ATP secretion in the male reproductive tract: essential role of CFTR

Abstract

Extracellular ATP is essential for the function of the epididymis and spermatozoa, but ATP release in the epididymis remains uncharacterized. We investigated here whether epithelial cells release ATP into the lumen of the epididymis, and we examined the role of the cystic fibrosis transmembrane conductance regulator (CFTR), a Cl⁻ and HCO₃⁻ conducting ion channel known to be associated with male fertility, in this process. Immunofluorescence labelling of mouse cauda epididymidis showed expression of CFTR in principal cells but not in other epithelial cells. CFTR mRNA was not detectable in clear cells isolated by fluorescence-activated cell sorting (FACS) from B1-EGFP mice, which express enhanced green fluorescent protein (EGFP) exclusively in these cells in the epididymis. ATP release was detected from the mouse epididymal principal cell line (DC2) and increased by adrenaline and forskolin. Inhibition of CFTR with CFTR_inh172 and transfection with CFTR-specific siRNAs in DC2 cells reduced basal and forskolin-activated ATP release. CFTR-dependent ATP release was also observed in primary cultures of mouse epididymal epithelial cells. In addition, steady-state ATP release was detected in vivo in mice, by measuring ATP concentration in a solution perfused through the lumen of the cauda epididymidis tubule and collected by cannulation of the vas deferens. Luminal CFTR_inh172 reduced the ATP concentration detected in the perfusate. This study shows that CFTR is involved in the regulation of ATP release from principal cells in the cauda epididymidis. Given that mutations in CFTR are a leading cause of male infertility, we propose that defective ATP signalling in the epididymis might contribute to dysfunction of the male reproductive tract associated with these mutations.

Key points

• CFTR is expressed in principal cells but not clear cells in mouse cauda epididymis.

• Inhibition or knockdown of CFTR inhibits ATP release from mouse epididymal principal cells.

• Inhibition of CFTR reduces ATP release into the lumen of cauda epididymis in mice in vivo.

• These results show the involvement of CFTR in the regulation of ATP release from epithelial principal cells in the cauda epididymidis.

• Defective ATP signalling in the epididymis might contribute to dysfunction of the male reproductive tract associated with CFTR mutations.

Introduction

While ATP serves as a major energy source within cells, it can also be secreted in response to physiological stimuli (Schwiebert & Zsembery, 2003; Burnstock, 2006). Extracellular ATP is a signalling molecule that regulates a variety of functions such as neurotransmission (Burnstock, 2007), muscle contraction (Ralevic, 2009) and water/electrolyte transport (Rieg & Vallon, 2009). In the male reproductive system, extracellular ATP is involved in several critical processes including spermatogenesis (Loir, 1999), steroidogenesis (Foresta et al. 1996), sperm motility acquisition (Edwards et al. 2007), epithelial ion secretion (Wong, 1988; Chan et al. 1995) and smooth muscle contraction (Ruan et al. 2008). In the epididymis, a small organ located downstream of the testis where sperm acquire their fertilizing abilities, extracellular ATP regulates transepithelial electrolyte and water transport (Wong, 1988; Zhou et al. 2007). The epididymal epithelium consists of different cell types including principal, apical, narrow, clear and basal cells, which all contribute to creating an optimal luminal environment for sperm maturation and storage (Robaire & Viger, 1995; Shum et al. 2009; Hinton & Cooper, 2010). We recently demonstrated that luminal ATP and its hydrolyte, adenosine, induce vacuolar H⁺-ATPase (V-ATPase) accumulation in the apical membrane of clear cells in the cauda epididymidis (Belleannée et al. 2010). V-ATPase-dependent H⁺ secretion leading to luminal acidification contributes to keeping sperm dormant during their maturation and storage in the epididymis (Carr et al. 1985; Shum et al. 2009; Vidarsson et al. 2009) and facilitates the acquisition of new sperm proteins during epididymal transit (Sullivan et al. 2007).

The biological activities of extracellular ATP have been well studied, but the pathways involved in ATP release remain incompletely characterized. Multiple transport mechanisms have been proposed including exocytosis of ATP-containing vesicles (Bodin & Burnstock, 2001; Schwiebert & Zsembery, 2003) and permeation through conductive pores (Sugita et al. 1998; Braunstein et al. 2001; Sabirov & Okada, 2005; Praetorius & Leipziger, 2009; Ransford et al. 2009; D’Hondt et al. 2011; Li et al. 2011). In addition, cystic fibrosis transmembrane conductance regulator (CFTR), the cAMP activated anion channel vital to many organ systems (Sheppard & Welsh, 1999; Chan et al. 2009), was shown to regulate ATP release from airway epithelial cells (Schwiebert et al. 1995; Sugita et al. 1998; Taylor et al. 1998; Braunstein et al. 2001; Schwiebert & Zsembery, 2003), cardiomyocytes (Lader et al. 2000), kidney cells (Sugita et al. 1998) and retinal pigment epithelial cells (Reigada & Mitchell, 2005). While CFTR was originally considered to be an ATP channel (reviewed in Praetorius & Leipziger, 2009 and Sabirov & Okada, 2005), it was proposed later that CFTR regulates a separate ATP pathway, in airway and kidney cells for example (Reddy et al. 1996; Sugita et al. 1998; Braunstein et al. 2001).

