|Home | About | Journals | Submit | Contact Us | Français|
Myosalpinx contractions are critical for oocyte transport along the oviduct. A specialized population of pacemaker cells—oviduct interstitial cells of Cajal—generate slow waves, the electrical events underlying myosalpinx contractions. The ionic basis of oviduct pacemaker activity is unknown. We examined the role of a new class of Ca2+-activated Cl− channels (CaCCs)—anoctamin 1, encoded by Tmem16a—in oviduct slow wave generation. RT-PCR revealed the transcriptional expression of Tmem16a-encoded CaCCs in the myosalpinx. Intracellular microelectrode recordings were performed in the presence of two pharmacologically distinct Cl− channel antagonists, anthracene-9-carboxylic acid and niflumic acid. Both of these inhibitors caused membrane hyperpolarization, reduced the duration of slow waves, and ultimately inhibited pacemaker activity. Niflumic acid also inhibited propagating calcium waves within the myosalpinx. Slow waves were present at birth in wild-type and heterozygous oviducts but failed to develop by birth in mice homozygous for a null allele of Tmem16a (Tmem16atm1Bdh/tm1Bdh). These data suggest that Tmem16a-encoded CaCCs contribute to membrane potential and are responsible for the upstroke and plateau phases of oviduct slow waves.
Contractions of the oviduct smooth muscle, known as the myosalpinx, are essential for normal transport of eggs along the oviduct . Electrical slow waves underlie the mechanical contractions of the myosalpinx . Several investigators have reported the existence of slow wave activity in the oviducts of mammalian species [1–5]. The origin of these slow wave events in mice has recently been established and determined to lie within a specialized population of KIT-positive pacemaker cells called the oviduct interstitial cells of Cajal (ICC-OVI) . Morphological studies have revealed that ICC-OVI are also present in human oviducts (fallopian tubes) [6, 7]. ICC networks and their associated pacemaker activity are sensitive to inflammatory conditions and infections such as postoperative ileus in the gastrointestinal tract  or chlamydia in the female reproductive tract .
The ionic basis of pacemaker activity in the oviduct is unknown. Previous studies of other visceral smooth muscle tissues have revealed that ICC pacemaker activity is dependent on [Ca2+]o (extracellular Ca2+ concentration) [9–12]. We have recently demonstrated that ICC-OVI pacemaker activity is also dependent on [Ca2+]o because its reduction from 2.5 mM to nominally free [Ca2+]o inhibited slow waves and myosalpinx contractions . ICC-OVI pacemaker activity is also abolished by the sarcoplasmic and endoplasmic reticulum calcium ATPase (SERCA) pump inhibitor cyclopiazonic acid, revealing that functional intracellular Ca2+ stores are essential for pacemaker activity . Based on these findings, it seems likely that a depolarizing Ca2+-dependent conductance is involved in ICC-OVI-mediated pacemaker activity.
It has been shown that the pacemaker activity of ICCs in several visceral smooth muscle tissues is sensitive to Cl− channel antagonists, including 4,4′-diisothiocyanostilbene-2,2′-disulphonic acid, antracene-9-carboxylic acid (9-AC), and niflumic acid (NFA) [10, 11, 14, 15]. Additionally, reduced [Cl−]o has been shown to inhibit pacemaker activity in the antrum and urethra [10, 11]. Intracellular Cl− concentration is known to vary from one smooth muscle to the next but is generally thought to lie in the range of 40–110 mM with equilibrium potentials for Cl− in the range of −20 to −30 mV . Because these potentials are positive to the resting membrane potential (RMP) of the oviduct myosalpinx (RMP average = −58 ± 1 mV; ), activation of Cl− channels would give rise to efflux of Cl−, resulting in membrane depolarization. Recently, it has been demonstrated that expression of the transmembrane protein anoctamin-1 (ANO1, also known as DOG1, and encoded by Tmem16a) yields Ca2+-activated Cl− channel (CaCC) currents, providing evidence that the ANO1/TMEM16 family of proteins are the molecular correlates of CaCCs [17–19]. Immunohistochemical studies have revealed the expression of ANO1 specifically within ICCs (and not in mast cells) in gastrointestinal smooth muscle tissues [15, 20]. Studies into the function of ANO1 in gastrointestinal ICCs have indicated a fundamental role for these channels in the generation of pacemaker activity [15, 21]. Therefore, we hypothesized that oviduct pacemaker activity may be similarly dependent on CaCCs.
