Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

LncRNA FOXP4-AS1 is activated by PAX5 and promotes the growth of prostate cancer by sequestering miR-3184-5p to upregulate FOXP4

LncRNA FOXP4-AS1 is activated by PAX5 and promotes the growth of prostate cancer by sequestering... Prostate cancer (PCa) is one of the major men malignancies worldwide. Long noncoding RNAs (lncRNAs) have been reported as essential regulators in human cancers, including PCa. In the present study, lncRNA forkhead box P4 antisense RNA 1 (FOXP4-AS1) was found to be highly expressed in TCGA PCa samples. Upregulation of FOXP4-AS1 was further validated in 64 PCa tissues and predicted poor prognosis in patients with PCa. Functionally, high FOXP4-AS1 level was associated with increased cell proliferation and decreased cell apoptosis, indicating that FOXP4-AS1 exerted oncogenic functions in the tumorigenesis of PCa. Furthermore, FOXP4-AS1 was located in the cytoplasm of PCa cell lines and positively regulated FOXP4. LncRNAs can exert their functions by cooperating with their nearby genes. Mechanistically, FOXP4-AS1 post-transcriptionally regulated FOXP4 by acting as a competing endogenous RNA (ceRNA) in PCa to sponge miR-3184-5p. Considering the upregulation of both FOXP4-AS1 and its nearby gene FOXP4, we further detected the coactivator of FOXP4-AS1 and FOXP4. Mechanism analysis indicated that paired box 5 (PAX5) transcriptionally activated FOXP4-AS1 and FOXP4 in PCa. Collectively, we determined that PAX5-induced upregulation of FOXP4-AS1/FOXP4 axis promoted tumorigenesis of PCa. Introduction Recent years, long noncoding RNAs (lncRNAs) that are Prostate cancer (PCa) is a commonly diagnosed malig- mainly transcribed from the genomic intergenic regions nant male cancer and is one of the prime reasons for have become a research focus in cancers transcriptome . cancer-related death . Lacking of the biomarker in early They have been demonstrated to be participants in var- 3,4 diagnosis resulted in low overall survival of patients with ious biological processes by different mechanisms . The PCa. Although numerous efforts have been made in oncogenic or tumor-suppressive role of lncRNAs has been 5–7 exploring the novel therapeutic strategies, the prognosis well characterized in tumorigenesis . To date, some of patients with advanced PCa remains unfavorable. lncRNAs have been defined to be PCa specific, such as 8 9 10 11 Exploring the molecular mechanism associated with the PCGEM1 , PRNCR1 , PCAT1 , and SChLAP1 . tumorigenesis of PCa is of great significance for the early Nevertheless, investigating the function and mechanism diagnosis or treatment of PCa. of novel lncRNAs is necessary to find functional bio- markers in PCa progression. Searching from online database TCGA, we found top 30 upregulated lncRNAs in PCa samples. Among these candidates, lncRNA forkhead Correspondence: Weigang Yan ([email protected]) Department of Urology, Peking Union Medical College Hospital, Chinese box P4 antisense RNA 1 (FOXP4-AS1) has been reported 12 13 Academy of Medical Sciences, Beijing, China in osteosarcoma and colorectal cancer . However, it is Department of Pathology, Peking Union Medical College Hospital, Chinese unclear whether FOXP4-AS1 exerted similar function in Academy of Medical Sciences, Beijing, China Edited by A. Stephanou © The Author(s) 2019 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to theCreativeCommons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/. Official journal of the Cell Death Differentiation Association 1234567890():,; 1234567890():,; 1234567890():,; 1234567890():,; Wu et al. Cell Death and Disease (2019) 10:472 Page 2 of 14 PCa tumorigenesis. Thus, we conducted in vitro and the Declaration of Helsinki. According to the median of in vivo experiments to determine the function of FOXP4- the expression level of FOXP4-AS1, miR-3184-5p or AS1 in PCa growth. FOXP4, 64 PCa tissues were classified into high- To detect the potential mechanism pattern of lncRNA expression group (n = 32) and low-expression group FOXP4-AS1 in PCa, we detected the cellular localization (n = 32). of FOXP4-AS1 in PCa cells and determined that FOXP4- Human PCa cell line (PC-3, DU145, VCaP, and LNCaP) and human normal prostate epithelial cell line (RWPE-1) AS1 was predominantly located in the cytoplasm of PCa cells, indicating that FOXP4-AS1 may regulate gene were bought from the American Type Culture Collection expression at post-transcriptional level. In term of post- (Rockville, MD, USA). Cell culture was conducted in transcriptional regulation, lncRNAs have been acknowl- Dulbecco’s Modified Eagle Medium containing 10% fetal edged to be competing endogenous RNAs (ceRNAs) by bovine serum (HyClone, Logan, UT, USA), 1% penicillin- competitively binding to miRNA response elements to streptomycin (HyClone) at 37 °C with 5% CO . 14–16 upregulate mRNAs . On this basis, we carried out bioinformatics analysis and mechanism experiments to Quantitative real-time PCR determine downstream genes of FOXP4-AS1. It was Total RNA was extracted from tissue samples using found that FOXP4-AS1 can regulate its nearby gene TRIzol reagent (Thermo Fisher Scientific, Waltham, MA, FOXP4 by sequestering miR-3184-5p. USA) following the user guide. RNA was converted into Upregulation of genes in human cancers may be cDNA using random primers and Reverse Transcription 17–19 induced by transcription activation . Here, we found Kit (Toyobo, Osaka, Japan). Nanodrop 1000 spectro- several transcription factors in the common transcrip- photometer (Thermo Fisher Scientific, Waltham, MA, tional region of FOXP4-AS1 and FOXP4. Further USA) and 1.5% denaturing agarose gels were used to mechanism investigation was made to validate the effect measure RNA concentration and purity. The primers are of paired box 5 (PAX5) on the transcription activation of shown in Supplementary Table 1. To evaluate the both FOXP4-AS1 and FOXP4. Collectively, the focus of expression of lncRNA or mRNA, qRT-PCR was con- this study is to explore the mechanism of PAX5-induced ducted using SYBR Green PCR Master Mix (Takara, FOXP4-AS1/FOXP4 axis in PCa tumorigenesis. Kyoto, Japan) with Roche Real-Time PCR system (Roche, USA). Relative expression was normalized to GAPDH or −ΔΔCt Materials and methods U6 and was determined by 2 method. Bioinformatics analysis Top 30 upregulated lncRNAs in PCa samples and the Cell transfection expression profile of FOXP4-AS1 in PCa samples were For cell transfection, LNCaP and PC-3 cell lines at acquired from TCGA dataset (http://gepia.cancer-pku.cn/ 70–80% confluence were planted in six-well plates. For index.html). UCSC (http://genome.ucsc.edu/) was applied overexpression of genes, pcDNA3.1 vectors (Genechem to search the nearby gene of FOXP4-AS1. DIANA tools Company, Shanghai, China) were subcloned with FOXP4- (http://carolina.imis.athena-innovation.gr/diana_tools/ AS1, FOXP4, or PAX5 (pcDNA-FOXP4-AS1, pcDNA- web/index.php?r=site%2Ftools) and the starbase database FOXP4, or pcDNA-PAX5). Empty pcDNA3.1 vector was (http://starbase.sysu.edu.cn/) were used to predict the used as negative control. The short hairpin RNAs miRNAs that had complementary base paring with both (shRNAs) against FOXP4-AS1, FOXP4, and PAX5 (sh- FOXP4-AS1 and FOXP4. The DNA motif of PAX5 and FOXP4-AS1#1/2/3, sh-FOXP4#1/2/3, and sh-PAX5#1/2/ five putative binding sites of PAX5 in FOXP4-AS1 or 3) were designed and synthesized by GenePharma FOXP4 promoter were obtained from JASPAR (http:// (Shanghai, China). To overexpress miR-3184-5p, miR- jaspar.genereg.net/). 3184-5p mimics and miR-NC (GenePharma) were trans- fected into PCa cell lines. Whereas miR-3184-5p inhibitor Patient specimens and cell culture and its negative control anti-miR-NC (GenePharma) were Sixty-four matched PCa and adjacent nontumor pros- transfected into PCa cells for silencing of miR-3184-5p. tate tissues were collected from patients with PCa who Similarly, STAT5A, EBF1, RAD21, and RUNX3 were received treatment at The Fourth Hospital of Harbin overexpressed by subcloning the whole sequence of them Medical University from May 2012 to June 2017. After into pcDNA3.1 vectors (Genechem). shRNAs targeting resection, all tissues were snap-frozen in liquid nitrogen STAT5A, EBF1, RAD21, and RUNX3 were synthesized by and stored at −80 °C. The written informed consents were Genechem were used for knockdown of them. Lipo- provided by all the participants. Our study was approved fectamine 2000 (Invitrogen, Carlsbad, CA, USA) was used by the Research Ethics Committee of the The Fourth for transfection following the standard method. After Hospital of Harbin Medical University and complied with 48 h, cells were reaped for subsequent experiments. Official journal of the Cell Death Differentiation Association Wu et al. Cell Death and Disease (2019) 10:472 Page 3 of 14 Cell proliferation assays In vivo experiment and immunohistochemical staining A total of 5 × 10 transfected LNCaP and PC-3 cell lines Male BALB/C athymic nude mice (aged 6 weeks) were were harvested and put into 96-well plates. Cell viability bought from the National Laboratory Animal Center was analyzed using 10 μL of Cell Counting Kit-8 reagent (Beijing, China) and preserved in an SPF-grade pathogen- (CCK-8, DOJINDO, Kumamoto, Japan). The absorbance free animal laboratory. Animal experiment was conducted at 450 nm was evaluated by a microplate reader when the strictly in light with the protocol approved by the Animal reagent reacted for 24, 48, 72 and 96 h at room tem- Research Ethics Committee of The Fourth Hospital of perature. All experimental procedures were conducted in Harbin Medical University and the principles of the triplicate. Declaration of Helsinki. A total of 1 × 10 transfected For colony formation assay, LNCaP and PC-3 cell lines LNCaP and PC-3 cell lines were injected subcutaneously were plated in six-well plates at 500 cells per well for around the left flank of nude mice (three mice per group). 2 weeks. After fixation in 4% paraformaldehyde for Tumor volumes were calculated with the following 10 min, cells were subjected to 0.1% crystal violet for equation: 0.5 × length × width every 4 days. Four weeks 30 min. Finally, colonies were counted and recorded. later, mice were killed. Tumors were excised, weighed, Experiments were repeated at least three times. and snap-frozen at −80 °C for subsequent analysis. For EdU incorporation assay, LNCaP and PC-3 cell lines on sterile coverslips were seeded in 24-well plates. Cell Ki-67 staining proliferation was detected by the EdU incorporation assay The excised tumor tissue specimens were put in for- kit (Ribobio, Guangzhou, China). The stained cells were malin, fixed and paraffin-embedded. After deparaffinating visualized by laser confocal microscopy (FV300, Olympus, and rehydrating, tissues were incubated with antibody Tokyo, Japan). Nucleus was double-stained with EdU and against Ki-67 (Santa Cruz Biotechnology, Dallas, TX, 4′,6-diamidino-2-phenylindole (DAPI, Beyotime, Shang- USA) at 4 °C overnight. Tissues were stained with dia- hai, China) as positively proliferative cells. All experi- minobenzidine and observed under light microscopy. ments were repeated in triplicate. Western blotting JC-1 fluorescence analysis Protein samples from tissues or cells were acquired Cell apoptosis was assessed by examine the mitochon- using RIPA lysis buffer (Beyotime, Beijing, China) and drial transmembrane potential (ΔΨ). LNCaP and PC-3 separated on 10% SDS-PAGE. After transferring onto cell lines were seeded in 96-well plates (1 × 10 PVDF membranes, proteins were blocked with 5% solu- cells/well) and incubated overnight. Afterward, cells were subjected tion of nonfat milk for 2 h. Subsequently, samples were to Cis and/or mdivi-1, with or without 2 mM NAC or treated with primary antibodies against FOXP4 0.5 mM TEMPOL in serum-free medium for 18 h. The (ab251688, Abcam, Cambridge, UK) and GAPDH cationic, lipophilic dye 5,5′,6,′-tetrachloro-1,1′,3,3′ tetra- (ab8245, Abcam) along with the corresponding secondary ethylbenzimidazolyl carbocyanine iodide (JC-1) was antibodies conjugated to horseradish peroxidase. The commercially obtained