OR Oncology Reports 1021-335X 1791-2431 D.A. Spandidos 10.3892/or.2018.6766 or-40-06-3171 Articles Downregulation of miR-224-5p in prostate cancer and its relevant molecular mechanism via TCGA, GEO database and in silico analyses GanBin-Liang 1 ZhangLi-Jie 2 GaoLi 3 MaFu-Chao 1 HeRong-Quan 1 ChenGang 3 MaJie 1 ZhongJin-Cai 1 HuXiao-Hua 1 Department of Medical Oncology, First Affiliated Hospital of Guangxi Medical University, Nanning, Guangxi Zhuang Autonomous Region 530021, P.R. China Department of Ultrasound, First Affiliated Hospital of Guangxi Medical University, Nanning, Guangxi Zhuang Autonomous Region 530021, P.R. China Department of Pathology, First Affiliated Hospital of Guangxi Medical University, Nanning, Guangxi Zhuang Autonomous Region 530021, P.R. China Correspondence to: Professor Xiao-Hua Hu, Department of Medical Oncology, First Affiliated Hospital of Guangxi Medical University, 6 Shuangyong Road, Nanning, Guangxi Zhuang Autonomous Region 530021, P.R. China, E-mail: gxmuhxh@163.com 122018 03102018 40 6 3171 3188 19022018 31072018 Copyright: © Gan et al. 2018 This is an open access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made.

The function of the expression of microRNA (miR)-224-5p in prostate adenocarcinoma (PCa) remains to be elucidated, therefore, the present study aimed to investigate the clinical significance and potential molecular mechanism of miR-224-5p in PCa. Data on the expression of miR-224-5p in PCa were extracted from The Cancer Genome Atlas (TCGA), Gene Expression Omnibus (GEO), ArrayExpress and previous literature, and meta-analyses with standardized mean difference (SMD) and summary receiver operating characteristic (sROC) methods were performed for statistical analyses. The prospective target genes of miR-224-5p were collected by overlapping the differentially expressed mRNAs in TCGA and GEO, and target genes predicted by miRWalk2.0. Subsequently, in silico analysis was performed to examine the associated pathways of miR-224-5p in PCa. The expression of miR-224-5p was markedly lower in PCa; the overall SMD was −0.562, and overall sROC area under the curve was 0.80. In addition, Kyoto Encyclopedia of Genes and Genomes analysis revealed that the prospective target genes of miR-224-5p were largely enriched in the amino sugar and nucleotide sugar metabolism signaling pathway, and three genes [UDP-N-acetylglucosamine pyrophosphorylase 1 (UAP1), hexokinase 2 (HK2) and chitinase 1 (CHIT1)] enriched in this pathway showed higher expression (P<0.05). In addition, key genes in the protein-protein interaction network analysis [DNA topoisomerase 2-α (TOP2A), ATP citrate lyase (ACLY) and ribonucleotide reductase regulatory subunit M2 (RRM2)] exhibited significantly increased expression (P<0.05). The results suggested that the downregulated expression of miR-224-5p may be associated with the clinical progression and prognosis of PCa. Furthermore, miR-224-5p likely exerts its effects by targeting genes, including UAP1, HK2, CHIT1, TOP2A, ACLY and RRM2. However, in vivo and in vitro experiments are required to confirm these findings.

 microRNA-224-5p prostate cancer molecular mechanism The Cancer Genome Atlas Gene Expression Omnibus in silico Introduction

According to cancer statistics from 2017, prostate adenocarcinoma (PCa) was ranked highest for newly diagnosed cases of cancer in men in the USA, and was considered the 3rd leading cause of cancer-associated mortality in men (1). Although the rates of incidence and mortality vary in different nations and regions, PCa poses a health risk to men due to its high incidence and mortality rates (2–4). PCa is a heterogeneous type of cancer; patients with PCa are prone to a relapse and metastasis, and the prognosis of PCa is associated with the age of patients (5,6). The majority of patients are diagnosed with PCa in its intermediate or terminal stage, which presents challenges to treatment and recovery of patients. Fortunately, the diagnosis and treatment of PCa have been modified owing to medical advances. For example, screening prostate-specific antigen in the early stage is considered an effective approach for early diagnosis and immediate treatment of PCa (7). Despite this, the incidence and mortality rates of PCa remain high, therefore, further investigations are urgently required to clarify the onset and mechanism of progression of PCa (8).

Based on previous studies, in addition to obesity, gene expression is closely associated with the onset and development of PCa (9,10). MicroRNAs (miRNAs) are small, non-coding RNA molecules of ~ 22 nucleotides, which perform functions in the post-transcriptional regulation of gene expression (11–15). miRNAs act as oncogenes or antioncogenes in tumorigenesis, thus influencing the onset and development of tumors (16–20). Previous studies have shown that the downregulated expression of miRNA may be essential in the onset and development of PCa (21–23). In addition, increased expressed of miRNA was likely to regulate the expression of target genes and exert its influences on the progression of PCa (24,25).

Currently, investigations on the role of miRNA-224-5p (miR-224-5p) in PCa have been limited. Mavridis et al performed a case-control study involving 73 cases of PCa and 66 cases of benign prostatic hyperplasia, demonstrating that the expression of miR-224-5p was downregulated in PCa tissues and, with disease progression, the expression of miR-224-5p was decreased further; patients with the lower expression of miR-224-5p were also more likely to have a poorer prognosis (26). Fu et al investigated the targeting association between miR-224-5p and calcium/calmodulin-dependent protein (CAMKK2) in tissues from 20 cases of PCa and non-cancerous counterpart tissues; it was found that the expression of miR-224-5p was markedly lower in PCa tissues, and patients with a lower expression of miR-224-5p tended to have a poorer survival rate; additionally, miR-224-5p inhibited the proliferation of tumor cells by targeting CAMKK2 (27). Having applied reverse transcription-quantitative polymerase chain reaction analysis to examine tissues from 36 cases of PCa and 14 non-cancerous tissues, Kristensen et al confirmed that the expression of miR-224-5p was decreased in PCa tissues and predicted an unsatisfactory outcome for patients (28). These previous studies offer important insights into the effects of miR-224-5p on PCa tissues, however, they were performed with a small sample size (n<80) and the results were not confirmed with a larger sample size, which may reduce the convincingness of the results. Furthermore, the studies mentioned above failed to perform bioinformatics analyses, which may assist in identifying more prospective target genes of miR-224-5p in PCa. Therefore, further verification is required of the expression of miR-224-5p and its clinical significance in PCa. In addition, the mining of multiple databases and bioinformatics analyses is required to examine the prospective molecular mechanism underlying the role of miR-224-5p in PCa.

The present study aimed to verify the expression of miR-224-5p in PCa using data from different databases, including The Cancer Genome Atlas (TCGA), Gene Expression Omnibus (GEO), ArrayExpress, and previous literature. In addition, prospective target genes of miR-224-5p were collected using online prediction tools and differentially expressed genes in TCGA and GEO. Bioinformatics analyses were also used to further examine the signaling pathways of miR-224-5p in PCa (Fig. 1A).