Because electrical slow waves are the pacemaker events underlying oviduct phasic contractions that are essential for oocyte transport along the oviduct, a better understanding of the ionic basis of slow waves in this tissue may aid in the development of future therapies for infertility. Using genetic and pharmacological approaches, we investigated the involvement of CaCCs, possibly encoded by Tmem16a, in the generation of oviduct pacemaker activity.
Adult BALB/c mice (30–60 days old; The Jackson Laboratory, Bar Harbor, MN) and neonatal Tmem16a−/− (Tmem16atm1Bdh/tm1Bdh), Tmem16a+/+, or Tmem16a+/− (Tmem16a+/tm1Bdh) mice (provided by B.D. Harfe, University of Florida, Gainesville, FL) (see  for details on the production of these animals) were utilized in the described studies. Mice were humanely killed by isoflurane (Baxter, Deerfield, IL) inhalation followed by cervical dislocation.
The animals were maintained and the experiments performed in accordance with the National Institutes of Health Guide for the Care and of Laboratory Animals. All the procedures were approved by the Institutional Animal Use and Care Committee at the University of Nevada.
Oviducts were removed from mice via an abdominal incision immediately after cervical dislocation and submerged in Krebs-Ringer bicarbonate solution (KRB; see below for composition and pH). Individual oviducts were cut free from the ovary and uterine horn and uncoiled by sharp dissection to yield single intact oviduct preparations comprising the infundibulum, ampulla, isthmus, and intramural segments.
For RT-PCR experiments, oviducts were cut open longitudinally and the mucosa was carefully removed. The remaining preparation consisted of the myosalpinx and serosal layers. These preparations were cut into thirds, which were assumed to approximate the infundibulum/ampulla, isthmus, and intramural segments.
Tissues were maintained and perfused with KRB containing (mmol/L): NaCl, 120.35; KCl, 5.9; NaHCO3, 15.5; NaH2PO4, 1.2; MgCl2, 1.2; CaCl2, 2.5; and glucose, 11.5. KRB was bubbled with a mixture of 97% O2–3% CO2 and warmed to a physiologically relevant temperature of 37°C ± 0.5°C. Under these conditions, the pH of the KRB was maintained between 7.3 and 7.4. High [K+]o KRB contained identical salt concentrations as those listed above for normal KRB except that KCl was elevated to 10 mM and NaCl was reduced to 116.25 mM to compensate for the charge difference and to maintain isotonicity. Antracene-9-carboxylic acid and NFA were purchased from Sigma-Aldrich (St. Louis, MO) and were dissolved in the manufacturer's recommended solvent to make stock solutions. These solutions were then diluted to the appropriate concentration in KRB and applied to recording chambers containing oviduct preparations at a constant perfusion rate of 3 ml/min.
Intact oviduct preparations were pinned to the base of a Sylgard elastomer (Dow Corning, Midland, MI)-lined recording chamber using fine tungsten pins (120-μm diameter) to restrict movement of the muscle and extend the length of time the impalement could be maintained. Tissues were allowed to equilibrate for at least 1 h prior to the commencement of experiments. After this time, individual smooth muscle cells within the myosalpinx, at locations approximately 50% along the length of the oviduct (i.e., in the isthmus), were impaled using sharp glass microelectrodes that had resistances of 80–120 MΩ and were filled with 3 M KCl. The potential difference across the smooth muscle cell membrane was recorded using a high impedance electrometer (Axoclamp 2B; Axon Instruments/Molecular Devices, Sunnyvale, CA). This data was digitized using a Digidata 1322A (Axon Instruments) and recorded onto a computer using Axoscope 9.2 (Axon Instruments) software.
Intracellular recordings were analyzed using Clampfit 9.0 (Axon Instruments) software. Several parameters of slow wave activity were measured with this program, including RMP, frequency, amplitude, half-maximal duration, and maximum rate of rise of the upstroke (dV/dt max). Statistical significance was calculated by running either a Student t-test or an ANOVA using SigmaStat (Systat Software Inc., Chicago, IL) software; P values of <0.05 were considered to represent significant changes. Results are expressed in the form of mean ± SEM as well as the number of oviducts used in the various experiments (n), with each oviduct coming from a different animal. Final figures were prepared from digitized data using Corel Draw 13.0 (Corel Corp., Ontario, Canada).
