The relative contributions of Ca2+-induced Ca2+ release (CICR) versus Ca2+ influx through voltage-dependent Ca2+ channels (VDCCs) to excitation-contraction coupling has not been defined in most smooth muscle cells (SMCs). The present study was undertaken to address this issue in mouse urinary bladder (UB) smooth muscle cells (UBSMCs). Confocal Ca2+ images were obtained under voltage- or current-clamp conditions. When UBSMCs were activated by a 30-ms depolarization to 0 mV, intracellular Ca2+ concentration ([Ca2+]i) increased in several small, discrete areas just beneath the cell membrane. These Ca2+ “hot spots” then spread slowly through the myoplasm as Ca2+ waves, which continued even after repolarization. Shorter depolarizations (5 ms) elicited only a few Ca2+ sparks, which declined quickly. The number of Ca2+ sparks, or hot spots, was closely related to the depolarization duration in the range of ∼5–20 ms. There was an apparent threshold depolarization duration of ∼10 ms within which to induce enough Ca2+ transients to spread globally and then induce a contraction. Application of 100 μM ryanodine to the pipette solution did not change the resting [Ca2+]i or the VDCC current, but it did abolish Ca2+ hot spots elicited by depolarization. Application of 3 μM xestospongin C reduced ACh-induced Ca2+ release but did not affect depolarization-induced Ca2+ events. The addition of 100 μM ryanodine to tissue segments markedly reduced the amplitude of contractions triggered by direct electrical stimulation. In conclusion, global [Ca2+]i rise triggered by a single action potential is not due mainly to Ca2+ influx through VDCCs but is attributable to the subsequent two-step CICR.
- Ca2+-induced Ca2+ release
- Ca2+-activated K+ current
- voltage-dependent Ca2+ channel
in smooth muscle cells (SMCs), Ca2+ influx through voltage-dependent Ca2+ channels (VDCCs) plays a crucial role in the regulation of membrane excitability and myogenic tone (37). Although changes in global intracellular Ca2+ concentration ([Ca2+]i) primarily determine SMC contraction and relaxation, the localized Ca2+ transients or Ca2+ sparks also serve as key elements in a negative feedback mechanism in the regulation of smooth muscle tone (4, 33, 40). Ca2+ sparks caused by spontaneous opening of a cluster of ryanodine receptors (RyRs) on the sarcoplasmic reticulum (SR) membrane (40) give rise to spontaneous transient outward currents (STOCs) due to adjacent Ca2+-activated K+ channels (KCa channels; mainly large-conductance KCa channels, referred to hereinafter as BK channels) (1, 40). STOCs induce membrane hyperpolarization and reduction of VDCC activities, which eventually results in a decrease of global [Ca2+]i and relaxation (44). Ca2+ spark activity (amplitude and frequency) increases with the activation of VDCCs resulting from depolarization (3, 34).
In contrast to this functional coupling between RyR and BK channels, the experimental analysis of the contribution of Ca2+-induced Ca2+ release (CICR) via RyR activation to contraction and excitation-contraction (E-C) coupling in SMCs remains incomplete. Initial experiments suggested that the Ca2+ requirement for CICR in SMCs was too high to participate in contraction (24). In portal vein myocytes, Ca2+ influx through VDCCs, as opposed to CICR, was suggested to be the major factor responsible for elevating [Ca2+]i to induce a contraction (36). On the other hand, an essential contribution of CICR to E-C coupling has been suggested in urinary bladder (UB) (7, 13, 19, 31), vas deferens (31), ureter (6), and coronary artery (14). In UB myocytes from the guinea pig, the increase in [Ca2+]i upon depolarization was substantially reduced by 10 μM ryanodine, suggesting an important contribution of CICR via RyRs in E-C coupling in this type of SMC (13).
Even in UB smooth muscle cells (UBSMCs), however, the importance of a functional contribution of CICR to E-C coupling and contraction has been assessed differently by different research groups. Collier et al. (7) reported that the coupling of VDCC and RyR in UBSMCs of the rabbit is weak and may not be effective in inducing contraction. The application of 10 or 50 μM ryanodine did not reduce, but instead enhanced, spontaneous contractions in guinea pig UB tissue preparations (18, 23). In contrast, Buckner et al. (5) reported that 10 μM ryanodine suppressed spontaneous contractions in the UB of the pig by ∼50%.
The present study was undertaken in an attempt to resolve the contribution of RyR and CICR to E-C coupling in UBSMCs. Our results, which were obtained using simultaneous measurement of fast confocal Ca2+ imaging and membrane currents, clearly show that two distinct steps of CICR are essential for E-C coupling triggered by an action potential in UBSMCs of the mouse.
MATERIALS AND METHODS
C57BL/6 mice (male and female, 8–15 wk old, and 20–30 g body wt) were used in the experiments. Single SMCs were enzymatically isolated from the UB using a previously described method (29) with slight modification. All experiments were performed in accordance with the guiding principles for the care and use of laboratory animals of the Science and International Affairs Bureau of the Japanese Ministry of Education, Science, Sports, and Culture and with the approval of the Ethics Committee at Nagoya City University. In brief, mice were anesthetized with ether and killed by performing exsanguination. The UB was dissected out and freed from other tissues in Ca2+-free Krebs solution. The tissue was immersed in Ca2+-free Krebs solution for 30–60 min in a test tube at 37°C. Subsequently, the solution was replaced with Ca2+-free Krebs solution containing 0.2–0.3% collagenase (Amano Enzyme, Nagoya, Japan). After 10- to 15-min treatment, the solution was replaced with Ca2+-free, collagenase-free Krebs solution. The tissue was gently triturated using a glass pipette to isolate cells. At the start of each experiment, a few drops of the cell suspension were put into a recording chamber. The bath was continuously perfused with HEPES-buffered solution at a flow rate of 5 ml·min−1.
