Compared to other animals, the anatomy of the human epididymis appears unusual. The caput epididymis is composed mostly of efferent ducts with an undefined initial segment. The present study investigates the regionalization of c-ros in human epididymis by real-time quantitative RT–PCR, in situ hybridization and immunohistochemistry studies. C-ros gene encodes a receptor-type protein tyrosine kinase that is expressed in adult mice exclusively in the epithelial cells of the initial segment of the epididymis. Transgenic mice targeted for the c-ros gene lack the initial segment of the epididymis and are infertile. Real-time PCR analysis showed that c-ros mRNA is expressed all along the human epididymis with the exception of the proximal caput epididymidis, where c-ros transcript was undetectable. In situ hydridization revealed that c-ros transcript was strongly expressed by principal cells and to a lower level by basal cells. Immunohistochemical studies showed that c-ros protein distribution was similar to the transcript. These results showed that c-ros expression in the human epididymis differs from that in mice suggesting that the unusual morphology of the human epididymis may reflect species differences in gene expression along the excurrent duct.
The epididymis consists of a single, involuted tubule that connects the testicular efferent ducts to the vas deferens. Epididymal functions include sperm protection and transport as well as sperm maturation and storage. The epididymis in most mammals can be roughly divided into four major regions: (i) a small initial segment; (ii) the proximal segment or caput; (iii) a narrow central region or corpus; and (iv) the cauda bulbous terminal segment (Robaire and Hermo 1988). These four regions can be distinguished by their epithelial cell morphology (Hermo et al.1994) and by their specific patterns of gene expression and enzyme activities (Kirchhoff, 1999).
The composition of the epididymal luminal compartment varies from one segment to the next. This results from reabsorption of constituents of testicular origin, transcytosis of serum components, and local secretion of protein by epithelial cells bordering the epididymal lumen. The complexity of the epididymal fluids underlies the well-orchestrated sequential modifications of the male gamete that lead to acquisition of fertilizing ability (Cooper, 1998). Thus, transcriptional activity will vary from one segment to the other to create segment-specific intraluminal microenvironments.
Cellular oncogenes were first detected as the homologues of the transforming genes of RNA tumour viruses (Bishop, 1985). Many oncogenes are mutated or activated analogues of cellular genes that normally function in signal transduction pathways. The c-ros proto-oncogene, which was isolated by a transfection-tumorigenicity assay, encodes a transmembrane protein with a sequence typical of tyrosine kinases (Birchmeier et al., 1986). C-ros is expressed frequently in cell lines established from glioblastomas, a type of human tumour. This proto-oncogene is also expressed in a small number of epithelial cell types, including those of the mouse epididymis where it is restricted to the initial segment. C-ros has a physiological function in differentiation and regionalization of the epididymal epithelium of mice. Dysregulation of epididymal initial segment differentiation in ros−/− mice leads to male infertility (Sonnenberg-Riethmacher et al., 1996). The infertility is associated with an inability of sperm to pass through the utero-tubal junction and the sperm are defective in osmoregulation of their intracellular volume (Yeung et al., 1998, 1999). Sperm volume regulation also occurs in human sperm and its blockage leads to failure in mucus penetration (Yeung and Cooper, 2001; Yeung et al., 2003).
The gross anatomy of the human epididymis is unusual when compared to other mammals. Most of the caput epididymidis is composed of efferent ducts with an undefined initial segment (Yeung et al., 1991). The cauda epididymis does not show the typical bulbous appearance, suggesting that the sperm reservoir capacity of the human excurrent duct is limited (Bedford, 1994; Turner, 1995). In view of the anatomical particularities of the human excurrent duct and the fact that murine c-ros mRNA is segregated in the initial segment in the mouse epididymis, the pattern of expression of this proto-oncogene along the human epididymis was investigated. Real-time RT–PCR, in situ hybridization and immunohistochemistry showed that human c-ros mRNA and protein are expressed all along the human epididymis except for the most proximal section. Here, we report new qualitative and quantitative data concerning the mRNA and protein expression profiles of c-ros in human epididymis. Our results suggest that the pattern of gene expression along the excurrent duct varies from one species to the other.
