Long Noncoding RNA TCONS_00027385 Acts as a miR-874-5p Sponge to Suppress the Progression of Prostate Cancer Through Regulating ASCC2 Expression

Background: A novel pyrrolo indole alkaloids, named Robustanoids A, was isolated from Coffea canephora beans, and it inhibits proliferation of prostate cancer (PCa) cells. However, the molecular mechanism linking Robustanoids A to the tumorigenesis of PCa is not yet clear. Methods: We investigated the expression of lncRNAs in PCa cells with Robustanoids A and control group by microarray analysis. The expression level of TCONS_00027385 in PCa tissues and cell lines was detected by qRT-PCR. Additionally, we conducted functional experiments to investigate the biological effects of TCONS_00027385 on the development of PCa both in vitro and in vivo. Furthermore, bioinformatic analysis, luciferase reporter experiment, RIP assay, pulldown assay, and protein chip were performed to investigate the oncogenic molecular mechanisms of TCONS_00027385. Results: In our current study, we focused on TCONS_00027385, which was up-regulated in PCa tissues and cell lines. The high expression of TCONS_00027385 was related to the progression of PCa. Function assays revealed that silencing TCONS_00027385 inhibited PCa cell proliferation and induced apoptosis, while over-expression of TCONS_00027385 remarkably played an opposite role. A deeper investigation showed that TCONS_00027385 acted as a sponge for hsa-miR-874-5p in PCa, and ASCC2 was a target of miR-874-5p in the downstream. Moreover, a positive association between TCONS_00027385 with ASCC2 and a negative relationship between miR-874-5p and TCONS_00027385 (or ASCC2) were also founded. According to the rescue assay, inhibiting ASCC2 could partially suppress the oncogenic effect on cell proliferation and apoptosis in PCa caused by the overexpression of TCONS_00027385. Conclusion: TCONS_00027385 acted as a competing endogenous RNA (ceRNA) for miR-874-5p to regulate the expression of ASCC2. Transfection eciency of overexpression Lentivirus in 22RV1 and DU-145 cells were evaluated by qRT-PCR. d Colony formation assays. e-f The effects of TCONS_00027385 knockdown on the proliferation of 22RV1 and DU-145 cells were examined by MTT assay. g Flow cytometry was performed to determine the effect of TCONS_00027385 on apoptosis by ow cytometry analysis. h EdU assays were used to detect the proliferation rate of 22RV1 and DU-145 cells after TCONS_00027385 knockdown and overexpression. i TUNEL assays were used to detect the apoptosis rate of 22RV1 and DU-145 cells after TCONS_00027385 knockdown and overexpression.


Background
Prostate cancer (PCa) is the one of most frequent malignancy involving uncontrolled cell growth, and its high mortality imposes a heavy burden to patients, family and society (1)(2)(3). Furthermore, in some geographical regions such as North America, Africa and East Asia, the mortality rate of PCa has increased for at least two decades (4)(5)(6). Despite signi cant advance has been made in early diagnosis, it is often unable to be detected in a timely manner. In addition, a high relapse rate after androgen deprivation therapy and chemotherapy also resulted a high mortality in advanced PCa (7). Therefore, effective treatment strategies for PCa are urgently needed. Recently, emerging researches has emphasized that long noncoding RNAs (lncRNAs) might play as new biomarkers for cancer diagnosis as well as potential targets for cancer treatment (8).
LncRNAs have been found to play important roles in various physiological as well as pathological processes (9,10). Today, thousands of highly expressed lncRNAs have already been identi ed in humans, which are characterized by high tissue-speci cation and stable structures (11)(12)(13). Expression disturbance of lncRNA expression might lead to abnormal expression of genes and promote tumorigenesis and carcinogenesis (14)(15)(16). Recently, emerging researches suggested that lncRNAs might also participate in PCa, including cell proliferation, metastasis, migration, and invasion (17,18), however, the detailed associations between lncRNAs with PCa are largely unknown and need to be addressed.
Our previous research discovered a novel pyrrolo indole alkaloid (Robustanoids A) from C. canephora (a kind of robusta coffee beans) (19), which repressed proliferation and promoted apoptosis of PCa cells.
However, the antitumor mechanism triggered by Robustanoids A in PCa cells has never been elucidated. Therefore, based on our experimental results, we rst analyzed the expression pro les of lncRNAs by Robustanoids A in PCa cell through microarrays. We focused on hsa-TCONS_00027385 located on chromosome 19, and founded that the expression of TCONS_00027385 increased in PCa tissues and cell lines, which was related to the proliferation and apoptosis of PCa cells in vitro and tumor growth in vivo. Mechanistically, we discovered that TCONS_00027385 might act as a sponge of miR-874-5p, which could further upregulate the expression of activating signal cointegrator complex 2 (ASCC2). Our current evidence above suggested that TCONS_00027385 might play the role of a potential biomarker in the occurrence of PCa and provided a novel target in PCa clinical treatment.

