http://mcr.aacrjournals.org/content/15/7/789.full#ref-list-1

Citing articles This article has been cited by 3 HighWire-hosted articles. Access the articles at: http://mcr.aacrjournals.org/content/15/7/789.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at Subscriptions [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://mcr.aacrjournals.org/content/15/7/789. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.