Krebs solution contained (in mM) 112 NaCl, 4.7 KCl, 2.2 CaCl2, 1.2 MgCl2, 25 NaHCO3, 1.2 KH2PO4, and 14 glucose. Ca2+-free Krebs solution was prepared by omitting Ca2+ from the Krebs solution. Standard and Ca2+-free Krebs solutions were equilibrated with a 95% O2-5% CO2 mixture. For electrophysiological recording, HEPES-buffered solution of the following composition was used as the external solution (in mM): 137 NaCl, 5.9 KCl, 2.2 CaCl2, 1.2 MgCl2, 14 glucose, and 10 HEPES. The pH was adjusted to 7.4 with NaOH. For simultaneous recordings of Ca2+ current (ICa) and [Ca2+]i, 30 mM tetraethyl ammonium was substituted for equimolar NaCl in the HEPES-buffered solution, and 1 mM 4-aminopyridine (4-AP) was added to the solution.
The pipette solution contained (in mM) 140 KCl, 1 MgCl2, 2 Na2ATP, 10 HEPES, and 0.1 fluo-4 or indo-1. The pH of this solution was adjusted to 7.2 with KOH. To examine the effects of 100 μM extracellular ryanodine on ICa (see Fig. 4B, right), a pipette solution containing the following composition was used (in mM): 120 CsCl, 20 tetraethylammonium (TEA), 1 MgCl2, 10 HEPES, 5 EGTA, and 2 Na2ATP. The pH of the solution was adjusted to 7.2 with CsOH.
The whole cell patch-clamp technique was applied to single cells (17) using a CEZ-2400 amplifier (Nihon Kohden, Tokyo, Japan). Pipette resistance ranged from 2 to 4 MΩ when filled with the pipette solution. The seal resistance was ∼30 GΩ. Series resistance was between 4 and 8 MΩ and partly compensated. Data were stored and analyzed as described previously (29). Briefly, membrane currents were monitored using an oscilloscope (VC-6041; Hitachi, Tokyo, Japan) and stored on videotape after being digitized using a pulse code modulation recording system (modified to acquire a direct current signal, PCM 501ES; Sony, Tokyo, Japan). Data on tape were later downloaded to an IBM-AT-compatible computer using an analog-to-digital (A/D) converter (DT2801A; Data Translation, Marlboro, MA). Data acquisition and analysis were performed using AQ/Cell-Soft software, which was developed in the laboratory of Dr Wayne Giles (University of Calgary, Calgary, AB, Canada). All current recordings were filtered at 750 Hz and performed at room temperature (24 ± 1°C).
Measurement of fluo-4 and indo-1 signals from single myocytes.
Ca2+ images were obtained using a fast laser-scanning confocal microscope (RCM 8000; Nikon, Tokyo, Japan) and Ratio3 software (Nikon) in the same manner as reported previously (31). Each selected myocyte was loaded with 100 μM fluo-4 or indo-1 by diffusion from the recording pipette. For measurements in which fluo-4 was used, 488-nm excitation from an argon ion laser was delivered through a water-immersion lens objective (×40 magnification, 1.15 numerical aperture; Nikon Fluo). Emitted light of >515 nm was detected using a photomultiplier. Fluorescence intensity (F) in a selected area was measured as an average of pixels included within the area. Data are shown as ΔF/F0 ratios in which F0 is the average fluorescence intensity of five images of the whole cell area during resting conditions and ΔF is the increase in fluorescence intensity from F0. It took 33 ms to scan one full frame (512 × 512 pixels). Using 1/2- and 1/4-band scan modes, we obtained frames that corresponded to areas of 170 × 55 μm or 170 × 27.5 μm every 16.5 or 8.25 ms, respectively. For indo-1 measurements, excitation of 355 nm and emission at 405 and 485 nm, respectively, were applied and detected, and fluorescence images were obtained at 66-ms intervals. The resolution of the microscope was ∼0.4 × 0.3 × 1.2 μm (x, y, and z-axes, respectively). The confocal plane through the cell was set so that the width of the cell was largest 2–3 μm from its lowest point. Recordings were started ∼5 min after rupturing the patch membrane to allow time for the Ca2+ indicator and drugs to diffuse into the cell.
Ca2+ images were stored on an optical disk cartridge (LM-A410; Panasonic, Osaka, Japan) using a rewritable optical disk recorder (LQ-4100A; Panasonic). The images on the optical disks were replayed later and analyzed using Ratio3 software (Nikon). Some analyses were performed using GLOBAL LAB Image software (Data Translation) on an IBM-AT-compatible computer. During the recording of fluorescence images, the cell shape was monitored and recorded on videotape using red or infrared light with a wavelength range of >600 nm and an infrared charge-coupled device video camera module (XC-77BB; Sony), which was attached to the microscope. Video image capture and analysis of cell shape changes were performed later on an IBM-AT-compatible computer using an A/D translation board (DT-55; Data Translation) and GLOBAL LAB Image software.
Calibration of [Ca2+]i.
[Ca2+]i values were calculated from ratios of fluorescent signals of indo-1 at 405 nm compared with those at 485 nm (R) using the following equation (38): where Kd is the dissociation constant of indo-1 (250 nM), Sf2 is fluorescence intensity at 485 nm without Ca2+, and Sb2 is fluorescence intensity with saturating Ca2+. Rmax is R in the absence of Ca2+, and Rmin is R in saturating Ca2+. Rmax and Rmin values were obtained using a method similar to that of Ganitkevich and Isenberg (12) in which we used indo-1-loaded UBSMCs under a whole cell patch clamp. At the end of some experiments, the cell membrane was made leaky by stepping the holding potential to −300 mV. After ∼2–3 min, the holding potential was set to 0 mV and then stepped to −300 mV for 200 ms at 500-ms intervals to maintain the myocyte membrane in a leaky state. Rmax and Rmin were obtained at a holding potential of 0 mV in 2.2 mM Ca2+-containing solution and in Ca2+-free solution containing 10 mM EGTA, respectively.
Measurement of cell volume and estimation of increase in [Ca2+]i from Ca2+ influx.
When [Ca2+]i and ICa were recorded simultaneously, estimated increases in [Ca2+]i were calculated using following equation (14, 36): where ∫− ICadt is total charge entry, F is the Faraday constant, and V is cell volume. ICadt was obtained by integration of the area under the ICa trace. V was calculated by assuming that the shape of UBSMCs consisted of two cones connected base to base. The V value calculated from length and width of resting UBSMCs was 12.8 ± 1.4 pl (n = 25).
Measurement of contraction from tissue preparation.
