Renal Physiology

Presence of the Ca2+-activated chloride channel anoctamin 1 in the urethra and its role in excitatory neurotransmission

Maria Sancho, Angeles García-Pascual, Domingo Triguero


We investigated the cellular distribution of the calcium-activated chloride channel (CaCC), anoctamin 1, in the urethra of mice, rats, and sheep by both immunofluorescence and PCR. We studied its role in urethral contractility by examining the effects of chloride-free medium and of several CaCC inhibitors on noradrenergic and cholinergic excitatory responses, and on nitrergic relaxations in urethral preparations. In all species analyzed, CaCC played a key role in urethral contractions, influencing smooth muscle cells activated by increases in intracellular calcium, probably due to calcium influx but with a minor contribution by IP3-mediated calcium release. The participation of CaCC in relaxant responses was negligible. Strong anoctamin 1 immunoreactivity was detected in the smooth muscle cells and urothelia of sheep, rat, and mouse urethra, but not in the interstitial cells of Cajal (ICC) in any of these species. RT-PCR confirmed the expression of anoctamin 1 mRNA in the rat urethra. This anoctamin 1 in urethral smooth muscle probably mediates the activity of chloride in contractile responses in different species, However, the lack of anoctamin 1 in ICCs challenges its proposed role in regulating urethral contractility in a manner similar to that observed in the gut.

  • noradrenergic neurotransmission
  • interstitial cells of Cajal

membrane potential is usually defined as the balance between small cations asymmetrically distributed across the cell membrane. Chloride, the only permeable anion, is present at very high concentrations on both sides of the membrane and it is usually considered in a second plane when considering other important ions. The complex role of chloride in maintaining the membrane potential is reflected by the wide variety of chloride channels known to exist (some of which are no longer included in this group). The recently cloned and characterized anoctamin 1 (ANO1; also known as TMEM16A) was identified as the main calcium-activated chloride channel (CaCC) (6, 26, 34). Only some members of the anoctamin family can produce measurable Ca2+-dependent Cl currents and since ANO-1 can elicit currents of much larger amplitude than other anoctamine paralogs (19), it is considered to be a reliable indicator of CaCC activity in excitable tissues.

The expression and function of ANO1 have been analyzed in different tissues from normal and ANO1-deficient mice (16). While ANO-1 is absent from smooth muscle cells in the gastrointestinal tract, it has been localized to a conspicuous network of interstitial cells of Cajal (ICC) (16) thought to mediate autorhythmicity and neurotransmission (6, 24). Thus, ANO-1 has been proposed as a specific marker of ICCs in the smooth muscle in the gut (12). The lack of ANO-1 in both smooth muscle and ICCs of the mouse urethra suggests differences between gastrointestinal ICCs and those of urinary tissues (16).

The pacemaker activity of isolated rabbit urethral ICCs is defined by the ionic balance of chloride. Moreover, specific CaCC inhibitors have been proposed as tools for their “pharmacological ablation” in ICCs, based on the abolition of chloride currents in smooth muscle cells (30). However, the rabbit urethra appears to reflect the exception to the rule as chloride-dependent currents have been described in muscle preparations of the lower urinary tract of several other species (bladder and urethra), in which they are implicated in the control of smooth muscle contraction (7).

Due to the aforementioned species and tissue differences, it is difficult to draw generalized conclusions regarding ICC function. In the present study, we analyzed the distribution of ANO1 in the urethra of sheep, rat, and mouse, using conventional PCR and immunohistochemistry. In addition, the role of chloride in the contractile and relaxant activity of the urethra mediated by norepinephrine (NE), acetylcholine (ACh), and nitric oxide (NO) was investigated in functional studies. These mediators were either released by intrinsic nerves via electrical field stimulation (EFS) or added as exogenous agonists. Our results demonstrate that chloride plays a key role in excitatory urethral contraction by acting directly on smooth muscle, having a negligible influence on relaxation. Moreover, no significant differences in chloride activity were detected between species. Intense ANO1 immunolabeling was consistently observed in the smooth muscle layers and the urothelium, but not in the ICC, strongly suggesting that ANO1 is not specific to ICCs in the urethra, in contrast to the gastrointestinal tract (12).


Drugs and solutions.

ACh, atropine sulphate, 2-aminoethoxydiphenyl borate (2-APB), guanethidine monosulphate, lG-nitro-l-arginine, norepinephrine bitartrate salt (NE), d-tubocurarine hydrochloride, [Arg8] vasopressin acetate salt (AVP), niflumic acid, anthracene 9-carboxylate (9-AC), and 4-acetamido-4′-isothiocyanato-2,2′-stilbenedisulfonic acid disodium salt hydrate (SITS) were all obtained from Sigma (Steinheim, Germany). Tetrodotoxin (with citrate, TTX) was obtained from Alomone Labs (Jerusalem, Israel). All drugs were dissolved in distilled water, except for niflumic acid and 9-AC, which were dissolved in DMSO. Stock solutions were stored at −20°C and working dilutions were prepared in 0.9% NaCl.