In the epididymis, no direct evidence has been shown as to whether ATP is released from the epithelium, and how this process is regulated. Of note, CFTR is expressed in the epididymal epithelium (Ruz et al. 2004; Hihnala et al. 2006; Kujala et al. 2007; Pietrement et al. 2008), where it functions as an anion channel responsible for Cl⁻ and HCO₃⁻ secretion (Chan et al. 1996; Leung et al. 1996). The CFTR gene has been long associated with male fertility (Wong, 1998; Chan et al. 2009). Most male patients with cystic fibrosis (CF), the recessive genetic disease caused by some CFTR mutations, are infertile due to congenital bilateral absence of the vas deferens (CBAVD) or epididymal morphological defects (Cuppens & Cassiman, 2004). While these abnormalities have been reproduced in CFTR knock-out (KO) pigs and ferrets (Pierucci-Alves et al.; Sun et al. 2010), CFTR KO male mice have apparently intact reproductive organs and were reported to be fertile (Snouwaert et al. 1992; O’Neal et al. 1993). However, subsequent more detailed studies showed that CFTR KO mice are in fact sub-fertile and that sperm from CFTR deficient mice have significantly lower fertilizing capacity (Reynaert et al. 2000; Xu et al. 2007). In humans, a higher prevalence of CFTR mutations is found in infertile men who are otherwise healthy (van der Ven et al. 1996; Schulz et al. 2006). Infertility in these men is often due to dysfunctional sperm and is not accompanied by any apparent morphological abnormalities of the male reproductive organs. Altogether, these studies suggest that, in addition to causing morphological alterations of the male excurrent duct, defects in CFTR might also be associated with spermatozoa that have a reduced fertilizing capacity, a function that is partially acquired in the epididymis.

In the present study, we determined whether epithelial cells participate in ATP release in the epididymis, and we examined the potential role of CFTR in this process. To do so, we used complementary procedures, including luminal perfusion of the epididymis in vivo, a mouse epididymal epithelial cell line and primary epithelial cultures, together with confocal microscopy, CFTR-specific siRNA knockdown, and a luciferase-based ATP secretion assay. Our data show that ATP is released from epithelial cells into the lumen of the epididymis, and that CFTR is involved in this process.

Methods

Chemicals and antibodies

Forskolin, adrenaline, CFTR inhibitor 172 and 5α-di-hydrotestosterone, trypsin and collagenase type II were purchased from Sigma-Aldrich (St Louis, MO, USA). A commercial rabbit affinity purified antibody against amino acid residues 1468–1480 of human CFTR and its competitive peptide were purchased from Alomone Labs (Jerusalem, Israel). A previously characterized affinity-purified chicken polyclonal antibody raised in our laboratory against the V-ATPase B1 subunit was used (Pietrement et al. 2006; Shum et al. 2011). A rat monoclonal antibody against ZO-1 (R40/76) was kindly provided by Dr Eveline Schneeberger (Department of Pathology, Massachusetts General Hospital). A mouse anti pan-actin antibody was purchased from Chemicon. Goat anti-rabbit IgG coupled to FITC or CY3, donkey anti-mouse IgG coupled to FITC or CY3, and donkey anti-rat IgG coupled to CY3 (Jackson ImmunoResearch Laboratories, West Grove, PA, USA) were used as secondary antibodies.

Animals

Male C57BL/6 mice were purchased from Charles River Laboratories. Previously generated V-ATPase B1-EGFP (B1-EGFP) mice were used (Miller et al. 2005). Animals were anaesthetized with sodium pentobarbital at doses listed below, and were killed with an overdose of sodium pentobarbital at the end of the experiments. All animal studies were approved by the Massachusetts General Hospital (MGH) Subcommittee on Research Animal Care, in accordance with National Institutes of Health, Department of Agriculture, and Accreditation of Laboratory Animal Care requirements.

Tissue fixation and immunofluorescence

Adult male mice were anaesthetized with an intraperitoneal (i.p.) injection of sodium pentobarbital at a dose of 75 mg (kg body weight)⁻¹, and perfused via the left ventricle with phosphate buffered saline (PBS) followed by a fixative containing 4% paraformaldehyde (PFA), 10 mm sodium periodate, 75 mm lysine, and 5% sucrose in 0.1 m sodium phosphate buffer, as described previously (Belleannée et al. 2010). Epididymis was cryoprotected in 30% sucrose in PBS, mounted for cryosectioning in Tissue-Tek OCT compound 4583 (Sakura Fintek, Torrance, CA, USA), and frozen. Sections were cut at a thickness of 5–10 μm using a Leica 3050 cryostat (Leica Microsystems, Bannockburn, IL, USA) and picked up onto Superfrost/Plus microscope slides (Fisher Scientific, Pittsburgh, PA, USA). Sections were hydrated, heated by microwave in an alkaline buffer (10 mm Tris buffer, 1 mm EDTA, pH 9.0) 3–4 times with 5 min interval, cooled down to room temperature, and treated with 1% SDS in PBS for 4 min. For blocking, 1% bovine serum albumin in PBS was applied for 15 min. Sections were incubated with primary antibodies in a moist chamber for 90 min at room temperature, or overnight at 4°C. The CFTR antibody was used at a concentration of 24 μg ml⁻¹ and the V-ATPase B1 antibody was used at a dilution of 1:100. Sections were washed in high salt PBS (PBS containing 2.7% NaCl) twice for 5 min and once in normal PBS. Fluorophore-conjugated secondary antibodies were applied for 1 h, followed by washes as for the primary antibodies.

Cell culture

The immortalized epididymal cell line, DC2, was kindly provided by Dr Marie-Claire Orgebin-Crist, and cultured in Iscove's modified Dulbecco's medium (IMDM, Invitrogen,Vanderbilt University School of Medicine) supplemented with 5α-di-hydrotestosterone (DHT, 1 nm) and 10% fetal bovine serum (FBS, Invitrogen) at 33°C as previously described (Araki et al. 2002). These cells are representative of epididymal principal cells and do not contain clear cells. For immunofluorescence labelling, cells were fixed with 4% PFA in PBS for 20 min and permeablized with 1% SDS in PBS for 4 min. Primary and secondary antibodies were applied as described for tissue sections. The CFTR antibody was used at a concentration of 24 μg ml⁻¹ and the ZO1 antibody was used at a dilution of 1:10 of the hybridoma cell supernatant.