After the UB was dissected and freed from the vesicle trigon, the ventral wall was opened longitudinally in Ca2+-free Krebs solution. The mucosal layer was then removed, and tissue segments 3–4 mm long and 0.8–1.2 mm wide were prepared. One end of the tissue segment was pinned to a rubber plate at the bottom of the organ bath (∼5 ml), and the other end was connected to a force transducer, which was domestically made (32). Segments were equilibrated at a resting load of ∼1 mN. To elicit contractions, a tissue segment was stimulated for 2 s using a train of 3-ms pulses of 300 mA at 5 Hz in Krebs solution that contained the following neurotransmitter antagonists (in μM): 1 atropine, 1 phentolamine, 1 propranolol, 1 TTX, and 10 suramin. Contractile responses were recorded using a strain gauge transducer on an ink-writing direct current servorecorder (Toa Electronics, Kobe, Japan). All of the experiments were performed at 36 ± 1°C. Measurement of twitch contractions was performed just before the addition of ryanodine and at the end of ryanodine treatment. The increase in resting tone induced by ryanodine was measured at the peak of the tone.
Pooled data are means ± SE. Statistical significance between two or multiple groups was determined using Student's t-test or Scheffé's test after one-way ANOVA, respectively. P < 0.05 and 0.01 indicate statistical significance.
Most pharmacological reagents were obtained from Sigma (St. Louis, MO). Ryanodine, xestospongin C, TEA-Cl−, 4-AP, and CdCl2 were purchased from Wako (Osaka, Japan), and EGTA, HEPES, and indo-1 were obtained from Dojin (Kumamoto, Japan). Fluo-4 was purchased from Molecular Probes (Eugene, OR).
Effects of ryanodine on Ca2+ images during action potentials in UBSMCs.
Ca2+ images obtained during and immediately after an action potential were obtained from single UBSMCs of the mouse using a fast laser-scanning confocal microscope in current-clamp mode (Fig. 1Aa). A myocyte was loaded with 100 μM fluo-4, which was added to a pipette filling solution. A 50-ms, 10-pA current injection elicited an action potential with an amplitude of 62.8 mV from a resting potential of −44.5 mV. An afterhyperpolatrization of 10.7 mV was also observed (Fig. 1Ab). The averaged resting membrane potential, action potential overshoot, and afterhyperpolarization values in UBSMCs were −50.1 ± 1.3, 15.0 ± 2.0, and 10.1 ± 1.0 mV (n = 9), respectively. The increase in [Ca2+]i initially occurred at approximately the same time as the peak action potential ∼40 ms after the start of current injection. This change in [Ca2+]i occurred exclusively at localized spots (termed Ca2+ “hot spots”) within the cell (e.g., α and β in Fig. 1A). The averaged fluorescence intensity in Ca2+ hot spots (2 μm in diameter) and in the whole cell area were measured in each frame at an interval of 8.25 ms. The resulting changes in the ratio of F to F measured before depolarization (ΔF/F0) were plotted against time. The ΔF/F0 ratio in Ca2+ hot spots reached a peak at ∼100–150 ms (Fig. 1B) and then progressively declined. The rise of [Ca2+]i spread slowly from these hot spots to other areas, and the ΔF/F0 ratio in the whole cell area reached peak values at ∼300 ms. The locations of initial Ca2+ hot spots were exactly the same in each cell when action potentials were elicited repetitively, even at intervals as long as 3–5 min (data not shown). Essentially identical images of Ca2+ dynamics, which started in some discrete local sites in superficial areas as Ca2+ hot spots and then spread slowly into the entire myoplasm, were obtained in response to action potentials in all cells in which Ca2+ images and action potentials were measured simultaneously (n = 4). After an action potential, cell length was reduced significantly beginning at ∼500 ms. Peak shortening (up to ∼20%) was recorded at ∼2–3 s, and relaxation developed within ∼10 s (data not shown).
The effects of 100 μM ryanodine on this pattern of changes in [Ca2+]i during an action potential were examined (Fig. 1B). Because ryanodine causes RyRs to remain open at low concentrations and simply blocks them at high concentrations (46), we applied 100 μM ryanodine together with fluo-4 from the recording pipettes to ensure complete blockage of RyRs. The resting membrane potential in cells loaded with ryanodine was depolarized by 8–20 mV compared with control (−30.3 ± 3.5 mV, n = 3; P < 0.01 vs. control). F0 at rest tended to be higher in ryanodine-loaded cells (33.6 ± 5.5; n = 3) than in control cells (19.5 ± 0.8, n = 4) (P > 0.05). Action potentials were elicited by current injection in a ryanodine-loaded cell (Fig. 1B) in a manner similar to that of the control recordings. The amplitude of action potential overshoot and afterhyperpolarization were 18.5 ± 1.9 mV (n = 3; P > 0.05 vs. control) and 7.2 ± 2.5 mV (P > 0.05 vs. control), respectively. During an action potential, Ca2+ hot spots were never detected and the increase in global cell area ΔF/F0 ratio during action potential was much smaller in ryanodine-loaded cells. Cell shortening upon an action potential was not observed in ryanodine-loaded cells (n = 3).
Image analysis of Ca2+ mobilization during depolarization in UBSMCs.
Figure 2Aa shows fluorescent images obtained every 8.25 ms from a cell that was depolarized under voltage-clamp conditions to 0 mV from a holding potential of −60 mV for preselected durations ranging from 5 to 30 ms. When the duration of the depolarization pulse was 5 ms, local Ca2+ increase occurred in only one spot (α) in subsarcolemmal areas. This [Ca2+]i change did not spread to other areas as a Ca2+ spark (Fig. 2A). An increase in clamp pulse duration to 10 ms elicited an additional spot β. When the cell was depolarized for 30 ms, the number of Ca2+ spots increased to about five in a single confocal plane. In addition, the Ca2+ spots spread to other areas and also increased global [Ca2+]i. This induced a contraction that started ∼500 ms after repolarization (data not shown). It is notable that the Ca2+ increase did not occur uniformly along the sarcolemma but did appear in the same Ca2+ hot spot sites, such as α and β, with repetitive application of depolarization of different durations. Figure 2B shows changes in F/F0 in hot spots (α and β) and global area corresponding to those measured from the images in Fig. 2A. At the duration of 5 ms, the rise in F/F0 was observed only in hot spot α but not in β and global area. The ΔF/F0 ratio in global area was increased as the duration was lengthened to >10 ms. The time courses of global ΔF/F0 ratio were replotted and superimposed in Fig. 2C for a longer time. The marked increase in ΔF/F0 ratio occurred in an all-or-none manner as the depolarization duration was changed. The threshold duration for the switching from local to global Ca2+ events appeared to be between 5 and 10 ms in this particular cell. In addition, the time to peak global ΔF/F0 ratio also depended on the duration of depolarization.