from Cayman Chemical (Ann signals of membranes were measured by ECL Substrate Arbor, MI, USA). After centrifugation, cells were loaded (Pierce, Rockford, IL). GAPDH was used as an internal in JC-1 dye for half an hour, following washing and reference. incubation in assay buffer. ΔΨ was detected by a fluor- escent plate reader. A fluorescence microscope was used Subcellular fractionation assay to capture Gemini XPS and images. Each experimental The subcellular fractionation assay was carried out by procedure was repeated at least three times the use of PARIS Kit (Invitrogen, USA) following the supplier’s suggestion. LNCaP and PC-3 cell lines were Caspase-3 activity analysis washed in prechilled PBS for three times and lysed in cell To detect caspase-3 activity, an EnzCheck Caspase-3 fractionation buffer. The supernatant was collected. The Assay kit#1 was purchased from Molecular Probes cell lysates were placed in cell fractionation buffer and (Eugene, OR, USA). The rinsed cells were lysed in cell centrifuged. Cell disruption buffer was added to lyse cell lysis buffer on ice for 10 min, followed by centrifuga- nuclei. The isolated RNA was assessed by quantitative tion. The supernatant was transferred into microplate real-time PCR. GAPDH and U6 were used as cytoplasmic wells and subjected to Z-DEVD-AMC as caspase-3 or nuclear fractionation indicators. substrate. Gemini XPS fluorescent plate reader was applied to measure fluorescent signals at a wavelength Fluorescence in situ hybridization (FISH) of 405 nm. Each apoptotic assay was performed at least After fixation with 4% formaldehyde and washing with three times. PBS, PC-3 cells were treat with Pepsin (1% in 10 mmol/l Official journal of the Cell Death Differentiation Association Wu et al. Cell Death and Disease (2019) 10:472 Page 4 of 14 HCl) and was dehydrated by ethanol. Then, dried cells Chromatin immunoprecipitation assay (ChIP) were mixed in a hybridization buffer by using 40 nmol/l of The ChIP assay was performed by the use of Sim- the FISH probe (FOXP4-AS1 lncRNA) and incubated at pleChIP Enzymatic Chromatin IP Kit (CST, Danvers, 80 °C for 120 s. After being stayed at 55 °C for 120 min, MA, USA) following the supplier’s protocol. The cross- the slides were washed and dehydrated again. Finally, the linked chromatin DNA was broken to 200 to 1000 bp slides were detected by Prolong Gold Antifade Reagent through enzymatic digestion. Then, the chromatin was with DAPI. Fluorescence-conjugated FOXP4-AS1 probes immunoprecipitated with antibodies against PAX5 and were designed and synthesized by Ribobio Company IgG. Precipitated chromatin DNA was recovered by (Guangzhou, China). magnetic beads and analyzed by quantitative real-time PCR. All samples were prepared in triplicate. RNA immunoprecipitation assay (RIP) For Ago 2-RIP assay, the Magna RIP RNA-Binding Statistical analysis Protein Immunoprecipitation Kit was purchased from All experimental data were presented as the mean ± Millipore (Billerica, MA, USA) in line with the user guide. standard deviation of three or more repetitive experi- LNCaP and PC-3 cell lines were washed in precold PBS ments. SPSS 13.0 (SPSS, Inc., Chicago, IL, USA) and and lysed in RIP buffer at 4 °C for half an hour. Cell lysates GraphPad PRISM 6 (GraphPad, San Diego, CA, USA) were treated with magnetic beads conjugated to anti- were used for statistical analysis. Kaplan–Meier analysis bodies against Ago 2 (Millipore) or normal mouse was utilized to estimate the overall survival rate of patients immunoglobulin G (IgG; Millipore). Immunoprecipitated with PCa. Correlation was analyzed with Pearson’s cor- RNA was extracted for quantitative real-time PCR. For relation method. Differences between groups were ana- MS2-RIP assay, cells treated with pMS2-GFP were lyzed by Student's t-test or one-way ANOVA. Differences cotransfected with pcDNA3.1-MS2 and pcDNA3.1- were considered to be significant when P < 0.05. FOXP4-AS1-MS2 for 48 h. Thereafter, cells were incu- bated with GFP antibody (Roche Diagnostics GmbH, Results Mannheim, Germany) and the Magna RIP RNA-Binding Upregulation of FOXP4-AS1 in PCa samples is correlated Protein Immunoprecipitation Kit in accordance with the with unfavorable patients’ prognosis guidebook for user. After collecting and lysing, cell sus- Based on the TCGA database, top 30 upregulated pension was cultured with magnetic beads. The antibody lncRNAs in PCa samples were searched out and illu- was added and incubated overnight. After RNA purifica- strated in Fig. 1a. Among which, FOXP4-AS1 has been tion, the isolated RNA was detected by quantitative real- reported as an oncogene in colorectal cancer and osteo- time PCR to quantify the presence of the binding targets. sarcoma. However, it is unknown that whether FOXP4- AS1 exerted function in PCa. The expression profile of Luciferase reporter assay FOXP4-AS1 in TCGA PCa samples were detected and The amplified FOXP4-AS1 or 3′-UTR of FOXP4 was shown (Fig. 1b). Moreover, the expression level of subcloned into the firefly plasmids in pmirGLO luciferase FOXP4-AS1 was examined in PCa samples and adjacent vector (GeneChem, Shanghai, China). The wild type of samples collected from 64 patients with PCa. As expected, FOXP4-AS1 or FOXP4 3′-UTR (FOXP4-AS1-WT or FOXP4-AS1 was expressed at a higher level in PCa FOXP4-WT) was constructed. The site-directed mutation samples compared to adjacent normal samples (Fig. 1c). of miR-3184-5p binding sites in FOXP4-AS1 or FOXP4 To analyze the prognostic potential of FOXP4-AS1 in 3′-UTR (FOXP4-AS1-MUT or FOXP4-MUT) was gen- patients with PCa, Kaplan–Meier surviving curves were erated using the GeneTailor™ Site-Directed Mutagenesis generated. After analysis, we determined that high level of System (Invitrogen, Carlsbad, CA, USA). LNCaP and PC- FOXP4-AS1 was closely associated with the low overall 3 cell lines were separately cotransfected with aforemen- survival rate of patients with PCa (Fig. 1d). Furthermore, tioned plasmids and miR-3184-5p mimics or miR-NC. At the relative higher level of FOXP4-AS1 was observed in last, a Dual-Luciferase Reporter Assay System (Promega) patients in advanced stage rather than those in early stage was used to evaluate luciferase activity. For FOXP4-AS1 (Fig. 1e). Consistently, the expression level of FOXP4-AS1 or FOXP4 promoter luciferase assay, cells were placed was found to be higher in PCa cell lines compared to into 24-well plates and cotransfected with the plasmids human normal prostate epithelial cell RWPE-1 (Fig. 1f). containing the putative binding sites of PAX5 to FOXP4- These data indicated that FOXP4-AS1 might be a parti- AS1 or FOXP4 promoter. All these plasmids were cipant in tumorigenesis of PCa. simultaneously subcloned into pGL3 vector (Promega Corporation, Fitchburg, WI, USA) and established. Each FOXP4-AS1 facilitated PCa cell growth in vitro procedure of these experiments was repeated indepen- To explore the role of FOXP4-AS1 in PCa cellular dently for three times. function, we designed functional assays in two different Official journal of the Cell Death Differentiation Association Wu et al. Cell Death and Disease (2019) 10:472 Page 5 of 14 Fig. 1 Upregulation of FOXP4-AS1 in PCa samples is correlated with unfavorable patients’ prognosis. a Thirty lncRNAs that are upregulated in TCGA PCa samples were listed. b FOXP4-AS1 expression prolife in PCa and normal tissues was obtained from the TCGA database. c The expression level of FOXP4-AS1 was examined in PCa samples and adjacent samples collected from 64 patients with PCa. d Overall survival of 64 patients with PCa with high or low FOXP4-AS1 level was analyzed. e The expression level of FOXP4-AS1 in patients with PCa with early or advanced tumor stage. f Expression level of FOXP4-AS1 in PCa cell lines and one human normal prostate epithelial cell RWPE-1. *P < 0.05, **P < 0.01 PCa cell lines. According to the data in Fig. 1g, LNCaP cell addition, we tested caspase-3 activity in two PCa cells that exhibited the relative lowest expression level of FOXP4- were separately transfected with FOXP4-AS1 expression AS1, while PC-3 cell presented the highest level of vector or sh-FOXP4-AS1#1.Itwas foundthatFOXP4-AS1 FOXP4-AS1 expression. Thus, we overexpressed FOXP4- negatively regulated caspase-3 activity in PCa cells (Fig. 2e). AS1 in LNCaP cell but silenced it in PC-3 cell (Supple- Therefore, we confirmed that ectopic expression of FOXP4- mentary Fig. 1A). At first, we performed the CCK-8 assay AS1 affected PCa proliferation and apoptosis. to detect the viability of cells with high or low level of FOXP4-AS1. The results indicated that overexpression of FOXP4-AS1 promoted PCa tumor growth in vivo FOXP4-AS1 efficiently enhanced cell viability, whereas Furtherly, animal study was carried out to demonstrate downregulation of FOXP4-AS1 led to the decreased cell the role of FOXP4-AS1 in regulating PCa growth. LNCaP viability (Fig. 2a). Cell viability was suppressed more cell stably transfected with pcDNA-FOXP4-AS1 or empty efficient when cells were transfected with sh-FOXP4- vector and PC-3 cell stably transfected with sh-FOXP4- AS1#1. Therefore, sh-FOXP4-AS1#1 was chosen for AS1#1 or sh-NC were separately injected into nude mice. subsequent experiments. Through colony formation and After 28 days, tumors were removed and calculated. As EdU assays, we determined that overexpression of presented in Fig. 3a, tumors derived from cell transfected FOXP4-AS1 markedly promoted cell proliferation, while with pcDNA-FOXP4-AS1 were bigger than those derived knockdown of FOXP4-AS1 led to an opposite result from cell transfected with empty vector. In addition, (Fig. 2b, c). Then, we explored whether FOXP4-AS1 tumors in sh-FOXP4-AS1#1 group were smaller than regulated cell proliferation by inducing cell apoptosis. those in sh-NC group (Fig. 3a). Consistently. Same ten- According to JC-1 staining, we supposed that high expres- dencies were observed in tumor volume and tumor weight sion level of FOXP4-AS1 correlated with increased mem- (Fig. 3b, c). Finally, immuohistochemical staining indi- brane potential, indicating the inhibitory effect of FOXP4- cated that ki-67 positivity in FOXP4-AS1 overexpression AS1 on the early apoptosis of PCa cells (Fig. 2d). In group was significantly higher in NC group. However, sh- Official journal of the Cell Death Differentiation Association Wu et al. Cell Death and Disease (2019) 10:472 Page 6 of 14 Fig. 2 FOXP4-AS1 facilitated PCa cell growth in vitro. a The CCK-8 assay was used to detect cell viability after overexpression or knockdown of FOXP4-AS1. b, c Proliferative ability of LNCaP cell transfected with FOXP4-AS1 expression vector or PC-3 cell transfected with sh-FOXP4-AS1 was assessed by the colony formation assay and EdU assay. Scale bar for EdU staining: 100 μm. d Membrane potential in FOXP4-AS1-overexpressed or -downregulated PCa cells was measured by JC-1 staining. Scale bar = 200 μm. e Caspase-3 activity was tested in two PCa cells that were separately transfected with FOXP4-AS1 expression vector or sh-FOXP4-AS1#1. *P < 0.05, **P < 0.01 FOXP4-AS1#1 group exhibited low ki-67 positivity com- was upregulated in PCa samples (Fig. 4c), which was pared to sh-NC group (Fig. 3d). consistent with FOXP4-AS1 (Fig. 4d). Similarly, we made survival analysis to determine whether FOXP4 was cor- FOXP4 was positively regulated by FOXP4-AS1 and exerted related with patients’ survival. It was found that patients oncogenic property in PCa in FOXP4 high-expression group had lower overall sur- Using online bioinformatics analysis tool UCSC, we vival rate than those in FOXP4 low-expression group found that FOXP4 is a nearby gene of FOXP4-AS1 (Fig. (Fig. 4e). The relative high level of FOXP4 was detected in 4a). FOXP4 has been reported in human cancers due to its PCa cells compared with the normal control cell (Fig. 4f). oncogenic property. Through qRT-PCR and western To investigate the potential of FOXP4 in regulating PCa blotting analyses, we found that FOXP4-AS1 could posi- cell growth, we overexpressed or silenced it in indicated tively regulate both mRNA and protein level of FOXP4 in PCa cells by transfecting with FOXP4 