The isolation protocol for primary cultures of epididymal epithelial cells was modified from previous studies (Araki et al. 2002). Pre-pubertal male mice (∼4 weeks old) were used to avoid interference from sperm. Epididymides were dissected from anaesthetized mice and connective and fat tissues were removed. Each epididymis was cut into two portions: proximal (initial segments and caput) and distal (corpus and cauda) portions. Tissues were minced into small pieces and transferred to PBS containing trypsin (4.5 mg ml⁻¹). After incubation at 32°C in a water-bath shaker (150 r.p.m.) for 30 min, cells/tissues were collected by centrifugation at 800 g for 5 min. The supernatant was discarded and the cells/tissues were re-suspended in collagenase type II (1 mg ml⁻¹) in PBS and incubated at 32°C in a water-bath shaker (150 rpm) for another 30 min. IMDM containing 10% FBS was added to stop enzymatic activity. Cells were centrifuged at 800 g for 5 min and the supernatant was discarded. The collected cell pellet was re-suspended in IMDM containing DHT (1 nm) and FBS (10%), and was incubated at 32°C to allow fibroblasts and smooth muscle cells to settle down at the bottom of the Petri dish. After a 4 h incubation, unattached epithelial cells were collected and cultured in IMDM containing DHT (1 nm) and FBS (10%) at 32°C.

Isolation of epididymal clear cells from B1-EGFP transgenic mice

Adult male B1-EGFP mice, in which enhanced GFP is expressed in all B1-positive cells including epididymal clear cells (Miller et al. 2005), were used. Immediately after dissection, cauda epididymides were minced with scissors in RPMI 1640 medium (Invitrogen) containing 1.0 mg ml⁻¹ collagenase type I (Invitrogen) and 1.0 mg ml⁻¹ collagenase type II (Sigma-Aldrich). Tissue dissociation was performed at 37°C for 35 min in a shaking (1000 r.p.m.) thermo-bath. Cell preparations were then passed through a cell strainer with 70 μm nylon mesh to remove undigested material, and washed once with RPMI 1640 medium and once with calcium-free PBS. Cells were then passed through a 35 μm nylon mesh. Populations of EGFP-positive (EGFP+) cells were isolated immediately by fluorescence-activated cell sorting (FACS) based on their green fluorescence intensity. Sorting was performed at the MGH flow cytometry core facility using a modified FACSVantage cell sorter (BD Biosciences, San Jose, CA, USA). EGFP+ and EGFP-negative (EGFP−) cell samples were collected in PBS and used without delay for RNA isolation.

RNA extraction and reverse transcription

RNA was isolated using the RNeasy Mini Kit (Qiagen, Valencia, CA, USA) or RNeasy Micro Kit (Qiagen) according to the manufacturer's protocol. Genomic DNA contamination was eliminated with RNase-free DNase set (Qiagen). The isolated RNA was denatured at 70°C in the presence of random hexamers (1.25 μm) and oligo dT (1.25 μm) for 10 min. The reaction tubes were then immediately placed on ice to stop denaturation, and RNA was converted into first-strand cDNA for 1 h at 42°C in 10 mm Tris-HCl, pH 8.3, 50 mm KCl, 5 mm MgCl₂, 1 mm each dATP, dCTP, dGTP and dTTP, with 1 U μl⁻¹ RNase inhibitor and 2.5 U μl⁻¹ murine leukemia virus reverse transcriptase (all reagents were from Applied Biosystems).

Polymerase chain reaction (PCR)

PCR was performed using a 25 μl reaction buffer containing 1.25 U Taq DNA polymerase (Invitrogen), 200 μm each of dATP, dCTP, dGTP and dTTP, 1.25 mm MgCl₂, 50 mm KCl, 10 mm Tris-HCl, pH 8.3, 0.5 μm forward primer, and 0.5 μm reverse primer. PCR was performed in a thermal cycler (Bio-Rad, Hercules, CA, USA) with the following parameters: 8 min at 95°C to activate the polymerase, followed by 35 cycles of melting for 30 s at 95°C, annealing for 30 s at 60°C, extension for 45 s at 72°C, and a final extension for 10 min at 72°C. PCR products were resolved in a 2% agarose gel containing GelStar stain (Lonza, Rockland, ME, USA). Sequences of primers are: mouse CFTR primers 1: 5′-GAGTGAGGAGGACAGGGATG-3′ and 5′-GGCCGTCTTAACCTTTGGT-3′ (153bp); mouse CFTR primers 2: 5′-GTGGGAGTGGCTGACACTTT-3′ and 5′-ACATAGGGGCGTGAAGAATG-3′ (127bp); mouse B1 primers: 5′-CCCTACGATTGAGCGGATCAT-3′ and 5′-TATATCCAGGAAAGCCACGGC-3′ (182bp).

Protein extraction and immunoblotting

DC2 cells were lysed for 30 min on ice in buffer containing 1% Triton X-100, 160 mm NaCl, 10 mm Hepes, pH 7.5, 1 mm EDTA, 1 mm EGTA, 0.1% IGEPAL CA-630 and Complete protease inhibitors (Roche). Centrifugation was performed at 16,000 g for 30 min at 4°C, and the supernatant was collected. Protein concentration was determined using the bicinchoninic acid assay (BCA). Samples were prepared in Laemmli sample buffer containing 2.5%β-mercaptoethanol and incubated for 30 min at room temperature. Proteins were separated by electrophoresis using 4–12% NuPAGE gels (Invitrogen). After SDS-PAGE separation, proteins were transferred onto Immun-Blot polyvinylidene difluoride (PVDF) membranes (Bio-Rad). Membranes were blocked in Tris-buffered saline (TBS) containing 5% non-fat dry milk and then incubated overnight at 4°C with the primary antibody diluted in TBS containing 2.5% milk. The CFTR antibody was used at a concentration of 0.8 μg ml⁻¹ and the pan-actin antibody at a concentration of 0.5 μg ml⁻¹. After three washes in TBS containing 0.1% Tween 20 (TBST), and a 15 min block in 5% milk/TBS, membranes were incubated with goat anti-rabbit IgG conjugated to horseradish peroxidase for 1 h at room temperature. After five further washes, antibody binding was detected with the Western Lightning Chemiluminescence reagent (Perkin Elmer Life Sciences, Waltham, MA, USA) and Kodak imaging films.

