Region-specific microRNA signatures in the human epididymis

Abstract

The epithelium of the human epididymis maintains an appropriate luminal environment for sperm maturation that is essential for male fertility. Regional expression of small noncoding RNAs such as microRNAs contributes to segment-specific gene expression and differentiated functions. MicroRNA profiles were reported in human epididymal tissues but not specifically in the epithelial cells derived from those regions. Here, we reveal miRNA signatures of primary cultures of caput, corpus, and cauda epididymis epithelial cells and of the tissues from which they were derived. We identify 324 epithelial cell-derived microRNAs and 259 tissue-derived microRNAs in the epididymis, some of which displayed regionalized expression patterns in cells and/or tissues. Caput cell-enriched miRNAs included miR-573 and miR-155. Cauda cell-enriched miRNAs included miR-1204 and miR-770. Next, we determined the gene ontology pathways associated with in silico predicted target genes of the differentially expressed miRNAs. The effect of androgen receptor stimulation on miRNA expression was also investigated. These data show novel epithelial cell-derived miRNAs that may regulate the expression of important gene networks that are responsible for the regionalized gene expression and function of the epididymis.

INTRODUCTION

The epididymis epithelium supports a luminal environment that promotes sperm maturation, and each region of the duct (caput, corpus, and cauda) has a unique role in the process. These region-specific functions are maintained by distinct gene expression signatures,1,2,3 which are coordinated by a network of transcription factors and also by noncoding RNAs. MicroRNAs (miRNAs) are one family of small noncoding RNAs that regulate gene expression posttranscriptionally, generally by binding to specific motifs in the 3’-untranslated regions (3’UTRs) of target genes (reviewed by Bartel4). miRNAs coordinate diverse biological processes including stem cell maintenance,5 development, metabolism,6 proliferation,7 differentiation,8 and apoptosis.9 In the epididymis, miRNAs may have tissue-specific roles and also be released from the epididymis epithelium in exosomes, which can be taken up by transiting sperm.10,11 Conditional deletion of Dicer, a critical component of the mature miRNA processing pathway, from epididymal principal cells of the mouse has a dramatic effect on the epithelium and impairs male fertility.12,13 Androgens are required for normal epididymal structure and function (reviewed by Robaire and Hamzeh14) and have been shown to regulate miRNA expression in rodents.15,16 Region-specific miRNA profiles were characterized in rodents17,18 and in whole human epididymis tissue by microarray analysis19 and RNA-seq.20 We recently described the transcriptome of human caput, corpus, and cauda epididymis tissues and primary epithelial cell cultures derived from each region,3 though miRNAs were not discussed. Here, we describe the region-specific expression of microRNAs in the caput, corpus, and cauda epididymal cells and tissues. Further, we use in silico prediction methods to identify candidate targets of several abundant miRNAs, which may directly impact regional functions of the epididymal epithelium.

PATIENTS AND METHODS

Preparation of primary cultures

Human epididymis tissue was obtained with the institutional review board permission from institutions listed in the author affiliations, and with informed consent from patients undergoing inguinal radical orchiectomy for a clinical diagnosis of testicular cancer. None of the epididymides had extension of the testicular cancer, and no donors were receiving hormone or drug treatments before surgery. Efferent ducts were removed and the three anatomical regions of the epididymis: caput, corpus, and cauda, were separated and segments of each were either snap frozen in liquid nitrogen or epithelial cells were isolated and established in culture as described previously.21 For the experiments to test androgen receptor (AR) function, cells were cultured in phenol-red-free CMRL-1066 medium containing 10% fetal bovine serum (FBS), hormone depleted with dextran coated-charcoal (C6241; Sigma, St. Louis, MO, USA), for 72 h before stimulation with vehicle or the synthetic androgen R1881 (1 nmol l⁻¹, methyltrienolone, NLP005005MG; PerkinElmer, Waltham, MA, USA) for a further 16 h.