expression vector or PCa cells (Fig. 4b). Furthermore, we found that FOXP4 specific shRNA (Supplementary Fig. 1B). Then, CCK-8 Official journal of the Cell Death Differentiation Association Wu et al. Cell Death and Disease (2019) 10:472 Page 7 of 14 Fig. 3 FOXP4-AS1 promoted PCa tumor growth in vivo. a Tumors derived from FOXP4-AS1-overexpressed LNCaP cell or FOXP4-AS1- downregulated PC-3 cell was resected and shown. b, c Tumor volume and tumor weight in different groups were quantified and illustrated. d Immuohistochemical staining of ki-67 positivity in tumors of different groups. Scale bar = 50 μm. **P < 0.01 and caspase-3 activity test were used to evaluate the (Fig. 5c). Based on bioinformatics analysis, we searched out importance of FOXP4 in PCa cell proliferation and five miRNAs that had complementary base paring with apoptosis. Interestingly, overexpression of FOXP4 pro- both FOXP4-AS1 and FOXP4 (Fig. 5d). Subsequently, the moted cell proliferation, while knockdown of FOXP4 had MS2-RIP assay was carried out to demonstrate the binding inhibitory effect on cell proliferation (Fig. 4g). In addition, of these five potential miRNAs to FOXP4-AS1. As a result, the caspase-3 activity was decreased in LNCaP cell miR-3184-5p and miR-423-5p showed the strongest affi- transfected with FOXP4 expression vector but was nity to FOXP4-AS1-MS2 beads (Fig. 5e). Then, we increased in PC-3 cell transfected with sh-FOXP4#1 examined the expression level of these two miRNAs in (Fig. 4h). These data showed that FOXP4 might be cells transfected with FOXP4-AS1 expression vector or cooperated with FOXP4-AS1 to exert function in PCa. shRNAs. The results showed that miR-3184-5p was effi- ciently downregulated by overexpressing FOXP4-AS1 but FOXP4-AS1 acted as the molecular sponge of miR-3184-5p was upregulated by silencing FOXP4-AS1 (Fig. 5f). The LncRNAs can exert functions by transcriptionally or predicted binding sequence between FOXP4-AS1 and post-transcriptionally activating their downstream genes. miR-3184-5p was shown (Fig. 5g). To change the expres- To investigate the regulatory pattern of FOXP4-AS1 in sion level of miR-3184-5p in PCa cells, miR-3184-5p PCa, we detected its cellular fractionation in two PCa cells inhibitor and miR-3184-5p mimics were separately trans- through the subcellular fractionation assay and FISH assay. fected into LNCaP and PC-3 cells. Luciferase activity Both experimental results indicated that FOXP4-AS1 was analysis revealed the effect of miR-3184-5p mimics on the predominantly located in the cytoplasm of PCa cells luciferase activity of reporter containing wild type FOXP4- (Fig. 5a, b). Therefore, we identified the post- AS1 (FOXP4-AS1-WT) or mutant type FOXP4-AS1 transcriptional regulation of FOXP4-AS1 in PCa. CeRNA (FOXP4-AS1-MUT). The results suggested that miR- network is a common post-transcriptional regulatory 3184-5p mimics efficiently decreased the luciferase activ- pattern of lncRNAs. Combining with above data, we ity of FOXP4-AS1-WT vector (Fig. 5h). All these experi- hypothesized that FOXP4-AS1 might upregulate FOXP4 mental results indicated that FOXP4-AS1 might act as a by sequestering a miRNA. Next, we performed the Ago 2- ceRNA to regulate miR-3184-5p and FOXP4. RIP assay in two PCa cells. The results showed that both FOXP4-AS1 and FOXP4 were enriched in Ago 2 con- FOXP4 is the target of miR-3184-5p in PCa taining beads, suggesting the probability of the involve- Likewise, we performed experiments to demonstrate ment of FOXP4-AS1 and FOXP4 in ceRNA network the interaction between miR-3184-5p and FOXP4. At Official journal of the Cell Death Differentiation Association Wu et al. Cell Death and Disease (2019) 10:472 Page 8 of 14 Fig. 4 FOXP4 was positively regulated by FOXP4-AS1 and exerted oncogenic property in PCa. a UCSC data revealed that FOXP4 is a nearby gene of FOXP4-AS1. b The mRNA and protein level of FOXP4 in cells transfected with pcDNA-FOXP4-AS1 or sh-FOXP4-AS1 were measured by qRT- PCR and western blotting. c Upregulation of FOXP4 in PCa samples. d Positive expression association between FOXP4-AS1 and FOXP4 in PCa samples was analyzed by Pearson correlation analysis. e Kaplan–Meier analysis of patients with PCa with high or low expression level of FOXP4. f Relative high level of FOXP4 was detected in PCa cells and one normal control cell. g Cell viability was examined in response to the overexpression or knockdown of FOXP4 by the CCK-8 assay. h Caspase-3 activity was increased in LNCaP cell transfected with FOXP4 expression vector but was increased in PC-3 cell transfected with sh-FOXP4#1. **P < 0.01 Official journal of the Cell Death Differentiation Association Wu et al. Cell Death and Disease (2019) 10:472 Page 9 of 14 Fig. 5 FOXP4-AS1 acted as the molecular sponge of miR-3184-5p. a, b FOXP4-AS1 was predominantly located in the cytoplasm of PCa cells. Scale bar = 200 μm. c Ago 2-RIP revealed that both FOXP4-AS1 and FOXP4 were enriched in Ago 2 containing beads. d Five miRNAs that had complementary base paring with both FOXP4-AS1 and FOXP4 were predicted from starBase and DIANA. e Enrichment of five miRNAs in MS2 with or without FOXP4-AS1. f miR-3184-5p and miR-423-5p were detected in cells transfected with FOXP4-AS1 expression vector or sh-FOXP4-AS1#1. g The predicted binding sequence between FOXP4-AS1 and miR-3184-5p. h The luciferase reporter assay was conducted in cells transfected with miR- 3184-5p mimics or miR-NC to examine the luciferase activity of FOXP4-AS1-WT and FOXP4-AS1-MUT. *P < 0.05, **P < 0.01 first, the binding sequence of miR-3184-5p to FOXP4 3′ correlation between miR-3184-5p expression and the UTR was predicted and obtained (Fig. 6a). Then, we overall survival rate of patients with PCa (Fig. 6e). examined the luciferase activity of reporter containing Compared with normal cell line, PCa cell lines exhib- wild-type FOXP4 (FOXP4-WT) or mutant-type FOXP4 ited relative low expression level of miR-3184-5p (Fig. (FOXP4-MUT) in cells transfected with miR-3184-5p 6f).Functionally,wefound that inhibition of miR-3184- mimics. The results suggested that the luciferase 5p had a positive effect on cell proliferation but had a activity of FOXP4-WT vector but not FOXP4-MUT negative effect on cell apoptosis. On the other hand, vector was decreased by miR-3184-5p mimics (Fig. 6b). upregulation of miR-3184-5p led to growth inhibition Subsequently, we tested the expression level of miR- by suppressing cell proliferation and promoting 3184-5p in PCa samples. Unsurprisingly, miR-3184-5p cell apoptosis (Fig. 6g, h). Taken together, FOXP4 exhibited relative low level in PCa samples, which was exerted oncogenic function in PCa and participated in a negative with FOXP4-AS1 or FOXP4 (Fig. 6c, d). ceRNA pathway by cooperating with FOXP4-AS1 and Kaplan–Meier analysis further revealed the positive 0FOXP4. Official journal of the Cell Death Differentiation Association Wu et al. Cell Death and Disease (2019) 10:472 Page 10 of 14 Fig. 6 FOXP4 is the target of miR-3184-5p in PCa. a The binding sequence of miR-3184-5p to FOXP4 3′UTR. b Luciferase activity of reporters containing wild type FOXP4 (FOXP4-WT) or mutant type FOXP4 (FOXP4-MUT) in cells transfected with miR-3184-5p mimics. c MiR-3184-5p expression in PCa samples and normal samples. d Expression association between miR-3184-5p and FOXP4-AS1 or FOXP4 in PCa samples. e Kaplan–Meier analysis of patients with PCa with high or low expression level of miR-3184-5p. f MiR-3184-5p expression in PCa cell lines and normal cell line. g, h. The effect of miR-3184-5p inhibitor or mimics on cell proliferation or apoptosis. **P < 0.01, ***P < 0.001 FOXP4 involved in FOXP4-AS1-mediated PCa cell growth overexpression on LNCaP cell proliferation. Moreover, Then, we analyzed the expression level of FOXP4 in inhibition of miR-3184-5p expression or overexpression indicted PCa cells. It was found that mRNA and protein of FOXP4 attenuated the impact of sh-FOXP4-AS1 on level of FOXP4 were negatively regulated by miR-3184- PC-3 cell proliferation (Fig. 7c, d). Caspase-3 activity was 5p. And the effect of miR-3184-5p inhibitor or mimics also tested in cells transfected with indicated plasmids. on FOXP4 expression was attenuated by FOXP4-AS1 The experimental results proved that upregulation of (Fig. 7a, b). These data further demonstrated the reg- miR-3184-5p or knockdown of FOXP4 reversed the ulatory network of FOXP4-AS1/miR-3184-5p/FOXP4. inhibitory effect of pcDNA-FOXP4-AS1 on cell apoptosis. According to above data, we confirmed that FOXP4-AS1 However, the positive effect of silenced FOXP4-AS1 on can exert function in PCa by sequestering miR-3184-5p to cell apoptosis was attenuated in PC-3 cell cotransfected upregulate its nearby gene FOXP4. To strengthen our with miR-3184-5p inhibitor or FOXP4 expression vector viewpoint, we conducted rescue assays in two PCa cells. (Fig. 7e). Collectively, we confirmed that FOXP4-AS1 Cell proliferation assays demonstrated that miR-3184-5p promoted PCa growth by regulating miR-3184-5p/FOXP4 mimics or sh-FOXP4 reversed the effect of FOXP4-AS1 axis. Official journal of the Cell Death Differentiation Association Wu et al. Cell Death and Disease (2019) 10:472 Page 11 of 14 Fig. 7 FOXP4 involved in FOXP4-AS1-mediated PCa cell growth. a, b The effect of miR-3184-5p inhibitor or mimics on FOXP4 expression was attenuated by FOXP4-AS1. c, d Cell proliferation was examined after transfected with indicated plasmids. Scale bar = 100 μm. e The effect of miR- 3184-5p or FOXP4 on FOXP4-AS1-mediated cell apoptosis. *P < 0.05, **P < 0.01 PAX5 transcriptionally activated FOXP4-AS1 and FOXP4 in was expressed at a high level in PCa samples (Fig. 8a), PCa cells which was consistent with both FOXP4-AS1 and FOXP4 Transcription activation is an important reason for the (Fig. 8b). dysregulation of genes. In this regard, we detect whether Then, we performed loss- or gain-of-function assays upregulation of FOXP4-AS1 and FOXP4 was caused in to demonstrate the effect of PAX5 on PCa cell pro- this manner. According to the ChIP-seq results of UCSC, liferation and apoptosis. As presented in Supplementary we found five potential transcription factors for both Fig. 3A, B, upregulation of PAX5 enhanced cell pro- FOXP4-AS1 and FOXP4. To analyze the regulatory liferation, while silencing of PAX5 led to the opposite potential, we overexpressed and silenced them in two PCa results. In addition, cell apoptosis was negatively regu- cells (Supplementary Fig. 2A, B). Then, we examined the lated by PAX5 (Supplementary Fig. 3C). These findings expression change of FOXP4-AS1 and FOXP4 in above revealed the oncogenic property of PAX5 in PCa. Fur- two cell lines. The results suggested that PAX5 could therly, we investigated whether there is a regulatory positively regulate the expression of both FOXP4-AS1 and relationship between PAX5 and miR-3184-5p. And we FOXP4 in PCa cells (Supplementary Fig. 2C, D). PAX5 found that PAX5 and miR-3184-5p could not regulate Official journal of the Cell Death Differentiation Association Wu et al. Cell Death and Disease (2019) 10:472 Page 12 of 14 Fig. 8 PAX5 transcriptionally activated FOXP4-AS1 and FOXP4 in PCa cells. a PAX5 expression in paired PCa samples and normal samples. b Positive expression correlation between PAX5 and FOXP4-AS1 or FOXP4 in PCa tissues. c DNA motif of PAX5 and predicted five putative binding sites of PAX5 in FOXP4-AS1 or FOXP4 promoter. d The luciferase reporter assay demonstrated that site 1 of FOXP4-AS1 promoter (from −1475 to −1493) and site 5 of FOXP4 promoter (from −409 to −427) were potentially responsible for the binding of PAX5 to FOXP4-AS1 or FOXP4 promoter. e The ChIP assay showed that there was a strong affinity of PAX5 to FOXP4-AS1 or FOXP4 promoter. f Luciferase activity analysis revealed that the luciferase activity of reporter vector containing FOXP4-AS1 promoter or FOXP4 promoter was not changed by PAX5 after mutations in site 1 or site 5. *P < 0.05, **P < 0.01 Official journal of the Cell Death Differentiation Association Wu et al. Cell Death and Disease (2019) 10:472 Page 13 of 14 each other (Supplementary Fig. 3D, E). To demonstrate demonstrated that FOXP4-AS1 could