Intracellular pH (pH_i) measurement

DC2 cells were grown on 35 mm² dishes with glass bottom at low density for 24–48 h. For pH_i measurements, cells were loaded with the pH-sensitive fluorescent dye, 2′,7′-bis-(2-carboxyethyl)-5-(and-6)-carboxyfluorescein (BCECF-AM, 5 μm, Invitrogen) for 30 min. The dish with dye-loaded cells was mounted onto a perfusion chamber (Warner Instruments, Hamden, CT, USA) and transferred to an inverted fluorescence microscopy (Eclipse Ti, Nikon). The cells were continuously perfused with a Cl⁻ free buffer containing: 140 mm sodium-d-gluconate, 5 mm potassium-d-gluconate, 2 mm calcium-d-gluconate, 1 mm MgSO₄, 10 mm d-glucose and 10 mm Hepes with pH adjusted to 7.4. To apply CO₂ and HCO₃⁻, a perfusion buffer gassed with 5% CO₂ contained: 115 mm sodium-d-gluconate, 5 mm potassium-d-gluconate, 2 mm calcium-d-gluconate, 1 mm MgSO₄, 10 mm d-glucose and 25 mm NaHCO₃ (pH 7.4). All solutions were adjusted to an osmolality of 290–300 mosmol (kg H₂O)⁻¹ by adding raffinose, a cell-impermeant osmolyte. BCECF was alternately excited at 440 nm and 495 nm and the emitted fluorescence at 530 nm was recorded. Changes of the fluorescence ratio (F₄₉₅/F₄₄₀) reflected changes in pH_i. Calibration was performed by equilibrating cells in high potassium solutions containing 10 μm of nigericin (Sigma-Aldrich) at pH 6, 6.5, 7.0, 7.5 and 8.0.

ATP bioluminescence assay

DC2 cells were grown in 24-well culture plates to 80–90% confluency. PBS (0.5 ml) containing the vehicle, agonists or inhibitors was gently applied to each well. After a 10 min incubation period at 33°C, the cell-free PBS was collected and incubated at 99°C for 2 min to inactivate ectonucleotidases. A luciferin–luciferase based ATP bioluminescence assay kit (Sigma-Aldrich) was used according to the manufacturer's manual and the luminescence was recorded using a luminometer (Berthold Technologies, Oak Ridge, TN, USA). All drugs used were tested and showed no significant effect on the assay. Cells in each well were lysed and protein concentration was determined using the bicinchoninic acid assay.

siRNA interference

Three sets of Stealth RNAi duplexes targeting mouse CFTR (siRNA__CFTR), Low GC Stealth RNAi negative control duplex (siRNA__NC), Lipofectamine 2000 and Opti-MEM Medium were purchased from Invitrogen. The sequences of siRNAs are: siRNA__CFTR_1: 5′-GACAACUUGUUAGUCUUCUUUCCAA-3′ and 5′-UUGGAAAGAAGACUAACAAGUUGUC-3′; siRNA__CFTR_2: 5′-GGAAGAGUUUCAUUCUGCUCUCAAU-3′ and 5′-AUUGAGAGCAGAAUGAAACUCUUCC-3′; siRNA__CFTR_3: 5′-GAGAUUGAUGGUGUCUCAUGGAA UU-3′ and 5′-AAUUCCAUGAGACACCAUCAAUCUC-3′. Lipofecatamine 2000 and siRNAs were diluted with OPTi-MEM. Each siRNA (100 nm) was transfected into DC2 cells with Lipofectamine 2000 using a reversed transfection procedure. Cells were used 60–72 h after transfection.

In vivo perfusion of cauda epididymidis

Adult wild-type (WT) mice were anaesthetized by injection of sodium pentobarbital (50 mg (kg body weight)⁻¹ i.p.) and perigenital incisions were made to expose the cauda epididymidis and vas deferens. A small incision was made in a portion of the tubule located in the proximal cauda epididymidis to insert a small catheter, which was connected to a syringe pump perfusion system (KD Scientific, Holliston, MA, USA). A physiological solution mimicking the epididymal luminal fluid (containing (mm): 50 NaCl, 50 potassium gluconate, 1.2 MgSO₄, 0.6 CaCl₂, 4 sodium acetate, 1 Na₃-citrate, 6.4 NaH₂PO₄ and 3.6 Na₂HPO₄ and adjusted to pH 6.8, and ∼350–360 mosmol (kg H₂O)⁻¹ with raffinose) was perfused through the lumen at a rate of 0.25 ml h⁻¹. A micro-cannula was inserted into the vas deferens lumen to collect the perfusate. After flushing out the sperm and a stabilizing period of 1 h, the perfusate was collected every 5 min and stored at 4°C for up to 2 h before ATP measurements using the luciferase/luciferin assay.

Statistical analysis

The effect of treatments between two groups was determined by Student's unpaired t test. Multigroup comparisons were determined by one-way ANOVA followed with Bonferroni's post hoc test. All tests were two-tailed and statistical significance was set at P < 0.05.