 Materials and methods <sec> <title>Collection of PCa data from TCGA

Data of Illumina HiSeq level 3 were acquired from the Launch ata portal of TCGA (https://cancergenome.nih.gov/). Following summarizing of the raw count data, raw data on miRNA expression in tissues of 500 cases of PCa and 52 normal prostatic tissues were obtained, from which reads per million data of pre-miR224 were extracted, including tissues from 498 cases of PCa and 51 normal PCa tissues. In these 500 cases, the age of the patients ranged between 41 and 78 years, with a mean age of 61 years. Additional clinicopathological parameters are listed in Table I.

 Screening of differentially expressed microarrays of miRNAs in PCa

In the microarray GEO (https://www.ncbi.nlm.nih.gov/gds/) and ArrayExpress (https://www.ebi.ac.uk/arrayexpress/) databases, a search was performed using the following key words: (parastata OR prostatic gland OR prostate gland OR prostat* AND (cancer OR carcinoma OR adenocarcinoma OR tumor OR malignan* OR neoplas*) AND (miR OR miRNA OR microRNA OR miR OR miRNA OR microRNA OR ‘miR’ OR ‘miRNA’ OR ‘microRNA’). Subsequently, the differentially expressed microarrays of miRNAs in PCa were screened and downloaded. Finally, studies were included which fulfilled the following criteria: i) compared the cancerous tissues with the controls; ii) contained microarrays of miRNA expression in PCa tissues, biofluids and cell lines; and iii) had more than 3 samples in each microarray. The procedures for the search are shown in Fig. 1B.

 Literature search

From the PubMed (https://www.ncbi.nlm.nih.gov/pubmed), Embase (https://embase.com/) and Web of Science (http://apps.webofknowledge.com/UA_General Search_input.do?product=UA&search_mode=GeneralSearch&SID=8ErJvGamhqk4GR7zNPu&preferencesSaved=) databases, studies on differentially expressed miR-224-5p in PCa tissues and in non-cancerous controls were collected. The key words used for the search included the following: (parastata OR prostatic gland OR prostate gland OR prostat*) and (cancer OR carcinoma OR adenocarcinoma OR tumor OR malignan* OR neoplas*) and (miR-224 OR miRNA-224 OR microRNA-224 OR miR224 OR miRNA224 OR microRNA224 OR ‘miR 224’ OR ‘miRNA 224’ OR ‘microRNA 224’ OR miR-224-5p OR miRNA-224-5p OR microRNA-224-5p). The studies were included if they met the following standards: i) involved the comparison of PCa with non-cancerous controls in tissues, biofluids and cell lines; and ii) provided the mean ± standard deviation (SD) or diagrams from which data extraction was possible. The procedures for the literature screening are shown in Fig. 1C.

 Collection of differentially expressed genes of PCa in TCGA

Gene expression profiling interactive analysis (GEPIA; http://gepia.cancer-pku.cn/), a visualized website based on TCGA database developed by Peking University (Beijing, China), contains various functional analyses, including the comparison of differentially expressed genes in cancerous and non-cancerous tissues (29). Data on the differential genes of PCa on GEPIA were retrieved, and differentially expressed genes calculated using the Linear Models for Microarray Data (LIMMA) package (http://gepia.cancer-pku.cn/detail.php?gene=) were downloaded. The genes were selected for further analysis if log2 fold change (FC)>1.

 Selection of microarrays of differentially expressed genes of PCa

In the GEO and ArrayExpress databases, a search was performed using the aforementioned key words for prostate and cancer. The relevant microarrays were included in the study if they conformed to the following criteria: i) investigation of differentially expressed mRNA based on post-transcriptional miR-224-5p; and ii) comparison of PCa cell lines and normal cell lines. If there existed numerous similar samples in one microarray, the intersections were obtained. When dealing with different microarrays, the unions were obtained. Further analyses were performed on all results.

 Prediction of potential target genes of miR-224-5p

The microRNA-mRNA prediction was performed with the miRWalk2.0 (http://zmf.umm.uni-heidelberg.de/apps/zmf/mirwalk2/) online prediction tools involving 12 prediction tools, namely, DIANA microT v4, RNA22, Pictar2, miRWalk, miRNAMap, RNAhybrid, mirBridge, TargetScan, miRMap, miRanda, PITA and miRDB. Genes that were predicted by three tools qualified for the present study. In order to acquire the potential target genes with accuracy, the overexpressed genes in TCGA, mRNAs expressed at low levels following miR-224-5p transcription and the predicted microarrays were combined, and the unions were obtained. Bioinformatics analysis was performed on these results.

 In silico analysis

Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses were performed on the genes that appeared in TCGA, GEO and target genes prediction tools on Database for Annotation, Visualization and Integrated Discovery (DAVID; http://david.abcc.ncifcrf.gov/). Bingo on Cytoscape 3.5.0 (http://www.cytoscape.org/) was applied to construct network analysis of GO terms, and the ClueGO and CluePedia plugins were used to establish the KEGG network. For genes enriched in significant pathways, the data in TCGA were used to verify their expression levels in PCa. In addition, protein-protein interaction (PPI) analysis was performed, a PPI network was constructed and the interactive associations between proteins were confirmed on the Search Tool for the Retrieval of Interacting Genes 0.5 (https://string-db.org/cgi/input.pl?sessionId=Ce1Dx9pYDluc&input_page_show_search=on) (30–37) database. Based on TCGA data, the mRNA expression of key genes in the PPI network were also confirmed, and the mechanism of miR-224-5p in PCa was further examined.

 Statistical analysis

SPSS 23.0 (IBM Corp., Armonk, NY, USA) was used for statistical analysis of the expression of miR-224-5p in PCa. An independent t-test was applied to evaluate the differentials of miR-224-5p between PCa tissues and non-cancerous tissues, and the results are presented as the mean ± SD. The standardized mean difference (SMD) was used to combine all the included studies on STATA 2.0 (StataCorp, College Station, TX, USA), and to calculate the expression trend of miR-224-5p in PCa. In addition, a receiver operating characteristic (ROC) was used to analyze the sensitivity and specificity of each study, and their cut-off value was calculated. Subsequently, the cut-off value was applied to identify the true positive, false positive, false negative and true negative of each study, following which a diagnostic test four-fold table was produced. STATA 12.0 was then used to confirm the expression of miR-224-5p in PCa, and the summary ROC (sROC) was used to measure its credibility. In addition, in order to examine the expression trend in each study, scatter diagrams were produced to show the expression of miR-224-5p in PCa tissues and adjacent tissues via GraphPad Prism 5.0 (GraphPad Software, Inc., La Jolla, CA, USA). P<0.05 was considered to indicate statistically significant difference.

 Results <sec> <title>Expression of miR-224-5p in PCa

According to the data in TCGA, the expression of pre-miR-224 in PCa tissues was 5.0255±1.1945 (PCa, vs. normal, 498, vs. 52), which was significantly lower than that of normal adjacent tissues (5.4773±0.8645, FC=0.9175, P=0.001; Fig. 2A). The area under the curve (AUC) of the downregulated pre-miR-224 in PCa tissues was 0.614 (95% CI, 0.542, 0.686. P=0.007; Fig. 2B). It was found that pre-miR-224 tended to exhibit lower expression with the progression of clinical staging by comparing T3-4 and T1-2 (4.9052±1.3362 vs. 5.1306±1.2187, FC=0.9561, P=0.061), M1 and M0 (3.9179±1.7677 vs. 4.9999±1.2525, FC=0.7836, P=0.087; Table I). However, no clear associations were found between its expression and prognosis or other types of staging.