Materials and methods
Human epididymal tissues were obtained as previously described (Boué et al., 1996; Légaré et al., 1999). Briefly, reproductive tissues were recovered from three donors of 47, 48 and 52 years of age registered in our local organ transplant program. These donors were victims of accidental death and did not have pathologies that could affect the reproductive function. These procedures were approved by our local ethics committee. Tissues were collected while artificial circulation was maintained to preserve organs and tissues assigned for transplantation. Tissues piece from proximal and distal portions of caput, corpus and cauda epididymis, dissected according to Légaré et al. (1999), were prepared under optimal conditions for RNA extraction. Other pieces were fixed overnight in 4% paraformaldehyde in phosphate-buffered saline (PBS) at 4°C and then included in OCT (Optimal Cutting Temperature; Canemco Supplies, Canada) for in situ hybridization and immunohistochemistry.
Total RNA from frozen epididymis sections was isolated using Absolutely RNA RT–PCR Miniprep Kit following the manufacturer's directions (Stratagene, USA). Residual DNA was removed by treatment with 5 IU of DNAse1 at 37°C for 15 min. The RNA concentrations were estimated by spectrophotometry at 260 nm. A PCR product was generated to evaluate the presence of c-ros in human epididymis. Five micrograms of total RNA from caput epididymis was used for first strand cDNA synthesis, using 100 IU of SuperScript (Gibco-BRL, USA), RT buffer (50 mmol/l Tris, pH 8.3, 75 mmol/l KCl, 3 mmol/l MgCl2), 10 mmol/l dithiothreitol, 200 μmol/l dNTP and 10 μmol/l random primer p(dN)6 (Roche Diagnostics) in a final volume of 20 μl. The cDNA was used as the template in the PCR reaction mixture. The primers 5′-GGTGACAGTGCTTATAAACG-3′ (sense) and 5′-AAGGTTGGAATGCGCTGGATA-3′ (anti-sense), were derived from nucleotide sequences of human c-ros cDNA. Reactions were run with 5 μl of RT template or negative control and 1.5 IU of Taq DNA polymerase (Pharmacia, Canada) in a final volume of 50 μl. PCR amplifications were achieved following 30 cycles. An 820 base pair (bp) PCR amplification product was generated and cloned in pGEM-T (Promega, USA). All nucleotide sequences were determined by the dideoxynucleotide termination method (Sanger) using T7 Sequenase v 2.0 kit (Amersham, Canada).
Quantitative real-time RT–PCR
Total RNA of each human epididymal segment was extracted as described above. Using SuperScript II (Gibco-BRL) and random primer in final volume of 20 μl, cDNA was synthesized from 5 μg of RNA. To amplify the cDNA, 5 μl aliquots of reverse-transcribed cDNA (diluted 1:12.5) were amplified by PCR in 20 μl containing a final concentration of 3 mmol/l MgCl2, 50 ng of each primer (5′-CATCTGATGAGCAAATTTAAT-3′; 5′-CTACACACAGGTCTACAAGG-3′) and 2 μl of ready-to-use LightCycler DNA master SYBRGreen I (Roche Diagnostics; 10×, containing TaqDNA polymerase, reaction buffer, dNTP mix with dUTP instead of dTTP, SYBRGreen I dye, and 10 mmol/l MgCl2). The plasmid containing the 820 bp c-ros fragment was used to validate the LightCycler reaction and to determine the quantification range (calibration curve). The reaction conditions were as follows: initial denaturation at 95°C for 10 min followed by 35 cycles of denaturation at 95°C for 1 s, annealing at 57°C for 5 s, and extension at 72°C for 20 s. The temperature ramp was 20°C/s, except when heating to 72°C, when it was 2°C/s. At the end of the extension step, fluorescence of each sample was measured to allow quantification of the RNA. After amplification, a melting curve was obtained by heating at 20°C/s to 95°C, cooling at 20°C/s to 60°C, and slowly heating at 0.1°C/s to 95°C with fluorescence data collected at 0.1°C intervals. To control for the recovery of intact RNA and for the uniform efficiency of each RT reaction, a GAPDH fragment was amplified by real-time RT–PCR using these primers: 5′-GAAGACTGTGGATGGCCCCTC-3′ and 5′-GTTGAGGGCAATGCCAGCCCC-3′.