Total RNA was isolated from BALB/c oviduct (ampulla, isthmus, and intramural segments following removal of the epithelial lining by sharp dissection) and brain (control) using TRIzol reagent (Invitrogen, Carlsbad, CA), and the RNA was treated with 1 unit/μl DNase I (Promega, Madison, WI). First-strand cDNA was prepared from 1 μg RNA using SuperScript II reverse transcriptase (Invitrogen) in a 20-μl reaction containing 25 ng oligo dT(12–18) primer, 1 μl of 10 mM dNTP, 5× first-strand buffer (250 mM Tris-HCl, pH 8.3, 375 mM KCl, 15 mM MgCl2), and 0.1 M dithiothreitol. PCR was performed with specific primers for Tmem16a on 2 μl cDNA using AmpliTaq Gold PCR Master Mix (Applied Biosystems, Foster City, CA). The PCR primers used are described in Table 1. All the primers were designed to span intronic sequences to eliminate amplification of contaminating genomic DNA in the source RNA. Amplification of glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as a control for assessing cDNA integrity. PCR reactions without a template served as controls for primer contamination. PCR reactions were performed in a GeneAmp 2700 thermal cycler (Applied Biosystems, Foster City, CA). The amplification profile was 95°C for 10 min to activate the AmpliTaq polymerase, then 35 cycles of 95°C for 15 sec, 60°C for 20 sec, and 72°C for 30 sec, followed by a final step of 72°C for 5 min. RT-PCR amplification fragments were analyzed by size analysis on 2% agarose gels alongside a 100-base pair (bp) marker.
Oviducts were collected as described above and treated with 0.7 mg/ml collagenase type F and 2 mg/ml bovine serum albumin (Sigma) in 2 ml KRB solution for 4 min at 37°C to facilitate dye loading into the tissues. Oviducts were washed out and pinned to the Sylgard elastomer-lined base of a recording chamber. After equilibration (1 h), oviducts were loaded with 25 μg FluoroPure-AM (Fluo-4; Molecular Probes, Eugene, OR) in a solution of 0.02% dimethyl sulfoxide and 0.01% nontoxic detergent Cremophor EL for 20 min at 25°C. After incubation, the preparation was perfused with the warm KRB (30 min) to allow for deesterification of the dye. Calcium imaging was performed through the serosa. We could readily visualize muscle bundles using this approach and did not detect nonspecific signals from the epithelial.
Tissues were imaged using an Eclipse E600FN microscope (Nikon Inc., Melville, NY) equipped with Nikon Plan Fluor lenses (20 and 40×). The indicator was excited at 488 nm (Polychrome IV; TILL Photonics, Gräfelfing, Germany), and the fluorescence emission (>515 nm) was detected using a cooled, interline transfer CCD-camera system (Imago QE; TILL Photonics) using 4 × 4 binning. Image sequences (344 × 260) were collected at 12.4 frames-per-sec (fps) for 40–60 sec using TILLvisION software (TILL Photonics).
Calcium activity in oviducts was analyzed using custom software (Volumetry G7mv; written by Grant W. Hennig). Calcium waves in the longitudinal axis of the oviduct that consisted of activation of smooth muscle cells and ICC-OVI from the ovarian to uterine pole were measured using spatio-temporal maps (ST maps) . The frequency and velocity of each Ca2+ wave was calculated. Due to the image acquisition rate, maximum velocities that could be resolved at 20× magnification were ~4 mm/sec. The average background fluorescence was subtracted in ST maps to reveal dynamic changes in Ca2+ fluorescence. Sequences of images showing the spread of Ca2+ waves were differentiated (Δt = 0.162 sec) to better visualize the spread of the Ca2+ wave front.