Based on the inclusion criteria, a total of 17 microarrays were eventually considered eligible for the present study, which were categorized into three subtypes: Tissues, biofluids and cell lines (Fig. 3). In the subgroup of tissues, 12 microarrays were included, among which the expression of miR-224-5p was notably lower in PCa tissues in GSE76260, GSE21036, GSE34932, GSE60371 and GSE36802. In GSE54516 and GSE64318, the expression of miR-224-5p was upregulated. In terms of SMD, a low expression of miR-224-5p was identified in PCa tissues: Sub-SMD (95% CI)=−0.304 (−0.452, −0.156) (Table II and Fig. 3A; P<0.001) PCa, vs. normal=421, vs. 253); sROC AUC (95% CI)=0.80 (0.77, 0.84) (Fig. 4). The optimum sensitivity and specificity were 0.82 (95% CI: 0.65, 0.91) and 0.61 (95% CI: 0.34, 0.83, Fig. 4), respectively. The scatter diagram and ROC curve of expression of each microarray are shown in Figs. 5 and 6. When analyzing the biofluids subgroup, four microarrays were included (GSE39314, GSE61741, GSE24201 and GSE49298). The SMD result indicated that Sub-SMD (95% CI)=−0.216 (−0.544, −0.111), P=0.195 (Table III and Fig. 3B) PCa, vs. normal, 92, vs 62. sROC analysis revealed the following: AUC (95% CI)=0.81 (0.78, 0.84), sensitivity (SENS; 95% CI)=0.71 (0.34, 0.92), specificity (SPEC; 95% CI)=0.78 (0.58, 0.90) (Fig. 7). The scatter diagrams and ROC curves are shown in Figs. 8 and 9. The expression of miR-224-5p was markedly lower in PCa biofluids despite no clear significance. In the subgroup of cell lines, only one microarray was included: Mean ± SD: 9.8705±0.1835 (P=0.013, PCa, vs. normal, 9, vs. 3; ROC AUC (95% CI)=1.000 (1.000, 1.000), P=0.013 (Fig. 10). The expression of miR-224-5p was significantly lower in the PCa cell lines, and exhibited higher specificity.

Furthermore, in the literature search, three studies were retrieved providing a mean ± SD (Fig. 1C) (15–17). The results also suggested that the expression of miR-224-5p was downregulated in PCa tissues: SMD (95% CI)=−2.245 (−2.587, −1.904), P<0.001 (Table IV and Fig. 3D; PCa, vs. normal, 129, vs. 100). The results of all studies provided the following results: Overall SMD (95% CI)=−0.562 (−0.687, −0.436; P<0.001); overall sROC AUC (95% CI)=0.80 (0.76, 0.83); SENS (95% CI) 0.79 (0.63, 0.89); SPEC (95% CI)=0.66 (0.46, 0.81) (Table V and Figs. 3 and 11A). This provided a more reliable conclusion that the expression of miR-224-5p was reduced in PCa.

 Prospective target genes of miR-224-5p in PCa

The prospective target genes of moR-224-5p in PCa were determined based on the results of TCGA, GEO and prediction tools. Based on TCGA data, 3,019 differentially expressed genes in PCa tissues were acquired from GEPIA, and the FC was used as a measure (log2FC>1.0). Finally, 687 overexpressed genes were obtained from the PCa tissues (Fig. 12). In addition, from GEO two cell lines were retrieved that were transfected with PC3 and DU145, and the microarrays of differentially expressed mRNA (GSE51053 and GSE56243) were examined. According to the unions in which the gene expression value was below −0.1, a total of 3,616 (GSE51053) and 3,326 (GSE56243) genes with low expression were obtained. Following combining the results of two microarrays and eliminating the duplicates, 6,038 genes with low expression that had been transfected with miR-224-5p were obtained. The top 100 genes are shown in the heat-map in Fig. 13. In addition, 102,240 potential target genes of miR-224-5p were accumulated via miRWalk2.0. Genes that appeared in at least three prediction tools were selected, with 9,876 genes acquired. A total of 82 overlapped mRNA genes qualified for further analysis following combining the results of TCGA, GEO and prediction tools (Fig. 11B).

 In silico analysis

GO and KEGG analyses were performed for the 82 overlapped target genes using DAVID, which indicated that the GO terms were enriched in the following pathways: Intracellular transport, vesicle-mediated transport, protein transport of biological processes; Golgi apparatus, membrane fraction, insoluble fraction of cellular components; protein homodimerization activity, protein dimerization activity, identical protein binding of molecular functions (Table VI and Figs. 14 and 15). KEGG analysis demonstrated that these 82 genes were simply enriched in two pathways: Amino sugar and nucleotide sugar metabolism, and Vibrio cholerae infection. Significance was only found in the amino sugar and nucleotide sugar metabolism pathway (P<0.05; Table VI and Fig. 16A), in which UDP-N-acetylglucosamine pyrophosphorylase 1 (UAP1), hexokinase 2 (HK2) and chitinase 1 (CHIT1) were notably upregulated in PCa tissues (P<0.05; Fig. 16B). The PPI analysis suggested that DNA topoisomerase 2-α (TOP2A), ATP citrate lyase (ACLY) and ribonucleotide reductase regulatory subunit M2 (RRM2) were key genes for protein interaction, which had higher expression levels in PCa tissues (P<0.05; Table VII and Fig. 17). miR-224-5p may have negative axial associations with the expression of UAP1, HK2, CHIT1, TOP2A, ACLY and RRM2 in PCa, which requires further experiments for confirmation (Fig. 18).

 Discussion

In the present study, referring to data in the TCGA, GEO, ArrayExpress and previous literature, it was confirmed that the expression of miR-224-5p was notably downregulated in PCa tissues, biofluids and cell lines, and its downregulated expression may be associated with the progression of PCa. In addition, prospective target genes of miR-224-5p in PCa were identified via prediction tools and microarrays of differentially expressed mRNAs in TCGA and GEO. Bioinformatics analysis was subsequently performed on these potential genes. GO and KEGG analyses revealed that miR-224-5p mediated the expression of UAP1, HK2 and CHIT1 in order to regulate the amino sugar and nucleotide sugar metabolism signaling pathway, thus exerting its effects in PCa. miR-224-5p may also be vital in PCa by being involved in the protein interaction through TOP2A, ACLY and RRM2.