Tissue preparation.

Studies were carried out on the lower urinary tracts from female sheep (3 to 6 mo old), Wistar rats (6 to 8 wk old, weighing 200–300 g), and Swiss mice (6 to 8 wk old, weighing 25–30 g). Sheep urinary tracts were collected at the local abattoir shortly after death and transported to the laboratory in cold Krebs solution, composed of (in mM) 119 NaCl, 4.6 KCl, 1.5 CaCl2, 1.2 MgCl2, 15 NaHCO3, 1.2 KH2PO4, 0.01 EDTA, and 11 glucose. Rats and mice were obtained from Harland Ibérica and housed on a 12:12-h light-dark cycle with ad libitum access to food and water. Rats and mice were killed by cervical dislocation and then exsanguinated, before the abdomen was opened to remove the entire lower urinary tract. All procedures were approved by the Ethical Committee at the Complutense University and performed in accordance with European guidelines (Council Directive 86/609/EEC). The urethra was dissected out from each urinary tract, and cleaned of fat and connective tissue, taking care not to damage the mucosal and serosal layers of the rat and mouse preparations, and to maintain the integrity of the tissue. The mucosa was removed from the sheep preparations. Subsequently, transverse strips (∼3 mm wide and 5 mm long) or rings (3 mm wide in rats and mice) were taken from the proximal urethra and used to study the relaxant and contractile responses, respectively.

Recording of isometric tension.

Urethral preparations (strips or rings) were mounted between two stainless steel hooks in 5-ml organ baths containing Krebs solution at 37°C, and they were bubbled with a mixture of 95% O2-5% CO2 (pH 7.4). The isometric tension was recorded with Grass FT03C transducers (Grass Instruments, Quincy, MA) and displayed on a MacIntosh computer with a MacLab analog-to-digital converter v5.5 (AD Instruments, Hastings, East Sussex, UK). Preparations were equilibrated at a resting tension of 15 mN (sheep) or 5 mN (rat and mice) for 60 min, and their viability was tested by twice inducing the contractile response by exposure to a high concentration of external K+ (120 mM).

EFS was applied using a Grass S-48 stimulator (Grass Instruments) that was connected to platinum electrodes placed parallel to the preparation and coupled to a Med-Lab stimulus splitter (Med-Lab Instruments, Loveland, CO). Square-wave pulses (0.8 ms) at a supramaximal voltage (current strength, 200 mA) and a train of 5-s pulses at different frequencies were delivered at 2-min intervals.

Basal urethral excitatory contractile responses were assessed by EFS (1 to 50 Hz, 5-s train duration, 2-min intervals) in the presence of lG-nitro-l-arginine (0.1 mM) to prevent interference by the NO released from nerves, and of d-tubocurarine (0.1 mM) to prevent stimulation of the somatic nerves that innervate the striated muscle of the urethra in rats and mice. After being washed, the preparations were pretreated for 30 min with the IP3 receptor inhibitor 2-APB (50 μM) in chloride-free Krebs solution (in mM: 119 C2H5COONa, 3.5 C6H11KO7, 1.5 CaCl2, 1.2 MgSO4, 15 NaHCO3, 1.2 NaH2PO4, 0.01 EDTA, and 11 glucose) or with one of the following chloride channel inhibitors: niflumic acid (0.1 mM), SITS (0.1 mM), or 9-AC (1 mM). The stimulation protocol was then repeated.

The relative contribution of NE and ACh to EFS-induced contractions was assessed by testing the effects of atropine (1 μM) before and after the addition of phentolamine (10 μM). The final addition of TTX (1 μM) served to test the nerve dependency of the contraction. Cumulative dose-response curves for NE (0.01 to 100 μM) and ACh (0.01 μM to 10 mM) were obtained and the effects of chloride channel inhibitors were evaluated by repeating the dose-response curve after a 30-min treatment.

The effect of niflumic acid on EFS-induced nitrergic relaxation was also analyzed. Accordingly, strip preparations were precontracted with AVP (0.1 μM) in the continuous presence of atropine (1 μM) and guanethidine (10 μM) to avoid cholinergic and adrenergic excitatory influences. d-Tubocurarine (10 μM) was also added to both rat and mouse preparations. To construct relaxant frequency-response curves, EFS were delivered at 2-min intervals in trains of 5 s at frequencies that ranged from 0.5 to 12 Hz. Long-train duration (60 s) single relaxations at a frequency of 2 and 35 Hz were subsequently performed at 1-min intervals. All responses were completely blocked by TTX (1 μM) and lG-nitro-l-arginine (0.1 mM), highlighting their nitrergic origin (4). Niflumic acid (0.1 mM) was added 30 min before the next contraction (AVP)-relaxation (EFS) protocol. Control preparations were run in parallel with experimental preparations, receiving the same volume of drug solvents and subjected to the same protocols.