The corresponding membrane current traces and the time course of ΔF/F0 ratios of the global area and hot spots from the same cell are shown in Fig. 2B. The whole cell capacitance of mouse UBSMCs was 64.6 ± 3.4 pF (n = 37). ICa was observed at 5 ms, but only a small increase in global ΔF/F0 ratio was detected. Marked increases in global ΔF/F0 ratio were consistently observed at clamp durations of 10 ms or longer. The activation of a large outward current and the pronounced increase in global ΔF/F0 ratio were recorded when the clamp depolarization lasted 30 ms or longer.
When the depolarization duration was >10 ms, most of the Ca2+ hot spots spread to other areas in the myoplasm as described above. Fig. 2Da shows the profile of the ΔF/F0 ratio along a section (x-y in Fig. 2A; 30 ms) crossing a hot spot (β). The elevation of ΔF/F0 ratio started in hot spot β just beneath the cell membrane and spread inside the cell as a Ca2+ wave. The time courses of Ca2+ waves at points in hot spot β (0 μm from cell edge) and 2.1, 4.3, and 6.4 μm inside the cell along the section (x-y) are shown in Fig. 2Db. Note that the Ca2+ wave spread and increased the ΔF/F0 ratio even after repolarization. These findings suggest that Ca2+ influx is not required for Ca2+ wave propagation. Moreover, the Ca2+ wave is unlikely to be simply the diffusion of Ca2+ from hot spot sites; it must include a Ca2+ release chain reaction across the myocyte, because [Ca2+]i at the top of the Ca2+ wave increased dramatically during propagation and was greater than [Ca2+]i in the original hot spot.
Figure 3 shows data from five cells demonstrating the relationship between the duration of depolarization and the peak increase in global ΔF/F0 ratio after clamp depolarization. The peak of global ΔF/F0 ratio induced by a 30-ms pulse was taken as unity in each cell, whereas values in response to shorter depolarizations were considered relative ΔF/F0 ratios (Fig. 3Aa). Although not all cells responded in an all-or-none manner, the global ΔF/F0 ratio was always close to the maximum value when the depolarization duration was >10 ms. The number of Ca2+ hot spots observed in a single confocal plane also increased with progressive lengthening of the duration of depolarization. When three or more Ca2+ hot spots appeared in a single confocal plane adjusted to near the middle of the cell's height, Ca2+ hot spots spread to other areas as waves (Fig. 3, Aa and Ab). If only one or two Ca2+ hot spots occurred, the hot spots often disappeared within ∼80 ms as shown for Ca2+ sparks in Fig. 3B. The increase in ΔF/F0 ratio after depolarization was totally inhibited when ICa through VDCCs was blocked by the addition of 100 μM Cd2+ (data not shown).
Effects of ryanodine on Ca2+ hot spots.
The effects of 100 μM ryanodine on Ca2+ images during depolarization for 30 ms from −60 to 0 mV were examined in the next series of experiments (Fig. 4). Because ryanodine stabilizes RyRs in a subopen state at low concentrations and blocks them at high concentrations (46), we applied 100 μM ryanodine from recording pipettes to block RyRs completely. The presence of 100 μM ryanodine in the pipette solution did not significantly affect the inward current through VDCCs (Fig. 4B) but markedly reduced the elevation of the ΔF/F0 ratio during depolarization (Fig. 4, A and B). Ca2+ hot spots were not observed in the presence of ryanodine (Fig. 4Ba). The three-dimensional images (Fig. 4C) and their profile analysis (Fig. 4D) clearly indicated that the rise of fluorescence intensity in the presence of ryanodine occurred almost uniformly along the cell membrane, whereas the rise was much smaller in Ca2+ hot spots in the control cell.
The changes in ΔF/F0 ratio from the whole cell area upon depolarization for 30 ms in the absence (control) and presence of 100 μM ryanodine are shown in a superimposed image in Fig. 5A, which represents relatively long time scale recordings. In addition to a large decrease in the global ΔF/F0 ratio, the time to peak global ΔF/F0 ratio also changed in the presence of 100 μM ryanodine (162.4 ± 24.8 ms and 46.7 ± 3.3 ms; n = 7 and n = 5, respectively; P < 0.01). It is particularly notable that the global ΔF/F0 ratio in the presence of ryanodine started to decline just after repolarization. In contrast, the global ΔF/F0 ratio in the control cells further increased and reached its peak ∼200 ms after repolarization. The relationship between the duration of depolarization and peak global ΔF/F0 ratio was examined in the presence of 100 μM ryanodine by changing the duration of depolarization in the manner shown in Fig. 2 (see also Fig. 5B). When the duration of depolarization was increased from 5 to 50 ms, the ΔF/F0 ratio was increased in a sigmoid fashion in the control cells. The global ΔF/F0 ratio in the presence of ryanodine also was increased by lengthening the duration of depolarization, but the increase was much smaller. The application of 100 μM ryanodine reduced the relative intensity to ∼20% of the control value at 50 ms. When the duration was increased to 150 ms, the ΔF/F0 ratio in the presence of ryanodine increased markedly because of the sustained influx of Ca2+ through VDCCs.
Although the results obtained using fluo-4 as a Ca2+ indicator allowed us to analyze fast changes in Ca2+ dynamics such as Ca2+ hot spots, fluo-4 cannot be used to assess changes in absolute [Ca2+]i. It is expected that resting [Ca2+]i may be changed by the application of 100 μM ryanodine, and this could significantly affect the ΔF/F0 ratio after depolarization. To evaluate the absolute [Ca2+]i, we performed similar experiments using indo-1. Ca2+ images were obtained every 66 ms. The resting [Ca2+]i level in the presence of 100 μM ryanodine tended to be slightly higher than that in control cells, but this difference was not significant (146.5 ± 21.6 and 157.1 ± 17.1 nM, n = 5 and n = 6, respectively; P > 0.05). The peak global [Ca2+]i levels after depolarization from −60 to 0 mV for 30 ms were 394.6 ± 44.4 and 249.2 ± 15.2 nM in the absence and presence of ryanodine, respectively (n = 5 and n = 6; P < 0.05) (data not shown). In the presence of 100 μM ryanodine, the increase in [Ca2+]i level induced by depolarization was reduced to 37% of control (Δ[Ca2+]i, 248.2 ± 4.25 and 92.1 ± 10.9 nM, n = 5 and 6; P < 0.05).
