Skip to main content
NCBI home page
Cell Cycle logoLink to Cell Cycle
. 2023 Dec 5;22(19):2194–2209. doi: 10.1080/15384101.2023.2280170

Epididymal segment-specific miRNA and mRNA regulatory network at the single cell level

Tong Chen a,*, Liangyu Yao a,*, Wen Liu b,c, Jiaochen Luan a, Yichun Wang a, Chao Yang a, Xiang Zhou a, Chengjian Ji a, Xuejiang Guo d, Zengjun Wang a,✉, Ninghong Song a,✉
PMCID: PMC10732646  PMID: 37982230

ABSTRACT

Spermatozoa released from the testis cannot fertilize an egg before becoming mature and motile in the epididymis. Based on three bulk and one single-cell RNA-seq (scRNA-seq) data series, we compared mRNA or miRNA expression between epididymal segment-specific samples and the other samples. Hereby, we identified 570 differentially expressed mRNAs (DE-mRNAs) and 23 differentially expressed miRNAs (DE-miRNAs) in the caput, 175 DE-mRNAs and 15 DE-miRNAs in the corpus, 946 DE-mRNAs and 12 DE-miRNAs in the cauda. In accordance with respective DE-miRNAs, we predicted upstream transcription factors (TFs) and downstream target genes. Subsequently, we intersected target genes of respective DE-miRNAs with corresponding DE-mRNAs, thereby obtaining 127 upregulated genes in the caput and 92 upregulated genes in cauda. Enriched upregulated pathways included cell motility-related pathways for the caput, smooth muscle-related pathways for the corpus, and immune-associated pathways for the cauda. Protein–protein interaction (PPI) network was constructed to extract key module for the caput and cauda, followed by identifying hub genes through cytohubba. Epididymis tissues from six mice were applied to validate hub genes expression using qRT-PCR, and 7 of the 10 genes displayed identical expression trends in mice caput/cauda. These hub genes were found to be predominantly distributed in spermatozoa using scRNA-seq data. In addition, target genes of DE-miRNAs were intersected with genes in the PPI network for each segment. Subsequently, the miRNA and mRNA regulatory networks for the caput and cauda were constructed. Conclusively, we uncover segment-specific miRNA-mRNA regulatory network, upstream TFs, and downstream pathways of the human epididymis, warranting further investigation into epididymal segment-specific functions.

KEYWORDS: Epididymal segment, expression distribution, single cell level

Introduction

The epididymis is a tubular structure bridging the testis and the vas deferens. Since the pioneering work conducted by OrgebinCrist [1] and Bedford [2] in 1967, interest in the epididymis has started to gain traction. The past half-century has witnessed a growing appreciation for the role that the epididymis plays in reproductive physiology. It is in this duct-like organ that spermatozoa acquire forward motility properties, fertilizing ability, and temporary storage prior to ejaculation. Also, the epididymal compartment is responsible for absorbing and releasing ionic, antioxidative, and epididymosomal components. Through interactions with these components of the extracellular environment, spermatozoa undergo a series of biochemical and functional modifications. Along with the rapid technological developments of assisted reproductive technologies (ART), sperm retrieval from the epididymis or testis has become routine in men with obstructive azoospermia [3], which makes it more necessary to focus on the role of the epididymis in human sperm physiology. From a histological and ultrastructural point of view, human epididymis can be subdivided into the caput, corpus, and cauda segments, and functional diversity of each segment has been preliminarily investigated. Overall, the caput may provide the spermatozoa with motility and fertilizing ability [4], whereas the corpus and cauda are involved in sperm storage and fertility preservation [5]. In laboratory animals, one recent study demonstrated a significant decrease of implantation rates and 100% of embryonic death for the caput spermatozoa-derived embryos using intracytoplasmic sperm injection (ICSI) [6], while other researches reported successful caput spermatozoa-derived embryos using ICSI [7,8]. For infertile patients with complete asthenozoospermia, whether testicular or ejaculated spermatozoa result in more successful pregnancy outcomes has yet to be determined [9], a question that might be answered, in part, by the physiology and pathophysiology of the epididymis.

While epididymal functions are substantially documented in laboratory rodents and some domestic animals, knowledge of the human epididymis remains poorly documented. One major factor contributing to this phenomenon is the difficulty of obtaining human tissue for research. Furthermore, unlike in rodents, the different segments of the human epididymis are not easily distinguishable from an anatomical perspective, making functional analysis even more challenging. With regard to well-established inter-species variation in anatomy and physiology of the epididymis, one should be cautious when attempting to extrapolate the epididymal functions obtained using laboratory species to humans [10]. Gratifyingly, several messenger RNA (mRNA) transcriptomic studies of the human epididymis revealed segment-specific differentially expressed genes (DEGs) and biological processes [11,12]. In addition, regional expression of small non-coding RNA (ncRNAs) such as microRNA (miRNA) contributes to segment-specific gene expression and differentiated functions [13]. Despite considerable effort, these previous studies were limited by a relatively small sample size and/or lacking an in-depth analysis. The molecular mechanism behind the segment-specific gene expression of human epididymis involves a complex network of regulatory controls and remains to be fully clarified. To integrate available resources and increase the statistical power, our current study conducted a comprehensive bioinformatic survey of three bulk and one single-cell transcriptomic data series regarding human epididymis tissues. Pursuant to this, we sought to address segment-specific miRNA and mRNA regulatory network, upstream transcription factors (TFs), and downstream biological pathways of the human epididymis, to further refine a complete picture of epididymal functions in male reproduction.