Tmem16atm1Bdh/tm1Bdh mice were generated by replacing exon 12 of Tmem16a with a PGK-neomycin cassette by homologous recombination in embryonic stem cells as previously described . Genomic DNA was isolated from transgenic mice tails using standard methods. DNA (0.5 μl) was amplified in each PCR reaction to determine transgenic mice genotypes. A 393-bp PCR fragment was amplified from the Tmem16a+/+ allele with primers that bind within and span exon 12. The Tmem16a–/– allele (350 bp) was amplified with primers that bind to the PGK-neomycin cassette.
Several studies have suggested that CaCCs are important for pacemaker activity in a variety of smooth muscle tissues [11, 12, 14, 23–25]. We performed RT-PCR to examine the expression of Tmem16a-encoded CaCCs in oviduct myosalpinx. Tmem16a transcripts were found in all the segments of the oviduct (Fig. 1).
Electrical slow waves occurred spontaneously in the oviduct myosalpinx. These events consisted of 41 ± 1 mV amplitude depolarizations with a half-maximal duration of 1.5 ± 0.2 sec, occurring from membrane potentials of −61 ± 1 mV at a frequency of 10.4 ± 0.7 cycles/min (n = 17), which are similar to the slow wave characteristics published previously . The peak membrane potential, reached during the upstroke phase of slow waves, was −20 ± 1 mV (n = 17). This potential is near the equilibrium potential predicted for Cl− in vas deferens smooth muscle tissues (−24 mV) . On this basis, we considered Cl− conductance to be responsible for the inward current-producing depolarization during the oviduct slow wave upstroke. Chloride channel-blocking drugs are known to have nonspecific side effects. To overcome this, we investigated the effects of 9-AC and NFA, two Cl− channel blockers with different chemical structures .
Antracene-9-carboxylic acid (1 mM) hyperpolarized RMP from a control averaging −64 ± 2 mV to −69 ± 1 mV (n = 6, P = 0.017; Fig. 2, A–E). At larger hyperpolarized potentials, a distinct inflection was noted in the upstroke phase of the slow waves (Fig. 2, C and D). This inflection resembles the transition between the primary and secondary components of slow waves described in some gastric smooth muscles . A secondary component was also apparent during the development of the effects due to 9-AC, but after several minutes of exposure, the secondary component was abolished, leaving only the primary component (Fig. 2, C and D). Antracene-9-carboxylic acid also caused significant reduction in slow wave frequency, from an average of 11.7 ± 1.4 cycles/min under control conditions to 5.3 ± 1.8 cycles/min in the presence of 9-AC (P = 0.004; Fig. 2, A–C and F). Slow wave half-maximal duration also decreased from 1.2 ± 0.1 sec to 0.6 ± 0.2 sec during 9-AC exposure (P = 0.011; Fig. 2G). Slow waves were completely inhibited in two of six preparations. Control activity returned upon washing out the 9-AC from the bathing solution.
The effect of NFA (10, 30, and 50 μM) on oviduct electrical activity was also tested. Niflumic acid caused concentration-dependent hyperpolarization of tissues, from −61 ± 1 mV (n = 15) to −68 ± 2 mV at 50 μM (n = 11, P = 0.021; Fig. 3, A–C and E). Niflumic acid (50 μM) abolished slow waves in every experiment (Fig. 3C). The half-maximal duration of slow waves was also reduced by NFA in a concentration-dependent manner. Under control conditions, the half-maximal duration was 1.8 ± 0.2 sec, and this decreased to 1.1 ± 0.1 sec just prior to the blockade of slow waves (Fig. 3, D and F). To investigate whether the abolition of slow waves in NFA (50 μM) was simply a factor of membrane hyperpolarization, cells were repolarized to control levels using elevated [K+]o in the continued presence of 50 μM NFA (n = 4; Fig. 3G). Spontaneous slow waves were not restored by repolarization of membrane potential to control levels. Control activity returned upon washing out the NFA from the bathing solution.
The spread of pacemaker activity in networks of ICC-OVI was monitored using the Ca2+ indicator Fluo-4 (Fig. 4). Calcium waves that propagated along the length of the oviduct were observed in six out of seven preparations (Fig. 4, A–C). These Ca2+ waves occurred at a frequency of 10.4 ± 2.0 cycles/min (interval = 5.8 ± 1.1 sec) and propagated at a velocity of 3.3 ± 0.4 mm/sec (n = 6), although this likely represents the lower end of the velocities (maximum resolvable velocity = 4 mm/sec using 20× lens at 12.4 fps).