Effects of 100 μM ryanodine on VDCCs also were carefully examined in UBSMCs. When cells were activated by depolarization from −60 mV to positive potentials in 10-mV steps, the maximum amplitude of peak inward currents was obtained at 0 mV in both the absence and presence of ryanodine in the pipette solution (data not shown). The inward current induced by depolarization to 0 mV was blocked by 100 μM Cd2+ or 100 nM nicardipine, indicating that the current was almost completely through L-type VDCCs. The densities of peak ICa at 0 mV were 7.8 ± 1.3 pA·pF−1 (n = 7) and 5.4 ± 1.0 pA·pF−1 (n = 5; P > 0.05 vs. control) in the absence and presence of ryanodine in the pipette solution, respectively (Fig. 5C). The addition of 100 μM ryanodine to the bathing solution did not significantly change the peak amplitude of ICa at 0 mV (8.8 ± 1.0 pA·pF−1 in control cells and 7.9 ± 1.0 pA·pF−1 in the presence of ryanodine, n = 4 for each; P > 0.05 vs. control). These findings show that the marked decrease in the rise of [Ca2+]i level upon depolarization by 100 μM ryanodine was not due to the decrease in Ca2+ influx through VDCC.
When 100 μM ryanodine was added to the pipette solution, outward currents as well as global [Ca2+]i level were substantially reduced (Fig. 6A). The current density of peak outward current of 150-ms depolarization was reduced from 15.4 ± 4.4 to 2.9 ± 0.4 pA·pF−1 (n = 5 and n = 6; P < 0.05) (Fig. 6B). The remaining outward current in the presence of ryanodine was not significantly affected by the addition of 100 nM iberiotoxin or 1 μM paxillin, whereas the current in the absence of ryanodine was markedly reduced by these specific BK channel blockers. These observations indicate that the outward current component reduced by ryanodine was due mainly to BK channel activity. Consistent with this finding, the STOCs induced by activation of BK channels by spontaneous Ca2+ release through RyRs located in the superficial SR were observed at a holding potential of −30 mV in the control but not in cells loaded with 100 μM ryanodine in the pipette solution (Fig. 6C).
Possible contribution of Ca2+ release from IP3-sensitive store sites to depolarization-induced Ca2+ wave.
The data shown in Figs. 2–6 demonstrate that Ca2+ waves from Ca2+ hot spots were essential for the increase in global [Ca2+]i level to induce a contraction in E-C coupling in UBSMCs. In some types of SMC, Ca2+ release from intracellular Ca2+ store sites, which are sensitive to ryanodine, appear to be independent from those sensitive to inositol 1,4,5-trisphosphate (IP3) (10). On the other hand, an interaction or functional overlap between stores sensitive to ryanodine and stores sensitive to IP3 also has been reported in some types of SMC. Moreover, it has been suggested that Ca2+ release initially mediated by RyRs may be amplified by Ca2+ release from IP3 receptors (IP3Rs) (15, 16, 48). Therefore, the possibility that the spread of the Ca2+ wave from hot spots in UBSMCs involves IP3-induced Ca2+ release (IICR) was examined using ACh and xestospongin C to induce IICR and block IP3R, respectively.
Figure 7 shows the responses to 10 μM ACh after the stimulation by voltage-clamp depolarization was repeated twice in the absence or presence of 100 μM ryanodine in the pipette solution. The ΔF/F0 ratio was elevated approximately twofold by 30-ms depolarization from −60 to 0 mV at the interval of 100 s, and the ΔF/F0 ratio was significantly smaller in the presence of ryanodine than in its absence as shown in Fig. 5. ACh (10 μM) was applied in the presence of 5 mM Ni2+ in the bathing solution to prevent Ca2+ influx though both VDCCs and receptor-operated Ca2+ channels. ACh elicited a large increase in the ΔF/F0 ratio in both control and ryanodine-loaded cells, but this increase was significantly smaller in ryanodine-loaded cells at a holding potential of −60 mV. These results clearly indicate that a large part of IP3-sensitive Ca2+ store sites remains intact in cells loaded with 100 μM ryanodine. These results also suggest that the Ca2+ rise via IICR during the response to ACh may in addition activate RyRs to induce CICR, at least in part. Alternatively, ryanodine at the high concentration of 100 μM may nonselectively interact with IP3Rs and slightly suppresses IICRs.
In the next series of experiments, xestospongin C was used as a blocker of IP3Rs. Initially, the efficacy of xestospongin C to block IICRs induced by 1 μM ACh was examined (Fig. 8). The Ca2+ indicator indo-1 and 3 μM xestospongin C were applied from the recording pipette. The resting [Ca2+]i level at a holding potential of −60 mV was not significantly affected by the presence of 3 μM xestospongin C. Application of 1 μM ACh markedly increased [Ca2+]i in a biphasic manner, with a large transient rise in [Ca2+]i level as well as a smaller, sustained one. The transient rise of [Ca2+]i was significantly reduced by 39% in the presence of xestospongin C (control, 620.5 ± 123.1 nM; xestospongin C, 242.2 ± 32.3 nM, n = 5 for each) (P < 0.05). On the other hand, neither the sustained rise of [Ca2+]i (381.7 ± 70.1 and 316.5 ± 82.7 nM, n = 5 for each; P > 0.05) nor the sustained inward current (33.1 ± 3.1 pA, n = 5, 29.6 ± 6.2 pA, n = 5; P > 0.05) was significantly affected by xestospongin C. These results indicate that the application of 3 μM xestospongin C from patch pipettes blocked IICR but not Ca2+ influx through receptor-operated cation channels.