Materials and methods

Data sources

We downloaded epididymis-associated data from Gene Expression Omnibus (GEO, https://www.ncbi.nlm.nih.gov/geo/) using the keyword “epididymis”. Data series were considered for inclusion only if each following criterion was met without exception: (1) sample type was classified as human epididymis tissue; (2) either miRNA or mRNA transcriptomic data was contained; (3) full data were provided in the data series. After screening, we chose two microarray-based mRNA data series (GSE7808 and GSE141568), one microarray-based miRNA data series (GSE35522), and one single-cell RNA-seq (scRNA-seq) data series (GSE148963) for further analysis. The basic information of the aforementioned four data series is depicted in Table S1.

GSE7808 [11], consisting of three caput samples, three corpus samples, and three cauda samples, was analyzed via GPL570 platform (Affymetrix Human Genome U133 Plus 2.0 Array). Besides, GSE141568 [12], comprising of nine efferent duct samples, nine caput samples, three corpus samples, and three cauda samples, was analyzed through GPL23159 platform (Affymetrix Clariom S Assay). To ensure a better representation of the caput, intermediate caput samples were selected. General information on selected samples for GSE141568 is shown in Table S2. GSE35522 [13] was an miRNA expression profiling data series, which included three caput samples, three corpus samples, and three cauda samples, and was analyzed through GPL8786 (Affymetrix Multispecies miRNA-1 Array). In addition, GSE148963 [14] was an scRNA-seq data series with three caput samples, which was analyzed through GPL24676 (Illumina NovaSeq 6000).

Screening differentially expressed mRNAs (DE-mRNAs) and miRNAs (DE-miRNAs)

During data preprocessing, we performed data normalization via the quantile normalization method to guarantee the validity of downstream analysis [15]. Furthermore, we assessed whether the data series were on a logarithmic scale, and performed log-base 2 transformation as appropriate. We mapped the probe to the gene symbol. According to the correspondence, the mean expression value of several probes was assigned the same gene symbol. Principal component analysis (PCA) was applied for obtaining the distribution of the samples. We subsequently achieved DE-mRNAs and DE-miRNAs through the limma package [16]. Statistical significance was determined if an adjusted P < 0.05 and an absolute log2FC > 1. We fulfilled the adjustment of a value of P through Benjamini – Hochberg (BH) method. Based on a comparison between specific epididymis segment samples and the other samples, we obtained DE-mRNAs or DE-miRNAs.

Enrichment analysis

We obtained aberrant hallmark pathways for each epididymis segment through two methods. Firstly, we performed gene set variation analysis (GSVA) via the package GSVA [17]. Gene sets “c2.cp.kegg.v7.1” and “h.all.v7.1” were achieved from the Molecular Signatures Database. Secondly, gene set enrichment analysis (GSEA) was performed via the package fgsea. Subsequently, hallmark pathways from the two methods were intersected.

Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses were conducted using the clusterProfiler package [18]. Significant enrichment was determined if the adjusted P < 0.05, which was adjusted via BH method.

Putative TFs and target genes of DE-miRNAs

TransmiR v2.0 [19] was utilized to predict the TFs of DE-miRNAs. A better visualization of the correlation of TF and miRNA was implemented via Cytoscape software [20]. Furthermore, we predicted target genes of DE-miRNAs via miRNet database [21].

Protein – protein interaction (PPI) network

In order to depict the associations of DE-miRNAs, PPI network was constructed through the STRING database [22]. Significant PPI pair was determined if the combined score ≥ 0.4 and further visualized through Cytoscape software [20]. Maximal clique centrality (MCC) method of Cytoscape plugin CytoHubba [23] was utilized to obtain top five hub genes in the PPI network. Based on the miRNA and mRNA pairs, we constructed potential segment-specific miRNA and mRNA networks along the human epididymis.

Tissue sampling and quantitative real-time polymerase chain reaction

Epididymis tissue samples were obtained from six male 10- to 12-week-old C57/BL6 mice. The sampling side of each mouse was randomly selected. Epididymis tissue samples were further divided into six caput samples, six corpus samples, and six cauda samples. When caput or cauda samples were regarded as experimental group, samples of the other two epididymal segments were set as the control group. Following sampling, the tissues were kept at − 80°C. Total RNAs were isolated from the tissues using TRIzol reagent (Life Technologies, USA). The reverse transcription was achieved through TaKaRa reverse transcription kit (TaKaRa Bio, Japan). We utilized SYBR Green fluorescence system (Roche, USA) and carried out mRNA qRT-PCR through TaKaRa quantitative mRNA kit (TaKaRa Bio, Japan). In accordance with the 2−△△Ct method, relative value of mRNA was normalized to Gapdh mRNA levels. All primers were synthesized by Tsingke Biological Technology (Nanjing, China). The primer sequences are shown in Table S8.

ScRNA-seq data analysis

The cell–unique molecular identifier (UMI) matrix was converted to a Seurat object through the package Seurat v3 [24]. The criteria for the filtration of the matrix were a minimum number of cells equal to 3 and a minimum number of features equal to 200. Both the median and third quantiles of mitochondrial reads ratio were below 0.05. We merged the three caput samples and detected a total of 5,299 single-cell transcriptomes after filtration. Subsequent batch effect correction was applied through the standard Seurat v3 integration workflow. We detected neighbors through 10 dimensions. As for the unsupervised clustering, a low resolution parameter (0.06) was selected. A total of 30 PCA dimensions were reduced via Uniform Manifold Approximation and Projection (UMAP). Besides, feature plots were generated through the package Seurat v3.