Quantitative analysis of the LightCycler data was performed using the LightCycler analysis software. The data analysis was divided into two parts: specificity control of the amplification reaction using the melting curve program of the LightCycler software followed by use of the quantification program. The SYBR Green I signal of each sample was plotted against the number of cycles. Using the LightCycler analysis software, background fluorescence was removed by setting a noise band. This fluorescence threshold was used to determine cycle number that correlated inversely with the log of the initial template concentration. To this end, the log–linear portions of the amplification curves were identified and best-fit lines calculated. The crossing points were the intersections between the best-fit lines of the log–linear region and the noise band. These crossing points correlated inversely with the log of the initial template concentration (LightCycler operator's Manual, Version II). The crossing points determined for c-ros mRNA were normalized to those of GAPDH to compensate for variability in RNA amount. Levels of c-ros and GAPDH mRNA were determined three times on each epididymal segment from each donor. Results are expressed as mean of c-ros/GAPDH±SEM.
In situ hybridization
cDNA insert was generated by RT–PCR using poly(A)RNA from normal human caput epididymidis. The oligonucleotide sequences used for primers were 5′-GGTGACAGTGCTTATAAACG and 5′-AAGGTTGGAATGCGCTGGATA. PCR product was subcloned into pGEM-T (Promega, USA). All nucleotide sequences were determined by the dideoxynucleotide termination method (Sanger) using T7 Sequenase v 2.0 kit (Amersham, Canada).
RNA probes (corresponding to nucleotides 5891–6710 in the human c-ros cDNA) were transcribed using the Digoxygenin (DIG) RNA labelling technique for in vitro transcription (Roche Diagnostics, Canada). Briefly, plasmid was digested with appropriate restriction endonucleases downstream from the target DNA insert. mRNA was transcribed using SP6 and T7 RNA polymerase (Roche Diagnostics) in the presence of DIG-UTP.
Fixation and pretreament of sections
Epididymis cryosections were fixed with freshly prepared 4% (w/v) paraformaldehyde in PBS for 5 min at room temperature, incubated for 10 min in 95% ethanol/5% acetic acid at −20°C, and rehydrated in successive baths of decreasing concentrations of ethanol diluted with diethylpyrocarbonate (DEPC)-treated water. Target RNA were unmasked by enzymatic digestion with 10 μg/ml proteinase K (Roche Diagnostics) in PBS for 10 min at 37°C, followed by a 5 min incubation in 0.2% glycine. Sections were post-fixed for 5 min with 4% paraformaldehyde in PBS, acetylated with 0.25% acetic anhydride, 0.1 mol/l triethanolamine pH 8.0 for 10 min, and finally washed with PBS.
Tissues were prehybridized for 2 h at 42°C, with 250 μg/ml salmon sperm DNA preheated in a hybridization solution (0.3 mol/l NaCl, 0.01 mol/l Tris–HCl pH 7.5, 1 mmol/l EDTA, 1 × Denhardt's solution [0.2% (w/v) Ficoll 400, 0.2% (w/v) polyvinylpyrrolidone, 0.2% (w/v) bovine serum albumin (BSA), 5% dextran sulphate, 0.02% sodium dodecyl sulphate and 50% formamide]. Sections were then incubated overnight at 42°C, under coverslips, with 25 μl of 10 μg/ml heat-denatured antisense or sense c-ros cRNA probed with DIG (Roche Diagnostics) according to the supplier's instructions. Sections were washed twice in 2× standard saline citrate (SSC) at room temperature, followed by two 10 min washes at 42°C in 2 × SSC, 1 × SSC and 0.2 × SSC.