promoted PCa the transcriptional regulation of PAX5 on FOXP4-AS1 growth. These data indicated that FOXP4-AS1 exerted andFOXP4, wesearchedDNA motifofPAX5and oncogenic function in the tumorigenesis of PCa. predicted five putative binding sites of PAX5 in FOXP4- Numerous evidence has shown that lncRNAs can exert AS1 or FOXP4 promoter (Fig. 8c). Through the luci- function in human cancers by regulating their nearby 29–31 ferase reporter assay, we found that site 1 of FOXP4- genes . According to bioinformatics analysis, we AS1 promoter (from −1475 to −1493) and site 5 of determined that FOXP4 is the nearby gene of FOXP4- FOXP4 promoter (from −409 to −427) were potentially AS1. To analyze the potential relationship between responsible for the binding of PAX5 to FOXP4-AS1 or FOXP4-AS1 and FOXP4, we examined the expression of FOXP4 promoter (Fig. 8d). To further validate the above FOXP4 in response to the overexpression or knockdown results, we separately cloned these putative binding sites of FOXP4-AS1. We determined the positive regulation of into pGL3 vector. As illustrated in Supplementary Fig. FOXP4-AS1 on FOXP4. More importantly, FOXP4 was 3F. PAX5 overexpression or knockdown affected the also upregulated in PCa samples and cell lines and pre- luciferase activity of site 1 of FOXP4-AS1 promoter and dicted poor outcome in patients with PCa. Mechan- site 5 of FOXP4 promoter. The ChIP assay showed that istically, we determined that FOXP4-AS1 was located in there was a strong affinity of PAX5 to FOXP4-AS1 or the cytoplasm of PCa cells, indicating post-transcriptional FOXP4 promoter (Fig. 8e). In order to validate that site regulatory potential of FOXP4-AS1 in PCa. It is widely 1 of FOXP4-AS1 promoter and site 5 of FOXP4 pro- acknowledged that lncRNAs can regulate gene expression moter were actually responsible for the binding of PAX5 by sequestering miRNAs . In our present study, we to their promoters, we separately mutated these binding investigated whether FOXP4-AS1 can regulate gene sites. Further luciferase activity analysis revealed that expression in this manner. Combining with all these data, the luciferase activity of reporter vector containing we analyzed whether FOXP4-AS1 positively regulated FOXP4-AS1 promoter or FOXP4 promoter was not FOXP4 in PCa by serving as a miRNA sponge. Both changed by PAX5 after mutations in site 1 or site 5 bioinformatics analysis and mechanism investigation (Fig. 8f). Collectively, PAX5 acted as a transcription revealed that FOXP4-AS1 and FOXP4 could interact with activator for both FOXP4-AS1 and FOXP4 in PCa. miR-3184-5p to form a ceRNA network. Through rescue assays, we determined that FOXP4-AS1/miR-3184-5p/ Discussion FOXP4 axis is an oncogenic pathway in PCa by promoting cell growth. Recently, increasing number of researches has revealed 20–22 the crucial role of lncRNAs in tumorigenesis . Dys- Upregulation of FOXP4-AS1 and FOXP4 in PCa may be regulation of lncRNAs is closely correlated with the caused by other mechanism, such as transcriptional reg- overall survival of patients with cancer. For instance, ulation. In the current study, we found that PAX5 is a lncRNA MNX1-AS1 and LINC00346 are upregulated in transcription factor of FOXP4-AS1 and FOXP4. Con- 23,24 gastric cancer and predicted poor outcome; upregu- sistent with FOXP4-AS1 and FOXP4, PAX5 was also lation of lncRNA EGFR-AS1 indicates low overall survival upregulated in PCa tissues. Further mechanism investi- of patients with renal cancer . LncRNA FOXP4-AS1 has gation proved that PAX5 enhanced the transcription been demonstrated to be a prognostic indicator in activity of FOXP4-AS1 and FOX4 by binding to their osteosarcoma. In this present study, we identified that promoter regions. Therefore, we confirmed that PAX5- FOXP4-AS1 was expressed at a high level TCGA PCa induced upregulation of FOXP4-AS1 and FOXP4 con- samples and collected PCa samples. Similarly, we analyzed tributed to tumorigenesis of PCa. whether high expression of FOXP4-AS1 is correlated with In conclusion, our present study demonstrated that the overall survival of patients with PCa. Interestingly. FOXP4-AS1 and FOXP4 were upregulated in PCa tissues High expression of FOXP4-AS1 is closely correlated with and cell lines and indicated poor outcome. FOXP4-AS1 the prognosis of patients with PCa. upregulated FOXP4 by sequestering miR-3184-5p. MiR- Functionally, upregulated lncRNAs can promote cell 3184-5p was downregulated in PCa samples and predicted proliferation and inhibit apoptosis, thereby facilitating poor overall survival. Upregulation of FOXP4-AS1/ 26–28 tumorigenesis . In our current study, we investigated FOXP4 axis was induced by the transcription activation of whether FOXP4-AS1 is a regulator in tumorigenesis of PAX5. Thus, this study revealed a novel oncogenic PCa. Both gain- or loss-of-function assays were carried pathway in PCa tumorigenesis (Fig. 9). All our findings out in two PCa cell lines. As expected, high expression of may contribute to investigate molecular mechanism FOXP4-AS1 promoted cell proliferation and suppressed associated with PCa tumorigenesis and will provide new cell apoptosis, indicating the oncogenic property of thought in exploring the novel diagnostic or therapeutic FOXP4-AS1 in PCa cells. In vivo animal study further biomarker for PCa. Official journal of the Cell Death Differentiation Association Wu et al. Cell Death and Disease (2019) 10:472 Page 14 of 14 12. Yang, L. et al. FOXP4-AS1 participates in the development and progression of osteosarcoma by downregulating LATS1 via binding to LSD1 and EZH2. Biochem. Biophys. Res. Commun. 502, 493–500 (2018). 13. Li, J. et al. Long non-coding RNA FOXP4-AS1 is an unfavourable prognostic factor and regulates proliferation and apoptosis in colorectal cancer. Cell Prolif., 50, https://doi.org/10.1111/cpr.12312 (2017). 14. Yan, X. et al. Mesenchymal stem cells promote hepatocarcinogenesis via lncRNA-MUF interaction with ANXA2 and miR-34a. Cancer Res., https://doi.org/ 10.1158/0008-5472.can-17-1915 (2017). 15. Chen, X. et al. Long noncoding RNA LINC01234 functions as a competing endogenous RNA to regulate CBFB expression by sponging miR-204-5p in gastric cancer. Clin. Cancer Res., https://doi.org/10.1158/1078-0432.ccr-17-2376 (2018). 16. Chen, D. L. et al. Long non-coding RNA UICLM promotes colorectal cancer liver metastasis by acting as a ceRNA for microRNA-215 to regulate ZEB2 expression. Theranostics 7,4836–4849 (2017). 17. Liang, Y. et al. LncRNA CASC9 promotes esophageal squamous cell carcinoma Fig. 9 A graphical abstract was created to illustrate the core of metastasis through upregulating LAMC2 expression by interacting with the this study CREB-binding protein. Cell Death Differ., https://doi.org/10.1038/s41418-018- 0084-9 (2018). 18. Sun, Z. et al. YAP1-induced MALAT1 promotes epithelial-mesenchymal tran- Acknowledgements sition and angiogenesis by sponging miR-126-5p in colorectal cancer. The authors sincerely appreciate all lab members. This study was supported by Oncogene, https://doi.org/10.1038/s41388-018-0628-y (2018). the Beijing Natural Science Foundation (No. 7192159) 19. Wang, Y. et al. TGF-β-induced STAT3 overexpression promotes human head and neck squamous cell carcinoma invasion and metastasis through malat1/ miR-30a interactions. Cancer Lett., https://doi.org/10.1016/j.canlet.2018.08.009 Conflict of interest (2018). The authors declare that they have no conflict of interest. 20. DeOcesano-Pereira, C. et al. Long non-coding RNA INXS is a critical mediator of BCL-XS induced apoptosis. Nucleic Acids Res. 42,8343–8355 (2014). 21. Lin, C. et al. Functional role of a novel long noncoding RNA TTN-AS1 in Publisher’s note esophageal squamous cell carcinoma progression and metastasis. Clin. Cancer Springer Nature remains neutral with regard to jurisdictional claims in Res., https://doi.org/10.1158/1078-0432.ccr-17-1851 (2017). published maps and institutional affiliations. 22. Kong, Y. et al. Pseudogene PDIA3P1 promotes cell proliferation, migration and invasion, and suppresses apoptosis in hepatocellular carcinoma by regulating Supplementary Information accompanies this paper at (https://doi.org/ the p53 pathway. Cancer Lett., https://doi.org/10.1016/j.canlet.2017.07.031 10.1038/s41419-019-1699-6). (2017). 23. Zhang, W. et al. Upregulated expression of MNX1-AS1 long noncoding RNA Received: 14 March 2019 Revised: 29 April 2019 Accepted: 2 May 2019 predicts poor prognosis in gastric cancer. Bosn. J. Basic Med. Sci., https://doi. org/10.17305/bjbms.2019.3713 (2019). 24. Xu, T. P. et al. KLF5 and MYC modulated LINC00346 contributes to gastric cancer progression through acting as a competing endogeous RNA and indicates poor outcome. Cell Death Differ., https://doi.org/10.1038/s41418-018- References 0236-y (2019). 1. Jemal, A. et al. Global cancer statistics. CA Cancer J. Clin. 61,69–90 (2011). 25. Wang, A. et al. Long noncoding RNA EGFR-AS1 promotes cell growth and 2. Prensner, J. R. & Chinnaiyan, A. M. The emergence of lncRNAs in cancer metastasis via affecting HuR mediated mRNA stability of EGFR in renal cancer. biology. Cancer Discov. 1,391–407 (2011). Cell Death Dis. 10, 154 (2019). 3. Kapranov, P. et al. RNA maps reveal new RNA classes and a possible function 26. Shi, Y. et al. The long noncoding RNA SPRY4-IT1 increases the proliferation of for pervasive transcription. Science 316,1484–1488 (2007). human breast cancer cells by upregulating ZNF703 expression. Mol. Cancer 4. Huarte, M. & Rinn, J. L. Large non-coding RNAs: missing links in cancer? Hum. 14, 51 (2015). Mol. Genet. 19,R152–R161 (2010). 27. Zhao, Y. et al. The lncRNA MACC1-AS1 promotes gastric cancer cell metabolic 5. Liu, Q. et al. LncRNA loc285194 is a p53-regulated tumor suppressor. Nucleic plasticity via AMPK/Lin28 mediated mRNA stability of MACC1. Mol. Cancer 17, Acids Res. 41, 4976–4987 (2013). 69 (2018). 6. Peng, L. et al. Super-enhancer-associatedlongnon-codingRNA HCCL5is 28. Su, W. et al. Silencing of long non-coding RNA MIR22HG triggers cell survival/ activated by ZEB1 and promotes the malignancy of hepatocellular carcinoma. death signaling via oncogenes YBX1, MET, and p21 in lung cancer. Cancer Res., Cancer Res., https://doi.org/10.1158/0008-5472.can-18-0367 (2018). https://doi.org/10.1158/0008-5472.can-18-0222 (2018). 7. Sun, M. et al. The pseudogene DUXAP8 promotes non-small-cell lung cancer 29. Su, W. et al. Long noncoding RNA ZEB1-AS1 epigenetically regulates the cell proliferation and invasion by epigenetically silencing EGR1 and RHOB. Mol. expressions of ZEB1 and downstream molecules in prostate cancer. Mol. Ther., https://doi.org/10.1016/j.ymthe.2016.12.018 (2017). Cancer 16, 142 (2017). 8. Pauli, A., Rinn, J. L. & Schier, A. F. Non-coding RNAs as regulators of embry- 30. Wang, C. et al. A cancer-testis non-coding RNA LIN28B-AS1 activates driver ogenesis. Nat. Rev. Genet. 12,136–149 (2011). gene LIN28B by interacting with IGF2BP1 in lung adenocarcinoma. Oncogene, 9. Prensner, J. R. et al. Transcriptome sequencing across a prostate cancer cohort https://doi.org/10.1038/s41388-018-0548-x (2018). identifies PCAT-1, an unannotated lincRNA implicated in disease progression. 31. Sun, J. et al. Long noncoding RNA FGFR3-AS1 promotes osteosarcoma Nat. Biotechnol. 29,742–749 (2011). growth through regulating its natural antisense transcript FGFR3. Mol. Biol. Rep. 10. Gibb, E. A., Brown, C. J. & Lam, W. L. The functional role of long non-coding 43,427–436 (2016). RNA in human carcinomas. Mol. Cancer 10, 38 (2011). 32. Lv, J. et al. Long non-coding RNA Unigene56159 promotes epithelial- 11. Taft, R. J.,Pang, K. C.,Mercer, T. R., Dinger, M. &Mattick,J.S.Non-codingRNAs: mesenchymal transition by acting as a ceRNA of miR-140-5p in hepatocellular regulators of disease. J. Pathol. 220,126–139 (2010). carcinoma cells. Cancer Lett. 382,166–175 (2016). Official journal of the Cell Death Differentiation Association http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Cell Death & Disease Springer Journals

LncRNA FOXP4-AS1 is activated by PAX5 and promotes the growth of prostate cancer by sequestering miR-3184-5p to upregulate FOXP4

Download PDF
14 pages

Loading next page...