Construction of miRNA and mRNA regulatory network

The mirwalk database was utilized to explore the target genes of DE-miRNAs for the caput and cauda, respectively. Then, we intersected the target genes with genes in the PPI network for each segment. The miRNA and mRNA regulatory networks were subsequently constructed for the caput and cauda, which were visualized through Cytoscape.

Statistical analysis

Data were analyzed and visualized through R software (version 4.2.1). The moderated t-test was utilized for a comparison of mRNA or miRNA expression values of specific epididymal segment with those of the other epididymal segments. Statistical significance was determined if an absolute log2FC > 1 and an adjusted P value <  0.05.

Results

DE-mRNAs detection

Expression values of samples in the GSE7808 and GSE141568 data series before and after normalization are visualized in Figure S1A and Figure S1B, respectively. Also, PCA results before and after batch effect removal are shown in Figures S2A and Figure S2B. We identified a total of 311 upregulated DE-mRNAs and 259 downregulated DE-mRNAs in the caput, 108 upregulated DE-mRNAs and 67 downregulated DE-mRNAs in the corpus, 567 upregulated DE-mRNAs and 379 downregulated DE-mRNAs in the cauda, which are visualized in Figures 1a–c, respectively. Apparently, more upregulated genes with higher significance were detected in the cauda when compared with the other two segments. The detailed information of the upregulated and downregulated DE-mRNAs in each epididymis segment is listed in Table S3 and Table S4, respectively.

Figure 1.

Figure 1.

Open in a new tab

DE-mRNAs analysis of the GSE7808 and GSE141568 data series. The volcano plots visualizing (a) DE-mRNAs between caput and the other samples, (b) DE-mRNAs between corpus and the other samples, and (c) DE-mRNAs between cauda and the other samples. DE‐mRNAs: differentially expressed messenger RNAs.

Enrichment for hallmark pathways in each epididymis segment

In order to enrich hallmark pathways for each epididymis segment, relationships between hallmark pathway expression and epididymis segment samples were computed using GSVA (Figure S3A). In addition, the dysregulated hallmark pathways in the caput, corpus, and cauda are visualized through lollipop plots (Figures S3B,C,D). GSEA was also utilized, and aberrant hallmark pathways in each epididymis segment are shown via ridgeplots (Figures S4A,D,G). In addition, the enriched genes for top four hallmark pathways in each epididymis segment are plotted through cnetplots (Figures S4B,E,H), while the top five significant hallmark pathways for each epididymis segment are shown using gseaplots (Figures S4C,F,I). Subsequently, significant hallmark pathways achieved from the two methods were intersected, contributing to one upregulated and 13 downregulated hallmark pathways in the caput (Figure 2a), one upregulated and nine downregulated hallmark pathways in the corpus (Figure 2b), 21 upregulated and four downregulated hallmark pathways in the cauda (Figure 2c). Apparently, most enriched hallmark pathways were found to be upregulated in the cauda.

Figure 2.

Figure 2.

Open in a new tab

The intersection of the predicted hallmark pathways in each epididymis segment through GSEA and GSVA. The intersection of the hallmark pathways in (a) caput, (b) corpus, or (c) cauda through GSEA and GSVA. GSEA, gene set enrichment analysis; GSVA, gene set variation analysis.

Enrichment of upregulated genes in each epididymis segment

We performed GO analysis in accordance with upregulated DE-mRNAs in each epididymis segment. Enriched GO terms for the caput contained cell-substrate adhesion, positive regulation of cell motility, regulation of cell activation (Figure 3a). As for the corpus, extracellular matrix organization, muscle tissue development, and cilium organization were identified as markedly enriched GO terms (Figure 3c). Furthermore, enriched GO terms for the cauda included positive regulation of MAPK cascade, myeloid cell activation involved in immune response, and leukocyte degranulation (Figure 3e).

Figure 3.

Figure 3.

Open in a new tab

GO terms and KEGG pathway enrichment analyses of the upregulated DEGs in each epididymis segment. The enriched (a) GO terms and (b) KEGG pathways in caput based on the upregulated DEGs; the enriched (c) GO terms and (d) KEGG pathways in corpus based on the upregulated DEGs; the enriched (e) GO terms and (f) KEGG pathways in cauda based on upregulated DEGs. DEGs: differentially expressed genes; GO: gene Ontology; KEGG: Kyoto Encyclopedia of genes and genomes.

Moreover, upregulated DE-mRNAs in each epididymis segment were also adopted for KEGG analysis. Enriched pathways for the caput included oxytocin signaling pathway, HIF-1 signaling pathway, cGMP-PKG signaling pathway (Figure 3b). Enriched pathways for the corpus identified Wnt signaling pathway, vascular smooth muscle contraction, and Apelin signaling pathway as markedly enriched pathways (Figure 3d). In addition, Relaxin signaling pathway, human cytomegalovirus infection, and human papillomavirus infection were markedly enriched for the cauda (Figure 3f).

DE-miRNAs detection

Expression values of epididymis samples within GSE35522 prior to and after normalization are shown (Figures S5A,B). The result of PCA is presented in Figure S6C. We obtained nine upregulated and 14 downregulated DE-miRNAs for the caput, 13 upregulated and 2 downregulated DE-miRNAs for the corpus, as well as 5 upregulated and 7 downregulated DE-miRNAs for the cauda, which are visualized through respective volcano plot (Figures 4 a,b,c). Correspondingly, upregulated and downregulated DE-miRNAs for each epididymis segment are shown in Table S5 and Table S6, respectively.