RNA sequencing

RNA was extracted using TRIzol (Life Technologies, Carlsbad, CA, USA) as per the manufacturer's protocol. RNA quality was confirmed by NanoDrop (NanoDrop™ One, Thermo Fisher Scientific, Waltham, MA, USA) measurement of OD 260/280 and 260/230 ratios, and the RNA was stored at −80°C under ethanol. RNA integrity was verified by the Bioanalyzer (Agilent, Santa Clara, CA, USA), and RNA-seq libraries were prepared using the TruSeq RNA Sample Preparation Kit v2 as per the manufacturer's Low-throughput protocol (Illumina, San Diego, CA, USA). The libraries were sequenced on Illumina HiSeq2500 machines and generated 1.9 × 10⁷ – 3.9 × 10⁷ reads per library from the cultured cells (95%–99% mapping to the genome) and 1.4 × 10⁷ – 3. 9 × 10⁷ reads per library from tissues (84%–99% mapping to the genome). Data were analyzed using TopHat and Cufflinks.22 All data are deposited at GEO (http://www.ncbi.nlm.nih.gov/geo/GSE72986).

In silico analysis of miRNAs and target prediction

Putative mRNA targets of differentially expressed miRNAs were predicted using TargetScan 7.0 (http://www.targetscan.org/)23,24 and miRecords.25 Only data generated with TargetScan are presented here, as the miRecord database was incomplete with respect to miRNAs of interest. Gene ontology process enrichment analysis26,27 was performed to identify statistically significant biological processes associated with the miRNA targets (as shown by both P value and false discovery rate [FDR]).

RESULTS

Regional miRNA expression in the human epididymis

To identify the regional microRNA signature of the epididymis epithelium, we examined RNA-seq data from both cultured human epididymal epithelial (HEE) cells and tissues from the caput, corpus, and cauda segments.3 The majority of the 324 miRNAs identified in HEE cells (Supplementary Table 1 ^{(9.4MB, tif)} ) were expressed in more than one region, as were most of the 259 tissue miRNAs (Supplementary Table 2 ^{(7.5MB, tif)} ). Fifty-seven percent (185/324) of the HEE cell miRNAs were also present in the tissues they were derived from. Regionally restricted miRNAs were defined by differential expression of at least 2.5-fold change between caput, corpus, and cauda HEE cells and minimum gene expression levels of ≥0.3 fragments per kilobase of transcript per million mapped reads (FPKM; Table 1). The same parameters were used in a comparison of differential gene expression for the miRNAs of the caput, corpus, and cauda tissues (Table 2).

Table 1.

Differentially expressed miRNAs comparing the caput, corpus and cauda epididymis cells

Table 2.

Differentially expressed miRNAs comparing the caput, corpus and cauda epididymis tissues

Supplementary Table 1

List of microRNAs in primary human epididymal epithelial cells (fragments per kilobase of transcript per million mapped reads [FPKM] >0.1)

Supplementary Table 2

List of microRNAs in primary human epididymal tissues (fragments per kilobase of transcript per million mapped reads [FPKM] >0.1)

Caput-enriched miRNAs of the cells and tissues

Two microRNAs (miR-573 and miR-155) were significantly more abundant in caput cells than in corpus or cauda HEE cells (A and B in Table 1). miR-30c2 was also enriched in the caput compared to corpus cells, and miR-196A1 was enhanced in the caput over the cauda cells. The most significant differentially expressed miRNAs observed in intact tissues were distinct from those seen in HEE cells, with miR-1247 more abundant in the caput tissue compared to both the corpus and cauda (A and B in Table 2). Likewise, miR-4461 was enriched in the caput in comparison to the corpus and cauda tissues though it was abundant in all tissue regions (A and B in Table 2). Sixteen other miRNAs were more highly expressed in the caput than the cauda tissue (B in Table 2).