Materials And Methods
Cell culture and reagents The cell lines used in this study provided by Stem Cell Bank of Chinese Academy of Sciencesand were cultured at 37 ℃ with 5 % CO 2 and 95 % air, as the instructions from the manufacturer. Brie y, cells of WPMY-1, VCaP and DU-145 were maintained in Dulbecco's Modi ed Eagle's Medium (12800017, GIBCO). Cells of 22RV1 and LNCaP were maintained in RPMI 1640 medium (31800022, GIBCO) and PC-3 were cultured in F12K medium (21127022, GIBCO). Cells were cultured in the complete mediums containing the above mediums respectively, 10 % fetal bovine serum (10099141C, GIBCO), 100 U/ml penicillin, and 100 μg/ml streptomycin.

Robustanoids A treatment on 22RV1 cells
Robustanoids A (Fig. 1a) was isolated and puri ed from C. canephora. 22RV1 cells were plated in 6-well plates followed by treated with 10 μM of Robustanoids A for 48 hr. In the control group, cells received 0.1% DMSO, which was tantamount to the amount in the experimental group.

Microarray analysis
After Robustanoids A treatment, cells were frozen by liquid nitrogen as soon as possible. According to Arraystar's protocol (Rockville, MD, USA), samples were prepared and microarray hybridization were performed. Data were further remained after deletion of repeat sequences and noncoding RNAs (ncRNAs) less than 200 bp. Subsequently, hybridization was performed in Arraystar Human LncRNA Microarray V5.0, intended for comprehensive analysis of human lncRNAs and protein coding transcripts, while Agilent Scanner G2505C (Jamul, CA, USA) was used to scan data. LncRNAs which were meeting the condition of fold-changes ≥ 2 and P-values < 0.05 were considered to be signi cantly differentially expressed.

Patients and specimens
PCa clinical samples and the corresponding normal samples were provided by 100 PCa patients who accepted surgically resection in the Urology Department of the First A liated Hospital of Xinjiang Medical University in 2020 and 2021 (the ethical number: 20210301-92). The enrolled patients were pathologically con rmed and did not have androgen deprivation treatment, chemotherapy, radiotherapy, or other anticancer treatment before operation. All individuals participating in this study signed informed consents, in addition, all procedures were carried out in accordance with the ethical standards of the First A liated Hospital of Xinjiang Medical University.

Subcellular fractionation
Nuclei and cytoplasm were separated using the PARIS TM Kit (AM1921, Thermo Fisher Scienti c) based on guides from the manufacturer.