Hybridization reactions were detected by immunostaining with alkaline phosphatase-conjugated DIG antibodies (Roche Diagnostics). Non-specific staining was blocked by preincubation for 1 h with 5% (v/v) heat-inactivated sheep serum in Tris–HCl/NaCl buffer (0.2 mol/l Tris–HCl, 0.2 mol/l NaCl, 0.3% Triton X-100). Sections were then incubated for 2 h at room temperature with the alkaline phosphatase-conjugated anti-DIG antibodies diluted 1:1000 in blocking solution, washed with Tris–HCl/NaCl buffer, and incubated with 0.1 mol/l Tris–HCl, pH 9.5, 0.1 mol/l NaCl, and 0.01 mol/l MgCl2. The hybridization signal was visualized after a 10–15 min incubation period with the phosphatase substrate, nitroblue tetrazolium chloride and 5-bromo-4-chloro-3-indolylphosphate p-toluidine salt (Gibco-BRL, USA). Levamisole (2 mmol/l; Sigma) was added to the reaction mixture to inhibit endogenous alkaline phosphatase. Microscopic slides were immersed in 1 mmol/l EDTA, 10 mmol/l Tris–HCl, pH 7.5, washed 5 min in H2O, counterstained with Neutral Red, dehydrated through baths of ethanol, cleared in xylene, and mounted with Permount (Fisher Scientific, Canada). Epididymis sections were processed in parallel to allow comparison.
Rabbit polyclonal antibody against human c-ros was purchased from Cederlane Laboratories Ltd (Canada) and used at 10 μg/ml for immunohistochemistry. Biotinylated goat anti-rabbit secondary antibody was obtained from Dako Diagnostics (Canada) and used at 1/400 (vol/vol).
Cryosections of 10 μm were prepared from frozen epididymal tissues. Endogenous peroxidase activity was quenched with 3% H2O2 (v/v) in PBS for 30 min. Non-specific binding sites were then blocked with 10% goat serum in PBS for 1 h. The c-ros-specific antibodies were diluted in PBS and applied overnight at 4°C. In control sections, the primary antibodies were replaced by the corresponding non-specific IgG and processed in parallel. Sections were subsequently incubated with biotinylated goat anti-rabbit antibody for 60 min, and with ABC elite reagent for 30 min. Immunostaining was revealed using 3-amino-9-ethylcarbazole (AEC). Mayer's haematoxylin solution was used for counterstaining, and mounted under cover slips using an aqueous mounting medium (Sigma). Slides were observed under a Zeiss Axioskop2 Plus microscope (Canada) linked to a digital camera from Diagnostic Instruments (USA). Images were captured using the Spot software (Diagnostic Instruments).
Expression of c-ros mRNA in the human epididymis
To verify if c-ros mRNA is present in the human epididymis, a RT–PCR study was performed using caput epididymidal mRNA. To prevent the amplification of genomic DNA, PCR primers were designed to be located in exon 3 (sense) and exon 9 (anti-sense) of human c-ros. The 820 bp PCR amplification product showed a homology of 99% with two cDNA encoding a transmembrane protein kinase: human homologue of the avian v-ros oncogene, called ros1; and human oncogene mcf3 (rearranged ros1) (Figure 1). Of the 820 nucleotides sequenced, only nucleotides 6453 and 6698 differed from the published human ros1 sequence. These nucleotide differences do not modify the encoded amino acid sequence.