The effects of xestospongin C on CICR upon depolarization and after Ca2+ waves were examined next. The rise of [Ca2+]i after depolarization from −60 to 0 mV for 30 ms was measured using fluo-4 in the absence or presence of 3 μM xestospongin C in the pipette solution. The Ca2+ hot spots were elicited by depolarization in the presence of 3 μM xestospongin C in a manner similar to those observed in the control cells (Fig. 9A). K+ currents were blocked by 30 mM TEA and 1 mM 4-AP in the bathing solution. The amplitude of inward currents at 0 mV was not significantly affected by xestospongin C (8.6 ± 0.9 and 10.7 ± 2.0 pA·pF−1, n = 5 and n = 6, respectively; P > 0.05).
To examine whether xestospongin C affects the Ca2+ wave after depolarization, profile analysis of these Ca2+ images was performed along the x-y sections shown in Fig. 9A. Ca2+ waves spread from superficial Ca2+ hot spot sites to inside the myocytes, regardless of the presence of xestospongin C (Fig. 9B, top). The time course of Ca2+ waves at points in hot spot locations (0 μm) 2.9 and 5.8 μm inside along the sections (x-y) are shown in Fig. 9B, bottom. Ca2+ wave velocity was calculated from the time required for the ΔF/F0 ratio at points 2.9 and 5.8 μm inside the cell edge to reach 50% of maximum ΔF/F0 ratio at 0 μm inside the cell. The velocity of Ca2+ wave propagation in cells loaded with vehicle or xestospongin C was 127.3 ± 19.6 and 125.5 ± 25.7 μm·s−1 (n = 4 for each; P > 0.05), respectively, and was not affected by xestospongin C. The ΔF/F0 values in the hot spots and the whole cell area were not affected by xestospongin C (Fig. 9C). Similar experiments were performed using indo-1 as a Ca2+ indicator. The [Ca2+]i values after depolarization lasting 30 ms was measured at the peak and those measured 1.0 s after depolarization were not affected by xestospongin C (Fig. 9D).
Ca2+ influx during depolarization and after Ca2+ buffering by store sites.
The amount of Ca2+ influx through VDCC in response to depolarization from −60 to 0 mV for 30 ms can be calculated from the recordings of inward VDCC currents (∼20 pC) (Tables 1 and 2). The estimated increase in [Ca2+]i by the Ca2+ influx was ∼9 μM. This increase is 26.6 ± 4.5 times larger than that measured using indo-1 if intracellular Ca2+ buffering action, Ca2+ uptake by Ca2+-ATPase in the SR, and Ca2+ extrusion by the Na+/Ca2+ exchanger Ca2+ pump in the plasma membrane are ignored. A much larger dissociation between Ca2+ influx and the rise of real [Ca2+]i was observed when cells were loaded with 100 μM ryanodine (60.0 ± 6.3 times; P < 0.05 vs. control). The loading of cells with 3 μM xestospongin C from the pipette solution did not significantly affect the ratio of estimated and measured [Ca2+]i in the range of pulse durations from 30 to 150 ms (Table 2).
To inhibit the Ca2+ pump in the SR, 10 μM cyclopiazonic acid (CPA) was added to the bathing solution in the presence of 100 μM ryanodine in the pipette solution. The application of CPA resulted in a rise of the resting [Ca2+]i level from 84.9 ± 9.9 to 305.6 ± 31.7 nM (n = 4 for each) (Fig. 10, A and B). The increase in [Ca2+]i level in response to the first 30-ms depolarization pulse was similar in ryanodine- and ryanodine + CPA-treated cells (165.5 ± 14.8 and 167.6 ± 6.3 nM, respectively, n = 4 for each; P > 0.05). In the presence of CPA, depolarization of a myocyte to 0 mV for 30 ms elicited ICa, which was significantly smaller than that in the absence of CPA (10.7 ± 0.9 and 6.9 ± 0.4 pA·pF−1, n = 4 for each; P < 0.05) (Fig. 10C). The relative value of Δ[Ca2+]i versus the charge carried by Ca2+ through VDCC was significantly larger in the presence of CPA than in control, however, because ICa was markedly suppressed in the presence of CPA (Fig. 10D). This finding indicates that Ca2+ uptake by Ca2+-ATPase in the SR after a [Ca2+]i rise upon depolarization is an important factor that needs to be part of the explanation of the discrepancy between the amount of Ca2+ influx and directly measured [Ca2+]i. However, this factor can explain the discrepancy only in part.
Effects of ryanodine on contraction.
The effects of ryanodine on twitch contractions induced by either direct electrical stimulation or spontaneous contractions were examined using UB tissue segments (see materials and methods). Twitch contractions were induced by a train of 10 square pulses (3-ms duration and 200-ms interval) applied every 90 s in the presence of neurotransmitter antagonists (as described in materials and methods). When 10 μM ryanodine was applied, the resting tension increased substantially and the amplitude of twitch contraction initially increased slightly for ∼10 min and then gradually decreased to a level slightly lower than that before the application of ryanodine (Fig. 11Aa). Application of 30 μM ryanodine markedly increased the resting tension and initially enhanced twitch contraction significantly for ∼10 min and then declined to ∼30% of that in the absence of ryanodine (Fig. 11Ab). The increases in resting tension and twitch contraction were not significant when 100 μM ryanodine was applied, and the twitch contraction was strikingly suppressed by 90% (Fig. 11Ac). Figure 11Ba summarizes the relative amplitude of twitch contraction after the effect of ryanodine had reached a steady state, and Fig. 11Bb shows the maximum increase in resting tension induced by ryanodine. These results suggest that 10 and 30 μM ryanodine caused RyRs to be activated (in half-open state) and only partly blocked RyRs, whereas 100 μM ryanodine exhibited blocking effects as expected on the basis of previous work.
Because spontaneous rhythmic contractions were small or were not clearly detected in mouse UB segments [as pointed out previously by Petkov et al. (45)], they were elicited by elevating external K+ concentration to 15 mM in the presence of selected neurotransmitter antagonists (Fig. 11C). Application of 100 μM ryanodine increased tone slightly and blocked rhythmic activity. Similar observations were made in all of the preparations examined (n = 4).