Fluorescence in situ hybridization (FISH)
The FISH test was conducted in 22RV1 and DU-145 cells according to the manufacturers' instructions. The 5'FAM-labeled TCONS_00027385 probe (Additional le 1: S 1) in the current study was delegated to GenePharma (Shanghai, China) to design and synthesize. In brief, before permeabilization, cells were xed in 4% paraformaldehyde (PFA) for 20 min. Subsequentially, samples were incubated overnight in a corresponding probe at 37 ℃. Finally, DAPI (D9542, Sigma-Aldrich) was used to stain the cell nuclei. Observation and photography of staining results were conducted in the uorescence microscope (Zeiss LSM 880, Germany

MTT assay
The transfected cells were seeded in 96-well plates with a density of 1 × 10 5 cells/well. After the treatment, cells were added and incubated with 20 μl MTT solution (5 mg/ml) (ST1537, Beyotime) for another 4 hr. Subsequentially, after carefully removing supernatants, 100 μl DMSO was added for crystal dissolution. Finally, calculating the relative value of absorbance at 490 nm measured by a microplate reader (Bio-Rad, USA) to determine the proliferation of the cells.

EdU incorporation test
Using an EdU Apollo DNA in vitro kit (C0078L, Beyotime), cell proliferation was measured by ethynyl-2deoxyuridine incorporation test. In short, after the cells were transfected with corresponding vector and cultured, they were added and incubated with 50 μM EdU in 100 μl at 37 °C for 2 hr. Finally, cells observance were performed by a uorescence microscopy (Zeiss LSM 880, Germany). The result was based on at least three repeats.

TUNEL assay
Cells were seeded in dishes speci c for confocal microscope (801002, NEST Biotechnology), and cultured in serum-free medium for 24 hr. Subsequentially, xing cells in 4 % PFA at room temperature for 20 min.
Washing cells in 4 ℃ PBS and permeabilizing cells in 0.1 % Triton X-100 in 0.1 % sodium citrate for 10 min on ice, they were performed TUNEL staining mixture by the in situ Cell Death Detection kit, TMR red (11684817910, Roche) at 37 ℃ in the dark for 1 hr. Finally, cells were then rinsed in PBS and stained in DAPI solution for nuclei location. A LSM 880 (Zeiss, Germany) confocal microscope was used to observe uorescence.

Apoptosis detection
The cells were collected with trypsin without EDTA and resuspended in 490 μl binding buffer.
Subsequentially, samples were incubated in the dark with 5 μl Annexin V-FITC and 5 μl PI (BD 559763, BD Biosciences) for 20 mins, and then detected apoptosis stage using a FACSCalibur (BD, Biosciences, USA) within 1 hr.

Western blot detection
After cell proteins were extracted, they were separated on 10% SDS-PAGE gels, and then transferred to PVDF membranes (ISEQ00010, Millipore). Next, the membranes were blocked with 5% skimmed milk powder, and incubated with corresponding speci c antibody at 4 ℃ overnight. On the next day, after incubating with appropriate secondary antibody, the expression level of protein in the samples were detected by an ECL detection system (ChemiScope 3200 Mini, China), with GAPDH as a control. In Additional le 1: S4, we have presented all the antibodies used in this experiment.

Lentivirus-induced overexpression and knockdown
To achieve TCONS_00027385 knockdown (KD) in 22RV1 and DU-145 cells, lentiviral particles within shRNAs against TCONS_00027385 or the corresponding control were obtained from Genechem (Shanghai, China). According to a previous research (20), three potential shRNAs (TCONS_00027385-KD1, TCONS_00027385-KD2, and TCONS_00027385-KD3) and a control one (TCONS_00027385-con) were designed for synthesis; the detailed sequences were presented in additional le 1: Table 2. Using polybrene transfection reagent for lentivirus transduction, and using an enhanced green uorescent protein (EGFP) to verify and estimate the e ciency of transfection (additional le 1: Figure 1-3). Subsequentially, screening positive transformants with puromycin (MA0318, Meilun Biotechnology), the selected clones were ampli ed and the KD e ciency measured by qRT-PCR was analyzed. The details of experimental method for stably over-expressing cell lines can be found in additional le 1: S5 & Figure 4.

RIP test
Based on instruction for manufacturer, using the Magna RIP RNA-Binding Protein Immunoprecipitation Kit (Millipore, MA) to perform the RIP test. Lysing the cells and incubating with Ago2 and IgG. After the cell was mixed with anti-IgG and anti-Ago2 in RIP buffer (Millipore), the RNAs in precipitation were preserved to further sequential analyze.