The quantification of low abundance cellular transcripts requires sensitive techniques. Real-time quantitative RT–PCR (LightCycler system) was used to compare mRNA levels along the human excurrent duct. The expression of the housekeeping GAPDH mRNA was quantified with gene-specific primers for each segment of the epididymis. When the data were normalized to that of the housekeeping GAPDH, different levels of c-ros transcript expression were found all along the epididymis of the three donors with the exception of the proximal caput epididymiis (Figure 2). C-ros/GAPDH for a particular epididymal segment showed variation from one donor to the other.
Region-specific localization of c-ros transcripts
In situRNA hybridization was used to study the spatial distribution of c-ros transcript in human epididymis. The DIG-labelled antisense RNA probe is generated from a 820 bp fragment of human c-ros cDNA (Figure 1). It appears that c-ros mRNA was expressed all along the human epididymis (Figure 3) except for the most proximal segment of the epididymides, the ‘initial-segment-like’ portion characterized by a small diameter and a low epithelial height (Figure 3A). Throughout the epididymal duct, c-ros mRNA staining was localized in the cytoplasm of basal cells and principal cells. No signal was detected when a sense strand of c-ros cRNA was used as a negative control (Figure 3C*).
Cellular distribution of c-ros protein in human epididymis
Localization of c-ros protein to specific cell types was carried out by immunohistochemical studies. Figure 4 shows typical c-ros distribution on a corpus epididymis histological section. The protein was strongly expressed in the cytoplasm of basal cells in the corpus epididymis. Principal cells were also reactive for c-ros, especially in the supranuclear cytoplasmic compartment (Figure 4A). Faint staining in interstitial tissues was also noticed. The same pattern of distribution was revealed along the distal caput to the distal cauda epididymidis (data not shown). Immunostaining for c-ros was not observed in control sections incubated with preimmune serum (Figure 4B).
Male gamete functionality is acquired during transit through the epididymis, which provides a good milieu for sperm maturation and storage in the tubule lumen by absorptive and secretory activities of the epididymal epithelium. When compared with other mammalian species, the human epididymis is poorly differentiated and this affects human excurrent duct functions. Whereas the storage capacity of the human epididymis is rather limited due to limited cauda epididymal sperm reservoir capacity, its importance in sperm maturation is controversial. This is based on fertility data obtained in pathological situations, mainly excurrent duct obstruction or agenesis. The morphology of the proximal epididymis in humans further obscures its role. In laboratory animals, the initial segment appears to be a major player in the acquisition of both motility and fertilizing ability of the male gamete. The absence of such differentiated proximal epididymal region argues against the importance of epididymis in sperm maturation in humans.
Cellular Ros gene resembles several proto-oncogenes that encode for a transmembrane molecule which may function as a receptor for a cell growth or differentiation factor(s) (Ullrich et al., 1985; Ullrich et al., 1986; Wada et al., 1993). C-ros is selectively expressed in the initial segment and caput of the murine epididymis. The expression of c-ros appears to be essential for differentiation of the initial segment of the mouse epididymis (Sonnenberg-Riethmacher et al., 1996). Indeed, in c-ros knockout (KO) mice, which are healthy but infertile animals, the initial part of the epididymis is not differentiated during puberty into the characteristic initial segment. These mice deposit the normal number of motile sperm into the uterus during normal mating, but the ejaculated sperm are swollen into abnormal shapes in the uterus and fail to migrate into the oviduct (Yeung et al., 2000) because of defective cell volume regulation (Yeung et al., 1999, 2002). Apparently a sperm defect is caused by epididymal malfunction (Yeung et al., 1998).
Two partial human ros sequences have been reported previously. One, the cDNA sequence of the activated human ros1 gene, MCF3, encodes a truncated protein missing all but eight amino acids of the extracellular domain (Birchmeier et al., 1986). The other ros1 sequence was determined from placenta genomic DNA clone isolated by its homology to the chicken v-ros gene (Matsushime et al., 1986). There are no differences between the coding sequences of the MCF3 gene and the human placental ros1 gene. The partial sequence obtained by RT–PCR reaction on human caput epididymidis confirmed the presence of c-ros. Two nucleotides difference between the c-ros PCR fragment and the published human ros sequence are noted. These differences, however, have no consequences on the deduced amino acid sequence. The sequence deduced on human epididymal tissue is located in the phosphotyrosine kinase region, which is highly conserved between species (Sonnenberg et al., 1991).