Urethral preparations were analyzed by immunofluorescence as described previously (11). The tissue was first fixed in ice-cold 4% paraformaldehyde in 0.1 M phosphate buffer (PB; pH 7.4) and then incubated in paraformaldehyde solutions with increasing concentrations of sucrose in 0.1 M PB (10% sucrose for 90 min followed by 20% sucrose for 120 min). Cryoprotection was terminated by incubating overnight in 30% sucrose in PB at 4°C, after which the tissues were snap-frozen in liquid nitrogen-cooled isopentane and stored at −80°C for up to 15 days. Cryostat sections (10 μm) transverse to the mucosal surface were obtained from urethras embedded in Tissue-Tek OCT (CM1850 UV, Leica Microsystems, Barcelona, Spain) and they were thawed onto poly-l-lysine-coated slides. Consecutive sections were collected from each urethra on separate slides to obtain 10 to 15 similar serial sections from the same animal. The slides were air-dried at room temperature for 12 to 24 h and then processed directly or stored at −80°C for no more than 30 days.

The urethral sections were washed three times with PB (5 min each) and incubated for 2 h with 3% normal donkey serum (Chemicon International, Temecula, CA) containing 0.3% Triton X-100. The sections were then incubated in a humidified chamber for 24 h at 4°C with the primary antibody (or antibodies for dual labeling) diluted in 2% normal donkey serum and 0.3% Triton X-100. In both sheep and rat preparations, the following primary antibodies were used: rabbit polyclonal antiserum raised against ANO1 (1:100; Abcam, Cambridge, UK), mouse monoclonal anti-vimentin antibody (clone V9, 1:100; Chemicon International), and mouse monoclonal anti-α-smooth muscle actin (1:800; Sigma). Mouse preparations were analyzed using the same primary rabbit antiserum against ANO1 as above and a rabbit polyclonal antiserum against α-smooth muscle actin (1:100; Abcam). Alexa fluor 488 donkey anti-rabbit and Alexa fluor 594 donkey anti-mouse (both at 1:200; Molecular Probes, Eugene, OR) were used as the secondary antibodies. For mouse preparations, primary antibodies for ANO1 and actin (both of rabbit origin) were examined on separate slides containing alternate sections, which were exposed individually to Alexa fluor 488 and Alexa fluor 594 donkey anti-rabbit secondary antibodies, respectively (both at 1:200; Molecular Probes). The sections were incubated with the secondary antibodies for 2 h in the dark in a humidified chamber at room temperature. After three washes with PB (10 min each), the nuclei were counterstained with DAPI (10.9 mM for 30 min; Sigma), and the sections were washed again and mounted with ProLong Gold antifade reagent (Molecular Probes). In all cases, multiple controls were performed in which the specificity of the immunoreactions was established by omitting the primary antibodies.

Sections were examined on an Axioplan 2 fluorescence microscope (Carl Zeiss MicroImaging GmbH, Göttingen, Germany) equipped with the appropriate filter sets, photographed with a Spot-2 digital camera (Diagnostic Instruments, Sterling Heights, MI), and the images were stored digitally as 12-bit images using MetaMorph 6.1 software (MDS Analytical Technologies, Toronto, Canada). Digital images were subsequently processed using Adobe Photoshop 8.0 (San José, CA).

Data analysis.

Relaxation was normalized as a fraction of the tension induced by AVP immediately before stimulation. The contractile responses and normalized relaxations were expressed as a percentage of the maximum response obtained during the first stimulation, before the treatment of each preparation (% of control), and the results are presented as means ± SE of n experiments (from n different animals). One-way ANOVA was used for multiple comparisons followed by an unpaired Student's t-test (2-tailed). Data were compared using GraphPad Prism 5 software (GraphPad Software, San Diego, CA).