Quanti cation of proteins by antibody array
The high-throughput protein pro les were performed using the protein array platform (AAH-APO-1-8, RayBiotech Life) based on the manufacturer's guide. Using centrifugation to clarify cell samples (control group and sh-1 group), and before application to the arrays, a total protein concentration within the working range was obtained. Details for the experimental methods can be found in additional le 1: S6.

Zebra sh experiments
The zebra sh experiments were completed on the zebra sh platform of Zhejiang University School of Medicine (the ethical number: 2021-20515#). After fertilization, zebra sh (Danio rerio) eggs were cultured in Danieau's solution at 28 ℃ according to standard laboratory conditions. At 48 hr after fertilization, the membranes out of the embryos were carefully removed with forceps, and the embryos were anesthetized in 0.04 mg/ml tricaine. The microinjection was performed to the embryos after transferred to a modi ed agarose gel. Using a Pneumatic Picopump and a manipulator (WPI) to inject approximately 400 DU-145 cells on the ventral end of the Cuvier Duct, where which entered embryo's heart. The injections were repeated in at least 15 embryos in each group. The survival rate less than 85 % in the control group was regarded to the demarcation line for abandonment. After implantation, embryos were cultured at 33 ℃ (21). Every other day, the growth of tumor was observed through a SMZ18 uorescence microscope (Nikon, Japan) while green pixels data were quanti ed by NIS-Elements imaging software (D 4.50.00).

Tumor xenograft model
After 22RV1 cells (1×10 7 cells) with or without TCONS_00027385-KD1 (sh-1) were suspended in 200 μl PBS, they were injected subcutaneously into nude mice of 4-6 weeks old in each side. 30 days after injection, the mice were sacri ced for measurements of tumors in the terms of maximum (L) and minimum (W) length and weight. This animal experiment had been approved by the Animal Care and Use Committee of Zhejiang University (the ethical number: 2021-20515#).

Statistical analysis
Analyzing the data of at least three in-dependent tests with the GraphPad Prism 7.04 software (La Jolla, USA) and expressing them in the form of mean ± S.D. The student's t-test was performed to compare the differences between two groups, while the survival rate was calculated through Kaplan-Meier survival analysis. In addition, analysis of Cox proportional hazard model multivariate was conducted to measure the signi cance of TCONS_00027385 expression and clinicopathological characteristics on overall survival. Statistically signi cant was de ned as P-value < 0.05.

Results
Expression of TCONS_00027385 increased in PCa and predicted a poor prognosis Microarray analysis showed lncRNA expression variations in Robustanoids A (+) group and control group. The box plot, scatter plot, and volcano plot demonstrated the expression variations of lncRNA between two groups ( Fig. 1b-d). A total of 1366 differentially expressed lncRNAs were screened out, with a condition of the folding changes greater than 2.0, and the P-value < 0.05. Among them, 816 lncRNAs were upregulated while 550 were downregulated (Additional le 3). In addition, the cluster heat map showed the more than 2.5-fold changes in differentially expressed lncRNAs (Fig. 1e). Incidentially, Table 1 had listed the lncRNAs of top 10 up-regulated and down-regulated. Expressions of these 20 lncRNAs were veri ed by RT-PCR, and 18 lncRNAs were validated successfully. First, we found that the expression of TCONS_00027385 in Robustanoids A (+) group was downregulated by sevenfold (P < 0.001). Next, compared with the normal samples, TCONS_00027385 was signi cantly up-regulated in PCa tissues (Fig.  1f). Furthermore, the up-regulated expression of TCONS_00027385 was remarkably related to age, tumor size, and clinical stage (Fig. 1g-h), however, no obvious relationship was found between lymph node metastasis or tumor differentiation (The results were not shown). Kaplan-Meier survival curves showed that compared to the group with low TCONS_00027385 level, high TCONS_00027385 expression was associated with a lower overall survival rate (Fig. 1i). Additionally, compared to prostate stromal immortalized cell line, the TCONS_00027385 expression was upregulated remarkably of PCa cell lines (Fig. 1j). In general, the expression of TCONS_00027385 increased in PCa, which might be related to clinical progress and poor prognosis in PCa patients. Based on the human reference genome (GRCh37/hg19) in the UCSC genome database (http://genome.ucsc.edu), TCONS_00027385 is located at chr19:47742377-47747476. Thus, we measured and found the genomic length of the lncRNA TCONS_00027385 was 1152 bp (see the Additional le 2: S 1 for detail).