Figure 4.

Figure 4.

Open in a new tab

DE-miRNAs analysis of the GSE35522 data series. The volcano plots visualizing (a) DE-miRNAs between caput and the other samples, (b) DE-miRNAs between corpus and the other samples, and (c) DE-miRNAs between cauda and the other samples. DE‐miRNAs: differentially expressed microRNAs.

Predictive TFs and target genes based on DE-miRNAs

Predictive TFs for the upregulated DE-miRNAs in the caput contained MKL1, RREB1, JAG1, N1ICD, TP63, NKX2–5, TP73, DDX6, etc (Figure 5a). On the other hand, predictive TFs for downregulated DE-miRNAs in the caput contained TGFB1, CUX1, E2F6, SMAD3, NFKB1, JARID2, etc (Figure 5b). In addition, the predicted TFs on the basis of upregulated genes in the corpus included ZNF263, MAX, ERG, E2F6, ZNF143, CTCFL, etc (Figure 5c), while the predicted TFs on the basis of downregulated genes in the corpus included STAT2, KLF6, INTS13, and MAPK1 (Figure 5d). Moreover, the predicted TFs based on upregulated genes in the cauda included TP53, AP-1, SNAl1, FOXO3, and DNMT1 (Figure 5e), while the predicted TFs according to downregulated genes in the cauda included TLR4, TWIST1, BMP4, HES1, PCGF2, PKM, and TGFB1 (Figure 5f).

Figure 5.

Figure 5.

Open in a new tab

Putative TFs of the DE-miRNAs in each epididymis segment. Putative TFs for (a) upregulated or (b) downregulated DE-miRNAs in caput; Putative TFs for (c) upregulated or (d) downregulated DE-miRNAs in corpus; Putative TFs for (e) upregulated or (f) downregulated DE-miRNAs in cauda. DE‐miRNAs: differentially expressed microRNAs; TFs: transcription factors.

In addition to the predictive TFs, target genes of downregulated DE-miRNAs in the caput or cauda were predicted. Specifically, we detected 10,007 target genes of the downregulated DE-miRNAs in the caput (Figure 6a) and 2521 target genes of the downregulated DE-miRNAs in the cauda (Figure 6b). Then, target genes of downregulated miRNAs and corresponding upregulated DE-mRNAs were intersected, contributing to 127 upregulated genes for the caput (Figure 6c) and 92 upregulated genes for the cauda (Figure 6d), respectively. Detailed information regarding the intersection of upregulated DE-mRNAs and potential target genes of downregulated DE-miRNAs is also listed (Tables S7).

Figure 6.

Figure 6.

Open in a new tab

Target genes of downregulated DE-miRNAs in each epididymis segment. (a) the network of downregulated DE-miRNAs and target genes in caput; (b) the network of downregulated DE-miRNAs and target genes in cauda; (c) the intersection of upregulated DE-mRNAs and target genes of downregulated DE-miRNAs in caput; (d) the intersection of upregulated DE-mRNAs and target genes of downregulated DE-miRNAs in cauda. DE‐miRNAs: differentially expressed microRNAs; DE‐mRNAs: differentially expressed mRNAs.

PPI network, key modules, hub genes, and their related miRNAs

We constructed the PPI networks for the caput and cauda, respectively. Specifically, we achieved 93 node pairs for the upregulated genes in the caput (Figure 7a), and 172 node pairs for the upregulated genes in the cauda (Figure 7c). We further characterized the key module in the caput (MCODE score = 5.714; Figure 7b) and key module in cauda (MCODE score = 7.75; Figure 7d). Consequently, the majority of genes in key module for the caput (TTC25, FOXJ1, TEKTIN-2, RSPH1, and SPAG1) were associated with flagellum axoneme assembly, while all genes in key module for cauda (THBS1, SDC4, HSPG2, ITGA5, LAMC1, FBN1, SERPINE1, COL4A1, and COL1A2) were primarily responsible for extracellular matrix organization.

Figure 7.

Figure 7.

Open in a new tab

Construction of PPI network based on upregulated DE-mRNAs in caput and cauda. (a) PPI network of upregulated DE-mRNAs in caput; (b) the top upregulated genes in caput based on the node degree; (c) PPI network of upregulated DE-mRNAs in cauda; (d) the top hub genes of upregulated DE-mRNAs in cauda based on the node degree. DE‐mRNAs: differentially expressed mRNAs; PPI network: protein–protein interaction network.

Based on the MCC method in CytoHubba, RSPH1, TTC25, PIH1D2, CCDC113, and EFHC1 were identified as hub genes for the caput. In addition, COL1A2, COL4A1, ITGA5, THBS1, and FBN1 were deemed to be hub genes for the cauda. In accordance with the predictive miRNA and mRNA pairs, hub genes along with their related miRNAs were visualized for the caput and cauda, respectively (Figure 8).

Figure 8.

Figure 8.

Open in a new tab

Hub genes and their related miRNAs in the caput and cauda.

Validation of hub genes expression levels using qRT-PCR

In accordance with mice epididymis tissue samples (Figures 9 a,b), the mRNA expression of five hub genes for the caput and five hub genes for the cauda was validated via qRT-PCR. It was found that expression values of Efhc1, Pih1d2, Rsph1, and Ccdc113 were significantly higher in mice caput when compared with the other two segments, while expression levels of Thbs1, Fbn1, and Itga5 were significantly higher in mice cauda than those in the other two segments. However, the expression levels of Col1a2, Ttc25, and Col4a1 did not exhibit significant differences between mice caput/cauda and the other two segments (Figure 9c).