 
/lp/springer-journals/lncrna-foxp4-as1-is-activated-by-pax5-and-promotes-the-growth-of-Zt8zLjoZN0

References (36)

Publisher
Springer Journals
Copyright
Copyright © 2019 by The Author(s)
Subject
Life Sciences; Life Sciences, general; Biochemistry, general; Cell Biology; Immunology; Cell Culture; Antibodies
eISSN
2041-4889
DOI
10.1038/s41419-019-1699-6
Publisher site
See Article on Publisher Site

Abstract

Prostate cancer (PCa) is one of the major men malignancies worldwide. Long noncoding RNAs (lncRNAs) have been reported as essential regulators in human cancers, including PCa. In the present study, lncRNA forkhead box P4 antisense RNA 1 (FOXP4-AS1) was found to be highly expressed in TCGA PCa samples. Upregulation of FOXP4-AS1 was further validated in 64 PCa tissues and predicted poor prognosis in patients with PCa. Functionally, high FOXP4-AS1 level was associated with increased cell proliferation and decreased cell apoptosis, indicating that FOXP4-AS1 exerted oncogenic functions in the tumorigenesis of PCa. Furthermore, FOXP4-AS1 was located in the cytoplasm of PCa cell lines and positively regulated FOXP4. LncRNAs can exert their functions by cooperating with their nearby genes. Mechanistically, FOXP4-AS1 post-transcriptionally regulated FOXP4 by acting as a competing endogenous RNA (ceRNA) in PCa to sponge miR-3184-5p. Considering the upregulation of both FOXP4-AS1 and its nearby gene FOXP4, we further detected the coactivator of FOXP4-AS1 and FOXP4. Mechanism analysis indicated that paired box 5 (PAX5) transcriptionally activated FOXP4-AS1 and FOXP4 in PCa. Collectively, we determined that PAX5-induced upregulation of FOXP4-AS1/FOXP4 axis promoted tumorigenesis of PCa. Introduction Recent years, long noncoding RNAs (lncRNAs) that are Prostate cancer (PCa) is a commonly diagnosed malig- mainly transcribed from the genomic intergenic regions nant male cancer and is one of the prime reasons for have become a research focus in cancers transcriptome . cancer-related death . Lacking of the biomarker in early They have been demonstrated to be participants in var- 3,4 diagnosis resulted in low overall survival of patients with ious biological processes by different mechanisms . The PCa. Although numerous efforts have been made in oncogenic or tumor-suppressive role of lncRNAs has been 5–7 exploring the novel therapeutic strategies, the prognosis well characterized in tumorigenesis . To date, some of patients with advanced PCa remains unfavorable. lncRNAs have been defined to be PCa specific, such as 8 9 10 11 Exploring the molecular mechanism associated with the PCGEM1 , PRNCR1 , PCAT1 , and SChLAP1 . tumorigenesis of PCa is of great significance for the early Nevertheless, investigating the function and mechanism diagnosis or treatment of PCa. of novel lncRNAs is necessary to find functional bio- markers in PCa progression. Searching from online database TCGA, we found top 30 upregulated lncRNAs in PCa samples. Among these candidates, lncRNA forkhead Correspondence: Weigang Yan ([email protected]) Department of Urology, Peking Union Medical College Hospital, Chinese box P4 antisense RNA 1 (FOXP4-AS1) has been reported 12 13 Academy of Medical Sciences, Beijing, China in osteosarcoma and colorectal cancer . However, it is Department of Pathology, Peking Union Medical College Hospital, Chinese unclear whether FOXP4-AS1 exerted similar function in Academy of Medical Sciences, Beijing, China Edited by A. Stephanou © The Author(s) 2019 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to theCreativeCommons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/. Official journal of the Cell Death Differentiation Association 1234567890():,; 1234567890():,; 1234567890():,; 1234567890():,; Wu et al. Cell Death and Disease (2019) 10:472 Page 2 of 14 PCa tumorigenesis. Thus, we conducted in vitro and the Declaration of Helsinki. According to the median of in vivo experiments to determine the function of FOXP4- the expression level of FOXP4-AS1, miR-3184-5p or AS1 in PCa growth. FOXP4, 64 PCa tissues were classified into high- To detect the potential mechanism pattern of lncRNA expression group (n = 32) and low-expression group FOXP4-AS1 in PCa, we detected the cellular localization (n = 32). of FOXP4-AS1 in PCa cells and determined that FOXP4- Human PCa cell line (PC-3, DU145, VCaP, and LNCaP) and human normal prostate epithelial cell line (RWPE-1) AS1 was predominantly located in the cytoplasm of PCa cells, indicating that FOXP4-AS1 may regulate gene were bought from the American Type Culture Collection expression at post-transcriptional level. In term of post- (Rockville, MD, USA). Cell culture was conducted in transcriptional regulation, lncRNAs have been acknowl- Dulbecco’s Modified Eagle Medium containing 10% fetal edged to be competing endogenous RNAs (ceRNAs) by bovine serum (HyClone, Logan, UT, USA), 1% penicillin- competitively binding to miRNA response elements to streptomycin (HyClone) at 37 °C with 5% CO . 14–16 upregulate mRNAs . On this basis, we carried out bioinformatics analysis and mechanism experiments to Quantitative real-time PCR determine downstream genes of FOXP4-AS1. It was Total RNA was extracted from tissue samples using found that FOXP4-AS1 can regulate its nearby gene TRIzol reagent (Thermo Fisher Scientific, Waltham, MA, FOXP4 by sequestering miR-3184-5p. USA) following the user guide. RNA was converted into Upregulation of genes in human cancers may be cDNA using random primers and Reverse Transcription 17–19 induced by transcription activation . Here, we found Kit (Toyobo, Osaka, Japan). Nanodrop 1000 spectro- several transcription factors in the common transcrip- photometer (Thermo Fisher Scientific, Waltham, MA, tional region of FOXP4-AS1 and FOXP4. Further USA) and 1.5% denaturing agarose gels were used to mechanism investigation was made to validate the effect measure RNA concentration and purity. The primers are of paired box 5 (PAX5) on the transcription activation of shown in Supplementary Table 1. To evaluate the both FOXP4-AS1 and FOXP4. Collectively, the focus of expression of lncRNA or mRNA, qRT-PCR was con- this study is to explore the mechanism of PAX5-induced ducted using SYBR Green PCR Master Mix (Takara, FOXP4-AS1/FOXP4 axis in PCa tumorigenesis. Kyoto, Japan) with Roche Real-Time PCR system (Roche, USA). Relative expression was normalized to GAPDH or −ΔΔCt Materials and methods U6 and was determined by 2 method. Bioinformatics analysis Top 30 upregulated lncRNAs in PCa samples and the Cell transfection expression profile of FOXP4-AS1 in PCa samples were For cell transfection, LNCaP and PC-3 cell lines at acquired from TCGA dataset (http://gepia.cancer-pku.cn/ 70–80% confluence were planted in six-well plates. For index.html). UCSC (http://genome.ucsc.edu/) was applied overexpression of genes, pcDNA3.1 vectors (Genechem to search the nearby gene of FOXP4-AS1. DIANA tools Company, Shanghai, China) were subcloned with FOXP4- (http://carolina.imis.athena-innovation.gr/diana_tools/ AS1, FOXP4, or PAX5 (pcDNA-FOXP4-AS1, pcDNA- web/index.php?r=site%2Ftools) and the starbase database FOXP4, or pcDNA-PAX5). Empty pcDNA3.1 vector was (http://starbase.sysu.edu.cn/) were used to predict the used as negative control. The short hairpin RNAs miRNAs that had complementary base paring with both (shRNAs) against FOXP4-AS1, FOXP4, and PAX5 (sh- FOXP4-AS1 and FOXP4. The DNA motif of PAX5 and FOXP4-AS1#1/2/3, sh-FOXP4#1/2/3, and sh-PAX5#1/2/ five putative binding sites of PAX5 in FOXP4-AS1 or 3) were designed and synthesized by GenePharma FOXP4 promoter were obtained from JASPAR (http:// (Shanghai, China). To overexpress miR-3184-5p, miR- jaspar.genereg.net/). 3184-5p mimics and miR-NC (GenePharma) were trans- fected into PCa cell lines. Whereas miR-3184-5p inhibitor Patient specimens and cell culture and its negative control anti-miR-NC (GenePharma) were Sixty-four matched PCa and adjacent nontumor pros- transfected into PCa cells for silencing of miR-3184-5p. tate tissues were collected from patients with PCa who Similarly, STAT5A, EBF1, RAD21, and RUNX3 were received treatment at The Fourth Hospital of Harbin overexpressed by subcloning the whole sequence of them Medical University from May 2012 to June 2017. After into pcDNA3.1 vectors (Genechem). shRNAs targeting resection, all tissues were snap-frozen in liquid nitrogen STAT5A, EBF1, RAD21, and RUNX3 were synthesized by and stored at −80 °C. The written informed consents were Genechem were used for knockdown of them. Lipo- provided by all the participants. Our study was approved fectamine 2000 (Invitrogen, Carlsbad, CA, USA) was used by the Research Ethics Committee of the The Fourth for transfection following the standard method. After Hospital of Harbin Medical University and complied with 48 h, cells were reaped for subsequent experiments. Official journal of the Cell Death Differentiation Association Wu et al. Cell Death and Disease (2019) 10:472 Page 3 of 14 Cell proliferation assays In vivo experiment and immunohistochemical staining A total of 5 × 10 transfected LNCaP and PC-3 cell lines Male BALB/C athymic nude mice (aged 6 weeks) were were harvested and put into 96-well plates. Cell viability bought from the National Laboratory Animal Center was analyzed using 10 μL of Cell Counting Kit-8 reagent (Beijing, China) and preserved in an SPF-grade pathogen- (CCK-8, DOJINDO, Kumamoto, Japan). The absorbance free animal laboratory. Animal experiment was conducted at 450 nm was evaluated by a microplate reader when the strictly in light with the protocol approved by the Animal reagent reacted for 24, 48, 72 and 96 h at room tem- Research Ethics Committee of The Fourth Hospital of perature. All experimental procedures were conducted in Harbin Medical University and the principles of the triplicate. Declaration of Helsinki. A total of 1 × 10 transfected For colony formation assay, LNCaP and PC-3 cell lines LNCaP and PC-3 cell lines were injected subcutaneously were plated in six-well plates at 500 cells per well for around the left flank of nude mice (three mice per group). 2 weeks. After fixation in 4% paraformaldehyde for Tumor volumes were calculated with the following 10 min, cells were subjected to 0.1% crystal violet for equation: 0.5 × length × width every 4 days. Four weeks 30 min. Finally, colonies were counted and recorded. later, mice were killed. Tumors were excised, weighed, Experiments were repeated at least three times. and snap-frozen at −80 °C for subsequent analysis. For EdU incorporation assay, LNCaP and PC-3 cell lines on sterile coverslips were seeded in 24-well plates. Cell Ki-67 staining proliferation was detected by the EdU incorporation assay The excised tumor tissue specimens were put in for- kit (Ribobio, Guangzhou, China). The stained cells were malin, fixed and paraffin-embedded. After deparaffinating visualized by laser confocal microscopy (FV300, Olympus, and rehydrating, tissues were incubated with antibody Tokyo, Japan). Nucleus was double-stained with EdU and against Ki-67 (Santa Cruz Biotechnology, Dallas, TX, 4′,6-diamidino-2-phenylindole (DAPI, Beyotime, Shang- USA) at 4 °C overnight. Tissues were stained with dia- hai, China) as positively proliferative cells. All experi- minobenzidine and observed under light microscopy. ments were repeated in triplicate. Western blotting JC-1 fluorescence analysis Protein samples from tissues or cells were acquired Cell apoptosis was assessed by examine the mitochon- using RIPA lysis buffer (Beyotime, Beijing, China) and drial transmembrane potential (ΔΨ). LNCaP and PC-3 separated on 10% SDS-PAGE. After transferring onto cell lines were seeded in 96-well plates (1 × 10 PVDF membranes, proteins were blocked with 5% solu- cells/well) and incubated overnight. Afterward, cells were subjected tion of nonfat milk for 2 h. Subsequently, samples were to Cis and/or mdivi-1, with or without 2 mM NAC or treated with primary antibodies against FOXP4 0.5 mM TEMPOL in serum-free medium for 18 h. The (ab251688, Abcam, Cambridge, UK) and GAPDH cationic, lipophilic dye 5,5′,6,′-tetrachloro-1,1′,3,3′ tetra- (ab8245, Abcam) along with the corresponding