TCONS_00027385 decreased proliferation and facilitated apoptosis in PCa cells
In order to investigate whether TCONS_00027385 might participate in proliferation and apoptosis of cells, we conducted experiments on gain and loss of function. The utility of TCONS_00027385 downexpression (shRNA & siRNA) and TCONS_00027385 over-expression were con rmed in RT-PCR assays (Fig. 2a-c). Cell proliferation and apoptosis detection assays were also conducted. The MTT, colony formation assays and EdU assay indicated that knocking-down TCONS_00027385 reduced proliferation in 22RV1 and DU-145 cells, while over-expression of TCONS_00027385 remarkably played an opposite role (Fig. 2d-h). Moreover, TUNEL assays and ow cytometry were utilized to prove that knocking-down TCONS_00027385 signi cantly promoted cell apoptosis (Fig. 2i-j). For further con rmation, we used protein chip to measure the expression of apoptosis markers in PCa cells. As expected, chip analysis showed differential expression proteins (DEPs) variations in Robustanoids A (+) group and control group. A total of 19 differentially expressed proteins has been screened out, with the adjusted P-value less than 0.05 and the folding change more than 1.2 or less than 0.83 (absolute logFC > 0.263) ( Table 3). The scatter plot and volcano plot demonstrated the changes in expressions of protein between the two groups ( Fig. 3a-b). The PCA and heatmap were performed on all DEPs (Fig. 3c-d), with plotting the rst two principal components to show the difference between the two groups. Using R package "clusterPro ler", we conducted analysis on the protein function annotation Gene Ontology (GO) and KEGG pathway. The three subtypes in GO analysis containing biological process (Fig. 3e), molecular function (Fig. 3f), and cellular component (Fig. 3g). KEGG analysis relating genomic information to high-level functional information, was a systematic analysis of gene function (Fig. 3h). In summary, TCONS_00027385 decreased PCa cell proliferative and promoted apoptosis capacities.

TCONS_00027385 knockdown inhibited tumor growth
To further investigate the bio-functions of TCONS_00027385 on the growth of tumor, we performed experiments on subcutaneous xenograft tumor models and zebra sh. In our established zebra sh model, representative uorescence microscopic images of both the experimental and the control groups are presented in Fig. 3i. The sh-TCONS_00027385 (sh-1) sh bodies had a declining uorescence intensity, however, we observed an opposite result in the TCONS_00027385 over-expression group (Additional le 2: S 3-4). In addition, in the pectoral region of nude mice, 22RV1 cells were implanted subcutaneously, with eight mice in negative control group and sh-1 group, respectively. For the next four weeks, the volumes of tumors were measured every 7 days. As we expected, the silence of TCONS_00027385 remarkably inhibited tumor growth in vivo (Figure 3j-k). Therefore, the above results indicated that TCONS_00027385 might be able to regulate the progression of PCa.
TCONS_00027385 played a role of a sponge for miR-874-5p In order to investigate intrinsic mechanism of TCONS_00027385 on PCa carcinogenesis, we located the expression of TCONS_00027385 subcellularly. The results indicated that TCONS_00027385 was mainly dispersed in cytoplasm ( Fig. 4a-b), suggesting TCONS_00027385 might perform the biological function by sponging miRNA. Then, using Miranda software and Targetscan, we determined a candidate microRNA (miR-874-5p) as well as predicted the probable targets of TCONS_00027385 ( Fig. 4c-d) in the downstream. The results from luciferase reporter experiment veri ed the luciferase activity of WT-TCONS_00027385 was signi cantly reduced by miR-874-5p mimics, however, it did not change remarkably in Mut-TCONS_00027385 (Fig. 4 e). In the other hand, the RIP test further showed that compared to the IgG group, TCONS_00027385 and miR-874-5p were accumulated in beads linked to Ago2 (Fig. 4f). In addition, the overexpression of WT-TCONS_00027385, but not Mut-TCONS_00027385, reduced the expression of miR-874-5p in PCa cells (Fig. 4g). Furthermore, overexpressing TCONS_00027385 greatly reduced expression level of miR-874-5p in both 22RV1 and DU-145 cells, while silence of TCONS_00027385 resulted an increase in expression level of miR-874-5p, with shRNA NC group acting as an internal control (Fig. 4h-i). Subsequentially, we further found the level of miR-874-5p in PCa tissues was lower than which in normal condition, with a negative association between expression levels of TCONS_00027385 expression with miR-874-5p (Fig. 4j). To excavate more evidence, we conducted in vivo experiments. The expression level of miR-874-5p in the tumors from nude mice collected previously was higher in the TCONS_00027385 knock-down cell line, but in the cells overexpressing TCONS_00027385, the expression level was lower (see the Additional le 2: S 2 for detail). In addition, TCONS_00027385 encouraged the proliferation and apoptosis of cells at least in part through sponging miR-874-5p (Fig. 2e-f). Above all, TCONS_00027385 accelerated the progress of PCa by sponging miR-874-5p.