Upon removal, the female rat urethra and male rat prostate glands were immediately frozen in liquid nitrogen and total RNA was extracted using the Qiagen RNeasy Fibrous Tissue Mini Kit with a DNase digestion step. RNA quality was determined by agarose gel electrophoresis. Conventional RT-PCR was performed to amplify the ANO1 transcript in a 25-μl reaction volume using the Access RT-PCR system (Promega, Madison, WI) on a Perkin Elmer thermocycler (GeneAmp PCR System 2400). The RT-PCR protocol involved the following: an incubation at 45°C for 45 min and at 94°C for 2 min; 35 cycles at 94°C for 30 s, 60°C for 45 s, and 68°C for 45 s; and a final extension at 68°C for 7 min. Two sets of specific primers for ANO1 amplification were designed based on the published sequence (Acc. Number: NM_001107564): ANO1 set 1 gave a product of 637 bp, forward primer 5′-GCGCGTGCCAGTCACCTCTT-3′ (positions 1482 to 1501) and reverse primer was 5′-GCCGACCAACAAACCGGCCT-3′ (positions 2099 to 2118); ANO1 set 2 gave a product of 308 bp, forward primer 5′-TGGAGGAGTGTGCCCCAGGC-3′ (positions 2155 to 2174) and reverse primer 5′-TGGGGCCAGAGGGAAGGACG-3′ (positions 2443 to 2462). Positive controls included the amplification of rat prostate cDNAs, which are reported to strongly express ANO1 mRNA (25) and to rule out the possibility of contamination with genomic DNA, RT-PCR was performed without reverse transcription. The fragments amplified were verified in 2% agarose gel electrophoresis stained with ethidium bromide and visualized by Bio-Rad Fluor-S MultiImager (Hercules, CA).


Effects of chloride channel inhibition on excitatory responses.

When exposed to the anion transport inhibitor SITS (0.1 mM), the urethral contractions induced by both EFS and NE in sheep, rats, and mice remained unaffected at all frequencies (1 to 50 Hz) and NE doses (0.01 to 300 μM) tested (Figs. 1 and 2). However, in the presence of more specific inhibitors of fenamate and anthracene chloride channels, such as niflumic acid (0.1 mM) and 9-AC (1 mM), respectively (8), nerve-mediated contractions (Fig. 1) and the contractions elicited by cumulative addition of NE (Fig. 2) were inhibited, especially in rat and mouse urethras. Sheep urethra was more resistant to the effects of these chloride inhibitors (Figs. 1 and 2).

Fig. 1.

Effects of chloride channel inhibitors on frequency-dependent contractile responses. A: representative traces showing rat (left) and mouse (right) urethral contractions elicited by electrical field stimulation (EFS; 5 s, 1 to 50 Hz) at a basal tension in control conditions (top) and following a 30-min incubation with the following chloride channel inhibitors: 4-acetamido-4′-isothiocyanato-2,2′-stilbenedisulfonic acid disodium salt hydrate (SITS; 0.1 mM), niflumic acid (0.1 mM), and anthracene 9-carboxylate (9-AC; 1 mM; bottom). B: changes in the frequency-response contraction curves at basal tension (expressed as a percentage of their respective control curves) in sheep (left), rat (middle), and mouse (right) urethra in the absence (open symbols) or the presence (30-min pretreatment) of SITS (0.1 mM), niflumic acid (0.1 mM), or 9-AC (1 mM; filled symbols). The data are expressed as means ± SE (n = 6 in sheep, n = 5 in rats and mice). *P < 0.05, **P < 0.01, ***P < 0.001 when compared with controls (1-way ANOVA followed by Student's t-test for unpaired observations).

Fig. 2.

Effects of chloride channel inhibitors on concentration-dependent contractions induced by norepinephrine (NE). A: representative traces showing sheep (left) and mouse (right) urethral contractions induced by cumulative addition of NE (0.01 to 100 μM) at basal tension in control conditions (top) and following a 30-min incubation with the following chloride channel inhibitors: SITS (0.1 mM), niflumic acid (0.1 mM), or 9-AC (1 mM; bottom). B: changes in the dose-response contraction curves induced by NE on basal tension (expressed as the percentage of the respective control curves) in sheep (left), rat (middle), and mouse (right) urethra in the presence (30-min pretreatment: filled symbols) or the absence (open symbols) of SITS (0.1 mM), niflumic acid (0.1 mM), or 9-AC (1 mM) pretreatment. Data are expressed as means ± SE (n = 8 in sheep, n = 5 in rats and mice). *P < 0.05, **P < 0.01, ***P < 0.001 when compared with controls (1-way ANOVA followed by Student's t-test for unpaired observations).

In agreement with previous findings in sheep (10), phentolamine (10 μM) inhibited EFS-induced contraction in rats (Fig. 3A) and mice. Maximal responses at 50 Hz were decreased to 16 ± 3.4% of the control preparations (n = 9, P < 0.01), which were not reduced further by TTX (1 μM; not shown) or subsequent addition of atropine (1 μM; Fig. 3A), suggesting that NE is the main excitatory neurotransmitter in the urethra. However, when atropine (1 μM) was added before phentolamine, EFS-induced contractions were reduced significantly in both the rat (Fig. 3B) and mouse (maximal responses at 50 Hz were reduced to 54 ± 3.6% of control preparations; n = 9, P < 0.01), and the remaining responses were subsequently abolished by phentolamine (10 μM; Fig. 3B). ACh elicited dose-dependent contraction in rat preparations (Fig. 3C), which were significantly inhibited by both niflumic acid (0.1 mM) and 9-AC (1 mM).

