Identification of potential targets of microRNA-101-3p in prostate cancer by bioinformatics analysis

Background: MiR-101-3p, a tumor suppressor, has been implicated as a tumor suppressor miRNA in multiple primary malignancies including prostate cancer (PCa). This study aimed to explore target genes and relevant signaling pathways regulated by microRNA-101-3p (miR-101-3p) for further researches in PCa with bioinformatics analysis. Results: 565 target genes were appeared in all databases and enriched in positive regulation of transcription, which were mainly enriched in axon guidance and MAPK pathway. Two important modules were detected from PPI network. Ten hub genes were selected, including MAPK1, PIKFYVE, EGFR, SMARCA4, TOP2B, GSK3B, FOS, RAC1, BCL2 and TAF1. After thoroughly reviewing published literature, we found that 10 target genes and six signaling pathways were truly inhibited by miR-101-3p in various tissues or cells; some of these verified targets were in accordance with our present prediction. Conclusion: This study demonstrated that miR-101-3p target hub genes, including MAPK1, PIKFYVE, EGFR, SMARCA4, TOP2B, GSK3B, FOS, RAC1, BCL2 and TAF1, might promote the development of PCa. However, further experiments are still required to confirm potential functions of these miR-101-3p target genes and pathways in PCa.


Introduction
Prostate cancer (PCa) is the most common malignant cancer and the second leading cause of cancerrelated death in men worldwide [1]. In recent years, the number of PCa patients has been significantly increased in China, which may be due to the huge population, ageing, changes of environment and life styles [2]. In the past few decades, improvements in screening, diagnostics and treatment have led to a consistent decrease in PCa mortality and an increase in overall survival rate [3]. However, more than 650,000 men are diagnosed with PCa annually, and this constitutes almost 10% of all new cancer cases in men worldwide [4]. Thus, exploring novel target genes, molecular mechanisms and generating therapeutic approaches are urgently needed.
MicroRNAs (MiRNAs) are small, noncoding, single-stranded RNAs of ~ 22 nucleotides that negatively regulate gene expression at the posttranscriptional level, primarily through base pairing to the 3' untranslated region (UTR) of target mRNAs [5]. MiR-101-3p, a tumor suppressor, has been implicated as a tumor suppressor miRNA in a number of malignancies, such as non-small-cell lung cancer (NSCLC) [6], breast cancer [7], and glioma [8]. However, only few studies have assessed such a connection between miR-101-3p and its target genes in PCa [9]. Thus, it is important to identify comprehensive downstream target genes of miR-101-3p with bioinformatics analysis in PCa, which may be helpful for future therapy development of PCa. In the present study, we sought to unveil the role of miR-101-3p in PCa through identification of putative molecular targets by bioinformatics analysis. Meanwhile, we established PPI network of miR-101-3p target genes and picked out hub genes with high degree of connectivity. Besides, analysis of biological process (BP), molecular function (MF), cellular component (CC) and Kyoto Encyclopedia of Genes and Genome (KEGG) pathways of miR-101-3p target genes and two modules were performed. Furthermore, the direct downstream targets and relevant signaling pathways regulated by miR-101-3p in PCa were extracted in the published literature.

Results
Prediction of target genes for miR-101-3p As shown in Fig. 1, the number of predicted target genes of miR-101-3p in miRWalk, miRanda, RNA22, and Targetscan databases was 12106, 5426, 4250 and 10540, respectively. There were 10795 target genes supported by at least two databases, 4461 target genes predicted by at least three databases and 565 target genes appeared in all four databases. The target genes of miR-101-3p predicted in all four databases are listed in Table 1 and were used for further analyses. Table 1 The target genes of miR-101-3p predicted in all four databases databases. As shown in Fig. 2 and Table 2, GO analysis results showed that these target genes were particularly enriched in biological processes, including positive regulation of transcription, intracellular signal transduction, protein autophosphorylation, activation of MAPKK activity, and so on. For cell component, they were enriched within both nucleoplasm and cytoplasm. In addition, GO molecular function also displayed that miR-101-3p target genes were enriched in protein binding, transcription coactivator activity, transcription factor activity and sequence-specific DNA binding. Table 2 Gene ontology (GO) analysis for predicted target genes of miR-101-3p As shown Table 3, the top rankings were related to axon guidance, MAPK signaling pathway, nonsmall cell lung cancer, colorectal cancer, neurotrophin signaling pathway, pathways in cancer, proteoglycans in cancer, inositol phosphate metabolism and adherens junction (all P < 0.05). Among them, axon guidance, MAPK signaling pathway, pathways in cancer and adherens junction were well known to be associated with the pathogenesis of PCa. Figure 3 shows the rich factor, Q value, and gene number corresponding to each pathway term.  . Then, we made the PPI network of these top 10 hub genes with higher degree of connectivity (Fig. 4A). In addition, we also used MCODE plugin and selected top two modules in order to detect significant modules in PPI network of miR-101-3p target genes ( Fig. 4b and 4c). KEGG pathway enrichment analysis showed that these two modules were mainly associated with some proliferation-related pathways, such as PI3K-Akt signaling pathway, ErbB signaling pathway and Ras signaling pathway.