ASCC2 was a target of miR-874-5p in the downstream in PCa
Through base pairing with 3′UTR, the post-transcriptional effects of miRNAs often resulted protein synthesis inhibition (22). Thus, in order to determine the detailed regulation mechanism of TCONS_00027385 in PCa, we searched TargetScan database and found that ASCC2 might be the target of miR-874-5p in the downstream (Fig. 4k). In order to con rm this prediction, we conducted a luciferase report experiment, and the results turned out that ASCC2 acted as a direct target of miR-874-5p (Fig. 4l). Furthermore, the expression of ASCC2 was down-regulated by silence of TCONS_00027385, and miR-874-5p mimics could partially offset the corresponding increases of the expression of ASCC2 caused by the over-expression of TCONS_00027385 in PCa cells (Fig. 4m-p). Consistently, we observed the same result in the protein level ( Fig. 5a-b). The above observation indicated that TCONS_00027385 could in uenced the expression of ASCC2 through its interaction with the miR-874-5p.
We also tested the level of ASCC2 in tumor tissues. Comparing with neighboring normal tissues, ASCC2 was remarkably increased in PCa tissues, and we found a positive association between the expression level of ASCC2 with TCONS_00027385, and a negative relationship between the expression of ASCC2 and the level of miR-874-5p (Fig. 5c). Moreover, we tested the expression of ASCC2 in vitro, while the transfection e ciency has been determined previously (Fig. 5d-e). As expected, by knocking down TCONS_00027385, the expression of ASCC2 in 22RV1 and DU-145 cells was signi cantly reduced, while an increase in expression was observed in cells over-expressing TCONS_00027385 (Fig. 4m-n). So far, we have veri ed that ASCC2 acted as a direct target of miR-874-5p. Nevertheless, the effects of ASCC2 in prostate cancer has far from been completely clari ed.

ASCC2 inhibition suppressed the tumorigenesis effects of TCONS_00027385
At last, we carried out functional experiments to examine the effects of ASCC2 by exploring the effects of TCONS_00027385 mediated by ASCC2 in aspect of tumor growth promotion. To investigate whether the inhibition of ASCC2 could reverse the carcinogenic effect of TCONS_00027385, we conducted ASCC2 knocking-down in 22RV1 and DU-145 cells. According to rescue experiments, silence of ASCC2 partially reversed the effect on cell proliferation and apoptosis caused by overexpression of TCONS_00027385 ( Fig. 5f-i). In summary, these results indicated that TCONS_00027385 competitively binded to miR-874-5p and subsequentially resulted in an up-regulation of the expression of ASCC2 (Fig. 5j), becoming a key tumor-promoting factors for PCa.

