Screening, Identification and Interaction Analysis of Key Micrornas
Total Page:16
File Type:pdf, Size:1020Kb
1 Screening, identification and Interaction Analysis of Key MicroRNAs 2 and Genes in Asthenozoospermia 3 Liman Li1, Song Chen*2, 3 4 1Center for Translational Medicine, Key Laboratory of Birth Defects and Related 5 Diseases of Women and Children (Sichuan University), Ministry of Education, West 6 China Second University Hospital, Sichuan University, Chengdu, China 7 2Department of Urology, Zhongnan Hospital of Wuhan University, Wuhan, China 8 3Department of Thoracic and Cardiovascular Surgery, Zhongnan Hospital of Wuhan 9 University, Wuhan 430071, China 10 *Corresponding author: Dr. Song Chen, Email: [email protected], Tel: 11 +86-27-6781-3104, Fax: +86-27-6781-2892; 1 12 Abstract 13 Background Asthenozoospermia, one of the most common causes of male infertility, 14 is a complicate multifactorial pathological condition that genetic factors are involved 15 in. However, the epigenetic signature and mechanism of asthenozoospermia still 16 remain limited. Our study aimed to confirm the key microRNAs (miRNAs) and genes 17 in asthenozoospermia and demonstrate the underlying epigenetic regulatory 18 mechanisms. 19 Methods We screened out and pooled previous studies to extracted potential 20 differentially expressed miRNAs (DEMs). GSE22331 and a published profile dataset 21 were integrated to identify differentially expressed genes (DEGs). Pathway and gene 22 ontology analysis were performed using DAVID. A protein-protein network (PPI) 23 was constructed using STRING. The target genes of DEMs were predicted using 24 TargetScan and the miRNA-mRNA network was built. 25 Results We reported 3 DEMs and 423 DEGs by pooling included dataset and 26 published studies. Pathway analysis showed that these DEGs might participate in 27 signaling pathways regulating pluripotency of stem cells, Wnt signaling pathway and 28 Notch signaling pathway. 25 hub genes were identified, and the most significant gene 29 was BDNF. We screened out the overlapped DEGs between the predicted target genes 30 of 3 DEMs and the 423 DEGs. Finally, a potential miRNA-mRNA regulatory 31 network was constructed. 2 32 Conclusion This study firstly pooled several published studies and a GEO dataset to 33 determine the significance of potential miRNAs and genes, such as miR-374b, 34 miR-193a, miR-34b, BDNF, NTRK2, HNRNPD and EFTUD2 in regulating 35 asthenozoospermia and underscore their interactions in the pathophysiological 36 mechanism. Our results provided theoretical basis and new clues for potential 37 therapeutic treatment in asthenozoospermia. Validations in vivo and in vitro are 38 required in future studies. 39 Keywords Asthenozoospermia, miRNAs, Genes 3 40 Background 41 Male infertility is a severe problem in human reproduction health [1, 2]. Abnormal 42 sperm morphology, poor sperm motility and low sperm count are identified as 43 common causes of male infertility. According to the criteria reported by the World 44 Health Organization, asthenozoospermia is characterized by total motility < 40% and 45 progressive motility < 32% in fresh semen samples [3]. Several factors have been 46 confirmed to be associated with asthenozoospermia, as shown in Figure 1. 47 Chromosomal abnormality was correlated with sperm deficiency including 48 asthenozoospermia [4]. The multiple morphological abnormalities of the sperm 49 flagella (MMAF) is one of the most known causes of asthenozoospermia. Mutations 50 in the mitochondrial genome have been reported in MMAF, such as DNAH1, 51 CFAP43, CFAP44, CFAP69, and QRICH2 [5-8], and study have shown compelling 52 evidence that the loss-of-function mutations of these genes could induce MMAF and 53 lead to male infertility. Hormone could also modulate spermatogenesis and impact 54 fertilization. In a clinical trial, most of 786 subfertility patients were characterized as 55 idiopathic oligo or asthenozoospermia who improved their semen quality and 56 concentration by receiving hormonal treatment [9]. Sperm motility was under the 57 control of flagellum [10] and any structural or functional defects in flagellum would 58 lead to reduced motility and induce asthenozoospermia [11]. Moreover, the 59 anti-sperm antibodies played key roles in female and male immunological infertility 60 and sperm-immobilizing antibodies that bound to the surface of ejaculated sperm 61 could impact the sperm motility [12]. There were other exogenous sources could 4 62 result in asthenozoospermia, for example, drinking alcohol, smoking, toxicant and 63 physical activity also have effect on the quality of sperm [13-17]. However, the 64 epigenetic signatures which affect asthenozoospermia are not well understood and 65 require to be further investigated. 66 Currently, increasing genetic factors contributing to male infertility were reported. 