Previous studies have demonstrated that the abnormal expression of miR-224-5p was a crucial factor in the initiation and progress of tumors. For example, the high expression of miR-224-5p was likely to be involved in the onset of digestive tract malignancy (38); the expression of miR-224-5p was significantly reduced in mucinous breast cancer (39); and cell experiments and histologic examination performed by Zheng et al revealed that miR-224-5p exhibited lower expression in uveal melanoma (40). However, studies on the expression of miR-224-5p in PCa have been limited. Only a small number of studies with small sample sizes have found that miR-224-5p was expressed at a low level in prostatic cancer (26–28), and this was not confirmed by a second study involving a larger sample size. Therefore, these findings may, to a certain extent, contain errors. The present study, using data from TCGA, GEO and ArrayExpress databases, and previous literature, investigated the expression of miR-224-5p in PCa from PCa tissues, biofluids and cell lines. Taking advantage of the public data, it was confirmed that the expression of miR-224-5p was low in PCa. The results of a study by Fu et al indicated that downregulated miR-224-5p was closely associated with terminal clinical staging and metastasis (27). The present study, based on analysis of TCGA data, revealed that the expression of pre-miR-224 was decreased only in the T stage and M stage of PCa, although additional investigations are required to confirm its expression in other stages. Furthermore, when investigating the association between miR-224-5p and prognosis, Lin et al found that the low expression of miR-224-5p was associated with a poorer RFS rate (41). Wan et al examined the recurrence rate of patients with PCa following surgery, and found that downregulated miR-224-5p was closely associated with RFS rate (42). Mavridis et al investigated the survival rate of 58 patients with PCa, and confirmed that the lower the expression of miR-224-5p, the poorer the RFS rate of the patients (26). Fu et al examined 20 patients of PCa and demonstrated that those patients with upregulated miR-224-5p tended to have improved OS rates (27). In the present study, data on the survival rate of 500 patients with PCa was acquired from TCGA. No significant association was found between the expression of pre-miR224 and the OS rates of patients, with additional follow-up investigations required to confirm this.

Bioinformatics analysis was used in the present study to examine the molecular mechanism of miR-224-5p in PCa. KEGG analysis indicated that potential target mRNAs were significantly enriched in the amino sugar and nucleotide sugar metabolism signaling pathway. Previous studies have demonstrated that this pathway may be involved in carbohydrate metabolism in tumor tissues, affecting the onset and development of cancer (43,44). Genes enriched in this pathway included UAP1, HK2 and CHIT1, all of which exhibited notably high expression in PCa tissues. It was demonstrated that higher expression of UAP1 in PCa tissues accelerated the growth of cancer cells (45). In addition, a high expression of HK2 in PCa is essential in cell proliferation, apoptosis and carbohydrate metabolism (46–48). Although experiments have shown that the differentially expressed CHIT1 is correlated with colon carcinoma (49), its association with PCa has not been elucidated. Additionally, analysis of three key genes in the PPI network, TOP2A, ACLY and RRM2, revealed that the expression levels of these three genes were markedly elevated. A study by Schaeferklein et al in 2015 showed that the increased expression of TOP2A in PCa stimulated the growth and proliferation of PCa cells via androgen receptor (50). Similarly, a study by Shah et al in 2016 further elucidated the correlations of ACLY with androgen receptor gene expression, and the proliferation and apoptosis of PCa cells. Xin et al found that ACLY inhibited the generation of fat in PCa tissues via a targeting association, which may suppress the growth and metastasis of PCA cells (51,52). RRM2 has been shown to convert ribonucleotide into deoxynucleotide and be involved in multiple biological processes, including the synthesis of DNA and cell growth. In vitro experiments have demonstrated that upregulated RRM2 promotes the proliferation and metastasis of tumor cells in PCa (53,54). Several studies have examined miR-224-5p targeting mRNA. Wan et al confirmed that miR-224-5p targeted and upregulated APLN, and these were involved in the onset and progression of PCa (42). Fu et al identified CAMKK2 as the target gene of miR-224-5p, and noted the negative axial regulatory associations between them; these 2 genes were also found to collaborate to promote the progression of PCa (27). According to the results of the present study, the target genes of miR-224-5p in PCa may include UAP1, HK2, CHIT1, TOP2A, ACLY and RRM2. miR-224-5p is likely to regulate the expression of these genes, therefore performing its functions in the initiation and development of PCa. However, these results based on theoretical analysis require additional in vivo and in vitro experiments for confirmation.

In conclusion, the present study, based on data from TCGA, GEO and ArrayExpress databases, and previous literature, examined the expression of miR-224-5p in PCa and its clinical significance. It was confirmed that the expression of miR-224-5p was low in PCa and was associated with its clinical progression. In addition, miR-224-5p may exert its effects in PCa by mediating UAP1, HK2, CHIT1, TOP2A, ACLY and RRM2. The present study provides a theoretical foundation for future investigations on the oncogenesis of PCa.

 Acknowledgements

The authors would like to thank all members of the Molecular Oncology Group of the First Affiliated Hospital of Guangxi Medical University (Nanning, Guangxi Zhuang Autonomous Region 530021, China) for their professional suggestions. FCM, JM and JCZ reviewed and edited the manuscript.

 Funding

No funding was received.

 Availability of data and materials

The datasets used during the present study are available from the corresponding author upon reasonable request.

 Authors' contributions

BLG and XHH and GC conceived and designed the study. BLG, LJZ, LG and RQH performed the experiments. BLG and GC wrote the paper. BLG, LJZ, LG, RQH, GC and XHH reviewed and edited the manuscript. All authors read and approved the manuscript and agree to be accountable for all aspects of the research in ensuring that the accuracy or integrity of any part of the work are appropriately investigated and resolved.

 Ethics approval and consent to participate

This study does not involve animal or human related experiments.

 Patient consent for publication

Not applicable.

 Competing interests

The authors declare that they have no competing interests.