In the present study, we have demonstrated that during E-C coupling after action potential generation in UBSMCs, CICR initially occurred quickly in localized hot spot sites that corresponded to functional coupling between VDCCs and RyRs. Thereafter a secondary spread of Ca2+ waves to other Ca2+ store sites occurred much more slowly. The amount of Ca2+ influx during short depolarization was calculated to be large enough to increase [Ca2+]i by several micromolar levels. In vivo, however, Ca2+ that entered a myocyte was strongly buffered and global [Ca2+]i was increased only slightly. Thus the Ca2+ influx initiated CICR mainly in discrete hot spot sites from which Ca2+ waves spread to the entire myoplasm in the second step of Ca2+ release. The second depolarization step increased global [Ca2+]i and induced a contraction.
Ryanodine suppressed Ca2+ hot spots, Ca2+ waves, and contraction.
In mouse UBSMCs, short depolarization elicited Ca2+ hot spots at discrete local sites in subsarcolemmal areas. These hot spots developed 10 ms after the start of depolarization as previously reported in SMCs of the guinea pig vas deferens and UB (31, 42). The occurrence of Ca2+ hot spots required Ca2+ influx through VDCC. These hot spots were usually detected as frequently discharging sites of spontaneous Ca2+ sparks (42). In this study, we found the number of Ca2+ hot spots elicited by depolarization to be closely related to the duration of the voltage-clamp depolarization up to ∼20 ms. This finding presumably reflects the principle that the coupling efficacy between Ca2+ influx through VDCCs with the activation of RyRs required to induce CICR may vary widely between discrete Ca2+ hot spot sites. The loose coupling compared with that in cardiac myocytes has been suggested in rabbit UBSMCs (7). A short depolarization (5 ms) elicited only a few hot spots in well-defined, discrete sites, presumably indicating tight coupling. Larger Ca2+ influx induced by longer depolarization may activate discrete sites with weaker coupling efficacy. Once a sufficient number of Ca2+ hot spots occurred during a certain period of depolarization, Ca2+ waves developed progressively and spread slowly from hot spots through the entire myoplasm. This phenomenon continued even after repolarization and elevated [Ca2+]i to the extent that contraction was induced.
To our knowledge, this study represents the first reported demonstration of a threshold in the duration (∼10 ms) of depolarization required to evoke local Ca2+ sparks or change hot spots to a global [Ca2+]i increase in SMCs. Ca2+ hot spots function as the initiation sites of Ca2+ waves essential for global [Ca2+]i increases in SMCs. Although the mechanism by which transient Ca2+ sparks switch to Ca2+ hot spots and/or waves is not clear, the coalescence of neighboring hot spots may underlie this switching to global [Ca2+]i increases. Accordingly, the number of evoked Ca2+ sparks or hot spots by a given depolarization may be determining factor with regard to whether a contraction is elicited by the depolarization in UBSMCs. It is noteworthy that the spread of Ca2+ waves did not absolutely require Ca2+ influx through VDCCs because the wave spread up to ∼200 ms after the cessation of 30-ms depolarization even though the Ca2+ tail current lasted for ∼20 ms.
In this study, application of 100 μM ryanodine from the recording pipette abolished both Ca2+ hot spots and Ca2+ waves but did not affect ICa. Although the application of 10 μM ryanodine from the pipette may reduce CICR (13), this concentration of ryanodine increases the resting [Ca2+]i in guinea pig UB tissues (20), presumably by locking RyRs in a half-open state (46). On the other hand, a high concentration of ryanodine blocks RyRs completely (46). In the present study, we used 100 μM ryanodine to block RyRs completely. This blocking occurred almost immediately after the start of whole cell recording. Consequently, we did not observe a substantial increase in resting [Ca2+]i. Neither STOCs nor Ca2+ sparks were observed in UBSMCs loaded with 100 μM ryanodine, confirming that RyRs were blocked completely. In these myocytes, the increase in [Ca2+]i occurred slowly but substantially during long depolarization (150 ms). In summary, CICR was almost completely blocked by 100 μM ryanodine in UBSMs and the Ca2+ influx through VDCCs per se increased global [Ca2+]i only slightly during 30-ms depolarization.
Experimental results from studies that have addressed the susceptibility to contraction in UBSMCs treated with ryanodine are varied and somewhat controversial. Three groups have reported that spontaneous contractions in guinea pig and rat UBSM strips were not reduced, but rather enhanced, by ryanodine treatment (18, 23, 41). However, another group reported effective blockage of spontaneous contraction by ryanodine to ∼50% of control in UBSMCs of the pig (5). The [Ca2+]i rise induced by spontaneous action potentials in tissue preparations of the guinea pig was reduced to ∼60% of control by application of 10 μM ryanodine (20), and 10 μM ryanodine had no significant effect on the [Ca2+]i rise elicited by field stimulation in UBSMs of the rat (22). Suppression of CICR by 10 μM ryanodine, however, has been shown convincingly in single UBSMCs of the guinea pig (13). In the present study, the contraction induced by direct electrical stimulation was markedly reduced by ryanodine, and this inhibition was dose dependent in UBSMs of the mouse. The resting tone was significantly increased by ryanodine at 10 and 30 μM but not at 100 μM, consistent with the changes in resting [Ca2+]i levels induced by 10–100 μM ryanodine. The ICa in UBSMCs was not affected significantly by the addition of 100 μM ryanodine to the superfusing solution. Unlike guinea pig UBSMCs, large, spontaneous contractions were seldom observed in mouse UBSM tissue (45). Our results (see Fig. 11) also demonstrate that the spontaneous contraction observed in 15 mM K+ solution was susceptible to 100 μM ryanodine as well as the contraction elicited by direct electrical stimulation.
Although the reasons for the discrepancies concerning the susceptibility of contraction to ryanodine in previous studies are not completely clear, the contribution of RyRs and CICRs to E-C coupling in UBSMCs could be different between species. Moreover, the susceptibility of contraction to ryanodine was higher when conditions for direct electrical stimulation were moderate or weak (Morimura K, unpublished observations). It is plausible that the relative contribution of CICR versus that of Ca2+ influx to increase global [Ca2+]i during E-C coupling becomes smaller under conditions in which Ca2+ influx is larger. This hypothesis is consistent with the finding that the elevation in global [Ca2+]i by depolarization was large even in UBSMCs loaded with 100 μM ryanodine and when the duration of depolarization was long (150 ms).
Ca2+ buffering for Ca2+ influx through VDCCs is still mysterious.
