Genome-Wide microRNA Expression Profiling in Human Spermatozoa and Its Relation to Sperm Quality

Abstract

Background: Sperm samples are separated into bad and good quality samples in function of their phenotype, but this does not indicate their genetic quality.

Methods: Here, we used GeneChip miRNA arrays to analyze microRNA expression in ten semen samples selected based on high-magnification morphology (score 6 vs. score 0) to identify miRNAs linked to sperm phenotype.

Results: We found 86 upregulated and 21 downregulated miRNAs in good-quality sperm (score 6) compared with bad-quality sperm samples (score 0) (fold change > 2 and p-value < 0.05). MiR-34 (FC × 30, p = 8.43 × 10^-8), miR-30 (FC × 12, p = 3.75 × 10^-6), miR-122 (FC × 8, p = 0.0031), miR-20 (FC × 5.6, p = 0.0223), miR-182 (FC × 4.83, p = 0.0008) and miR-191 (FC × 4, p = 1.61 × 10^-6) were among these upregulated miRNAs. In silico prediction algorithms predicted that miRNAs upregulated in good-quality sperm targeted 910 genes involved in key biological functions of spermatozoa, such as cell death and survival, cellular movement, molecular transport, response to stimuli, metabolism, and the regulation of oxidative stress. Genes deregulated in bad-quality sperm were involved in cell growth and proliferation.

Conclusions: This study reveals that miRNA profiling may provide potential biomarkers of sperm quality.

Keywords: biomarkers; high magnification; male infertility; microRNA; spermatozoa.

Figure 1

Differences in the global miRNA expression profiles of S6 and S0 sperm samples. (A). Unsupervised 3D PCA representing the miRNA expression patterns of S6 spermatozoa (n = 5 samples) and S0 spermatozoa (n = 5 samples). Each sample was analyzed using the GeneChip^® miRNA 4.0 Array. Red dots, S6 samples; blue dots, S0 samples. (B). Hierarchical clustering of the samples using the differentially expressed miRNAs with the highest variation. S6 and S0 samples (n = 5/each group) are clustered in two distinct groups. (C). Heat map of the S6 and S0 miRNA signatures based on the 107 miRNAs that are differentially expressed between S6 and S0 samples. Each column corresponds to a specific miRNA, and each row represents a sperm sample. The color scale reflects the relative miRNA expression levels, with red indicating higher expression and blue indicating lower expression. (D). Violin plots showing the expression of the top 10 upregulated miRNAs in S6 samples based on the TAC analysis of the microarray data. S6: good quality samples, S0: bad quality samples.

Figure 2

Analysis of GO terms associated with S6-miRNA targets and their functions. (A). Analysis of significantly represented GO terms. Pathway enrichment analyses were carried out using the human gene names of S6-miRNA targets. The size of the blue dots reflects the degree of enrichment, with larger dots representing more significant p-values. (B). GSEA was conducted using the S6-miRNA targets. The heat map illustrates the clustering of genes within the leading-edge subsets, emphasizing the dynamic expression of genes associated with programmed cell death regulation, phosphorylation, positive regulation of cell proliferation, and metabolic processes. Genes are shown on the vertical bars colored from deep blue (top rank) to blank (lowest rank). (C). Bubble plot of the overlapping canonical pathways associated with S6-miRNA targets. The circle size reflects the number of genes involved in the pathway. The canonical pathways were categorized into various types based on the IPA database.

Figure 3

Top-ranked functional networks of the S6-miRNA target genes. Top networks identified by IPA of S6-miRNA target genes related cell growth and proliferation, cell cycle regulation, DNA replication and repair, system development and function, tissue morphology, reproductive system disorders, cell morphology, cellular assembly and organization, cellular function and maintenance, cell death and survival, and developmental disorders. Green nodes represent genes regulated by S6-miRNAs. Dashed lines represent indirect relationships, while solid lines indicate direct molecular interactions. Within each network, the edge types are defined as follows: a line without an arrowhead signifies binding only, a line ending with a vertical bar represents inhibition, and a line with an arrowhead indicates an “acts on” relationship. *: indicate that several gene identifiers in the dataset file correspond to a single gene in the Global Molecular Network.

Figure 4

Networks of the S6-miRNA target genes. The IPA tool was used to generate the networks based on the predicted miRNA–mRNA interactions. Pink nodes represent the miRNAs upregulated in S6 samples and green nodes represent the genes targeted by S6-miRNAs. Solid lines represent direct interactions and dashed lines indirect interactions. *: indicate that several gene identifiers in the dataset file correspond to a single gene in the Global Molecular Network.

Figure 5

The promoters of the predicted S6-miRNA target genes are not differentially methylated. Integrative Genome Viewer snapshots illustrating the methylation levels at individual CpG sites (0–100%) across the examined genes. Each promoter region (red arrow) overlaps with a CpG island (green box).

Figure 6

Enrichment of S6-miRNA targets in critical signaling pathways and their expression in testes. (A). Pathway analysis (KEGG pathway) using the Pathview server (https://pathview.uncc.edu/ (accessed on 17 June 2024)). Highlighted genes are pathway components identified as targets of S6 miRNAs. (B). Expression profile of candidate genes in various human tissues. Expression levels (in Log2 RPKM) of PDGFA, PDGFRA, GRB2, MECP2, MAP2K1, ARHGDIA, and MET in 30 tissues from GTEx. For each gene, the colored circle corresponding to each tissue represents the RPKM value averaged across all samples within that tissue. RPKM stands for reads per kilobase of transcript per million mapped reads.