 References SiegelRLMillerKDJemalACancer statistics, 2017CA Cancer J Clin67730201710.3322/caac.2138728055103 AnHTaoNLiJGuanYWangWWangYWangFdetection of prostate cancer metastasis by whole body magnetic resonance imaging combined with bone scintigraphy and PSA LevelsCell Physiol Biochem4010521062201610.1159/00045316127941343 ChenSLWangSCHoCJKaoYLHsiehTYChenWJWuPRKoJLLeeHSungWWProstate cancer mortality-to-incidence ratios are associated with cancer care disparities in 35 countriesSci Rep74000032017 PanXWGanSSYeJQFanYHHongΥChuCMGaoYLiLLiuXChenLSMC1A promotes growth and migration of prostate cancer in vitro and in vivoInt J Oncol4919631972201610.3892/ijo.2016.369727667360 OstPBossiADecaesteckerKDe MeerleerGGiannariniGKarnesRJRoachMIIIBrigantiAMetastasis-directed therapy of regional and distant recurrences after curative treatment of prostate cancer: A systematic review of the literatureEur Urol67852863201510.1016/j.eururo.2014.09.00425240974 PetterssonARobinsonDGarmoHHolmbergLStattinPAge at diagnosis and prostate cancer treatment and prognosis: A population-based cohort studyAnn Oncol29377385201810.1093/annonc/mdx74229161337 JemalAFedewaSAMaJSiegelRLinCCBrawleyOWardEMprostate cancer incidence and PSA testing patterns in relation to USPSTF screening recommendationsJAMA31420542061201510.1001/jama.2015.1490526575061 LiuJChenZWangTLiuLZhaoLGuoGWangDinfluence of four radiotracers in PET/CT on diagnostic accuracy for prostate cancer: A Bivariate Random-Effects Meta-AnalysisCell Physiol Biochem39467480201610.1159/00044563927383216 BarringtonWESchenkJMEtzioniRArnoldKBNeuhouserMLThompsonIMJrLuciaMSKristalARDifference in association of obesity with prostate cancer risk between US African American and Non-Hispanic White Men in the Selenium and Vitamin E Cancer Prevention Trial (SELECT)JAMA Oncol1342349201510.1001/jamaoncol.2015.0513261811844570268 VidalACHowardLESunSXCooperbergMRKaneCJAronsonWJTerrisMKAmlingCLFreedlandSJObesity and prostate cancer-specific mortality after radical prostatectomy: Results from the Shared Equal Access Regional Cancer Hospital (SEARCH) databaseProstate Cancer Prostatic Dis207278201710.1038/pcan.2016.4727698439 GambariRBrognaraESpandidosDAFabbriETargeting oncomiRNAs and mimicking tumor suppressor miRNAs: Νew trends in the development of miRNA therapeutic strategies in oncology (Review)Int J Oncol49132201610.3892/ijo.2016.3503 GaoYFengBHanSLuLChenYChuXWangRChenLMicroRNA-129 in human cancers: From tumorigenesis to clinical treatmentCell Physiol Biochem3921862202201610.1159/00044791327802440 PaulPChakrabortyASarkarDLangthasaMRahmanMBariMSinghaRSMalakarAKChakrabortySInterplay between miRNAs and Human Diseases: A ReviewJ Cell Physiol23320072018201710.1002/jcp.2585428181241 XieTHuangMWangYWangLChenCChuXMicroRNAs as regulators, biomarkers and therapeutic targets in the drug resistance of colorectal cancerCell Physiol Biochem406276201610.1159/00045252527842308 ZhangXTangWLiRHeRGanTLuoYChenGRongMDownregulation of microRNA-132 indicates progression in hepatocellular carcinomaExp Ther Med1220952101201610.3892/etm.2016.3613276986985038555 HuangWTWangHLYangHRenFHLuoYHHuangCQLiangYYLiangHWChenGDangYWLower expressed miR-198 and its potential targets in hepatocellular carcinoma: A clinicopathological and in silico studyOncoTargets Ther951635180201610.2147/OTT.S108828 PangCLiuMFangWGuoJZhangZWuPZhangYWangJMiR-139-5p is increased in the peripheral blood of patients with prostate cancerCell Physiol Biochem3911111117201610.1159/00044781927562849 van BeijnumJRGiovannettiEPoelDNowak-SliwinskaPGriffioenAWmiRNAs: Micro-managers of anticancer combination therapiesAngiogenesis20269285201710.1007/s10456-017-9564-7284742825519663 WuCZhuangYJiangSLiuSZhouJWuJTengYXiaBWangRZouXInteraction between Wnt/β-catenin pathway and microRNAs regulates epithelial-mesenchymal transition in gastric cancer (Review)Int J Oncol4822362246201610.3892/ijo.2016.348027082441 YangXPangYYHeRQLinPCenJMYangHMaJChenGDiagnostic value of strand-specific miRNA-101-3p and miRNA-101-5p for hepatocellular carcinoma and a bioinformatic analysis of their possible mechanism of actionFEBS Open Bio86484201710.1002/2211-5463.12349293219585757177 FengQHuangXUrologyDOExpression of miRNA-32, −196a,-218,-128 and let7i in serum of patients with prostate cancer and its clinical significanceGuangdong Yixue2016 PanYGanQJunLUHuangTMangKESerological determination and its clinical significance of miRNA-129 and miRNA-21 in prostate cancer patientsChinese Journal of Health Laboratory Technology2016 YaoCHYuanXCLiuCDengJPExpression and diagnostic value of miRNA-15a in prostate cancerChin J Immunol2016 LuSWangMSChenPJRenQBaiPmiRNA-186 inhibits prostate cancer cell proliferation and tumor growth by targeting YY1 and CDK6Exp Ther Med1333093314201710.3892/etm.2017.4387285874055450643 TianXMLuoYZHePLiJMaZWAnYInhibition of invasion and migration of prostate cancer cells by miRNA-509-5p via targeting MDM2Genet Mol Res16doi: 10.4238/gmr16019195 MavridisKStravodimosKScorilasADownregulation and prognostic performance of microRNA 224 expression in prostate cancerClin Chem59261269201310.1373/clinchem.2012.19150223136246 FuHHeHCHanZDWanYPLuoHWHuangYQCaiCLiangYXDaiQSJiangFNMicroRNA-224 and its target CAMKK2 synergistically influence tumor progression and patient prognosis in prostate cancerTumour Biol3619831991201510.1007/s13277-014-2805-025394900 KristensenHHaldrupCStrandSMundbjergKMortensenMMThorsenKOstenfeldMSWildPJArsovCGoeringWHypermethylation of the GABRE~miR-452~miR-224 promoter in prostate cancer predicts biochemical recurrence after radical prostatectomyClin Cancer Res2021692181201410.1158/1078-0432.CCR-13-264224737792 TangZLiCKangBGaoGLiCZhangZGEPIA: A web server for cancer and normal gene expression profiling and interactive analysesNucleic Acids Res45W98W102201710.1093/nar/gkx247284071455570223 ChenMWangJLuoYHuangKShiXLiuYLiJLaiZXueSGaoHIdentify Down syndrome transcriptome associations using integrative analysis of microarray database and correlation-interaction networkHum Genomics122201810.1186/s40246-018-0133-y293518105775600 HeXZhangCShiCLuQMeta-analysis of mRNA expression profiles to identify differentially expressed genes in lung adenocarcinoma tissue from smokers and non-smokersOncol Rep399299382018293284935802042 LiFShiWWanYWangQFengWYanXWangJChaiLZhangQLiMPrediction of target genes for miR-140-5p in pulmonary arterial hypertension using bioinformatics methodsFEBS Open Bio718801890201710.1002/2211-5463.12322292260755715273 LiangLZengJHWangJYHeRQMaJChenGCaiXYHuXHDown-regulation of miR-26a-5p in hepatocellular carcinoma: A qRT-PCR and bioinformatics studyPathol Res Pract21314941509201710.1016/j.prp.2017.10.00129113686 MiBLiuGZhouWLvHLiuYLiuJIdentification of genes and pathways in the synovia of women with osteoarthritis by bioinformatics analysisMol Med Rep17446744732018293446515802222 WangLPengZWangKQiYYangYZhangYAnXLuoSZhengJNDUFA4L2 is associated with clear cell renal cell carcinoma malignancy and is regulated by ELK1PeerJ5e4065201710.7717/peerj.4065291589915695248 WeiDA multigene support vector machine predictor for metastasis of cutaneous melanomaMol Med Rep1729072914201829257259 ZhangHLiuJFuXYangAIdentification of key genes and pathways in tongue squamous cell carcinoma using bioinformatics analysisMed Sci Monit2359245932201710.12659/MSM.905035292407235738838 ZhangLHuangLSChenGFengZBPotential targets and clinical value of miR-224-5p in cancers of the digestive tractCell Physiol Biochem44682700201710.1159/00048528129169158 ZhouFLiSMengHMQiLQGuLMicroRNA and histopathological characterization of pure mucinous breast carcinomaCancer Biol Med1022272013236914413643684 ZhengXTangHZhaoXSunYJiangYLiuYLong non-coding RNA FTH1P3 facilitates uveal melanoma cell growth and invasion through miR-224-5pPLoS One12e0184746201710.1371/journal.pone.0184746290958235667836 LinZYHuangYQZhangYQHanZDHeHCLingXHFuXDaiQSCaiCChenJHMicroRNA-224 inhibits progression of human prostate cancer by downregulating TRIB1Int J Cancer135541550201410.1002/ijc.2870724382668 WanYZengZCXiMWanSHuaWLiuYLZhouYLLuoHWJiangFNZhongWDDysregulated microRNA-224/apelin axis associated with aggressive progression and poor prognosis in patients with prostate cancerHum Pathol46295303201510.1016/j.humpath.2014.10.02725532941 CaoJLuXXLiYZhuLQYangCOuCTangYPApplying gene set enrichment analysis and meta-analysis to screen key genes controlling the development and progression of hepatic carcinomaWorld Chin J Digestology207542012(In Chinese)10.11569/wcjd.v20.i9.754 LiZLiBQJiangMChenLZhangJLiuLHuangTPrediction and analysis of retinoblastoma related genes through gene ontology and KEGGBiomed Res Int20133040292013239981223755425 ItkonenHMEngedalNBabaieELuhrMGuldvikIJMinnerSHohlochJTsourlakisMCSchlommTMillsIGUAP1 is overexpressed in prostate cancer and is protective against inhibitors of N-linked glycosylationOncogene3437443750201510.1038/onc.2014.30725241896 KudryavtsevaAVNyushkoKMZaretskyARExpression of HK2 gene is deregulated in Prostate cancerAsian J Pharm11S158S1612017 TaoTChenMJiangRGuanHHuangYSuHHuQHanXXiaoJInvolvement of EZH2 in aerobic glycolysis of prostate cancer through miR-181b/HK2 axisOncol Rep3714301436201710.3892/or.2017.5430281849355364858 TaoTXiangPHuangTExpression of HK2 in prostate cancer and its effect on malignant phenotype of prostate cancer cellsChinese Journal of Clinical & Experimental Pathology2017 LiFFYanPZhaoZXLiuZSongDWZhaoXWWangXSWangGYLiuSLPolymorphisms in the CHIT1 gene: Associations with colorectal cancerOncotarget739572395812016271535625129954 Schaefer-KleinJLMurphySJJohnsonSHVasmatzisGKovtunIVTopoisomerase 2 alpha cooperates with androgen receptor to contribute to prostate cancer progressionPLoS One10e0142327201510.1371/journal.pone.0142327265602444641711 ShahSCarriveauWJLiJCampbellSLKopinskiPKLimHWDaurioNTrefelySWonKJWallaceDCTargeting ACLY sensitizes castration-resistant prostate cancer cells to AR antagonism by impinging on an ACLY-AMPK-AR feedback mechanismOncotarget74371343730201610.18632/oncotarget.9666272483225190055 XinMQiaoZLiJLiuJSongSZhaoXMiaoPTangTWangLLiuWmiR-22 inhibits tumor growth and metastasis by targeting ATP citrate lyase: Evidence in osteosarcoma, prostate cancer, cervical cancer and lung cancerOncotarget74425244265201610.18632/oncotarget.10020273177655190093 AfrasiabiAFiujiFMirhafezRAvanARRM2 (ribonucleotide reductase M2)Atlas Genet Cytogenet Oncol Haematol19p32372015 HuangYLiuXWangYHYehSDChenCLNelsonRAChuPWilsonTYenYThe prognostic value of ribonucleotide reductase small subunit M2 in predicting recurrence for prostate cancersUrol Oncol3251.e951.e19201410.1016/j.urolonc.2013.08.002