Results

CFTR expression in the mouse cauda epididymidis

In the rat epididymis, while CFTR expression has been shown in principal cells (Ruz et al. 2004; Pietrement et al. 2006), CFTR was reported to be either absent (Pietrement et al. 2008; Shum et al. 2009) or present (Ruz et al. 2004) in clear cells. In the human epididymis, CFTR was shown to be expressed in apical cells (Kujala et al. 2007). Expression of CFTR in the mouse epididymis remained uncharacterized. In the present study, immunofluorescence labelling of PLP-fixed mouse cauda epididymidis cryostat sections showed a strong CFTR expression in the apical membrane of principal cells (Fig. 1A–C and G), while no CFTR labelling was detected in clear cells, identified by their positive labelling for the B1 subunit of the proton pumping V-ATPase (Miller et al. 2005; Pietrement et al. 2006; Da Silva et al. 2007). No labelling was detectable when the antibody was pre-incubated with the immunizing CFTR peptide (Fig. 1D–F). Expression of CFTR in principal cells and its absence from clear cells was also shown by 3D reconstruction confocal microscopy imaging (Fig. 1H). The absence of CFTR from clear cells was further confirmed by RT-PCR using B1-EGFP transgenic mice, which express EGFP in clear cells and not in other epididymal cell types (Miller et al. 2005). B1-EGFP-expressing clear cells were isolated by fluorescence-activated cell sorting (FACS) from the cauda epididymidis, and their Cftr mRNA content was compared with EGFP-negative cells, which represent all other cell types in the epididymis. EGFP-positive clear cells but not EGFP-negative cells showed V-ATPase B1 mRNA-specific transcripts, indicating successful isolation of clear cells (Fig. 1I). In contrast, Cftr mRNA was detectable in EGFP-negative cells only but not in EGFP-positive clear cells. Altogether, these data demonstrate that CFTR is expressed in principal cells but not clear cells in the mouse cauda epididymidis.

Figure 1. Expression of CFTR in mouse cauda epididymidis.

A–C, mouse cauda epididymidis was double-labelled with antibodies against the V-ATPase B1 subunit (B1) and CFTR. A strong CFTR labelling (A and C: red) was detected in the apical membrane of principal cells, but not in clear cells identified by their positive labelling for B1 (B and C: green). D–F, no CFTR staining was detected after pre-incubation of the CFTR antibody with the immunizing peptide (D and F, anti-CFTR + peptide: red; B1: green). Higher magnification image (G) and 3D reconstruction from Z-series confocal images (H) show the absence of CFTR from the apical membrane of V-ATPase-positive clear cells (CFTR: red, B1: green). Nuclei were labelled in blue with DAPI in panels C, F and G. Bars = 50 μm (A–F); 10 μm (G) and 5 μm (H). Arrows in panel H indicate the 3-dimension X (green), Y (red) and Z (blue) axes, and 3D spatial representation is illustrated by the cross-hatched sphere. I, RT-PCR analysis showed the presence of mRNA transcripts for B1 and absence of CFTR mRNA transcripts in EGFP-positive clear cells (+) isolated by FACS from transgenic mice. CFTR mRNA was detected in EGFP-negative cells, representing all other cell types in the cauda epididymis.

CFTR expression in epididymal epithelial cell line

To study the role of CFTR in ATP release from epithelial cells, we next used a previously established mouse epididymal principal cell line, DC2 (Araki et al. 2002). We first examined the expression of Cftr in these cells by RT-PCR (Fig. 2A). A strong Cftr mRNA signal was detected using two different CFTR primer sets (Primers 1 and Primers 2). In addition, Western-blot analysis revealed a ∼160 kDa band in DC2 cells, corresponding to the molecular size of CFTR (Fig. 2B, left panel). A similar band was also detected in mouse lung tissues (ML) used as a positive control. Pre-incubation of the CFTR antibody with the immunizing CFTR peptide abolished the signal detected in DC2 cells and lung tissues (Fig. 2B, right panel). A positive immunofluorescence labelling was obtained using the same antibody in DC2 cells (Fig. 2C, left panel). Double-labelling using an anti-ZO-1 antibody to detect tight junctions confirmed the epithelial nature of this cell line. Here again, the CFTR fluorescence labelling was abolished by pre-incubating the antibody with the immunizing peptide (Fig. 2C, right panel). These results show that DC2 cells express CFTR.

Figure 2. CFTR expression in DC2 cells.

A, RT-PCR analysis using two different sets of primers (Primers 1 and Primers 2) showed mRNA expression of CFTR in DC2 cells. B, left panel, Western blot detection of CFTR in mouse lung (ML) and DC2 cells (DC2). DC2 proteins (50 μg) and 15 μg of lung proteins were loaded onto each lane. A single band at approximately 160 kDa, corresponding to the expected size of CFTR, was detected in both samples. Right panel, pre-incubation of the anti-CFTR antibody with the immunizing peptide, at a 5:1 peptide:antibody ratio, completely abolished the signal in DC2 cells and lung extracts. Loading control was performed by re-blotting the membrane for actin. C, left panel, immunofluorescence double-labelling for ZO-1 (red), a marker of tight-junctions, and CFTR (green) in DC2 cells. Right panel, pre-incubation of the anti-CFTR antibody with the immunizing peptide completely abolished CFTR labelling in DC2 cells. Bars = 15 μm.

CFTR functions as an anion channel in DC2 cells

Previous electrophysiological studies on primary cultures showed that CFTR mediates both Cl⁻ and HCO₃⁻ secretion across the epididymal epithelium (Chan et al. 1996; Leung et al. 1996). We tested here whether CFTR is a functional plasma membrane protein in DC2 cells by monitoring CFTR-dependent HCO₃⁻ transport using the ratiometric pH dye, BCECF (Fig. 3A, top panels). The rate of intracellular pH (pH_i) recovery following an intracellular alkalinization induced by the removal of extracellular CO₂ and HCO₃⁻ was measured as an assessment of HCO₃⁻ efflux (Fig. 3A, bottom panel). To avoid possible participation of Cl⁻/HCO₃⁻ exchangers, DC2 cells were bathed in a Cl⁻ free solution. Under these conditions, a pH_i recovery was observed showing that DC2 cells secrete HCO₃⁻ (Fig. 3B, top trace, red line). Treating the cells with forskolin (Fsk, 10 μm), a known activator of CFTR (Huang et al. 1992,1994; Leung et al. 1998), accelerated the rate of pH_i recovery (Fig. 3B, middle). Moreover, the CFTR inhibitor, CFTR_inh172 (10 μm), abolished the stimulatory effect of Fsk (Fig. 3B, bottom). Quantification analysis showed that Fsk significantly increased HCO₃⁻ secretion as revealed by the enhanced rate of pH_i recovery (Fig. 3C, P < 0.01 versus control), and that CFTR_inh172 prevented its activation by Fsk (Fig. 3C, P < 0.01 Fsk + CFTR _inh172 versus Fsk). Thus, these results indicate that CFTR mediates HCO₃⁻ transport in DC2 cells, and show that these cells thus express functional CFTR in their plasma membrane.