Fig. 3.

Cholinergic participation in excitatory contractile responses and its inhibition by chloride channel inhibitors. Changes in the frequency-response (EFS) contraction curves at basal tension (expressed as a percentage of the respective control curves) in rat urethra in the presence (30-min pretreatment, filled symbols) or the absence (open symbols) of A: phentolamine (10 μM), followed by atropine (1 μM) or B: atropine, followed by phentolamine. C: changes in the dose-response contraction curves induced by ACh at basal tension (expressed as the percentage of the respective control curves) in rat urethra in the presence (30-min pretreatment, filled symbols) or the absence (open symbols) of niflumic acid (0.1 mM) or 9-AC (1 mM). Data are expressed as means ± SE (n = 8). *P < 0.05, **P < 0.01, ***P < 0.001 compared with controls (1-way ANOVA followed by Student's t-test for unpaired observations).

Significant fast contraction was induced when urethral preparations were exposed to chloride-free Krebs solution, which returned to the basal level of tension within 10 to 20 min (Fig. 4A). In both sheep and rat preparations, contractions of a similar shape and comparable magnitude were observed (40% of their respective preliminary responses to 120 mM K+), which were very similar to the first rapid component of the contraction induced by high external K+ concentrations (10). Indeed, these responses may reflect a shift in membrane potential to a more positive value, due to an increase in the outward Cl gradient that is not balanced by the impermeant anions that replace extracellular Cl. Exposure to chloride-free Krebs solution for 30 min resulted in a significant reduction in EFS-induced contractions at all the frequencies tested (1 to 50 Hz), with greater sensitivity in the rat rather than the sheep urethra (Fig. 4, B and C). However, in both species, this effect was much weaker than that observed with niflumic acid or 9-AC (0.1 and 1 mM, respectively; compare Figs. 2 and 4).

Fig. 4.

Effects of chloride-free Krebs solution on nerve-mediated contractile responses. A: representative traces (left) showing the transient urethral contraction elicited on switching to chloride-free Krebs (arrowhead). Right: bar diagrams showing the magnitude of contraction compared with that induced by 120 mM K+ solution in the same preparation. B: representative traces showing sheep (top) and rat (bottom) urethral contractions elicited by EFS (5 s, 1 to 50 Hz) at basal tension in control conditions (left) and following a 30-min incubation with the chloride-free Krebs solution (right). C: changes in the frequency-response contraction curves at basal tension (expressed as a percentage of the respective control curves) in sheep (left) and rat (right) urethra in chloride-free Krebs solution presence (pretreatment for 30 min, filled symbols) or in normal Krebs solution (open symbols). Data are expressed as means ± SE (n = 8 in sheep, n = 7 in rats). *P < 0.05, **P < 0.01, ***P < 0.001 when compared with controls (1-way ANOVA followed by Student's t-test for unpaired observations).

Effect of 2-APB on contractile responses.

In a model of pacemaker activity proposed previously (29), CaCC stimulation occurs following the rise in intracellular Ca2+ levels that results from the activation of intracellular IP3-sensitive calcium stores. When the effect of IP3 receptor inhibition with 2-APB was investigated, significant inhibition of EFS-induced contractions was only observed in sheep urethral preparations following a 30-min exposure to 2-APB (50 μM; Fig. 5).

Fig. 5.

Effects of the IP3 receptor inhibitor 2-APB on contractions induced by EFS. Changes in the frequency-response contraction curves (EFS, 5 s, 1 to 50 Hz) at basal tension (expressed as percentage of the respective control curves) in sheep (left) and rat (right) urethra in the presence (●) or absence (○) of 2-APB (60 μM). Data are expressed as means ± SE (n = 7 in sheep, n = 6–8 in rats). *P < 0.51, **P < 0.01, ***P < 0.001 compared with controls (1-way ANOVA followed by Student's t-test for unpaired observations).

Effect of niflumic acid on nitrergic relaxant responses.

We limited the study of the effects of niflumic acid to rat urethral preparations, as these were the most sensitive to the effects of chloride channel inhibitors. Niflumic acid was the most effective inhibitor of EFS-induced contractions, although pretreatment of rat urethral preparations with niflumic acid (50 μM) did not modify the magnitude or shape of nitrergic relaxations at any frequency or duration of EFS tested (0.5 to 12 Hz, 5 s; 2 and 35 Hz, 1 min; Fig. 6). Furthermore, pretreatment with 2-APB (50 μM, 30 min) did not modify EFS-induced rat urethral relaxations (data not shown).