67 Alterations were observed between fertile men and asthenozoospermia patients’ 68 transcriptomes, including pathways and genes which were associated with this disease. 69 SEMG1 mRNA and protein expression were upregulated in asthenozoospermia 70 patients compared with that in normozoospermia group [18]. CABYR and ROPN1 71 mRNA were significantly downregulated in asthenozoospermia patients’ samples and 72 a positive correlation was confirmed between the expression of the two genes, 73 indicating that the co-expression of CABYR and ROPN1 was a prerequisite for 74 normal sperm motility and flagellar function [19]. In addition, miRNAs also have 75 dysregulated expressions in testicular biopsy samples and sperms of infertile men. 76 The expression of miR-34b in the control group was significantly higher than that in 77 the asthenozoospermia group, demonstrating that miR-34b might serve as a novel 78 biomarker of male subfertility [20]. The downregulated expression of miR-525 and 79 upregulated expression of miR-151a were also associated with asthenozoospermia 80 and male infertility [18, 21]. 81 In this study, we explored the differential expressed miRNAs (DEMs) and 82 differentially expressed genes (DEGs) in asthenozoospermia patients through 5 83 reanalyzing the data from published studies and a GEO dataset. Then, we conducted 84 pathway and gene ontology analysis of DEGs using DAVID. The protein-protein 85 interaction (PPI) network was constructed by STRING and DEMs’ target genes were 86 predicted by TargetScan. Our results proposed a novel insight into the related 87 biological characteristics and molecular pathways of the DEMs and DEGs in 88 asthenozoospermia. A miRNA-mRNA network further exhibited the potential 89 mechanisms linked to asthenozoospermia. 6 90 Methods 91 Literature search of DEMs 92 Firstly, we searched GEO database for published miRNA profiling data of 93 ashenozoospermia, however, there was no available result. Then, we conducted a 94 systematic search in PubMed and Web of Science and found 4 studies, which 95 contained miRNA profiling results in ashenozoospermia [22-25]. All of samples used 96 in 4 studies were semen samples and all of them used normal fertile individuals as 97 controls. All studies identified the alteration trends of miRNAs (upregulated or 98 downregulated). In order to investigate the involvement of crucial miRNAs in the 99 modulation of ashenozoospermia, we pooled these 4 studies, and the overlapped 100 miRNAs were presented with Venn diagram, using an online tool, Draw Venn 101 Diagram (http://bioinformatics.psb.ugent.be/webtools/Venn/). Next, we chose the 102 miRNAs with consistent expression alteration trends in studies as DEMs in following 103 analysis. 104 Data processing of DEGs 105 Xiaoning Zhang’s study investigated the expression profiles of long noncoding RNA 106 (lncRNA) and mRNA in mammalian sperm. mRNA between the normozoospermic 107 and asthenozoospermic groups with q-value (FDR) ≤ 0.001 and fold-change (FC) ≥ 108 2 (|log2FC| > 1) was considered significantly changed, and we reanalyzed the 109 changed mRNA in our following study [26]. Next, we searched GEO database for 110 the available mRNA expression profile data using keywords “ashenozoospermia” 7 111 and “Homo sapiens”, and the GSE22331 dataset was downloaded. GSE22331 had 2 112 groups from normozoospermic and asthenozoospermic men. Each group was pooled 113 by 30 sperm samples to obtain enough total RNA. Then, we screened significantly 114 changed mRNA between normozoospermia and asthenozoospermia groups based on 115 FC ≥ 2. Since there was only 1 pooled sample in each group, the FDR could not be 116 calculated. Next, we obtained the intersected elements of significantly changed 117 mRNA between GSE22331 dataset and Xiaoning Zhang’s study which were used as 118 DEGs in following analysis. 119 KEGG and GO enrichment analysis of DEGs 120 We utilized DAVID (https://david.ncifcrf.gov/), an online biological information 121 database, to visualize the DEGs enrichment of biological processes (BP), cellular 122 components (CC), molecular function (MF) and biological pathways (P value < 123 0.05). The results of gene ontology and pathway analysis of DEGs were showed with 124 histograms. 125 Protein-protein interaction (PPI) network and module analysis 126 We utilized an online common software, STRING database (https://string-db.org), to 127 build PPI network of DEGs. Then Cytoscape (www.cytoscape.org), a public source 128 bioinformatics software platform, was applied to visualize and analyze the molecular 129 interaction networks. The plugin, Molecular Complex Detection (MCODE), was 130 applied to identify the most dense and significant module in PPI network based on 131 degree cut-off = 2, node score cut-off = 0.2, max