Figure 3. CFTR mediates HCO₃⁻ secretion from DC2 cells.

A, bicarbonate secretion from DC2 cells was measured using the ratiometric dye, BCECF. The rate of pH_i recovery observed upon the removal of extracellular CO₂ and HCO₃⁻ (from Cl⁻ free solution) was measured as an assessment of HCO₃⁻ secretion. a, b and c, pictures of DC2 cells taken at different time points as indicated on the trace. The colour from purple to red represents pH_i from the lowest to the highest values. The trace represents the average pH_i of 22 cells ± SEM, measured under control conditions. After an intracellular alkalinization induced by the removal of HCO₃⁻/CO₂ from the perfusate, a pH_i recovery was observed due to HCO₃⁻ secretion (dashed box). B, representative traces showing pH_i recovery measured under control conditions (top; Ctrl), in the presence of 10 μm forskolin (middle; Fsk), or in the presence of 10 μm forskolin after pretreatment with 10 μm CFTR_inh172 (bottom; CFTR_inh172+ Fsk). The red lines indicate the initial rate of pHi recovery (ΔpH_i min⁻¹). C, bar graph showing the averaged pH_i recovery rates obtained from several DC2 cell preparations (each containing at least 20 cells) under different conditions. Data are shown as means ± SEM (n = 7–9 preparations). *P < 0.01.

CFTR is involved in ATP release from DC2 cells

ATP secretion by DC2 cells was measured under different conditions using a luciferase–luciferin assay. As shown in Fig. 4, Fsk (10 μm) or adrenaline (Adr; 50 μm) – another CFTR activator (Huang et al. 1992, 1994) – increased the concentration of ATP detected in the extracellular solution after a 10 min incubation period compared to non-treated cells (Ctrl). Pre-treatment of the cells with CFTR_inh172 (10 μm) for 15–20 min decreased ATP concentration (CFTR_inh172 versus Ctrl), and abolished the increase induced by Fsk (CFTR_inh172+Fsk versus Fsk).

Figure 4. CFTR is involved in ATP release in DC2 cells.

Effect of adrenaline (Adr, 50 μm), forskolin (Fsk, 10 μm) and CFTR_inh172 (10 μm) on ATP release from DC2 cells measured using a luciferase–luciferin assay. ATP secretion was detected under control conditions (Ctrl), and was significantly increased after stimulation with Adr and Fsk. CFTR_inh172 significantly reduced baseline ATP secretion and prevented the activation induced by Fsk (CFTR_inh172+ Fsk). Data are shown as means ± SEM (n = 6–14). *P < 0.05, **P < 0.01, ***P < 0.001 versus Ctrl; ###P < 0.001 versus Fsk.

Effect of knocking down CFTR expression on ATP release

To further confirm the involvement of CFTR in ATP release, DC2 cells were transfected with small interfering RNAs (siRNAs) to knock down CFTR expression. Three different sets of CFTR-specific siRNAs (siRNA_{CFTR1,2, and 3}) were used and a set of siRNA of non-silencing sequence was applied as a negative control (siRNA_NC). As shown in Fig. 5A, 60–72 h after transfection, all CFTR-specific siRNAs decreased CFTR protein levels in DC2 cells with the maximal effect achieved by siRNA_CFTR2 (82 ± 6%) and siRNA_CFTR3 (72 ± 13%). siRNA_CFTR2 and siRNA_CFTR3 were then used to examine the role of CFTR in ATP release. The ATP concentration measured under baseline conditions in the incubation solution was reduced with siRNA_CFTR2 and siRNA_CFTR3 compared to siRNA_NC (Fig. 5B). Forskolin still induced an increase in ATP concentration in siRNA_NC-transfected cells (siRNA_NC+Fsk versus siRNA_NC) but did not increase ATP release in siRNA_CFTR2- and siRNA_CFTR3- transfected DC2 cells (siRNA_{CFTR2 or 3}+ Fsk versus siRNA_NC+Fsk. These results thus confirmed that CFTR participates in ATP release from DC2 cells.

Figure 5. Effect of CFTR knockdown on ATP release in DC2 cells.

A, Western blot showed that three different CFTR-specific siRNA constructs significantly reduced CFTR expression, with maximum effects obtained with siRNA_CFTR2 and siRNA_CFTR3, compared to cells transfected with a non-silencing construct (siRNA_NC). B, transfection with siRNA_CFTR2 and siRNA_CFTR3 significantly reduced the concentration of ATP measured in the extracellular solution bathing DC2 cells (light grey bars) and prevented the increase in ATP concentration induced by forskolin (black bars), compared to cells transfected with siRNA_NC. Data are shown as means ± SEM (n = 17). *P < 0.05 and **P < 0.01 versus siRNA_NC; ###P < 0.05, versus siRNA_NC+Fsk.

CFTR is involved in ATP-release from primary cultures of epididymal epithelial cells

Epithelial cells from the proximal (initial segments and caput) and distal (corpus and cauda) regions of mouse epididymis were isolated, cultured for 4 days and assessed for ATP release. In both primary cell cultures, baseline and forskolin-activated ATP release were detected (Fig. 6). CFTR_inh172 greatly reduced baseline and forskolin-stimulated ATP release in both proximal and distal primary cultures.

Figure 6. CFTR is involved in ATP release from primary cultured epididymal cells.