secondary ethylbenzimidazolyl carbocyanine iodide (JC-1) was antibodies conjugated to horseradish peroxidase. The commercially obtained from Cayman Chemical (Ann signals of membranes were measured by ECL Substrate Arbor, MI, USA). After centrifugation, cells were loaded (Pierce, Rockford, IL). GAPDH was used as an internal in JC-1 dye for half an hour, following washing and reference. incubation in assay buffer. ΔΨ was detected by a fluor- escent plate reader. A fluorescence microscope was used Subcellular fractionation assay to capture Gemini XPS and images. Each experimental The subcellular fractionation assay was carried out by procedure was repeated at least three times the use of PARIS Kit (Invitrogen, USA) following the supplier’s suggestion. LNCaP and PC-3 cell lines were Caspase-3 activity analysis washed in prechilled PBS for three times and lysed in cell To detect caspase-3 activity, an EnzCheck Caspase-3 fractionation buffer. The supernatant was collected. The Assay kit#1 was purchased from Molecular Probes cell lysates were placed in cell fractionation buffer and (Eugene, OR, USA). The rinsed cells were lysed in cell centrifuged. Cell disruption buffer was added to lyse cell lysis buffer on ice for 10 min, followed by centrifuga- nuclei. The isolated RNA was assessed by quantitative tion. The supernatant was transferred into microplate real-time PCR. GAPDH and U6 were used as cytoplasmic wells and subjected to Z-DEVD-AMC as caspase-3 or nuclear fractionation indicators. substrate. Gemini XPS fluorescent plate reader was applied to measure fluorescent signals at a wavelength Fluorescence in situ hybridization (FISH) of 405 nm. Each apoptotic assay was performed at least After fixation with 4% formaldehyde and washing with three times. PBS, PC-3 cells were treat with Pepsin (1% in 10 mmol/l Official journal of the Cell Death Differentiation Association Wu et al. Cell Death and Disease (2019) 10:472 Page 4 of 14 HCl) and was dehydrated by ethanol. Then, dried cells Chromatin immunoprecipitation assay (ChIP) were mixed in a hybridization buffer by using 40 nmol/l of The ChIP assay was performed by the use of Sim- the FISH probe (FOXP4-AS1 lncRNA) and incubated at pleChIP Enzymatic Chromatin IP Kit (CST, Danvers, 80 °C for 120 s. After being stayed at 55 °C for 120 min, MA, USA) following the supplier’s protocol. The cross- the slides were washed and dehydrated again. Finally, the linked chromatin DNA was broken to 200 to 1000 bp slides were detected by Prolong Gold Antifade Reagent through enzymatic digestion. Then, the chromatin was with DAPI. Fluorescence-conjugated FOXP4-AS1 probes immunoprecipitated with antibodies against PAX5 and were designed and synthesized by Ribobio Company IgG. Precipitated chromatin DNA was recovered by (Guangzhou, China). magnetic beads and analyzed by quantitative real-time PCR. All samples were prepared in triplicate. RNA immunoprecipitation assay (RIP) For Ago 2-RIP assay, the Magna RIP RNA-Binding Statistical analysis Protein Immunoprecipitation Kit was purchased from All experimental data were presented as the mean ± Millipore (Billerica, MA, USA) in line with the user guide. standard deviation of three or more repetitive experi- LNCaP and PC-3 cell lines were washed in precold PBS ments. SPSS 13.0 (SPSS, Inc., Chicago, IL, USA) and and lysed in RIP buffer at 4 °C for half an hour. Cell lysates GraphPad PRISM 6 (GraphPad, San Diego, CA, USA) were treated with magnetic beads conjugated to anti- were used for statistical analysis. Kaplan–Meier analysis bodies against Ago 2 (Millipore) or normal mouse was utilized to estimate the overall survival rate of patients immunoglobulin G (IgG; Millipore). Immunoprecipitated with PCa. Correlation was analyzed with Pearson’s cor- RNA was extracted for quantitative real-time PCR. For relation method. Differences between groups were ana- MS2-RIP assay, cells treated with pMS2-GFP were lyzed by Student's t-test or one-way ANOVA. Differences cotransfected with pcDNA3.1-MS2 and pcDNA3.1- were considered to be significant when P < 0.05. FOXP4-AS1-MS2 for 48 h. Thereafter, cells were incu- bated with GFP antibody (Roche Diagnostics GmbH, Results Mannheim, Germany) and the Magna RIP RNA-Binding Upregulation of FOXP4-AS1 in PCa samples is correlated Protein Immunoprecipitation Kit in accordance with the with unfavorable patients’ prognosis guidebook for user. After collecting and lysing, cell sus- Based on the TCGA database, top 30 upregulated pension was cultured with magnetic beads. The antibody lncRNAs in PCa samples were searched out and illu- was added and incubated overnight. After RNA purifica- strated in Fig. 1a. Among which, FOXP4-AS1 has been tion, the isolated RNA was detected by quantitative real- reported as an oncogene in colorectal cancer and osteo- time PCR to quantify the presence of the binding targets. sarcoma. However, it is unknown that whether FOXP4- AS1 exerted function in PCa. The expression profile of Luciferase reporter assay FOXP4-AS1 in TCGA PCa samples were detected and The amplified FOXP4-AS1 or 3′-UTR of FOXP4 was shown (Fig. 1b). Moreover, the expression level of subcloned into the firefly plasmids in pmirGLO luciferase FOXP4-AS1 was examined in PCa samples and adjacent vector (GeneChem, Shanghai, China). The wild type of samples collected from 64 patients with PCa. As expected, FOXP4-AS1 or FOXP4 3′-UTR (FOXP4-AS1-WT or FOXP4-AS1 was expressed at a higher level in PCa FOXP4-WT) was constructed. The site-directed mutation samples compared to adjacent normal samples (Fig. 1c). of miR-3184-5p binding sites in FOXP4-AS1 or FOXP4 To analyze the prognostic potential of FOXP4-AS1 in 3′-UTR (FOXP4-AS1-MUT or FOXP4-MUT) was gen- patients with PCa, Kaplan–Meier surviving curves were erated using the GeneTailor™ Site-Directed Mutagenesis generated. After analysis, we determined that high level of System (Invitrogen, Carlsbad, CA, USA). LNCaP and PC- FOXP4-AS1 was closely associated with the low overall 3 cell lines were separately cotransfected with aforemen- survival rate of patients with PCa (Fig. 1d). Furthermore, tioned plasmids and miR-3184-5p mimics or miR-NC. At the relative higher level of FOXP4-AS1 was observed in last, a Dual-Luciferase Reporter Assay System (Promega) patients in advanced stage rather than those in early stage was used to evaluate luciferase activity. For FOXP4-AS1 (Fig. 1e). Consistently, the expression level of FOXP4-AS1 or FOXP4 promoter luciferase assay, cells were placed was found to be higher in PCa cell lines compared to into 24-well plates and cotransfected with the plasmids human normal prostate epithelial cell RWPE-1 (Fig. 1f). containing the putative binding sites of PAX5 to FOXP4- These data indicated that FOXP4-AS1 might be a parti- AS1 or FOXP4 promoter. All these plasmids were cipant in tumorigenesis of PCa. simultaneously subcloned into pGL3 vector (Promega Corporation, Fitchburg, WI, USA) and established. Each FOXP4-AS1 facilitated PCa cell growth in vitro procedure of these experiments was repeated indepen- To explore the role of FOXP4-AS1 in PCa cellular dently for three times. function, we designed functional assays in two different Official journal of the Cell Death Differentiation Association Wu et al. Cell Death and Disease (2019) 10:472 Page 5 of 14 Fig. 1 Upregulation of FOXP4-AS1 in PCa samples is correlated with unfavorable patients’ prognosis. a Thirty lncRNAs that are upregulated in TCGA PCa samples were listed. b FOXP4-AS1 expression prolife in PCa and normal tissues was obtained from the TCGA database. c The expression level of FOXP4-AS1 was examined in PCa samples and adjacent samples collected from 64 patients with PCa. d Overall survival of 64 patients with PCa with high or low FOXP4-AS1 level was analyzed. e The expression level of FOXP4-AS1 in patients with PCa with early or advanced tumor stage. f Expression level of FOXP4-AS1 in PCa cell lines and one human normal prostate epithelial cell RWPE-1. *P < 0.05, **P < 0.01 PCa cell lines. According to the data in Fig. 1g, LNCaP cell addition, we tested caspase-3 activity in two PCa cells that exhibited the relative lowest expression level of FOXP4- were separately transfected with FOXP4-AS1 expression AS1, while PC-3 cell presented the highest level of vector or sh-FOXP4-AS1#1.Itwas foundthatFOXP4-AS1 FOXP4-AS1 expression. Thus, we overexpressed FOXP4- negatively regulated caspase-3 activity in PCa cells (Fig. 2e). AS1 in LNCaP cell but silenced it in PC-3 cell (Supple- Therefore, we confirmed that ectopic expression of FOXP4- mentary Fig. 1A). At first, we performed the CCK-8 assay AS1 affected PCa proliferation and apoptosis. to detect the viability of cells with high or low level of FOXP4-AS1. The results indicated that overexpression of FOXP4-AS1 promoted PCa tumor growth in vivo FOXP4-AS1 efficiently enhanced cell viability, whereas Furtherly, animal study was carried out to demonstrate downregulation of FOXP4-AS1 led to the decreased cell the role of FOXP4-AS1 in regulating PCa growth. LNCaP viability (Fig. 2a). Cell viability was suppressed more cell stably transfected with pcDNA-FOXP4-AS1 or empty efficient when cells were transfected with sh-FOXP4- vector and PC-3 cell stably transfected with sh-FOXP4- AS1#1. Therefore, sh-FOXP4-AS1#1 was chosen for AS1#1 or sh-NC were separately injected into nude mice. subsequent experiments. Through colony formation and After 28 days, tumors were removed and calculated. As EdU assays, we determined that overexpression of presented in Fig. 3a, tumors derived from cell transfected FOXP4-AS1 markedly promoted cell proliferation, while with pcDNA-FOXP4-AS1 were bigger than those derived knockdown of FOXP4-AS1 led to an opposite result from cell transfected with empty vector. In addition, (Fig. 2b, c). Then, we explored whether FOXP4-AS1 tumors in sh-FOXP4-AS1#1 group were smaller than regulated cell proliferation by inducing cell apoptosis. those in sh-NC group (Fig. 3a). Consistently. Same ten- According to JC-1 staining, we supposed that high expres- dencies were observed in tumor volume and tumor weight sion level of FOXP4-AS1 correlated with increased mem- (Fig. 3b, c). Finally, immuohistochemical staining indi- brane potential, indicating the inhibitory effect of FOXP4- cated that ki-67 positivity in FOXP4-AS1 overexpression AS1 on the early apoptosis of PCa cells (Fig. 2d). In group was significantly higher in NC group. However, sh- Official journal of the Cell Death Differentiation Association Wu et al. Cell Death and Disease (2019) 10:472 Page 6 of 14 Fig. 2 FOXP4-AS1 facilitated PCa cell growth in vitro. a The CCK-8 assay was used to detect cell viability after overexpression or knockdown of FOXP4-AS1. b, c Proliferative ability of LNCaP cell transfected with FOXP4-AS1 expression vector or PC-3 cell transfected with sh-FOXP4-AS1 was assessed by the colony formation assay and EdU assay. Scale bar for EdU staining: 100 μm. d Membrane potential in FOXP4-AS1-overexpressed or -downregulated PCa cells was measured by JC-1 staining. Scale bar = 200 μm. e Caspase-3 activity was tested in two PCa cells that were separately transfected with FOXP4-AS1 expression vector or sh-FOXP4-AS1#1. *P < 0.05, **P < 0.01 FOXP4-AS1#1 group exhibited low ki-67 positivity com- was upregulated in PCa samples (Fig. 4c), which was pared to sh-NC group (Fig. 3d). consistent with FOXP4-AS1 (Fig. 4d). Similarly, we made survival analysis to determine whether FOXP4 was cor- FOXP4 was positively regulated by FOXP4-AS1 and exerted related with patients’ survival. It was found that patients oncogenic property in PCa in FOXP4 high-expression group had lower overall sur- Using online bioinformatics analysis tool UCSC, we vival rate than those in FOXP4 low-expression group found that FOXP4 is a nearby gene of FOXP4-AS1 (Fig. (Fig. 4e). The relative high level of FOXP4 was detected in 4a). FOXP4 has been reported in human cancers due to its PCa cells compared with the normal control cell (Fig. 4f). oncogenic property. Through qRT-PCR and western To investigate the potential of FOXP4 in regulating PCa blotting analyses, we found that FOXP4-AS1 could posi- cell growth, we