Study design. (A) Flow chart of the study design. (B) Flow chart of microarray retrieval for differential expression of miR-224-5p in PCa. (C) Flow chart of literature retrieval for the differential expression of miR-224-5p in PCa. GO, Gene Ontology; KEGG, l TCGA, The Cancer Genome Atlas; GEO, Gene Expression Omnibus; PPI, protein-protein interaction DEGs, differentially expressed genes; miR, microRNA; PCa, prostate adenocarcinoma.

Expression of miR-224. (A) Scatter diagram of miR-224 precursor expression between PCa tissue and normal prostate tissue based on TCGA data. (B) ROC curve of miR-224-5p expression in PCa tissue. miR, microRNA; PCa, prostate adenocarcinoma; AUC, area under the curve; TCGA, The Cancer Genome Atlas.

Forest profile of SMD of the expression of microRNA-224-5P between prostate adenocarcinoma and non-cancerous (A) tissues, (B) body fluid, (C) cell line and (D) literature. SMD, standardized mean difference.

Combined sROC curve for the expression of miR-224-5P between prostate adenocarcinoma and normal prostate tissue. sROC, summary receiver operating characteristic; AUC, area under the curve; SENS, sensitivity; SPEC, specificity.

Scatter plot of levels of miR-224-5p between PCa tissues and normal prostate tissues based on microarray. miR, microRNA; PCa, prostate adenocarcinoma.

Receiver operating characteristic curves of microRNA-224-5p levels between prostate adenocarcinoma tissues based on microarray. AUC, area under the curve.

Combined sROC curve for the expression of microRNA-224-5p between prostate adenocarcinoma body fluids based on microarray. sROC, summary receiver operating characteristic; AUC, area under the curve; SENS, sensitivity; SPEC, specificity.

Scatter plot of the expression of miR-224-5p between PCa body fluid and non-cancerous body fluid based on microarray. miR, microRNA; PCa, prostate adenocarcinoma.

Receiver operating characteristic curves of the expression of microRNA-224-5p between prostate adenocarcinoma body fluid based on microarray. AUC, area under the curve.

Expression of miR-224-5p between PCa cell line and normal cell line based on microarray. (A) Scatter diagram. (B) Receiver operating characteristic curve. miR, microRNA; PCa, prostate adenocarcinoma; AUC, area under the curve.

(A) Combined sROC curve for the expression of miR-224-5p between PCa based on the overall microarray. (B) Venn diagram for the crossing potential target genes of miR-224-5p in PCa. sROC, summary receiver operating characteristic; AUC, area under the curve; SENS, sensitivity; SPEC, specificity; miR, microRNA; PCa, prostate adenocarcinoma; TCGA, The Cancer Genome Atlas; DEGs, differentially expressed genes; GEO, Gene Expression Omnibus.

687 overexpressed gene positions based on GRCh38.p2 (NCBI).

Heat-map of the top 100 of 6,038 genes. The depth of color represents the expression value of the gene.

(A) Network of biological processes of GO terms of miR-224-5p potential target genes in PCa was constructed with the Bingo plugin of Cytoscape v3.5.0. Each node represents a GO term, node color indicates the P-value of a GO term. (B) Network of cellular components of GO terms of miR-224-5p potential target genes in PCa was constructed with the Bingo plugin of Cytoscape v3.5.0. Each node represents a GO term, the node color indicates the P-value of a GO term. miR, microRNA; PCa, prostate adenocarcinoma; GO, Gene Ontology.