Effect of Fsk (10 μm) and CFTR_inh172 (10 μm) on ATP release from primary cultures of proximal (A) and distal (B) epididymal regions. A baseline ATP secretion was observed in both proximal and distal epididymis primary cultures (Ctrl). Forskolin (Fsk) increased ATP secretion in both preparations, although the activation did not reach statistical significance in the proximal primary culture. CFTR_inh172 inhibited baseline ATP secretion and prevented the Fsk-induced activation in both cultures. Data are shown as means ± SEM (n = 9). *P < 0.05, ***P < 0.001 versus Ctrl; ###P < 0.05 versus Fsk.

In vivo detection of ATP release from cauda epididymidis

ATP secretion into the lumen of the cauda epididymal tubule was assessed in vivo. To do so, the cauda region of the epididymis was perfused through the lumen as we have previously published (Shum et al. 2008; Belleannée et al. 2010). The luminal perfusate was collected by cannulation of the vas deferens. This procedure allows for a large portion of the cauda region to be perfused in vivo (Fig. 7A). The lumen was first washed free of spermatozoa, and the perfusate was then collected over consecutive periods of 5 min. ATP concentration was measured using the luciferase–luciferin assay. Under control conditions, a steady-state ATP release was detected (Fig. 7B and C). Addition of 10 μm CFTR_inh172 into the luminal perfusate induced a reduction in ATP concentration in the samples collected over a period of 30 min, compared to the concentration measured during the initial control period (Fig. 7B). The vehicle induced only a slight reduction in ATP concentration compared to the initial control period (Fig. 7C). Figure 7D shows the averaged effect of the vehicle and CFTR_inh172 on ATP concentration relative to the concentration that was measured during the initial control period. A significantly greater decrease was induced by CFTR_inh172 compared to vehicle. These results show the participation of CFTR in ATP release from the cauda epididymidis in vivo.

Figure 7. ATP release from the mouse cauda epididymis in vivo.

A, photograph of a cross-section from a mouse cauda epididymis after perfusion through the lumen with a physiological solution. The perfused region contains the segment of the tubule that was washed free of sperm. B and C, representative graphs showing the effect of 10 μm CFTR_inh172 (B) and vehicle (DMSO) alone (C) on the concentration of ATP measured in solutions collected every 5 min after perfusion through the epididymal tubule lumen. D, averaged effect of vehicle or CFTR_inh172 on ATP concentration measured in the sample collected after 20 min exposure, as a percentage of the concentration measured during the initial period. Data are means ± SEM (n = 6), ***P < 0.01.

Discussion

Extracellular ATP is a key modulator of epididymal function, regulating both transepithelial transport (Wong, 1988; Zhou et al. 2007; Belleannée et al. 2010) and sperm motility acquisition (Bhattacharyya & Pakrashi, 1993; Edwards et al. 2007). However, the cellular origin of ATP and the regulation of ATP release into the epididymal luminal compartment remained elusive. The present study shows that epididymal principal cells secrete ATP under basal conditions and that ATP release is enhanced by adrenaline and forskolin. Pharmacological inhibition of CFTR or knocking down CFTR expression by specific siRNAs in the epididymal principal cell line, DC2, reduced basal ATP release and completely prevented the activation induced by forskolin. ATP secretion from epithelial cells was also demonstrated in vivo by the detection of ATP in solutions perfused through the epididymal lumen of mouse cauda epididymis. ATP was detected after the lumen has been washed free of sperm, eliminating them as a potential source of ATP, and showing the participation of the epididymal epithelium in ATP release. The involvement of CFTR was shown in vivo by the lower ATP concentrations measured in perfusate containing CFTR_inh172. Taken together, using both in vitro and in vivo models, the present study shows that CFTR is involved in ATP secretion by epithelial cells in the epididymis. CFTR mutations are associated with male infertility (Wong, 1998; Cuppens & Cassiman, 2004; van der Ven et al. 1996; Schulz et al. 2006). Given the critical roles of ATP in sperm function as well as in creating an optimal epididymal luminal environment for their maturation and storage, the present study provides new foundations for a better understanding of the role of CFTR in the purinergic regulation of sperm function/maturation, and thus in the establishment of male fertility.

Results presented here are in agreement with previous reports showing that CFTR regulates ATP secretion in several epithelia, and further confirm the role of CFTR in purinergic regulation. It has been proposed that deficient ATP secretion contributes to the pathologies of the lung associated with CF (Schwiebert & Zsembery, 2003). In the epididymis, future studies will be required to determine whether ATP secretion is also affected by CFTR mutations and to identify the exact mechanism(s) by which CFTR regulates ATP release. As mentioned above, whereas the participation of CFTR as a regulator of ATP secretion was demonstrated in several epithelia, the notion that CFTR directly conducts ATP triggered much debate (Abraham et al. 1997) (for review see Praetorius & Leipziger, 2009 and Sabirov & Okada, 2005). In the airway and kidney, CFTR was shown to regulate a separate ATP pathway (Reddy et al. 1996; Sugita et al. 1998; Braunstein et al. 2001). Pannexin-1 (PANX-1), a known ATP channel (Li et al. 2011), is expressed in the epididymal epithelium (Turmel et al. 2011), and we also detected PANX-1 expression in DC2 cells (Supplementary Figure A). In addition we found that the pannexin inhibitor, carbenoxolone (CBX, 10 μm), reduced both basal and forskolin-activated ATP release in these cells (Supplementary Figure B). These effects were similar to the inhibition elicited by CFTR_inh172, indicating that CFTR might regulate ATP release by modulating PANX-1. However, the direct participation of CFTR in ATP release in epididymal cells cannot be ruled out in the present study.