In contrast to the mouse epididymis where c-ros mRNA expression is restricted to the caput region, c-ros transcript is present all along the human epididymis at different levels. The epididymal epithelium in man is organized as a pseudo-stratified columnar epithelium with basal cells. Human principal cells are unusual in that they contain many more inclusions, such as dense bodies and lysosomes, which tend to disrupt the organization of the cells (Moore et al., 1983). In mouse, the initial segment consists of tall columnar epithelia expressing high levels of c-ros mRNA. Distal tubules in the caput contain thinner columnar epithelia expressing lower levels of c-ros transcript. In corpus and cauda epididymal sections where the height of the epithelium is shorter, c-ros mRNA expression is absent. There is no such correlation between the epithelium height bordering the epididymal lumen and the expression of c-ros mRNA in humans. In c-ros knockout mice, the tall columnar cells are completely absent in the proximal part of the caput, the initial segment. Instead, distinct low columnar epithelial cells are found in the proximal epididymis of these mice. Transforming the initial segment into a distal caput-like structure directly results in changes in the sperm osmoregulatory functions, resulting in male infertility. The levels of c-ros mRNA along the human epididymis show inter-individual variability which does not seem to be correlated with the age of donors. However, the overall pattern of c-ros expression is consistent from one man to the other, as shown by almost undetectable c-ros mRNA transcript in the most proximal part of the epididymis. This pattern of expression greatly contrasts with what is known in the murine model.
Cellular ros mRNA expression was determined by in situ hybridization analysis using human c-ros-specific probe. Consistent with the results from real-time RT–PCR, c-ros transcript was detected in almost all segments of the epididymis. In all segments, c-ros mRNA was detected in the basal cells and principal cells surrounding the tubules of the epididymal duct. The human epididymal tubule bearing the epithelium of an initial segment is not recognizable macroscopically as a separate region, and does not constitute a continuous tubule like the initial segment in mice. In fact, most of the caput epididymis is made up of efferent ducts (Yeung et al., 1991). We denoted no hybridization signal for c-ros in the efferent ducts as well as in the most proximal segment of the human epididymis.
The localization of c-ros protein in murine epididymis has not been investigated. If c-ros were directly involved in initial segment differentiation and maintenance through interaction with luminal factor ligands, localization at the apical/luminal border of the epithelium would be anticipated (Cooper and Yeung, 2002). In humans, consistently with mRNA expression, the cellular localization of c-ros protein is selective for basal cells and principal cells. The origin and function of basal cells being obscured, the significance of c-ros expression in this epididymal cell type remains to be determined (Yeung et al., 1994). By its localization, C-ros may be involved in a more general function of the human epididymis than that hypothesized in the murine model.
In summary, c-ros is present in the human epididymis and its expression pattern is different from that described in the epididymis of mice. Transcription of c-ros in the human excurrent duct does not seem to be as highly segregated as it is in mice. Even though c-ros plays important regulatory roles in sperm maturation, its function in humans remains to be determined. The unusual morphology of the human epididymis may reflect species differences in gene expression patterns along the excurrent duct. Results from animal models should be used with caution when investigating the function of the human epididymis.
We thank Dr M.A.Sirard for providing access to the LightCycler, Mr Dominic Gagné and Mr Claude Robert for their technical advice. This work was supported by a grant to R.S. from the Canadian Institutes for Health Research.
1Centre de Recherche en Biologie de la Reproduction and 2Département d'Obstétrique–Gynécologie, Faculté de Médecine, Université Laval, Québec, Québec, Canada