Discussion
Recently, because of the high incidence and poor prognosis, PCa has received widespread attention (23)(24)(25). Emerging researches have demonstrated that lncRNAs acted an essential role in the occurrence and development of PCa (26-28). For example, lncRNA-PCAT1 disrupted the complex of PHLPP/FKBP51/IKKα and promoted AKT and NF-κB signaling pathways. The expression of lncRNA-PCAT1 was positively correlated with the progress of CRPC (29). The function of LINC00844 was trans, affecting the transcription of genes regulated by androgen, and preventing the migration and invasion of prostate cancer cell (30). As a new AR translational regulator, lnc-LBCS inhibited the castration resistance of prostate cancer by connected with hnRNPK (31). Our current study discovered that the expression of TCONS_00027385 was remarkably up-regulated in PCa tissues and cell lines, while it was related to the development and prognosis in PCa patients, indicating TCONS_00027385 might be involved in the progression of PCa. Additionaly, we founded that TCONS_00027385 promoted the proliferation and apoptosis of PCa cells in vitro and tumor growth in vivo. However, the basic molecular mechanisms hasn't been elucidated.
Recently, emerging researches have con rmed that the ceRNA mechanism took part in various diseases (32)(33)(34)(35)(36)(37), which has made great progress in PCa. For instance, the lncRNA UCA1 exerted carcinogenic activity through sponging miR-143 and then upregulating the expression of MYO6 in PCa (38). The lncRNA FOXP4-AS1 played a role as a ceRNA to sponge miR-3184-5p and the regulate FoxP4 in posttranscription (39). The lncRNA BLACAT1 could regulate proliferation and metastasis capabilities of prostate cancer cells, and might able to be the ceRNA to modulate the expression level of DVL3 through sponging miR-29a-3p (40). However, the impact of TCONS_00027385 on the development of PCa is still unexplored. Therefore, it is worth exploring whether the network of ceRNA involving miR-874-5p and TCONS_00027385 existed in PCa cells. In the current study, subcellular fractionation experiments showed that TCONS_00027385 was mainly existed in cytoplasm, providing the potential for TCONS_00027385 to function as a ceRNA in the course of disease in PCa. Then, the online database ENCORI (41), miRWalk (42,43) and TargetScan  The ceRNA mechanism pointed out that because of miRNAs competitively bind to lncRNAs, the expression of mRNA would increase. The upstream miRNA of ASCC2 in PCa was found by means of bioinformatics analysis. Although, ASCC2 was rst reported in the study of Lee YH (47), the researches on it in human cancers were still limited (48). In our study, ASCC2 was predicted to be a direct target of miR-874-5p through TargetScan. The luciferase reporter experiment and RIP test further enhanced the postulation of direct binding between TCONS_00027385 and miR-874-5p. ASCC2, as an oncogene, whose expression level had a positive relationship with TCONS_00027385, and a negative correlation with miR-874-5p. Moreover, according to rescue trials, silence of ASCC2 partially abolished the tumorigenic effects of TCONS_00027385.

Conclusions
In conclusion, our current research indicated that the new ceRNA network of TCONS_00027385/miR-874-5p/ASCC2 axis contributing to the development of PCa, which might reveal a novel biomarker of PCa and also provide new enlightenment for PCa treatment.

Consent for publication
Not applicable.

Availability of data and materials
All of the source data are available from the corresponding authors upon reasonable request. Microarray data was listed in Additional le 3.