Network of molecular function of GO terms of miR-224-5p potential target genes in PCa was constructed with the Bingo plugin of Cytoscape v3.5.0. Each node represents a GO term, the node color indicates the P-value of a GO term. miR, microRNA; PCa, prostate adenocarcinoma; GO, Gene Ontology.

(A) Signaling pathways of Kyoto Encyclopedia of Genes and Genomes-enriched miR-224-5p potential targets genes in PCa constructed using the ClueGO plugin. The color of circles represents the degree of enrichment of the pathway (P-value). (B) Potential target genes of miR-224-5p between PCa tissue and normal prostate tissue. Pink indicates the PCa group, green indicates the normal group (*P<0.05). miR, microRNA; PCa, prostate adenocarcinoma; UAP1, UDP-N-acetylglucosamine pyrophosphorylase 1; HK2, hexokinase 2; CHIT1, chitinase 1.

(A) Protein-protein interaction network analysis of miR-224-5p potential target genes in PCa. (B) Hub genes of the expression of miR-224-5p between PCa tissue and normal prostate tissue. Red indicates the cancer group, green indicates the normal group (*P<0.05). miR, microRNA; PCa, prostate adenocarcinoma; TOP2A, DNA topoisomerase 2-α; ACLY, ATP citrate lyase; RRM2, ribonucleotide reductase regulatory subunit M2.

Correlation between the targets (UAP1, HK2, CHIT1, TOP2A, ACLY, RRM2) and miR-224-5p in prostate adenocarcinoma. miR, microRNA; UAP1, UDP-N-acetylglucosamine pyrophosphorylase 1; HK2, hexokinase 2; CHIT1, chitinase 1; TOP2A, DNA topoisomerase 2-α; ACLY, ATP citrate lyase; RRM2, ribonucleotide reductase regulatory subunit M2.

Association between the expression of miR-224-5p between prostate cancer tissue and non-cancerous tissue based on The Cancer Genome Atlas data.

miR-224 expression T-test
Clinicopathological parameter Cases Mean SD FC T-value P-value
Group
  Normal adjacent 51 5.4773 0.86449 1.0000 −3.413 0.001
  Cancer 498 5.0255 1.19453 0.9175
Age (years)
  <60 204 4.9880 1.36730 1.0000 −0.086 0.931
  ≥60 296 4.9982 1.26126 1.0020
Pathological T stage
  T1-T2 188 5.1306 1.21866 1.0000 1.880 0.061
  T3-T4 305 4.9052 1.33624 0.9561
N stage
  N0 348 5.0312 1.28350 1.0000 1.347 0.179
  N1 79 4.8182 1.19763 0.9577
M stage
  M0 456 4.9999 1.25252 1.0000 1.715 0.087
  M1 4 3.9179 1.76728 0.7836
Gleason score
  ≤7 291 5.1057 1.13197 1.0000 1.143 0.254
  8≥ 203 4.9815 1.26629 0.9757
Gleason grade
  2 1 4.8074 1.048 0.371
  3 197 5.1616 1.12858
  4 248 5.0052 1.23926
  5 48 4.8763 1.16617
Recurrence
  No 373 5.0491 1.30796 1.0000 1.008 0.314
  Yes 58 4.8636 1.27528 0.9633
Clinical T stage
  T1-T2 351 5.0312 1.31167 1.0000 −0.338 0.735
  T3-T4 55 5.0945 1.14167 1.0126

miR, microRNA; SD, standard deviation; FC, fold-change.

Characteristics of the expression of microRNA-224-5p between studies of prostate adenocarcinoma and non-cancerous tissue based on Gene Expression Omnibus, ArrayExpress and TCGA datasets.

Patients Controls
Study N Mean SD N Mean SD T-value P-value
GSE54516 51 4.3588 0.0525 48 4.2649 0.0420 9.858 <0.001
GSE59156 24 3.6686 1.0369 18 4.0963 1.6407 −1.033 0.308
GSE64318 27 3.0859 0.9412 27 2.0776 1.7120 2.682 0.011
GSE76260 32 3.6162 0.0662 32 3.6723 0.0512 −3.794 <0.001
GSE14857 12 5.0417 1.3522 12 5.9110 0.5401 −2.068 0.051
GSE18671 14 6.6809 0.3943 6 6.4473 0.5782 1.057 0.305
GSE21036 114 5.2423 1.0451 28 6.6041 0.9873 −6.243 <0.001
GSE34932 8 2.1788 1.0875 8 3.5600 1.0803 −2.549 0.023
GSE23022 20 0.9755 0.2065 20 1.0201 0.1732 −0.740 0.464
GSE60371 56 6.3003 0.4778 21 6.6752 0.3432 −3.285 0.002
GSE36802 21 4.4416 0.9294 21 5.7362 1.1766 −3.957 <0.001
E-MTAB-408 42 3.2419 1.5142 12 3.7560 1.6093 −1.023 0.311
TCGA 498 5.0164 1.2154 52 5.4589 0.8661 −3.413 0.001
Total Standardized mean difference (95% CI)=−0.304 (−0.452, −0.156) P<0.001

TGCA, The Cancer Genome Atlas; SD, standard deviation.

Characteristics of the expression of microRNA-224-5p between prostate adenocarcinoma and non-cancerous body fluid studies based on the Gene Expression Omnibus dataset.

Patients Controls
Study N Mean SD N Mean SD T-value P-value
GSE39314 9 7.4306 3.6046 8 10.1763 6.1584 −1.104 0.293
GSE61741 65 4.8860 2.4333 35 5.4507 1.5656 −1.407 0.163
GSE24201 14 3.5766 1.8929 15 3.5814 2.1733 −0.006 0.996
GSE49298 4 5.1792 1.5787 4 4.9743 0.3945 0.252 0.810
Total Standardized mean difference (95% CI)=−0.216 (−0.544, 0.111) P=0.195

SD, standard deviation.

Characteristics of three studies selected from previous literature.

Control Prostate adenocarcinoma
Author, date Country N Mean ± SD N Mean ± SD Detection method Refs:PMID
Fu et al (27) China 20 2.171±0.564 20 1.344±0.358 RT-qPCR 25394900
Mavridis et al (26) Athens 66 8.470±2.090 73 4.228±0.809 RT-qPCR 23136246
Kristensen et al (28) Denmark 14 2.929±0.147 36 2.562±0.252 RT-qPCR 24737792
Total Standardized mean difference (95% CI)=−2.245 (−2.587, −1.904) P=0.195, P<0.001

RT-qPCR, reverse transcription-quantitative polymerase chain reaction; SD, standard deviation.

Characteristics of the expression of microRNA-224-5p between prostate adenocarcinoma and non-cancerous studies.