Figure 9.

Figure 9.

Open in a new tab

Verification of important genes expression levels for the caput/cauda in male mice. (a) photographs of the epididymis extracted from male mice; (b) Representative H&E-stained sections of the caput, corpus and cauda; (c) the comparison of important genes expression levels for the caput/cauda and the other two segments in mice epididymis tissues. *P value < 0.05; **P value < 0.01.

Expression distribution of critical genes in different cell types within the caput

In order to unveil the expression distribution of critical genes in different cell types, scRNA-seq data series of the caput was applied for further exploration. We identified up to eight clusters of cells, which were subsequently annotated as principal cells, apical and narrow cells, efferent duct cells, basal cells, stromal/muscle cells, spermatozoa, clear cells and immune cells (Figure 10a). TTC25, one intersected gene of genes in the key module and hub genes, was unexpectedly not detected in this scRNA-seq data series. Interestingly, it was found that the expression of genes in the key module (Figure 10b) and hub genes (Figure 10c), without exception, were mainly distributed in spermatozoa. In addition, 37.0% of spermatozoa-specific genes and 28.4% of principal cell-specific genes were found to be upregulated for the caput in bulk transcriptome data, while the percentages of other cell type specific genes that were also upregulated for the caput in bulk transcriptome data were no more than 10.0% (Figure S7).

Figure 10.

Figure 10.

Open in a new tab

Expression distribution of important genes for the caput in different cell types. (a) UMAP dimension reduction plot revealing eight distinct cell types. (b) feature dot plot indicating expression distribution of genes in key module for caput. (c) expression distribution of hub genes for each cell type through UMAP dimension reduction plot. UMAP: uniform manifold approximation and projection.

Establishment of miRNA-mRNA regulatory network

We predicted the target genes of 14 downregulated DE-miRNAs for the caput and 7 downregulated DE-miRNAs for the cauda, respectively. Then, target genes were intersected with genes in PPI network for each segment. Finally, based on the predictive miRNA and mRNA pairs, we constructed potential miRNA and mRNA interactive networks for the caput (Figure 11a) and cauda (Figure 11b) along the human epididymis. It was found that two hub genes for the caput (CCDC113 and EFHC1) and three hub genes for the cauda (THBS1, ITGA5 and FBN1) were also present in the networks.

Figure 11.

Figure 11.

Open in a new tab

The miRNA-mRNA networks for genes in the PPI network. The interactive network of miRNA and mRNA in the PPI network for (a) the caput or (b) the cauda. Two hub genes in the caput (CCDC113 and EFHC1) and three hub genes in the cauda (THBS1, ITGA5 and FBN1) were highlighted by yellow oval nodes in the networks.

Discussion

Spermatozoa released directly from the testis are not capable of accomplishing sperm-egg fusion before obtaining maturation in the epididymis. Although advances in ART have enabled bypassing epididymis, there is some evidence that ART can result in an offspring health-care cost [25]. A better understanding of potential molecular mechanisms and biological functions of the human epididymis may come with novel therapeutic strategies.

Our study revealed decreased levels of oxidative phosphorylation in the human corpus epididymis (Figures 2b,S3C), which further refined functional zone of the epididymis in terms of antioxidant defenses. The activity of superoxide anion was demonstrated to be detected within rat epididymis [26]. Long-term persistent hypoxia may compromise sperm production and ultimately lead to the initiation of Sertoli cell-only syndrome. Sufficient correction of the hypoxia may repair sperm production to some extent [27]. Our findings suggested the presence of low levels of hypoxia in caput and corpus (Figures 2a,b), which may protect spermatozoa from hypoxia-induced damage. Only one smooth muscle layer envelopes the epithelium of caput while two layers of the smooth muscles envelope the cauda [28]. In doing so, the most frequent smooth muscle contraction emerges in the caput and the largest magnitude of contraction emerges in the cauda. Indeed, smooth muscle-related GO terms were found to be upregulated throughout all three segments of the epididymis in this study. For instance, regulation of actin filament-based process and (striated) muscle tissue development were found to be upregulated in the corpus (Figure 3c). Additionally, effects of smooth muscle contractions, including positive regulation of cell motility, cellular components, and locomotion, served as the most significant GO terms for the cauda (Figure 3e). Based on KEGG annotation, MAPK signaling pathway was upregulated throughout the epididymis (Figures 3b,d,f). It was reported that MAPK pathway plays critical roles in the formation and maintenance of blood-epididymis barrier [29], and the orchestration of extracellular microenvironment for sperm maturation in the epididymis of mice [30]. Besides, we uncovered TNFA signaling via NFKB to be upregulated in cauda (Figure 2c). Similar to our finding, lipopolysaccharide-induced epididymitis was reported to be characterized by leukocyte infiltration and fibrosis in mouse cauda, and these inflammatory responses could be abolished in TNFA knockout mice [31].