The addition of NFA (100 μM) abolished (four of six preparations) or significantly decreased the amplitude and the frequency of propagating Ca2+ waves (two of six preparations: 7.6 ± 0.9 cycles/min, interval = 7.92 ± 1.0 sec; Fig. 4D). In one preparation in which uncoordinated intracellular Ca2+ transients were observed, NFA blocked all the activity.
The role of CaCCs in oviduct pacemaker activity was further investigated by performing intracellular recordings on oviducts from mice that were homozygous for a null allele of Tmem16a (Tmem16a−/−) and their wild-type littermates. Previous reports have found that 90% of Tmem16a−/− mice die within the first 9 days (P9) of birth . Therefore, in this study experiments were performed on oviducts from litters killed on P0. Functional pacemaker activity and slow waves have been reported previously in BALB/c P0 mouse oviducts , and this was again confirmed in a BL/6 background in the present study. The RMP in wild-type homozygote (Tmem16a+/+) and heterozygote (Tmem16a+/−) oviducts was −49 ± 3 mV. Slow waves occurred at a rate of 2.1 ± 0.3 cycles/min and had amplitudes of 21 ± 3 mV. The half-maximal duration was 8.0 ± 3.3 sec, and the maximum rate of rise of the upstroke was 165 ± 7 mV/sec (n = 7; Fig. 5, A and D). In contrast, although the RMP was within a normal range and not statistically different from wild-type controls (−47 ± 5 mV), slow waves were not recorded from Tmem16a−/− oviducts (n = 6; Fig. 5, B–D). These results suggest that CaCCs are essential for slow wave generation in the murine oviduct and may compete against a hyperpolarizing conductance (presumably a K+ conductance) during the plateau of slow waves, influencing their half-maximal duration.
Electrical slow waves, which underlie the rhythmic phasic contractions and egg propulsion in the oviduct, are generated by KIT-ICC-OVI . Key conductances involved in the pacemaker activity of ICC-OVI have not previously been determined. In the present study, we found that oviduct tissues expresses Tmem16a, which encodes ANO1, a CaCC recently identified in ICCs of the gastrointestinal tract. Anoctamin-1 is responsible for large inward pacemaker currents in ICC  and the generation of slow waves in gastrointestinal tissues . Inhibitors of CaCC blocked oviduct slow waves and propagation of slow waves within ICC-OVI networks. Thus, our findings are in basic agreement with an earlier study in which it was suggested that a Cl− conductance plays a role in the plateau component of guinea pig oviduct slow waves . In our experiments, NFA (50 μM) blocked slow waves and caused hyperpolarization in all the preparations studied. Hyperpolarization, per se, was not responsible for the inhibition of slow waves produced by NFA because repolarization with elevated [K+]o did not restore slow waves. Although Cl− channel-blocking drugs inhibited slow wave activity, these drugs have been reported to have nonspecific effects and rather high concentrations were necessary to elicit effects. Therefore, important confirmation of the importance of CaCC (due to Tmem16a expression) in oviduct pacemaker activity was provided by the observation that slow waves were absent in mice homozygous for a null allele of Tmem16a. Taken together, these findings suggest that ANO1 channels encoded by Tmem16a are responsible for spontaneous electrical activity in the oviduct and are critical in egg transport.
In other visceral smooth muscles, pacemaker activity has been found to coincide with transient increases in intracellular Ca2+ concentration [11, 29], suggesting that conductance(s) responsible for pacemaker current are Ca2+ dependent. In support of this hypothesis, pacemaker activity was abolished by reducing [Ca2+]o  or buffering [Ca2+]i with the membrane-permeable Ca2+ chelators 1,2-bis(o-aminophenoxy)ethane-N,N,N′,N′-tetraacetic acid tetra(acetoxymethyl) ester (BAPTA-AM) [9, 14] or 5,5′-dimethyl BAPTA-AM (also known as MAPTA-AM) . Calcium transients in gastrointestinal pacemaker cells appear to be dependent upon release from intracellular stores, and slow wave activity was abolished in gastric muscles of mice lacking inositol-1,4,5-trisphosphate (IP3)-type 1 receptors  and by pharmacological inhibitors of IP3 receptors such as 2-aminoethoxydiphenyl borate and xestospongin [29, 32]. Pacemaker activity in other smooth muscles was abolished by inhibitors of Cl− channels [10, 11, 14]. Our findings are consistent with these previous investigations and suggest that pacemaker activity in the oviduct may be linked to periodic release or entry of Ca2+ and activation of CaCC to generate inward current and slow wave depolarizations .