overexpressed or silenced it in indicated tively regulate both mRNA and protein level of FOXP4 in PCa cells by transfecting with FOXP4 expression vector or PCa cells (Fig. 4b). Furthermore, we found that FOXP4 specific shRNA (Supplementary Fig. 1B). Then, CCK-8 Official journal of the Cell Death Differentiation Association Wu et al. Cell Death and Disease (2019) 10:472 Page 7 of 14 Fig. 3 FOXP4-AS1 promoted PCa tumor growth in vivo. a Tumors derived from FOXP4-AS1-overexpressed LNCaP cell or FOXP4-AS1- downregulated PC-3 cell was resected and shown. b, c Tumor volume and tumor weight in different groups were quantified and illustrated. d Immuohistochemical staining of ki-67 positivity in tumors of different groups. Scale bar = 50 μm. **P < 0.01 and caspase-3 activity test were used to evaluate the (Fig. 5c). Based on bioinformatics analysis, we searched out importance of FOXP4 in PCa cell proliferation and five miRNAs that had complementary base paring with apoptosis. Interestingly, overexpression of FOXP4 pro- both FOXP4-AS1 and FOXP4 (Fig. 5d). Subsequently, the moted cell proliferation, while knockdown of FOXP4 had MS2-RIP assay was carried out to demonstrate the binding inhibitory effect on cell proliferation (Fig. 4g). In addition, of these five potential miRNAs to FOXP4-AS1. As a result, the caspase-3 activity was decreased in LNCaP cell miR-3184-5p and miR-423-5p showed the strongest affi- transfected with FOXP4 expression vector but was nity to FOXP4-AS1-MS2 beads (Fig. 5e). Then, we increased in PC-3 cell transfected with sh-FOXP4#1 examined the expression level of these two miRNAs in (Fig. 4h). These data showed that FOXP4 might be cells transfected with FOXP4-AS1 expression vector or cooperated with FOXP4-AS1 to exert function in PCa. shRNAs. The results showed that miR-3184-5p was effi- ciently downregulated by overexpressing FOXP4-AS1 but FOXP4-AS1 acted as the molecular sponge of miR-3184-5p was upregulated by silencing FOXP4-AS1 (Fig. 5f). The LncRNAs can exert functions by transcriptionally or predicted binding sequence between FOXP4-AS1 and post-transcriptionally activating their downstream genes. miR-3184-5p was shown (Fig. 5g). To change the expres- To investigate the regulatory pattern of FOXP4-AS1 in sion level of miR-3184-5p in PCa cells, miR-3184-5p PCa, we detected its cellular fractionation in two PCa cells inhibitor and miR-3184-5p mimics were separately trans- through the subcellular fractionation assay and FISH assay. fected into LNCaP and PC-3 cells. Luciferase activity Both experimental results indicated that FOXP4-AS1 was analysis revealed the effect of miR-3184-5p mimics on the predominantly located in the cytoplasm of PCa cells luciferase activity of reporter containing wild type FOXP4- (Fig. 5a, b). Therefore, we identified the post- AS1 (FOXP4-AS1-WT) or mutant type FOXP4-AS1 transcriptional regulation of FOXP4-AS1 in PCa. CeRNA (FOXP4-AS1-MUT). The results suggested that miR- network is a common post-transcriptional regulatory 3184-5p mimics efficiently decreased the luciferase activ- pattern of lncRNAs. Combining with above data, we ity of FOXP4-AS1-WT vector (Fig. 5h). All these experi- hypothesized that FOXP4-AS1 might upregulate FOXP4 mental results indicated that FOXP4-AS1 might act as a by sequestering a miRNA. Next, we performed the Ago 2- ceRNA to regulate miR-3184-5p and FOXP4. RIP assay in two PCa cells. The results showed that both FOXP4-AS1 and FOXP4 were enriched in Ago 2 con- FOXP4 is the target of miR-3184-5p in PCa taining beads, suggesting the probability of the involve- Likewise, we performed experiments to demonstrate ment of FOXP4-AS1 and FOXP4 in ceRNA network the interaction between miR-3184-5p and FOXP4. At Official journal of the Cell Death Differentiation Association Wu et al. Cell Death and Disease (2019) 10:472 Page 8 of 14 Fig. 4 FOXP4 was positively regulated by FOXP4-AS1 and exerted oncogenic property in PCa. a UCSC data revealed that FOXP4 is a nearby gene of FOXP4-AS1. b The mRNA and protein level of FOXP4 in cells transfected with pcDNA-FOXP4-AS1 or sh-FOXP4-AS1 were measured by qRT- PCR and western blotting. c Upregulation of FOXP4 in PCa samples. d Positive expression association between FOXP4-AS1 and FOXP4 in PCa samples was analyzed by Pearson correlation analysis. e Kaplan–Meier analysis of patients with PCa with high or low expression level of FOXP4. f Relative high level of FOXP4 was detected in PCa cells and one normal control cell. g Cell viability was examined in response to the overexpression or knockdown of FOXP4 by the CCK-8 assay. h Caspase-3 activity was increased in LNCaP cell transfected with FOXP4 expression vector but was increased in PC-3 cell transfected with sh-FOXP4#1. **P < 0.01 Official journal of the Cell Death Differentiation Association Wu et al. Cell Death and Disease (2019) 10:472 Page 9 of 14 Fig. 5 FOXP4-AS1 acted as the molecular sponge of miR-3184-5p. a, b FOXP4-AS1 was predominantly located in the cytoplasm of PCa cells. Scale bar = 200 μm. c Ago 2-RIP revealed that both FOXP4-AS1 and FOXP4 were enriched in Ago 2 containing beads. d Five miRNAs that had complementary base paring with both FOXP4-AS1 and FOXP4 were predicted from starBase and DIANA. e Enrichment of five miRNAs in MS2 with or without FOXP4-AS1. f miR-3184-5p and miR-423-5p were detected in cells transfected with FOXP4-AS1 expression vector or sh-FOXP4-AS1#1. g The predicted binding sequence between FOXP4-AS1 and miR-3184-5p. h The luciferase reporter assay was conducted in cells transfected with miR- 3184-5p mimics or miR-NC to examine the luciferase activity of FOXP4-AS1-WT and FOXP4-AS1-MUT. *P < 0.05, **P < 0.01 first, the binding sequence of miR-3184-5p to FOXP4 3′ correlation between miR-3184-5p expression and the UTR was predicted and obtained (Fig. 6a). Then, we overall survival rate of patients with PCa (Fig. 6e). examined the luciferase activity of reporter containing Compared with normal cell line, PCa cell lines exhib- wild-type FOXP4 (FOXP4-WT) or mutant-type FOXP4 ited relative low expression level of miR-3184-5p (Fig. (FOXP4-MUT) in cells transfected with miR-3184-5p 6f).Functionally,wefound that inhibition of miR-3184- mimics. The results suggested that the luciferase 5p had a positive effect on cell proliferation but had a activity of FOXP4-WT vector but not FOXP4-MUT negative effect on cell apoptosis. On the other hand, vector was decreased by miR-3184-5p mimics (Fig. 6b). upregulation of miR-3184-5p led to growth inhibition Subsequently, we tested the expression level of miR- by suppressing cell proliferation and promoting 3184-5p in PCa samples. Unsurprisingly, miR-3184-5p cell apoptosis (Fig. 6g, h). Taken together, FOXP4 exhibited relative low level in PCa samples, which was exerted oncogenic function in PCa and participated in a negative with FOXP4-AS1 or FOXP4 (Fig. 6c, d). ceRNA pathway by cooperating with FOXP4-AS1 and Kaplan–Meier analysis further revealed the positive 0FOXP4. Official journal of the Cell Death Differentiation Association Wu et al. Cell Death and Disease (2019) 10:472 Page 10 of 14 Fig. 6 FOXP4 is the target of miR-3184-5p in PCa. a The binding sequence of miR-3184-5p to FOXP4 3′UTR. b Luciferase activity of reporters containing wild type FOXP4 (FOXP4-WT) or mutant type FOXP4 (FOXP4-MUT) in cells transfected with miR-3184-5p mimics. c MiR-3184-5p expression in PCa samples and normal samples. d Expression association between miR-3184-5p and FOXP4-AS1 or FOXP4 in PCa samples. e Kaplan–Meier analysis of patients with PCa with high or low expression level of miR-3184-5p. f MiR-3184-5p expression in PCa cell lines and normal cell line. g, h. The effect of miR-3184-5p inhibitor or mimics on cell proliferation or apoptosis. **P < 0.01, ***P < 0.001 FOXP4 involved in FOXP4-AS1-mediated PCa cell growth overexpression on LNCaP cell proliferation. Moreover, Then, we analyzed the expression level of FOXP4 in inhibition of miR-3184-5p expression or overexpression indicted PCa cells. It was found that mRNA and protein of FOXP4 attenuated the impact of sh-FOXP4-AS1 on level of FOXP4 were negatively regulated by miR-3184- PC-3 cell proliferation (Fig. 7c, d). Caspase-3 activity was 5p. And the effect of miR-3184-5p inhibitor or mimics also tested in cells transfected with indicated plasmids. on FOXP4 expression was attenuated by FOXP4-AS1 The experimental results proved that upregulation of (Fig. 7a, b). These data further demonstrated the reg- miR-3184-5p or knockdown of FOXP4 reversed the ulatory network of FOXP4-AS1/miR-3184-5p/FOXP4. inhibitory effect of pcDNA-FOXP4-AS1 on cell apoptosis. According to above data, we confirmed that FOXP4-AS1 However, the positive effect of silenced FOXP4-AS1 on can exert function in PCa by sequestering miR-3184-5p to cell apoptosis was attenuated in PC-3 cell cotransfected upregulate its nearby gene FOXP4. To strengthen our with miR-3184-5p inhibitor or FOXP4 expression vector viewpoint, we conducted rescue assays in two PCa cells. (Fig. 7e). Collectively, we confirmed that FOXP4-AS1 Cell proliferation assays demonstrated that miR-3184-5p promoted PCa growth by regulating miR-3184-5p/FOXP4 mimics or sh-FOXP4 reversed the effect of FOXP4-AS1 axis. Official journal of the Cell Death Differentiation Association Wu et al. Cell Death and Disease (2019) 10:472 Page 11 of 14 Fig. 7 FOXP4 involved in FOXP4-AS1-mediated PCa cell growth. a, b The effect of miR-3184-5p inhibitor or mimics on FOXP4 expression was attenuated by FOXP4-AS1. c, d Cell proliferation was examined after transfected with indicated plasmids. Scale bar = 100 μm. e The effect of miR- 3184-5p or FOXP4 on FOXP4-AS1-mediated cell apoptosis. *P < 0.05, **P < 0.01 PAX5 transcriptionally activated FOXP4-AS1 and FOXP4 in was expressed at a high level in PCa samples (Fig. 8a), PCa cells which was consistent with both FOXP4-AS1 and FOXP4 Transcription activation is an important reason for the (Fig. 8b). dysregulation of genes. In this regard, we detect whether Then, we performed loss- or gain-of-function assays upregulation of FOXP4-AS1 and FOXP4 was caused in to demonstrate the effect of PAX5 on PCa cell pro- this manner. According to the ChIP-seq results of UCSC, liferation and apoptosis. As presented in Supplementary we found five potential transcription factors for both Fig. 3A, B, upregulation of PAX5 enhanced cell pro- FOXP4-AS1 and FOXP4. To analyze the regulatory liferation, while silencing of PAX5 led to the opposite potential, we overexpressed and silenced them in two PCa results. In addition, cell apoptosis was negatively regu- cells (Supplementary Fig. 2A, B). Then, we examined the lated by PAX5 (Supplementary Fig. 3C). These findings expression change of FOXP4-AS1 and FOXP4 in above revealed the oncogenic property of PAX5 in PCa. Fur- two cell lines. The results suggested that PAX5 could therly, we investigated whether there is a regulatory positively regulate the expression of both FOXP4-AS1 and relationship between PAX5 and miR-3184-5p. And we FOXP4 in PCa cells (Supplementary Fig. 2C, D). PAX5 found that PAX5 and miR-3184-5p could not regulate Official journal of the Cell Death Differentiation Association Wu et al. Cell Death and Disease (2019) 10:472 Page 12 of 14 Fig. 8 PAX5 transcriptionally activated FOXP4-AS1 and FOXP4 in PCa cells. a PAX5 expression in paired PCa samples and normal samples. b Positive expression correlation between PAX5 and FOXP4-AS1 or FOXP4 in PCa tissues. c DNA motif of PAX5 and predicted five putative binding sites of PAX5 in FOXP4-AS1 or FOXP4 promoter. d The luciferase reporter assay demonstrated that site 1 of FOXP4-AS1 promoter (from −1475 to −1493) and site 5 of FOXP4 promoter (from −409 to −427) were potentially responsible for the binding of PAX5 to FOXP4-AS1 or FOXP4 promoter. e The ChIP assay showed that there was a strong affinity of PAX5 to FOXP4-AS1 or FOXP4 promoter. f Luciferase activity analysis revealed that the luciferase activity of reporter vector containing FOXP4-AS1 promoter or FOXP4 promoter was not changed by PAX5 after mutations in site 1 or site 5. *P < 0.05, **P < 0.01 Official journal of the Cell Death Differentiation Association