Fig. 6.

Effects of niflumic acid on the frequency-dependent nitrergic relaxations. A: representative traces showing the relaxation elicited by EFS (5 s, 0.5 to 12 Hz, and 1-min duration, 2 and 35 Hz) in AVP (0.1 μM) precontracted rat urethral preparations under control conditions (top) and after 30-min pretreatment with 50 μM niflumic acid (bottom). B: changes in the frequency-response curves in precontracted rat urethras following short (5 s; left)- and long-duration EFS (1 min; right) in the presence (filled symbols and bars) or the absence (open symbols and bars) of 50 μM niflumic acid. Data are expressed as means ± SE (n = 7). No significant differences were found between the 2 groups (1-way ANOVA).

Expression and distribution of ANO1 chloride channels in the urethral wall.

ANO1 channel immunoreactivity (-ir) was exclusively detected in smooth muscle and urothelial cells in all three species studied (no such staining was detected in control preparations when the primary antibody was omitted: data not shown). The strong and clear ANO1 signal observed in smooth muscle colocalized with α-actin (Fig. 7, A to C and D to F). In mouse preparations, the restrictions imposed by the antibodies used (see materials and methods) precluded determination of ANO1 and α-actin in the same section. However, when immunoreactions were performed on contiguous sections, the profile of both markers revealed similar patterns (Fig. 7, G and H) to those seen in sheep and rat. Analysis of the colocalization of ANO1 with vimentin revealed ANO1 in vimentin-expressing ICCs in the lamina propia or between smooth muscle bundles (Fig. 8). Furthermore, ANO1 was not detected in any nervous structures, including nerve fibers and nerve trunks.

Fig. 7.

Colocalization of anoctamin 1 (ANO1) channels and α-actin in smooth muscle of the urethra. Representative photomicrographs showing immunolocalization of ANO1 (green: A, D, and G) and α-actin (red: B, E, and H) within the smooth muscle layer of rat (top), sheep (middle), and mouse (bottom) urethra. Intense ANO1-immunoreactivity (ir) can be observed colocalized with α-actin in smooth muscle cells (C and F). Faint but positive ANO1-ir was detected in the rat and mouse urothelium (A and G). Scale bars = 50 μm.

Fig. 8.

ANO1 channels and vimentin do not colocalize in urethral interstitial cells of Cajal (ICCs). Representative photomicrographs showing intense ANO1-ir (green) within the rat smooth muscle coat (A) and sheep urothelium (D). No ANO1-ir was detected in the submucosal layer, which exhibited abundant vimentin-positive ICCs (B, E). ANO1 (green) did not colocalize with vimentin (red), which was distributed in the submucosa (F) and between the muscle bundles (C). Scale bars = 50 μm.

ANO1 mRNA expression in the rat urethra.

RT-PCR was performed using two sets of specific primers designed from known ANO1 sequences, directed against different regions of the mRNA, and producing products of different sizes and compositions, to ensure accurate assessment of ANO1 gene expression. In both cases, each set of primers amplified a specific band of the predicted size from RNA obtained from both urethra and prostate tissue, the latter serving as a positive control. Indeed, the product amplified from urethral tissue produced a band of stronger intensity. No bands were detected in reactions lacking reverse transcriptase, discounting the possibility of genomic DNA contamination of the preparations (Fig. 9).

Fig. 9.

ANO1 mRNA expression in rat urethra. Amplification products of ANO1 visualized on 2% agarose gels stained with ethidium bromide. The mRNA transcripts of the expected sizes for this channel were amplified from rat urethral tissue [637 bp for the first set of ANO1 primers (top) and 308 bp for the second set of ANO1 primers (bottom)] and corresponded to the transcripts amplified from rat prostate tissue (positive control). Negative controls were performed without the reverse-transcriptase enzyme (-RT). Right lane: 100-bp ladder.


We performed a comparative study of the distribution of the recently cloned ANO1 in three species, considered to be the main CaCC (6, 26, 34). In addition, we assessed how the movement of chloride participated in both contractile and relaxant responses in the urethra, and we analyzed the modulatory role of ICCs in the urinary tract by studying the involvement of CaCC.

Excitatory urethral contractile responses are generally considered to be of adrenergic origin, as confirmed here. While other transmitters may contribute to evoked contractions in several species (3), their role has not yet been clearly established. Evoked contractions may have a cholinergic component, involving either direct action of ACh on discrete smooth muscle layers (14) or on the modulation of NE release (21). Since atropine partial inhibits nerve-mediated excitatory urethral contractions in all three species studied, there does indeed seem to be some cholinergic input to these events. However, blocking this effect with phentolamine suggests that ACh acts on presynaptic muscarinic receptors in adrenergic nerve terminals, supporting a modulatory effect of ACh on NE release (21). Indeed, the effect of CaCC inhibitors on cholinergic responses could not be distinguished from that of adrenergic inhibitors, pointing to NE as the true mediator of smooth muscle activation.