Patients Controls
Study N Mean SD N Mean SD T-value P-value
GSE54516 51 4.3588 0.0525 48 4.2649 0.0420 9.858 <0.001
GSE59156 24 3.6686 1.0369 18 4.0963 1.6407 −1.033 0.308
GSE64318 27 3.0859 0.9412 27 2.0776 1.7120 2.682 0.011
GSE76260 32 3.6162 0.0662 32 3.6723 0.0512 −3.794 <0.001
GSE14857 12 5.0417 1.3522 12 5.9110 0.5401 −2.068 0.051
GSE18671 14 6.6809 0.3943 6 6.4473 0.5782 1.057 0.305
GSE21036 114 5.2423 1.0451 28 6.6041 0.9873 −6.243 <0.001
GSE34932 8 2.1788 1.0875 8 3.5600 1.0803 −2.549 0.023
GSE23022 20 0.9755 0.2065 20 1.0201 0.1732 −0.740 0.464
GSE60371 56 6.3003 0.4778 21 6.6752 0.3432 −3.285 0.002
GSE36802 21 4.4416 0.9294 21 5.7362 1.1766 −3.957 <0.001
E-MTAB-408 42 3.2419 1.5142 12 3.7560 1.6093 −1.023 0.311
TCGA 498 5.0164 1.2154 52 5.4589 0.8661 −3.413 0.001
GSE39314 9 7.4306 3.6046 8 10.1763 6.1584 −1.104 0.293
GSE61741 65 4.8860 2.4333 35 5.4507 1.5656 −1.407 0.163
GSE24201 14 3.5766 1.8929 15 3.5814 2.1733 −0.006 0.996
GSE49298 4 5.1792 1.5787 4 4.9743 0.3945 0.252 0.810
GSE17317 9 9.8705 0.1835 3 12.2457 0.5820 −7.559 0.013
Kristensen et al (28) 36 2.5623 0.2523 14 2.9291 0.1469 <0.001
Fu et al (27) 20 1.3444 0.3578 20 2.1705 0.5643 <0.001
Mavridis et al (26) 73 4.2280 0.8090 66 8.4700 2.0900 <0.001
Total Standardized mean difference (95% CI)=−0.562 (−0.687, −0.436) P<0.001

SD, standard deviation; TCGA, The Cancer Genome Atlas.

Top 10 most significant GO terms of the potential target genes of microRNA-224-5-p in prostate adenocarcinoma.

Category Term Count P-value FDR
Biological process
  GOTERM_BP_FAT GO:0046907~intracellular transport 10 0.002874 4.327707
  GOTERM_BP_FAT GO:0016192~vesicle-mediated transport 9 0.004584 6.818167
  GOTERM_BP_FAT GO:0015031~protein transport 10 0.007548 10.992314
  GOTERM_BP_FAT GO:0045184~establishment of protein localization 10 0.007995 11.607428
  GOTERM_BP_FAT GO:0006163~purine nucleotide metabolic process 5 0.010383 14.821117
  GOTERM_BP_FAT GO:0006885~regulation of pH 3 0.010551 15.043556
  GOTERM_BP_FAT GO:0034613~cellular protein localization 7 0.011114 15.783945
  GOTERM_BP_FAT GO:0070727~cellular macromolecule localization 7 0.011491 16.275859
  GOTERM_BP_FAT GO:0008610~lipid biosynthetic process 6 0.016047 22.014591
  GOTERM_BP_FAT GO:0006605~protein targeting 5 0.016883 23.026892
Cellular component
  GOTERM_CC_FAT GO:0005794~Golgi apparatus 15 0.000193 0.241372
  GOTERM_CC_FAT GO:0005624~membrane fraction 14 0.000337 0.419916
  GOTERM_CC_FAT GO:0005626~insoluble fraction 14 0.000477 0.595260
  GOTERM_CC_FAT GO:0005783~endoplasmic reticulum 15 0.000517 0.644979
  GOTERM_CC_FAT GO:0042598~vesicular fraction 7 0.001999 2.470347
  GOTERM_CC_FAT GO:0000267~cell fraction 14 0.004767 5.799806
  GOTERM_CC_FAT GO:0031090~organelle membrane 14 0.005271 6.394020
  GOTERM_CC_FAT GO:0005768~endosome 7 0.006966 8.368649
  GOTERM_CC_FAT GO:0016021~integral to membrane 40 0.008019 9.576264
  GOTERM_CC_FAT GO:0005792~microsome 6 0.009016 10.705916
Molecular function
  GOTERM_MF_FAT GO:0042803~protein homodimerization activity 8 0.000927 1.226713
  GOTERM_MF_FAT GO:0046983~protein dimerization activity 9 0.003674 4.779822
  GOTERM_MF_FAT GO:0042802~identical protein binding 9 0.009776 12.255478
  GOTERM_MF_FAT GO:0043176~amine binding 4 0.014922 18.132938
  GOTERM_MF_FAT GO:0046982~protein heterodimerization activity 5 0.016408 19.761183
  GOTERM_MF_FAT GO:0046923~ER retention sequence binding 2 0.018664 22.176377
  GOTERM_MF_FAT GO:0031406~carboxylic acid binding 4 0.030136 33.449456
  GOTERM_MF_FAT GO:0016597~amino acid binding 3 0.032286 35.386583
  GOTERM_MF_FAT GO:0019899~enzyme binding 7 0.035546 38.224143
  GOTERM_MF_FAT GO:0015297~antiporter activity 3 0.040576 42.376479
KEGG category Term Count P-value Genes
  KEGG_PATHWAY hsa00520: Amino sugar and nucleotide sugar metabolism 3 0.038278 UAP1, HK2, CHIT1
  KEGG_PATHWAY hsa05110: Vibrio cholerae infection 3 0.059125 KDELR3, KDELR2 ATP6V0E2

GO, Gene Ontology; KEGG, Kyoto Encyclopedia of Genes and Genomes; FDR, false discovery rate.

Top 10 genes with combined scores in the protein-protein interaction network of potential target genes of microRNA-224-5p in prostate adenocarcinoma.

Node 1 Node 2 Node 1 string internal ID Co-expression Experimentally determined interaction Database annotated Automated text mining Combined score
TOP2A RRM2 1860923 0.880 0.000 0.000 0.673 0.960
MYO6 CDH1 1855148 0.000 0.360 0.900 0.214 0.945
CDH1 ADAM10 1844988 0.050 0.091 0.900 0.409 0.942
TMED10 KDELR2 1848494 0.310 0.137 0.900 0.105 0.939
KDELR3 TMED10 1859221 0.270 0.137 0.900 0.105 0.936
TOP2A KIAA0101 1860923 0.926 0.000 0.000 0.145 0.934
RAB3IP YWHAE 1862399 0.000 0.346 0.900 0.000 0.931
TOP2A TPX2 1860923 0.879 0.000 0.000 0.340 0.916
EPHA10 ADAM10 1855819 0.053 0.000 0.900 0.111 0.908
ADRB2 ARF6 1848724 0.000 0.000 0.900 0.112 0.907
TOP2A RRM2 1860923 0.880 0.000 0.000 0.673 0.960
MYO6 CDH1 1855148 0.000 0.360 0.900 0.214 0.945
CDH1 ADAM10 1844988 0.050 0.091 0.900 0.409 0.942
TMED10 KDELR2 1848494 0.310 0.137 0.900 0.105 0.939
KDELR3 TMED10 1859221 0.270 0.137 0.900 0.105 0.936