Corpus-enriched miRNAs of the cells and tissues

Although our analysis did not identify any corpus-specific cell-derived miRNAs, four miRNAs (miR-4730, miR-196a1, miR-let7d, and miR-3916) were enriched in the corpus over the cauda cells (C in Table 1). In the tissues, miR-662 and miR-3936 were more abundant in the corpus than either the caput or cauda. Three other miRNAs (miR-33b, miR-135b, and miR-3074) were enriched in the corpus over the caput tissues (A in Table 2). Multiple other tissue-derived miRNAs were enriched in the corpus over the cauda tissues, including the highly expressed miR-205, miR-141, and miR-3648-miR-3687 (C in Table 2).

Cauda-enriched miRNAs of the cells and tissues

In HEE cells, miR-1204 showed enhanced expression in the cauda compared to both the caput and corpus cells (Table 1). H19-miR-675 was also enriched in the cauda over the corpus cells. miR-770 and miR-let7i were higher in the cauda than the caput cells (B in Table 1). In the tissues, miR-146a was more abundant in the cauda compared to the caput and corpus (B in Table 2).

Also of note, but solely in the tissues, miR-135b and miR-3074 were more abundant in both the corpus and the cauda (A in Table 2) than in the caput (B in Table 2).

In silico analysis of microRNA-regulated processes of human epididymal epithelial cells

Putative target genes of the differentially expressed miRNAs were predicted using TargetScan 7.0 (http://www.targetscan.org/).23,24 We chose this in silico prediction tool since it was one of the few such resources that were recently updated and included all the differentially expressed miRNAs in the HEE cells. Moreover, it was a robust predictive tool in our recent work on other miRNAs.28,29,30 Target genes (<−0.2 total context score) were then subjected to gene ontology process enrichment analysis using the Database for Annotation, Visualization, and Integrated Discovery (DAVID).26,27 The top 10 predicted targets and their associated processes are shown in Table 3 and 4.

Table 3.

Predicted target genes of caput-enriched microRNAs (left) and their enriched processes (right)

Table 4.

Predicted target genes of cauda-enriched microRNAs (left) and their enriched processes (right)

MicroRNA-regulated processes of the caput

Processes enriched among the predicted targets of the caput cell-enriched miRNAs (miR-155 and miR-573) were identified first. For miR-155-3p and miR-155-5p, these included predominantly aspects of transcriptional regulation and membrane-enclosed lumens of both nuclei and intracellular organelles (A and B in Table 3). Processes associated with miR-573 included some relevant to modification-dependent protein catabolic processes and as for miR-155, to the lumens of nuclei and intracellular organelles including the mitochondrion (C in Table 3).

MicroRNA-regulated processes in the corpus and cauda

Although no corpus-specific HEE cell miRNAs were identified here, processes associated with the predicted targets of cauda cell-enriched miRNAs (miR-1204, miR-770, and miR-let7i) are shown in Table 4. miR-1204-associated processes include “small GTPase-mediated intracellular signaling” and “regulation of cell motion” (A in Table 4). The diverse processes associated with miR-770 included those relevant to metal ion binding (including zinc), transcription, and DNA (B in Table 4). Transcriptional regulation was also enriched in the miR-let7i-associated processes, together with multiple processes of less obvious relevance to HEE cell function including “regulation of neuron differentiation” and “blood vessel development” (C in Table 4).

Androgen-regulated miRNAs of the epididymis

To investigate whether any miRNAs were regulated by androgens in the human epididymis epithelium, caput HEE cells were treated with vehicle or R1881 (1 nmol l⁻¹) for 18 h and gene expression was analyzed by RNA-seq (Yang et al., manuscript in review). Eight miRNAs were differentially expressed following R1881 treatment. Of these, four miRNAs were downregulated (miR-137, miR-3074, miR-3190, and miR-3916) and four were upregulated (miR-4740, miR-506, miR-573, and miR-let7d; P = 5 × 10⁻⁵, q = 0.007, Table 5).

Table 5.