Although classically linked to the development and maintenance of contractile responses in smooth muscle structures (7), the physiology of chloride channels remains controversial, in part due to the lack of specific pharmacological tools, and their widespread distribution and variability. In our experimental conditions, the CaCC inhibitors niflumic acid and 9-AC effectively inhibit urethral contractile responses in all species tested.

The sheep urethra was slightly more resistant to the inhibitory action of these compounds than that of the rat and mouse. However, the putative CaCC antagonist SITS (36) consistently had no effect in any of the three species under each experimental condition, challenging its activity as a specific CaCC inhibitor. Indeed, previous studies reported no effect of SITS or 4,4′-diisothiocyanatostilbene-2,2′-disulfonic acid on intracellular chloride levels ([Cl]i) in sheep cardiac Purkinje fibers (32), or in guinea pig vas deferens (2) or ureter (1). It has been well-established that this stilbenedisulfonic acid derivative modifies [Cl]i in other systems (15), although its activity as an inhibitor of Cl/HCO3 exchange may also account for this effect (reviewed in Ref. 7).

While a discussion of the mechanisms regulating chloride movements is beyond the scope of the present study, some basic concepts should be considered. Physiological levels of [Cl]i in smooth muscle cells are higher than those predicted by the Donnan equilibrium, since Em > ECl and thus, Cl movements have a net depolarizing effect (7). Deviation from the equilibrium is augmented further when smooth muscle cells are immersed in artificial saline solutions (e.g., Krebs) containing much higher [Cl] than the extracellular milieu, thereby raising the [Cl]i even further (18). In these conditions, removal of external chloride is followed by an increase in Cl efflux across the membrane along a strong concentration gradient, leading to a membrane depolarization that is augmented when external Cl is replaced by impermeant anions (20). When depolarization is sufficient to activate voltage-gated Ca2+ channels, the increase in [Ca2+]i can induce contraction. As shown in vascular preparations (20), this membrane depolarization depends on the length of exposure to low extracellular chloride [Cl]o conditions, with the initially significant effect shortly after exposure progressively diminishing thereafter. By contrast, a low [Cl]o potentiates NE-induced vascular contraction through CaCCs that are activated by NE-mediated increases in [Ca2+]i (potentiation by low [Cl]o disappears at 0 Ca2+) (20). In the urethra of all three species studied here, contractions induced by both EFS and exogenous NE or ACh were dependent on Cl currents, as witnessed by the strong inhibitory effect of low [Cl]o, and their inhibition by 9-AC and niflumic acid. However, variation between species regarding the contribution of different Ca2+ sources to CaCC activity in NE-induced contraction cannot be ruled out. The Cl current in the rat and mouse urethra is most likely activated by Ca2+ influx, consistant with the failure of 2-APB to affect NE-induced contractions. By contrast, Ca2+ mobilization from IP3-sensitive stores triggered by NE receptor activation in the sheep urethra may be partly responsible for Ca2+-mediated CaCC activation.

Despite these differences, our functional results are consistent with chloride movements via chloride channels representing a major component of smooth muscle contractility in the urethra. The earlier failure to detect ANO1 in the mouse urethra (16) clearly contradicts our current results. While this may be due to differences in the immunohistochemical protocols used (the earlier study used methanol or paraformaldehyde as fixatives, with no cryoprotection), it is more likely that these conflicting findings are due to the use of different antibodies (i.e., a rabbit polyclonal antiserum generated in-house vs. a commercially available ANO1 rabbit antibody that produced similar results in the gastrointestinal tract) (12, 17, 35). Given the successful detection of ANO1 in independent studies, it is unlikely that our results are a mere artefact. Moreover, the specificity of the immunoreactivity was further confirmed by assessing the distribution of ANO1 mRNA using two different probes targeting separate regions of the mRNA sequence. In both cases, a clear, unique band of the predicted size was amplified and while we did not perform quantitative PCR, the apparent intensity of the bands amplified from the urethra was stronger than that amplified from the prostate, which was used as a positive control and strongly expresses ANO1 (surpassed only by the pancreas) (25). Thus, the results of PCR amplification are consistent with the ANO1 immunolabeling findings.