When it comes to the segment-specific miRNA and mRNA interactive networks, miR-17-5p, miR-500a-3p, miR-26a-5p, miR-362-5p, EFHC1, PIH1D2, TTC25, RSPH1, and CCDC113 constituted caput-specific regulatory network in the present study (Figure 8). In previous literature, fluoride was predicted and validated experimentally to mediate male reproductive toxicity through regulating miR-17-5p in mice [32]. In men with primary infertility, miR-26a-5p transcript expression levels in spermatozoa were positively associated with sperm motility and morphology [33]. EFHC1 serves as a calmodulin and encodes an EF-hand-containing calcium-binding protein which is related to sperm flagellar assembly in men [34]. PIH1D2 enables small GTPase binding activity and was identified as one of the male-biased DEGs associated with testis differentiation in the juvenile phase of the snakeskin gourami [35]. TTC25, localizing to the ciliary axons, was found to be related to spermatogenesis and at a persistent low-level in the trajectory of sperm development [36]. Furthermore, our results demonstrated that the expression of hub genes as well as genes in the key module for the caput was mainly distributed in spermatozoa. In addition, compared with the other cell type specific genes, spermatozoa-specific genes shared a higher percentage with upregulated genes for the caput in bulk transcriptome data. Previous transcriptome and proteome researches reported that the caput has a higher transcriptomic activity in comparison with the corpus and cauda [37]. The caput is well known for its function in acquiring sperm motility and fertilizing ability, while the corpus and cauda are critical for sperm storage and the maintenance of a sterile luminal environment [11,38,39]. Besides, the function of sperm storage could also be provided by the scrotal portion of the vas deferens and ampulla [40]. Collectively, our results further expanded the knowledge about the dominant role of the caput in male fertility.

In addition to the caput-specific interaction network, our work also demonstrated that miR-665, miR-205-5p, miR-892a, COL1A2, COL4A1, THBS1, FBN1, and ITGA5 comprised cauda-specific regulatory network (Figure 10). Indeed, establishment of miRNA-mRNA regulatory network is more conducive to investigate the physiological or pathological processes, which balanced the depth and breadth of the data [41]. MiR-892a belongs to a primate-specific cluster located on the X chromosome with specific expression in the epididymis [13]. Additionally, miR-892a expression was found to be dysregulated in men with idiopathic asthenozoospermia [42]. In immature mice, COL1A2 was reported to be enriched in seminiferous tubules. Also, components including COL1A2, COL1A1, and procollagen I Type A might play a potential role in maintaining spermatogonia in an undifferentiated state. During spermatogenesis, these components might mediate the detachment and migration of germ cells [43]. COL4A1 was found to be localized in epithelial basement membrane of the endometrium. Knockdown of COL4A1 impaired the adhesive ability of the endometrium, putatively resulting in implantation defects and infertile issues [44]. The expression of THBS1 was observed in seminal plasma, and the deposition of THBS1 in the female genital tract after insemination could alter the maternal–fetal interface during early pregnancy [45]. In male rats, FBN1 stimulated gonadotropin-releasing hormone expression on the hypothalamus, which in turn led to increased levels of LH and FSH in pituitary gland. As a consequence, serum levels of testosterone and sperm motility were elevated [46]. ITGA5 is a member of the integrin alpha-chain family and was found to be downregulated in female patients with polycystic ovary syndrome. The downregulation of ITGA5 was supposed to be associated with oocyte nuclear maturation [47].

In conclusion, we obtained five hub genes for the caput (RSPH1, TTC25, PIH1D2, CCDC113, and EFHC1) and five hub genes for the cauda (COL1A2, COL4A1, ITGA5, THBS1, and FBN1). RT-qPCR validation in mice showed that Efhc1, Pih1d2, Rsph1 and Ccdc113 were upregulated in the caput than in the other two segments, while Thbs1, Fbn1 and Itga5 were upregulated in the cauda than in the other two segments. In addition, compared with the other two segments, miR-17-5p, miR-500a-3p, miR-26a-5p, and miR-362-5p were determined to be downregulated in the caput, whereas miR-665, miR-205-5p, and miR-892a were determined to be downregulated in the cauda. Moreover, we identified segment-specific miRNA and mRNA regulatory network, upstream transcription factors (TFs), and downstream biological pathways involved in human epididymal physiology, which facilitate future treatment options for male infertility and the development of targeted male contraceptives.

Supplementary Material

Supplemental Material
Click here for additional data file. (9.2MB, zip)

Acknowledgements

We would like to acknowledge Linlin Tian from Nanjing Municipal Center for Disease Control and Prevention for the proof reading of this manuscript.

Funding Statement

This study was funded by the National Natural Science Foundation of China (81871151 and 82071638) and Jiangsu Funding Program for Excellent Postdoctoral Talent (2022ZB730).

Disclosure statement

No potential conflict of interest was reported by the author(s).

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/15384101.2023.2280170

Author contributions

Conceptualization: Ninghong Song and Zengjun Wang; Study design: Ninghong Song and Tong Chen; Analysis and visualization of data: Tong Chen, Liangyu Yao, Wen Liu, Jiaochen Luan and Xiang Zhou; Manuscript writing: Tong Chen, Yichun Wang and Chao Yang; Scientific review: Xuejiang Guo and Chengjian Ji.

Data availability statement

All data series used in this study are available in online repository. The names of the repository and accession number of the included data series can be found in Table S1.

Ethics statement

This study design was identified and approved by the medical research ethics committee of The First Affiliated Hospital of Nanjing Medical University and was therefore conducted according to ethical standards laid down in the 1964 Declaration of Helsinki and its amendments.