Differentially expressed microRNAs in caput human epididymal epithelial cells after R1881 treatment compared to vehicle

DISCUSSION

Mechanisms that control gene expression along the human epididymis epithelium are pivotal to coordinating its role in sperm maturation and male fertility. One aspect of this coordination likely involves noncoding RNAs including microRNAs. miRNAs were profiled previously in tissue samples from the human epididymis that contained many cell types.19,20,31 In order to focus on the specific functions of cells within the epithelial layer lining the epididymis, we established the HEE cell culture model.21 Here, we use RNA-seq analysis to reveal the miRNA signatures of caput, corpus, and cauda HEE cells and the tissues from which they were derived.

Tissue-derived miRNAs

Among the epididymis tissue-derived miRNAs identified in this study, we detected approximately 22.0% (116/527)20 and approximately 24.1% (81/336)19, respectively of the tissue-derived miRNAs previously reported by others. As previously noted by Belleannee et al.19 most of the tissue miRNAs were present in all three regions, with similar observations of HEE cell-derived miRNAs. Furthermore, most of the HEE cell-derived miRNAs were also evident in the tissues.

MicroRNA-regulated processes of the caput cells

miR-573 and miR-155, which are differentially expressed in caput HEE cells compared to corpus or cauda cells, were investigated previously in other contexts. miR-573 is downregulated by the inflammatory cytokine tumor necrosis factor (TNF)-alpha in primary human airway epithelial cells.32 More relevant to the male reproductive tract, miR-573 is downregulated in testicular tissue from patients with nonobstructive azoospermia.33 It is perhaps of relevance that we observed miR-573 among miRNAs that were upregulated in androgen-stimulated caput HEE cells. Gene ontology process enrichment analysis on the mRNA targets of miR-573 identified processes related to protein modification by small protein conjugation or removal and modification-dependent protein catabolic process, both of which might be relevant to maintenance of optimal luminal environment in the epididymis. The role of miR-155 in the male reproductive tract has yet to be explored, but it is well characterized as a pleiotropic regulator of both immunity and cancer.34,35,36

MicroRNA-regulated processes of the corpus/cauda cells

The role of miR-1204 was investigated in human carcinoma cell lines,37,38 though to date there is no information on its involvement in processes relevant to the male reproductive tract. miR-770 is downregulated in testicular tissue from patients with nonobstructive azoospermia.33 miR-let7i negatively regulates cardiac inflammation and fibrosis39 and is downregulated in the serum from ovarian cancer patients.40 However, deciphering its role in the epididymis will require further study.

Androgen-regulated miRNAs in the caput epididymis

Androgens are important for epididymal epithelial structure and function (reviewed by Robaire and Hamzeh14) and regulate their effects via the AR. We previously showed enrichment of AR protein in caput HEE cells and its nuclear accumulation in these cells in response to R1881.3,21 Hence, here, we also examined the effect of R1881 on miRNA expression in caput HEE cells. Among the miRNAs that were differentially regulated in response to R1881, miR-137 is of interest since this miRNA is differentially expressed after androgen treatment in the prostate adenocarcinoma cell line, LnCaP.41 Of note, in LnCaP cells, androgen treatment increases miR-137 in contrast to the repression observed here in HEE cells. However, the overamplification of AR in LnCaP cells compared to normal prostate epithelium and the strong context-dependence of AR cofactors could account for these differences. Another androgen-regulated miRNA in the caput epididymis is miR-506, which plays a complex role in cancer. It is oncogenic in melanomas,42 a tumor suppressor in ovarian cancer,43 and confers chemoresistance in colon cancer.44

In conclusion, our data suggest that further investigation of regionalized miRNA expression along the epididymis may contribute to the understanding of mechanisms controlling segment-specific epididymal epithelial function.

AUTHOR CONTRIBUTIONS

JAB, SHL, SEE, and AH acquired, analyzed, and interpreted data. JAB and AH wrote the manuscript. All authors read and approved the final manuscript.

COMPETING INTERESTS

All authors declared no competing interests.