Our structural data revealed the restriction of ANO1 to smooth muscle cells (expressing smooth muscle actin) and epithelial cells of the urothelium. Anoctamins have been described in almost all secretory epithelia and they are considered essential for the secretion process (see Ref. 19 for a recent review). However, the urothelium is generally considered not to have a secretory function, except for its role in Na+, K+, and Cl transport in the bladder, which is triggered by an increase in hydrostatic pressure (33). Interestingly, before its identification as a CaCC, ANO-1 was associated with tumoural cells and malignancy (measured as migratory ability), due to its participation in regulating cell volume during mechanical deformation (19). Indeed, ANO-1 itself has been directly described as a mediator of mechanosensitive secretion in biliary epithelium (9). While we found no significant changes in urethral contractions in preparations naturally devoid of urothelium (sheep strips) or in which it has been removed (rat rings, not shown), the participation of ANO-1 in mechanoreception by urothelium (routinely subject to large changes in both stretching and fluid pressure) appears plausible and should be analyzed further using different experimental approaches.

ANO1-ir was consistently absent from cells expressing vimentin, despite the distinct subtypes of ICC described in different layers of the urethra (11). These results appear surprising at first glance, as CaCCs are thought to be present in rabbit urethral ICCs, where they generate large-amplitude spontaneous Cl currents (28) and may act as “pacemakers” of urethral slow wave activity (27). Moreover, the use of specific CaCC inhibitors, such as 9-AC and niflumic acid, has been proposed for “pharmacological knock-out” of ICCs in the rabbit urethra (30), emphasizing the important role of CaCCs in the origin of this spontaneous oscillator.

ICCs were initially proposed to act as cytosolic Ca2+ oscillators in the gut, based on the observed activation of CaCC by Ca2+ released from IP3-dependent stores and the subsequent Ca2+ influx (5). However, this mechanism does not appear to persist in the urethra given the lack of correlation between Ca2+ transients in ICCs and smooth muscle cells in different regions of the urinary tract, and in different species (13). Furthermore, the ICC-Ca2+ oscillator model suggests electrical coupling between ICCs and smooth muscle cells via gap junctions (5). However, this direct coupling appears not to be relevant in urethral neurotransmission despite the existence of gap junctions between these cells, suggesting chemical rather than electrical communication (23). In agreement, the present findings do not support the ICC-Ca2+ oscillator model in whole urethral preparations from sheep, rats, or mice, as no spontaneous or slow wave activity was detected and IP3 inhibitors had no effect on EFS-induced urethral contractions.

The absence of ANO1 in urethral ICCs from the three species studied may be explained by the presence of a different and as-yet uncharacterized CaCC in these cells. Alternatively, ANO1 may only be expressed in rabbit urethral ICC. The lack of ANO1 channels in urethral ICC is consistent with the failure of CaCC inhibitors to modify EFS-induced nitrergic relaxation. Although significant increases in cGMP have been observed in a subset of urethral ICCs upon functional nitrergic stimulation (11), this appears to be at least partially mediated by a different ionic mechanism, involving cyclic nucleotide-gated (CNG) channels (31). CNG channels are thought to modify [Ca2+]i in different cellular microdomains, possibly leading to the release of some chemical mediators (22).

In conclusion, we detected strong ANO1 expression in the smooth muscle cells of the urethra in three different species, where it played a crucial role in the development and maintenance of excitatory contractile responses, but it did not participate significantly in urethral relaxation. Moreover, ANO1 was absent from all urethral ICC subtypes, in sharp contrast to the strong expression in ICCs of the gastrointestinal tract. Therefore, despite their morphological resemblance, the functional differences between distinct ICCs prevent us from drawing generalized conclusions regarding their function.


This work was supported by grants from the “Comunidad de Madrid–Universidad Complutense de Madrid” (UCMGR85/06-920307), UCM-Santander (GR35/10-A-920307), and Fundación Mutua Madrileña (FMM2011).


No conflicts of interest, financial or otherwise, are declared by the author(s).


Author contributions: M.S., A.G.-P., and D.T. performed experiments; M.S. and D.T. analyzed data; M.S. and D.T. prepared figures; M.S., A.G.-P., and D.T. edited and revised manuscript; M.S., A.G.-P., and D.T. approved final version of manuscript; A.G.-P. and D.T. conception and design of research; A.G.-P. and D.T. interpreted results of experiments; A.G.-P. and D.T. drafted manuscript.


The microphotographs were acquired and analyzed at the Microscopy and Cytometry Centre (Complutense University, Madrid, Spain). We thank Alfonso Cortés and Luis M. Alonso for technical assistance with the fluorescence microscopy.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
  22. 22.
  23. 23.
  24. 24.
  25. 25.
  26. 26.
  27. 27.
  28. 28.
  29. 29.
  30. 30.
  31. 31.
  32. 32.
  33. 33.
  34. 34.
  35. 35.
  36. 36.
View Abstract