References

  • [1].Orgebin-Crist MC. Sperm maturation in rabbit epididymis. Nature. 1967;216(5117):816–818. doi: 10.1038/216816a0 [DOI] [PubMed] [Google Scholar]
  • [2].Bedford JM. Effects of duct ligation on the fertilizing ability of spermatozoa from different regions of the rabbit epididymis. J Exp Zool. 1967;166(2):271–281. doi: 10.1002/jez.1401660210 [DOI] [PubMed] [Google Scholar]
  • [3].Shin DH, Turek PJ. Sperm retrieval techniques. Nat Rev Urol. 2013;10(12):723–730. doi: 10.1038/nrurol.2013.262 [DOI] [PubMed] [Google Scholar]
  • [4].Turner TT. On the epididymis and its role in the development of the fertile ejaculate. J Androl. 1995;16(4):292–298. doi: 10.1002/j.1939-4640.1995.tb00532.x [DOI] [PubMed] [Google Scholar]
  • [5].Jones RC, Murdoch RN. Regulation of the motility and metabolism of spermatozoa for storage in the epididymis of eutherian and marsupial mammals. Reprod Fertil Dev. 1996;8(4):553–568. doi: 10.1071/RD9960553 [DOI] [PubMed] [Google Scholar]
  • [6].Conine CC, Sun F, Song L, et al. Small RNAs gained during epididymal transit of sperm are essential for embryonic development in mice. Dev Cell. 2018;46(4):470–480.e3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [7].Suganuma R, Yanagimachi R, Meistrich ML. Decline in fertility of mouse sperm with abnormal chromatin during epididymal passage as revealed by ICSI. Hum Reprod. 2005;20(11):3101–3108. doi: 10.1093/humrep/dei169 [DOI] [PubMed] [Google Scholar]
  • [8].Zhou D, Suzuki T, Asami M, et al. Caput Epididymidal Mouse Sperm Support Full Development. Dev Cell. 2019;50(1):5–6. [DOI] [PubMed] [Google Scholar]
  • [9].Chen T, Fan D, Wang X, et al. ICSI outcomes for infertile men with severe or complete asthenozoospermia. Basic Clin Androl. 2022;32(1):6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [10].Sullivan R, Légaré C, Lamontagne‐Proulx J, et al. Revisiting structure/functions of the human epididymis. Andrology. 2019;7(5):748–757. [DOI] [PubMed] [Google Scholar]
  • [11].Thimon V, Koukoui O, Calvo E, et al. Region-specific gene expression profiling along the human epididymis. Mol Hum Reprod. 2007;13(10):691–704. [DOI] [PubMed] [Google Scholar]
  • [12].Légaré C, Sullivan R. Differential gene expression profiles of human efferent ducts and proximal epididymis. Andrology. 2020;8(3):625–636. doi: 10.1111/andr.12745 [DOI] [PubMed] [Google Scholar]
  • [13].Belleannée C, Calvo E, Thimon V, et al. Role of microRnas in controlling gene expression in different segments of the human epididymis. PLoS One. 2012;7(4):e34996. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [14].Leir SH, Yin S, Kerschner JL, et al. An atlas of human proximal epididymis reveals cell-specific functions and distinct roles for CFTR. Life Sci Alliance. 2020;3(11):e202000744. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [15].Yang Q, Li B, Tang J, et al. Consistent gene signature of schizophrenia identified by a novel feature selection strategy from comprehensive sets of transcriptomic data. Brief Bioinform. 2020;21(3):1058–1068. [DOI] [PubMed] [Google Scholar]
  • [16].Ritchie ME, Phipson B, Wu D, et al. Limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res. 2015;43(7):e47. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [17].Hänzelmann S, Castelo R, Guinney J. GSVA: gene set variation analysis for microarray and RNA-seq data. BMC Bioinf. 2013;14(1):7. doi: 10.1186/1471-2105-14-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [18].Yu G, Wang L-G, Han Y, et al. clusterProfiler: an R package for comparing biological themes among gene clusters. OMICS. 2012;16(5):284–287. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [19].Tong Z, Cui Q, Wang J, et al. TransmiR v2.0: an updated transcription factor-microRNA regulation database. Nucleic Acids Res. 2019;47(D1):D253–d258. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [20].Shannon P, Markiel A, Ozier O, et al. Cytoscape: a software environment for integrated models of biomolecular interaction networks. Genome Res. 2003;13(11):2498–2504. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [21].Fan Y, Siklenka K, Arora SK, et al. miRnet - dissecting miRNA-target interactions and functional associations through network-based visual analysis. Nucleic Acids Res. 2016;44(W1):W135–41. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [22].Szklarczyk D, Franceschini A, Wyder S, et al. STRING v10: protein–protein interaction networks, integrated over the tree of life. Nucleic Acids Res. 2015;43(Database issue):D447–52. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [23].Chin CH, Chen S-H, Wu H-H, et al. cytoHubba: identifying hub objects and sub-networks from complex interactome. BMC Syst Biol. 2014;4(Suppl 4):S11. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [24].Butler A, Hoffman P, Smibert P, et al. Integrating single-cell transcriptomic data across different conditions, technologies, and species. Nat Biotechnol. 2018;36(5):411–420. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [25].Chen M, Heilbronn LK. The health outcomes of human offspring conceived by assisted reproductive technologies (ART). J Dev Orig Health Dis. 2017;8(4):388–402. doi: 10.1017/S2040174417000228 [DOI] [PubMed] [Google Scholar]
  • [26].Zubkova EV, Robaire B. Effect of glutathione depletion on antioxidant enzymes in the epididymis, seminal vesicles, and liver and on spermatozoa motility in the aging brown Norway rat. Biol Reprod. 2004;71(3):1002–1008. doi: 10.1095/biolreprod.104.028373 [DOI] [PubMed] [Google Scholar]