Wu et al. Cell Death and Disease (2019) 10:472 Page 13 of 14 each other (Supplementary Fig. 3D, E). To demonstrate demonstrated that FOXP4-AS1 could promoted PCa the transcriptional regulation of PAX5 on FOXP4-AS1 growth. These data indicated that FOXP4-AS1 exerted andFOXP4, wesearchedDNA motifofPAX5and oncogenic function in the tumorigenesis of PCa. predicted five putative binding sites of PAX5 in FOXP4- Numerous evidence has shown that lncRNAs can exert AS1 or FOXP4 promoter (Fig. 8c). Through the luci- function in human cancers by regulating their nearby 29–31 ferase reporter assay, we found that site 1 of FOXP4- genes . According to bioinformatics analysis, we AS1 promoter (from −1475 to −1493) and site 5 of determined that FOXP4 is the nearby gene of FOXP4- FOXP4 promoter (from −409 to −427) were potentially AS1. To analyze the potential relationship between responsible for the binding of PAX5 to FOXP4-AS1 or FOXP4-AS1 and FOXP4, we examined the expression of FOXP4 promoter (Fig. 8d). To further validate the above FOXP4 in response to the overexpression or knockdown results, we separately cloned these putative binding sites of FOXP4-AS1. We determined the positive regulation of into pGL3 vector. As illustrated in Supplementary Fig. FOXP4-AS1 on FOXP4. More importantly, FOXP4 was 3F. PAX5 overexpression or knockdown affected the also upregulated in PCa samples and cell lines and pre- luciferase activity of site 1 of FOXP4-AS1 promoter and dicted poor outcome in patients with PCa. Mechan- site 5 of FOXP4 promoter. The ChIP assay showed that istically, we determined that FOXP4-AS1 was located in there was a strong affinity of PAX5 to FOXP4-AS1 or the cytoplasm of PCa cells, indicating post-transcriptional FOXP4 promoter (Fig. 8e). In order to validate that site regulatory potential of FOXP4-AS1 in PCa. It is widely 1 of FOXP4-AS1 promoter and site 5 of FOXP4 pro- acknowledged that lncRNAs can regulate gene expression moter were actually responsible for the binding of PAX5 by sequestering miRNAs . In our present study, we to their promoters, we separately mutated these binding investigated whether FOXP4-AS1 can regulate gene sites. Further luciferase activity analysis revealed that expression in this manner. Combining with all these data, the luciferase activity of reporter vector containing we analyzed whether FOXP4-AS1 positively regulated FOXP4-AS1 promoter or FOXP4 promoter was not FOXP4 in PCa by serving as a miRNA sponge. Both changed by PAX5 after mutations in site 1 or site 5 bioinformatics analysis and mechanism investigation (Fig. 8f). Collectively, PAX5 acted as a transcription revealed that FOXP4-AS1 and FOXP4 could interact with activator for both FOXP4-AS1 and FOXP4 in PCa. miR-3184-5p to form a ceRNA network. Through rescue assays, we determined that FOXP4-AS1/miR-3184-5p/ Discussion FOXP4 axis is an oncogenic pathway in PCa by promoting cell growth. Recently, increasing number of researches has revealed 20–22 the crucial role of lncRNAs in tumorigenesis . Dys- Upregulation of FOXP4-AS1 and FOXP4 in PCa may be regulation of lncRNAs is closely correlated with the caused by other mechanism, such as transcriptional reg- overall survival of patients with cancer. For instance, ulation. In the current study, we found that PAX5 is a lncRNA MNX1-AS1 and LINC00346 are upregulated in transcription factor of FOXP4-AS1 and FOXP4. Con- 23,24 gastric cancer and predicted poor outcome; upregu- sistent with FOXP4-AS1 and FOXP4, PAX5 was also lation of lncRNA EGFR-AS1 indicates low overall survival upregulated in PCa tissues. Further mechanism investi- of patients with renal cancer . LncRNA FOXP4-AS1 has gation proved that PAX5 enhanced the transcription been demonstrated to be a prognostic indicator in activity of FOXP4-AS1 and FOX4 by binding to their osteosarcoma. In this present study, we identified that promoter regions. Therefore, we confirmed that PAX5- FOXP4-AS1 was expressed at a high level TCGA PCa induced upregulation of FOXP4-AS1 and FOXP4 con- samples and collected PCa samples. Similarly, we analyzed tributed to tumorigenesis of PCa. whether high expression of FOXP4-AS1 is correlated with In conclusion, our present study demonstrated that the overall survival of patients with PCa. Interestingly. FOXP4-AS1 and FOXP4 were upregulated in PCa tissues High expression of FOXP4-AS1 is closely correlated with and cell lines and indicated poor outcome. FOXP4-AS1 the prognosis of patients with PCa. upregulated FOXP4 by sequestering miR-3184-5p. MiR- Functionally, upregulated lncRNAs can promote cell 3184-5p was downregulated in PCa samples and predicted proliferation and inhibit apoptosis, thereby facilitating poor overall survival. Upregulation of FOXP4-AS1/ 26–28 tumorigenesis . In our current study, we investigated FOXP4 axis was induced by the transcription activation of whether FOXP4-AS1 is a regulator in tumorigenesis of PAX5. Thus, this study revealed a novel oncogenic PCa. Both gain- or loss-of-function assays were carried pathway in PCa tumorigenesis (Fig. 9). All our findings out in two PCa cell lines. As expected, high expression of may contribute to investigate molecular mechanism FOXP4-AS1 promoted cell proliferation and suppressed associated with PCa tumorigenesis and will provide new cell apoptosis, indicating the oncogenic property of thought in exploring the novel diagnostic or therapeutic FOXP4-AS1 in PCa cells. In vivo animal study further biomarker for PCa. Official journal of the Cell Death Differentiation Association Wu et al. Cell Death and Disease (2019) 10:472 Page 14 of 14 12. Yang, L. et al. FOXP4-AS1 participates in the development and progression of osteosarcoma by downregulating LATS1 via binding to LSD1 and EZH2. Biochem. Biophys. Res. Commun. 502, 493–500 (2018). 13. Li, J. et al. Long non-coding RNA FOXP4-AS1 is an unfavourable prognostic factor and regulates proliferation and apoptosis in colorectal cancer. Cell Prolif., 50, https://doi.org/10.1111/cpr.12312 (2017). 14. Yan, X. et al. Mesenchymal stem cells promote hepatocarcinogenesis via lncRNA-MUF interaction with ANXA2 and miR-34a. Cancer Res., https://doi.org/ 10.1158/0008-5472.can-17-1915 (2017). 15. Chen, X. et al. Long noncoding RNA LINC01234 functions as a competing endogenous RNA to regulate CBFB expression by sponging miR-204-5p in gastric cancer. Clin. Cancer Res., https://doi.org/10.1158/1078-0432.ccr-17-2376 (2018). 16. Chen, D. L. et al. Long non-coding RNA UICLM promotes colorectal cancer liver metastasis by acting as a ceRNA for microRNA-215 to regulate ZEB2 expression. Theranostics 7,4836–4849 (2017). 17. Liang, Y. et al. LncRNA CASC9 promotes esophageal squamous cell carcinoma Fig. 9 A graphical abstract was created to illustrate the core of metastasis through upregulating LAMC2 expression by interacting with the this study CREB-binding protein. Cell Death Differ., https://doi.org/10.1038/s41418-018- 0084-9 (2018). 18. Sun, Z. et al. YAP1-induced MALAT1 promotes epithelial-mesenchymal tran- Acknowledgements sition and angiogenesis by sponging miR-126-5p in colorectal cancer. The authors sincerely appreciate all lab members. This study was supported by Oncogene, https://doi.org/10.1038/s41388-018-0628-y (2018). the Beijing Natural Science Foundation (No. 7192159) 19. Wang, Y. et al. TGF-β-induced STAT3 overexpression promotes human head and neck squamous cell carcinoma invasion and metastasis through malat1/ miR-30a interactions. Cancer Lett., https://doi.org/10.1016/j.canlet.2018.08.009 Conflict of interest (2018). The authors declare that they have no conflict of interest. 20. DeOcesano-Pereira, C. et al. Long non-coding RNA INXS is a critical mediator of BCL-XS induced apoptosis. Nucleic Acids Res. 42,8343–8355 (2014). 21. Lin, C. et al. Functional role of a novel long noncoding RNA TTN-AS1 in Publisher’s note esophageal squamous cell carcinoma progression and metastasis. Clin. Cancer Springer Nature remains neutral with regard to jurisdictional claims in Res., https://doi.org/10.1158/1078-0432.ccr-17-1851 (2017). published maps and institutional affiliations. 22. Kong, Y. et al. Pseudogene PDIA3P1 promotes cell proliferation, migration and invasion, and suppresses apoptosis in hepatocellular carcinoma by regulating Supplementary Information accompanies this paper at (https://doi.org/ the p53 pathway. Cancer Lett., https://doi.org/10.1016/j.canlet.2017.07.031 10.1038/s41419-019-1699-6). (2017). 23. Zhang, W. et al. Upregulated expression of MNX1-AS1 long noncoding RNA Received: 14 March 2019 Revised: 29 April 2019 Accepted: 2 May 2019 predicts poor prognosis in gastric cancer. Bosn. J. Basic Med. Sci., https://doi. org/10.17305/bjbms.2019.3713 (2019). 24. Xu, T. P. et al. KLF5 and MYC modulated LINC00346 contributes to gastric cancer progression through acting as a competing endogeous RNA and indicates poor outcome. Cell Death Differ., https://doi.org/10.1038/s41418-018- References 0236-y (2019). 1. Jemal, A. et al. Global cancer statistics. CA Cancer J. Clin. 61,69–90 (2011). 25. Wang, A. et al. Long noncoding RNA EGFR-AS1 promotes cell growth and 2. Prensner, J. R. & Chinnaiyan, A. M. The emergence of lncRNAs in cancer metastasis via affecting HuR mediated mRNA stability of EGFR in renal cancer. biology. Cancer Discov. 1,391–407 (2011). Cell Death Dis. 10, 154 (2019). 3. Kapranov, P. et al. RNA maps reveal new RNA classes and a possible function 26. Shi, Y. et al. The long noncoding RNA SPRY4-IT1 increases the proliferation of for pervasive transcription. Science 316,1484–1488 (2007). human breast cancer cells by upregulating ZNF703 expression. Mol. Cancer 4. Huarte, M. & Rinn, J. L. Large non-coding RNAs: missing links in cancer? Hum. 14, 51 (2015). Mol. Genet. 19,R152–R161 (2010). 27. Zhao, Y. et al. The lncRNA MACC1-AS1 promotes gastric cancer cell metabolic 5. Liu, Q. et al. LncRNA loc285194 is a p53-regulated tumor suppressor. Nucleic plasticity via AMPK/Lin28 mediated mRNA stability of MACC1. Mol. Cancer 17, Acids Res. 41, 4976–4987 (2013). 69 (2018). 6. Peng, L. et al. Super-enhancer-associatedlongnon-codingRNA HCCL5is 28. Su, W. et al. Silencing of long non-coding RNA MIR22HG triggers cell survival/ activated by ZEB1 and promotes the malignancy of hepatocellular carcinoma. death signaling via oncogenes YBX1, MET, and p21 in lung cancer. Cancer Res., Cancer Res., https://doi.org/10.1158/0008-5472.can-18-0367 (2018). https://doi.org/10.1158/0008-5472.can-18-0222 (2018). 7. Sun, M. et al. The pseudogene DUXAP8 promotes non-small-cell lung cancer 29. Su, W. et al. Long noncoding RNA ZEB1-AS1 epigenetically regulates the cell proliferation and invasion by epigenetically silencing EGR1 and RHOB. Mol. expressions of ZEB1 and downstream molecules in prostate cancer. Mol. Ther., https://doi.org/10.1016/j.ymthe.2016.12.018 (2017). Cancer 16, 142 (2017). 8. Pauli, A., Rinn, J. L. & Schier, A. F. Non-coding RNAs as regulators of embry- 30. Wang, C. et al. A cancer-testis non-coding RNA LIN28B-AS1 activates driver ogenesis. Nat. Rev. Genet. 12,136–149 (2011). gene LIN28B by interacting with IGF2BP1 in lung adenocarcinoma. Oncogene, 9. Prensner, J. R. et al. Transcriptome sequencing across a prostate cancer cohort https://doi.org/10.1038/s41388-018-0548-x (2018). identifies PCAT-1, an unannotated lincRNA implicated in disease progression. 31. Sun, J. et al. Long noncoding RNA FGFR3-AS1 promotes osteosarcoma Nat. Biotechnol. 29,742–749 (2011). growth through regulating its natural antisense transcript FGFR3. Mol. Biol. Rep. 10. Gibb, E. A., Brown, C. J. & Lam, W. L. The functional role of long non-coding 43,427–436 (2016). RNA in human carcinomas. Mol. Cancer 10, 38 (2011). 32. Lv, J. et al. Long non-coding RNA Unigene56159 promotes epithelial- 11. Taft, R. J.,Pang, K. C.,Mercer, T. R., Dinger, M. &Mattick,J.S.Non-codingRNAs: mesenchymal transition by acting as a ceRNA of miR-140-5p in hepatocellular regulators of disease. J. Pathol. 220,126–139 (2010). carcinoma cells. Cancer Lett. 382,166–175 (2016). Official journal of the Cell Death Differentiation Association

Journal

Cell Death & Disease – Springer Journals

Published: Jun 17, 2019

There are no references for this article.