Screening target genes and signaling pathways inhibited by miR-101-3p on PCa in published studies
A comprehensive electronic search of PubMed and Web of Science databases was performed until November 1, 2019, to obtain target genes and signaling pathways modulated by miR-101-3p in published studies. Finally, 13 papers including ten target genes and six signaling pathways inhibited by miR-101-3p were obtained; most of them focus on the functions of miR-101-3p suppressing tumor growth, migration, and invasion in PCa tissues and cells. The details were shown in Table 4. To date, there have been fewer studies concerning the characteristics of miR-101-3p in PCa.
In the present study, we further identified that candidate target genes for miR-101-3p were involved in the regulation of crucial biological processes in PCa, including MAPK1, PIKFYVE, EGFR, SMARCA4, TOP2B, GSK3B, FOS, RAC1, BCL2 and TAF1. MAPK1 is involved in a number of biochemical signals and cellular processes such as proliferation, differentiation, transcription regulation and development of various cancers [23][24][25]. Chen demonstrate that miR-378 inhibits PCa cell growth through directly suppresses of MAPK1 in vitro and in vivo [26]. PIKFYVE, a lipid kinase that converts PI(3)P into PI (3,5)P2 in the endocytic pathway [27], has been reported to promote several cancer cells migration and invasion [28,29]. To our knowledge, no studies have been done to evaluate possible involvement of the PIKFYVE gene in clinical PCa. For EGFR, which is aberrantly expressed in both androgen independent and metastatic PCa, are closely associated with aggressive phenotype, poor clinical prognosis, high Gleason scores, reduced survival rate, then contributing to castrate resistant PCa and progression to metastasis [30,31]. Shao find that SMARCA4 (also known as BRG1) expression is significantly higher in malignant tissues compared to their benign compartments, especially in highgrade PCa, suggesting increased SMARCA4 expression might promote cell growth and invasion in PCa [32]. TOP2B has been found mediated androgen-induced the double-strand breaks and prostate cancer gene rearrangements [33]. It has been suggested that higher levels of cytoplasmic GSK3B expression are associated with aggressive PCa [34,35]. Recently, Barrett and colleagues shown that the proto-oncogenes FOS is required for migration and invasion in PCa cells [36]. RAC1, a member of the Rho family GTPases, has been found hyperactivated in the metastatic PCa cells [37] and inhibition of RAC1 activity blocks the migration and invasion of PCa cells [38]. It has been reported an association of high BCL2 expression with higher Gleason scores and lower biochemical recurrencefree survival in patients with advanced PCa undergoing androgen deprivation therapy [39]. A previous study proved that TAF1, a coactivator of androgen receptor, increased expression is associated with progression of human PCa to the lethal castration-resistant state [40]. These results indicate that most of miR-101-3p hub genes were involved in the development of PCa.

Conclusions
Our bioinformatics analysis identified miR-101-3p target hub genes in PCa that might play a central role in the occurrence, development and prognosis of PCa. In order to get more accurate correlation results, further experiments are still required to confirm potential functions of these miR-101-3p target genes and pathways in PCa.

GO and KEGG pathway enrichment analysis
Integration Discovery (DAVID) software, version 6.8 (https://david.ncifcrf.gov/), was used to perform GO analysis to identify BP, CC, and MF of these target genes of miR-101-3p. [42] Meanwhile, the probable signaling pathways in which these target genes enriched were analyzed by KEGG database (http://www.genome.jp/kegg/). The P-value < 0.05 was considered significant.

Protein-protein interaction (PPI) network and module analysis
STRING database was used to predict the association between miR-101-3p and the target gene in the regulatory network analysis (https://string-db.org/), interactions with a combined score > 0.4 were selected as significant, and the PPI pairs were output to construct the PPI network using Cytoscape software version 3.6.0 (www.cytoscape.org/). Moreover, the Molecular Complex Detection (MCODE) app was utilized to screen modules of PPI network in Cytoscape with degree cutoff = 2, node score cutoff = 0.2, k-core = 2, and max. depth = 100. GO and KEGG pathway analysis were also made to explore the potential information.  Gene ontology (GO) enrichment analysis for predicted target genes of miR-101-3p. Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis for predicted target genes of miR-101-3p.