We have recently reported that oviducts display rhythmic Ca2+ oscillations that coincide with myosalpinx contractions . Furthermore, ICC-OVI pacemaker activity has been shown to be dependent on [Ca2+]o and intact intracellular stores because slow waves were inhibited by reducing [Ca2+]o from 2.5 mM to nominally free [Ca2+]o or by the addition of the SERCA pump inhibitor cyclopiazonic acid . It therefore seemed plausible that CaCCs could have a role in pacemaker activity in the oviduct, and this possibility was explored in the current study.
The equilibrium potential for Cl− has been reported to be approximately −24 mV in smooth muscle cells of the vas deferens . The peak depolarization of oviduct slow waves (−23 ± 0.8 mV) is close to this equilibrium potential for Cl− ions. Involvement of a Cl− conductance may explain why slow waves do not depolarize to 0 mV or overshoot 0 mV, as might be expected if the main inward current responsible for these events is due to a Ca2+, Na+, or nonselective cation conductance, which would be the other ionic candidates for inward currents.
The initiation of transient inward currents in ICCs leads to spontaneous transient depolarizations (or unitary potentials) [9, 33] that appear as an ongoing discharge of baseline noise in smooth muscle tissues. Inward currents activated sporadically in multiple cells convey net inward current in a tissue, where cells are connected by gap junctions to form a syncytium. This activity would tend to produce a depolarizing influence on neighboring smooth muscle cells and could contribute to tone or increased excitability. In addition to spontaneous pacemaker activity, channels encoded by Tmem16a appear to provide spontaneous inward currents and have a net depolarizing function in the oviduct because NFA also caused an ~7 mV membrane hyperpolarization. Perhaps regulation of this conductance, or [Ca2+]i, which is likely to determine the open probability of ANO1 channels, is a major means by which neurotransmitters and hormones regulate the excitability of the oviduct myosalpinx. It will be interesting in the future to investigate how hormones and nerves, known to regulate the motility of the oviduct and the force of propulsive contractions, possibly influence the ongoing activation of ANO1 channels.
We found previously that ICCs are lost in response to infections by Chlamydia, a major cause of tubal infertility factor . RT-PCR showed that transcripts of Tmem16a are abundant in all the segments of the mouse oviduct, and previous work on gastrointestinal ICCs shows localization of Tmem16a in ICCs. In fact, ANO1, the gene product of Tmem16a, is a more specific label for ICCs in gastrointestinal muscles than KIT protein, suggesting that this could be a common important protein conveying pacemaker activity in ICCs of different organs. It is possible that expression of Tmem16a, therefore, could be an important defining feature of the pacemaker phenotype of ICCs, and defects in pacemaker activity might precede complete loss of the ICC phenotype in oviduct infections. The relative expression of Tmem16a in infectious diseases may be an early biomarker for disease-related damage to oviduct function. Future experiments will evaluate Tmem16a expression as a function of the loss of pacemaker activity in host inflammatory responses to Chlamydia infection.
In summary, data from the present study suggests that a CaCC is involved in the generation of slow waves in the oviduct. At least a major component in the CaCC in oviduct is due to expression of Tmem16a, which encodes a CaCC (i.e., ANO1) and a prominent inward current responsible for slow waves in ICCs of the gastrointestinal tract [15, 21]. Our data provide the first evidence that ANO1 is essential for pacemaker activity in ICC-OVI.
The authors express sincere appreciation to Lauren E. Peri for excellent technical assistance. We also thank Kellan Bigley for his assistance in the analysis of the calcium-imaging data.
1Supported by National Institutes of Health grants DK57236 (S.M.W.) and DK41315 (S.M.W. and K.M.S.). G.W.H. was partially supported by P20RR018751, and F.C.B. was partially supported by HL091238.