  • [27].Gat Y, Gornish M, Perlow A, et al. Azoospermia and sertoli-cell-only syndrome: hypoxia in the sperm production site due to impairment in venous drainage of male reproductive system. Andrologia. 2010;42(5):314–321. [DOI] [PubMed] [Google Scholar]
  • [28].Elfgen V, Mietens A, Mewe M, et al. Contractility of the epididymal duct: function, regulation and potential drug effects. Reproduction. 2018;156(4):R125–r141. [DOI] [PubMed] [Google Scholar]
  • [29].Kim B, Breton S. The MAPK/ERK-Signaling pathway regulates the expression and distribution of tight junction proteins in the mouse proximal epididymis. Biol Reprod. 2016;94(1):22. doi: 10.1095/biolreprod.115.134965 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [30].Kim B, Breton S. The MAPK/ERK signaling pathway regulates the expression and localization of Cx43 in mouse proximal epididymis†. Biol Reprod. 2022;106(5):919–927. doi: 10.1093/biolre/ioac034 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [31].Wang F, Liu W, Jiang Q, et al. Lipopolysaccharide-induced testicular dysfunction and epididymitis in mice: a critical role of tumor necrosis factor alpha†. Biol Reprod. 2019;100(3):849–861. [DOI] [PubMed] [Google Scholar]
  • [32].Raghunath A, Jeyabaskar D, Sundarraj K, et al. In silico prediction of microRnas on fluoride induced sperm toxicity in mice. Food Chem Toxicol. 2016;98(Pt A):34–49. [DOI] [PubMed] [Google Scholar]
  • [33].Dorostghoal M, Galehdari H, Moramezi F, et al. Sperm miR-26a-5p and its target PTEN transcripts content in men with unexplained infertility. Andrology. 2020;8(5):1167–1173. [DOI] [PubMed] [Google Scholar]
  • [34].Li DY, Yang X-X, Tu C-F, et al. Sperm flagellar 2 (SPEF2) is essential for sperm flagellar assembly in humans. Asian J Androl. 2022;24(4):359–366. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [35].Boonanuntanasarn S, Jangprai A, Na-Nakorn U. Transcriptomic analysis of female and male gonads in juvenile snakeskin gourami (Trichopodus pectoralis). Sci Rep. 2020;10(1):5240. doi: 10.1038/s41598-020-61738-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [36].He H, Yu F, Shen W, et al. The novel key genes of non-obstructive azoospermia affect spermatogenesis: transcriptomic analysis based on RNA-Seq and scRNA-Seq data. Front Genet. 2021;12:608629. doi: 10.3389/fgene.2021.608629 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [37].Sullivan R, Mieusset R. The human epididymis: its function in sperm maturation. Hum Reprod Update. Human Reproduction Update. 2016;22(5):574–587. doi: 10.1093/humupd/dmw015 [DOI] [PubMed] [Google Scholar]
  • [38].Belleannée C, Thimon V, Sullivan R. Region-specific gene expression in the epididymis. Cell Tissue Res. Cell Tissue Res. 2012;349(3):717–731. doi: 10.1007/s00441-012-1381-0 [DOI] [PubMed] [Google Scholar]
  • [39].Browne JA, Leir S-H, Yin S, et al. Transcriptional networks in the human epididymis. Andrology. 2019;7(5):741–747. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [40].Nistal M, Santamaría L, Paniagua R. The ampulla of the ductus deferens in man: morphological and ultrastructural aspects. J Anat. 1992;180(Pt 1):97–104. [PMC free article] [PubMed] [Google Scholar]
  • [41].Mortezapour M, Tapak L, Bahreini F, et al. Identification of key genes in colorectal cancer diagnosis by co-expression analysis weighted gene co-expression network analysis. Comput Biol Med. 2023;157:106779. doi: 10.1016/j.compbiomed.2023.106779 [DOI] [PubMed] [Google Scholar]
  • [42].Qing X, Shi J, Dong T, et al. Dysregulation of an X-linked primate-specific epididymal microRNA cluster in unexplained asthenozoospermia. Oncotarget. 2017;8(34):56839–56849. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [43].He Z, Feng L, Zhang X, et al. Expression of Col1a1, Col1a2 and procollagen I in germ cells of immature and adult mouse testis. Reproduction. 2005;130(3):333–341. [DOI] [PubMed] [Google Scholar]
  • [44].Griffiths M, Van Sinderen M, Rainczuk K, et al. miR-29c overexpression and COL4A1 downregulation in infertile human endometrium reduces endometrial epithelial cell adhesive capacity in vitro implying roles in receptivity. Sci Rep. 2019;9(1):8644. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [45].Edwards AK, van den Heuvel MJ, Wessels JM, et al. Expression of angiogenic basic fibroblast growth factor, platelet derived growth factor, thrombospondin-1 and their receptors at the porcine maternal-fetal interface. Reprod Biol Endocrinol. 2011;9(1):5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [46].Keskin T, Erden Y, Tekin S. Intracerebroventricular asprosin administration strongly stimulates hypothalamic-pituitary-testicular axis in rats. Mol Cell Endocrinol. 2021;538:111451. doi: 10.1016/j.mce.2021.111451 [DOI] [PubMed] [Google Scholar]
  • [47].Hassani F, Oryan S, Eftekhari-Yazdi P, et al. Downregulation of extracellular matrix and cell adhesion molecules in cumulus cells of infertile polycystic ovary syndrome women with and without insulin resistance. Cell J. 2019;21(1):35–42. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplemental Material
Click here for additional data file. (9.2MB, zip)

Data Availability Statement

All data series used in this study are available in online repository. The names of the repository and accession number of the included data series can be found in Table S1.

Articles from Cell Cycle are provided here courtesy of Taylor & Francis

RESOURCES