P2 receptors require approximately 1–10 μm of ATP to reach full activation, but ATP concentrations in the nanomolar range have been reported in the extracellular compartment surrounding several cell types (Praetorius & Leipziger, 2010). Thus, our result showing a concentration of 100 nm in the solution collected after luminal perfusion of the epididymis is consistent with previously published data. However, this concentration is probably an underestimation of the actual concentration that is reached near the apical surface of the epithelium following ATP release from epithelial cells. Several factors might be involved in this underestimation. (1) In the non-perfused epididymis, the luminal flow is considerably lower than in the perfused epididymis. Thus it is likely that in the resting epididymis, where luminal flow would be relatively lower, the luminal concentration of ATP would be higher than the concentration that we measured in the perfused lumen. (2) We have previously shown that exogenous ATP perfused into the lumen of rat epididymis in vivo is rapidly degraded by nucleotidases (Belleannée et al. 2010). Thus, it is likely that mouse epididymal nucleotidases also contributed to reducing the concentration of extracellular ATP following partial degradation into ADP and adenosine. (3) In order to avoid any potential contribution of spermatozoa in ATP release, we perfused the epididymal lumen with a sperm-free solution. Normally, the epididymal lumen is filled with tightly packed spermatozoa, which occupy a large proportion of the luminal volume and reduce considerably the amount of fluid. Thus, our experimental sperm-free condition would contribute to diluting ATP compared to the in vivo condition. (4) Finally, the diameter of the cauda epididymis tubule is very large compared to the size of epithelial cells. Therefore, the concentration of ATP near the apical surface of the epithelium might be higher than the average concentration that we measured in the entire volume of the perfusate. Nevertheless, while it is difficult to assess the exact concentration of ATP that was reached locally, near the apical surface of the epithelium following secretion by epithelial cells, our results show the ability of the epididymal epithelium to secrete ATP in the intact tissue in vivo, and the participation of CFTR in the regulation of ATP release.

Using immunofluorescence confocal imaging and RT-PCR analysis from clear cells isolated by FACS, we show here that in the mouse cauda epididymidis, CFTR is exclusively expressed in principal cells and is absent from clear cells. In addition, the DC2 cell line is representative of principal cells, and does not contain clear cells. These results suggest that principal cells are responsible for CFTR-dependent ATP release. We have previously shown that V-ATPase-mediated H⁺ secretion in clear cells is regulated by luminal ATP (Belleannée et al. 2010). The present study, thus, indicates the participation of principal cells, via CFTR-dependent ATP release, in the regulation of luminal acidification by clear cells. ATP would serve here as a paracrine regulator of clear cells via crosstalk with principal cells. An acidic luminal environment in the epididymis keeps sperm in a quiescent state (Levine & Kelly, 1978; Shum et al. 2009) During sexual arousal, a transient CFTR-mediated HCO₃⁻ secretion (Chan et al. 1996; Wong, 1998) was proposed to ‘prime’ spermatozoa before ejaculation (Carlin et al. 2003). We have previously demonstrated that HCO₃⁻ induces the apical recruitment of V-ATPase in clear cells to activate H⁺ secretion, via activation of the cAMP-sAC pathway (Pastor-Soler et al. 2005; Pastor-Soler et al. 2008), providing a feedback mechanism for the re-establishment of the acidic ‘resting’ luminal pH. Therefore, we propose that CFTR plays a dual role in HCO₃⁻ and ATP secretion, and indirectly promotes luminal acidification by clear cells via paracrine regulation.

CFTR can be activated by several factors, including adrenaline, which triggers the cAMP signalling pathway (Huang et al. 1992, 1994; Cantiello et al. 1994; Chan et al. 1999; Lam et al. 2003). In addition, it has been reported that CFTR is activated by extracellular ATP in the epididymis (Chan et al. 1995) and in other tissues (Cantiello et al. 1994). Since ATP itself can activate ATP release (Anderson et al. 2004), it is possible that ATP could regulate its own release via activation of CFTR in principal cells, thereby providing an autocrine feedback mechanism. Interestingly, CFTR has recently been reported to be mechanically sensitive (Zhang et al. 2010) and ATP release can also be induced by mechanical stress (Furuya et al. 2004; Birder, 2006; Li et al. 2011). In the present study, ATP release was measured in DC2 cells that were subjected to identical mechanical disturbance under control and stimulated conditions, by gently adding into each dish a small volume of PBS containing either the vehicle or agonist. Under these conditions, we detected stimulation of ATP release by forskolin and adrenaline. However, addition of forskolin (10 μm) into the luminal perfusate failed to increase ATP release in epididymal tubules perfused in vivo (data not shown). This is probably due to the fact that the mechanical forces induced by the luminal perfusion might have already substantially activated ATP secretion so that no further activation by forskolin could be detected. Alternatively, CFTR may be more constitutively activated in the in vivo condition compared to the DC2 cell line. Physiologically, the cauda region of the epididymis is subjected to variations in luminal volume as sperm transit from more proximal regions, and epididymal epithelial cells are exposed to physical disturbances, via either expansive pressure or shear force (Johnson & Howards, 1975; Baltz et al. 1990; Turner, 2008). Being located in the apical membrane of the epithelium, CFTR is suitably positioned to respond to these mechanical stimuli by activating ATP release in the epididymis. Thus, the epididymis could serve as a powerful in vivo model system for the study of CFTR-dependent ATP secretion following mechanical stimuli, including stretch and flow rate variations, that might be applicable to other CFTR-expressing tissues.

In summary, using both in vitro and in vivo assays, we demonstrate here for the first time that ATP, an important purinergic regulator in the epididymis, is released from the epididymal epithelium and that CFTR, which is expressed exclusively in principal cells, is involved in ATP luminal secretion. This study, therefore, identifies defective purinergic signalling as a potential cause for the male infertility cases associated with CFTR mutations.

Glossary

CFTR

cystic fibrosis transmembrane conductance regulator

FACS

fluorescence-activated cell sorting

Author contributions

S.B. and Y.C.R. designed experiments. Y.C.R., W.W.S., C.B. and N.D.S performed experiments and data analysis. S.B. and Y.C.R. wrote the manuscript. All authors approved the final version.

Supplementary material

Supplemental Figure A

Supplemental Figure B