Competing interests
The authors declare that they have no competing interests.
Funding Figure 1 LncRNA expression pro le in PCa. a Robustanoids A structure. b Box Plot presents the normalized intensity value in LncRNA expression between control and B group samples. c The scatter plot was used for assessing the variation in LncRNA expression between control and Robustanoids A (+) group samples. d The volcano plot was constructed using fold-change values and P-values. e The cluster heat map showed the differentially expressed lncRNAs over 2.5-fold change. f TCONS_00027385 expression was detected in PCa tissue and adjacent normal tissue by qRT-PCR. g-h Associations between TCONS_00027385 expression and tumor size or clinical stage were detected by qRT-PCR. i Kaplan-Meier analysis was used to assess the relation between TCONS_00027385 expression level and overall survival in PCa patients. *p<0.05, **p<0.01, ***p<0.001. All experiments were repeated at least for three times and mean ± SD was used to represent the nal result. j qRT-PCR was applied to con rm the expression level of TCONS_00027385 in PCa cell lines and human normal prostate stromal immortalized cell line.   GO analysis includes: biological process (Fig. 3e), molecular function (Fig. 3f) (Fig. 3g). KEGG is systematic analysis of gene functions, linking genomic information with higher order functional information (Fig. 3h). i Zebra sh assays were used to detect the proliferation rate of DU-145 cells after TCONS_00027385 knockdown and overexpression. Tumor burden was quanti ed 1dpi (days post injection). Green = PCa cells. j-k The tumor volumes and weights of sh-TCONS_00027385 group compared with NC group were quanti ed. Tumor volumes were analyzed by ANOVA. Figure 4 TCONS_00027385 was a sponge for miR-28-5p. a The expression level of TCONS_00027385 in the subcellular fractions of 22RV1 and DU-145 cells were detected by qRT-PCR. U6 and GAPDH were used as nuclear and cytoplasmic markers, respectively. b The location of TCONS_00027385 (green) in 22RV1 and DU-145 cells were determined by RNA FISH assay. DAPI-stained nuclei are blue. c ceRNA analysis for TCONS_00027385. Cytoscape was used to visualize lncRNA TCONS_00027385-miRNA-target gene interactions. d Schematic diagram representing the predicted binding sites for miR-874-5p in TCONS_00027385. e-g Luciferase reporter assay and RIP assay was performed to demonstrate that miR-874-5p was a downstream target of TCONS_00027385. h-i The miR-874-5p expression levels under TCONS_00027385 silencing and TCONS_00027385 overexpression were evaluated in vitro. j The expression of miR-874-5p in PCa tissue and normal tissue were detected by qRT-PCR. k Schematic diagram representing the predicted binding sites for ASCC2 in miR-874-5p. l Luciferase reporter assay was performed to determine the association between miR-874-5p and ASCC2. m-n The ASCC2 mRNA expression levels under TCONS_00027385 silencing and TCONS_00027385 overexpression were evaluated in vitro. o-p After miR-874-5p mimics, the mRNA expression of ASCC2 was evaluated by RT-PCR and western blot in 22RV1 and DU-145 cells. (*P < 0.05, **P < 0.01, and ***P < 0.001) Figure 5 ASCC2 was a downstream target of miR-874-5p in PCa. a-b The correlation between ASCC2 and TCONS_00027385 as well as the correlation between ASCC2 and miR-874-5p were analyzed by Spearman's rank correlation test. c Relative ASCC2 expression in tumor tissue and normal tissue were detected by qRT-PCR. d The expression levels of ASCC2 in 22RV1 and DU-145 cells after transfection with si-NC or si-ASCC2 were detected by RT-PCR. e The expression levels of ASCC2 in 22RV1 and DU-145 cells after transfection with si-NC or si-ASCC2 were detected by western blot. f The TCONS_00027385 expression levels under ASCC2 silencing were evaluated in vitro. g Flow cytometry was performed to determine the effect of ASCC2 on apoptosis by ow cytometry analysis. h-i MTT assay was performed to detect proliferation of cells transfected with ASCC2 siRNA and cells co-transfected with ASCC2 siRNA and miR-874-5p mimics. j The schematic diagram of the oncogenic role of TCONS_00027385 in PCa cells. TCONS_00027385 functions as a miRNA sponge to positively regulate ASCC2 expression through sponging miR-874-5p and subsequently promotes malignant phenotypes of PCa cells, thus playing an oncogenic role in prostate cancer pathogenesis.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.