We observed a clear discrepancy between the amount of Ca2+ influx during short depolarization that mimicked an action potential and its direct contribution to the rise in global [Ca2+]i. As shown in Table 1, the estimate of [Ca2+]i increase on the basis of the amount of Ca2+ influx during depolarization for 30 ms from −60 to 0 mV was >25 times larger (so-called buffering power) than the peak [Ca2+]i measured directly using of indo-1. A similar discrepancy between estimated and measured [Ca2+]i values has been discussed in the following other SMC models: rat femoral artery buffering power, 130 times (35); guinea pig coronary artery, 150 times (14); rat portal vein, 114 times (36); equine airway, 15–50 times (9); and guinea pig UB, 46 times (11). When CICR triggered by depolarization was blocked by 100 μM ryanodine in this study, this buffering power was apparently much larger: 60 times. Such a large discrepancy indicates mechanisms of fast, strong buffering power for Ca2+ coming into the cytosol through VDCCs.
In many types of cells that express KCa channels, measurement of KCa channel current under whole cell clamp mode is a reliable monitor of [Ca2+]i levels in the subsarcolemmal regions. In UBSMCs from the mouse as well as in those from the guinea pig, the major component of outward current upon depolarization was K+ current though BK channels. The activation of BK channels is responsible for repolarization and afterhyperpolarization of an action potential (21, 28, 30). The ill-defined buffering mechanism for Ca2+ entering through VDCCs is functional in superficial areas just beneath the cell membrane in UBSMCs, because BK channel current activated by depolarization also was strongly suppressed in ryanodine-loaded cells in this study. In inside-out patch-clamp mode, the single-channel activity of BK channels in UBSMCs was not affected by 100 μM ryanodine (Morimura K et al., unpublished observations). Taken together, these findings show that any substantial Ca2+ entering though VDCCs does not elevate [Ca2+]i markedly and does not activate BK channels effectively, presumably because of strong and fast buffering of Ca2+ by uptake, binding, and extrusion from superficial areas. The results of the present study confirm that CICR through RyRs in the SR is required for the activation of BK channels as well as for the activation of the contractile apparatus during E-C coupling in UBSMCs.
The mechanisms by which Ca2+ that enters through VDCCs during an action potential is strongly buffered are not known in detail. It is likely that Ca2+ uptake to the SR by the Ca2+ pump (i.e., sarcoplasmic reticulum Ca2+-ATPase) is the largest component of this buffering. The addition of 10 μM CPA to the ryanodine-loaded cells resulted in a significant decrease in the buffering power from 60 to 34 (Table 1). However, this decrease in buffering power by the addition of CPA was not large enough to explain completely the difference between the measured and calculated [Ca2+]i levels after clamp depolarizations. Alternative or additive mechanisms for fast, strong buffering of entering Ca2+ may be Ca2+ extrusion by the Na+/Ca2+ exchanger and the Ca2+ pump in the plasma membrane, Ca2+ uptake by mitochondria, and/or trapping by cytosolic Ca2+-binding proteins. Further experiments are required to clarify the mechanisms underlying the mysteriously fast and strong buffering of the Ca2+ that enters through VDCCs during an action potential.
IP3 formation was not involved in Ca2+ waves from hot spots during E-C coupling.
The spread of Ca2+ waves from Ca2+ hot spots across the myoplasm is essential for the second step of [Ca2+]i increase in E-C coupling. This Ca2+ wave cannot be evaluated on the basis of simple diffusion of Ca2+ from hot spot sites to other areas, because Ca2+ release occurs progressively near the wave front. Our profile analysis (Fig. 4C) indicates that the ΔF/F0 ratio in the central area of the myocyte was increased by the Ca2+ wave to a higher level than that in the original hot spot site. This pattern of results was observed in all myocytes examined in the present study. Therefore, the Ca2+ wave spreads by chain reactions of Ca2+ release from hot spots to separate Ca2+ stores within a cell.
Ca2+ waves have been observed in many types of SMCs and analyzed in detail. In many cases, Ca2+ waves are mediated by Ca2+ release from IP3Rs (i.e., IICRs) after IP3 formation by agonist stimulation (26, 39). Cross talk between CICRs via RyRs and IICRs via IP3Rs has been reported in various types of cells, including SMCs (15). It has been reported that [Ca2+]i waves due to noradrenaline involve CICRs in addition to IICRs in the rat portal vein (2). The Ca2+-dependent activity of some types of PLC and IP3R activity shows bell-shaped dependence on [Ca2+]i and thus can exhibit positive feedback for IICR facilitation (25, 27). In a cultured neuron derived from the central nervous system, it also has been shown that Ca2+ influx through VDCCs induces IICRs, presumably via the formation of IP3 (43). In the present study, xestospongin C markedly reduced IICRs after the application of ACh but did not affect CICRs triggered by depolarization. The loading of UBSMCs with 100 μM ryanodine reduced IP3-sensitive Ca2+ stores by 40% but did not deplete it. These results strongly suggest the presence of separate stores for IICR and CICR, with some overlap (47). The mechanisms underlying Ca2+ waves from Ca2+ hot spots during E-C coupling remain to be determined but could be due simply to CICR or, alternatively, to the formation of cADP-ribose (8).
In conclusion, Ca2+ release during E-C coupling in UBSMCs occurs in two steps. Ca2+ influx during an action potential does increase [Ca2+]i significantly. It initiates CICRs in discrete hot spot sites via functional coupling between VDCCs and RyRs. In the second step, which also involves Ca2+ release, Ca2+ waves slowly spread to other Ca2+ store sites without mediating IP3R activation. A twitch contraction induced by action potentials triggered by direct electrical stimulation under moderate conditions is highly susceptible to ryanodine treatment. However, Ca2+ dynamics during E-C coupling triggered by transmitter release from autonomic nerve endings are more physiological than those observed in the present study. Notably, under those conditions, IICR is likely to also be involved in E-C coupling in UBSMCs.
This work was supported by a Grant-in-Aid for Scientific Research (B) from Japan Society for the Promotion of Science and also by a Grant-in-Aid for Research on Health Sciences focusing on Drug Innovation from Japan Health Sciences Foundation (to Y. Imaizumi).
We thank Dr. Wayne R. Giles (University of Calgary, Calgary, AB, Canada) for providing data acquisition and analysis programs and also for his reading of this manuscript.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2006 the American Physiological Society