Elementary purinergic Ca²⁺ transients evoked by nerve stimulation in rat urinary bladder smooth muscle

Abstract

The translation of nerve transmission to Ca²⁺ signals in urinary bladder smooth muscle (UBSM) is incompletely understood. Thus, we sought to characterize Ca²⁺ signals in strips of UBSM loaded with the Ca²⁺-sensitive fluorescent dye, fluo-4, using laser scanning confocal microscopy. Two types of Ca²⁺ signals occurred spontaneously and could be evoked with field stimulation: large, rapid, global Ca²⁺ transients termed ‘global Ca²⁺ flashes’, and much smaller, localized Ca²⁺ transients. Global Ca²⁺ flashes were inhibited by the L-type voltage-dependent Ca²⁺ channel (VDCC) inhibitor, diltiazem and with P2X receptor blockade. Simultaneous intracellular recordings and Ca²⁺ measurements indicated that these events are caused by Ca2+ influx through VDCCs during action potentials. Small, local Ca²⁺ transients occurred spontaneously, and their frequency could be elevated with field stimulation. Atropine, an inhibitor of muscarinic receptors, did not affect these local Ca²⁺ transients. However, the desensitizing P2X receptor agonist α,β-methylene ATP, and the purinergic antagonist, suramin, effectively inhibited the local Ca²⁺ transients. The frequency of these ‘purinergic Ca²⁺ transients’ was increased about 7-fold by a 10 s stimulus train (1 Hz). The amplitude, duration at one-half amplitude and the spatial spread of the evoked purinergic Ca²⁺ transients were F/Fo= 2.4 ± 0.13, 111.7 ± 9.3 ms and 14.0 ± 1.0 μm2, respectively. Tetrodotoxin inhibited evoked purinergic Ca²⁺ transients, indicating that they were dependent on nerve fibre activation. Purinergic Ca²⁺ transients were not dependent on VDCC activity. Neither 2-APB, an inhibitor of inositol 1,4,5-triphosphate (Ins(1,4,5)P3) (IP3)-induced Ca²⁺ release, nor ryanodine inhibited the purinergic Ca²⁺ transients. We have identified two novel Ca²⁺ signals in rat UBSM. Large, rapid, global Ca²⁺ flashes that represent Ca²⁺ influx through VDCCs during action potentials, and local, purinergic Ca²⁺ transients that represent Ca²⁺ entry through P2X receptors. Our results indicate that purinergic Ca²⁺ transients evoked by release of ATP from nerve varicosities are elementary signals in the process of nerve-smooth muscle communication.

The function of the urinary bladder is to relax and store urine during filling, and contract forcefully to empty the bladder during micturition. These seemingly disparate roles are achieved through complex interactions between the autonomic nervous system and urinary bladder smooth muscle (UBSM). UBSM in adult rats is well innervated, receiving approximately 16 000 afferent and efferent axons from ganglion neurones (Gabella, 1999). The bundles of nerve fibres branch repeatedly in UBSM and eventually become single fibres containing varicosities. Located within the varicosities are clear vesicles, or a combination of clear vesicles and dense-cored vesicles (Gabella, 1999) containing neurotransmitters, including acetylcholine (ACh) and ATP or a related purine (Hoyle & Burnstock, 1993; Gabella, 1995). Nerve stimulation evokes the release of ACh and ATP (Dowdall et al. 1974; Kasakov & Burnstock, 1982; Theobald & de Groat, 1989; Silinsky & Redman, 1996). ACh binds to muscarinic receptors, stimulating the production of inositol 1,4,5-triphosphate [Ins (1,4,5)P₃] (IP₃) and the release of Ca²⁺ from the sarcoplasmic reticulum. ATP binds to, and opens purinergic receptors allowing the influx of cations, including Ca²⁺ and Na⁺ (Isenberg et al. 1992; Evans et al. 1996).

Local Ca²⁺ signals in smooth muscle, in the form of Ca²⁺ waves, were first observed in electrically stimulated rat tail arteries (Iino et al. 1994). The release of noradrenaline from surrounding nerve fibres induced Ca²⁺ waves through the activation of (IP₃)-mediated Ca²⁺ release. Stationary, local Ca²⁺ signals (Ca²⁺ sparks), caused by local Ca²⁺ release through a cluster of ryanodine receptors (RyRs), were subsequently detected in smooth muscle (Nelson et al. 1995).

Transient increases in Ca²⁺ in smooth muscle have been recorded following the activation of purinerigic receptors. In UBSM, the activation of purinergic receptors induced excitatory junctional potentials (EJPs) and Ca²⁺ transients (Bramich & Brading, 1996; Hashitani et al. 2000). Recently, Ca²⁺ imaging techniques revealed fast Ca²⁺ events in smooth muscle following nerve stimulation. Field stimulation of the vas deferens evoked local, transient Ca²⁺ events in smooth muscle cells (neuroeffector Ca²⁺ events). These events were mediated by postsynaptic smooth muscle P2X receptors activated by ATP released at synaptic varicosities (Brain et al. 2002, 2003). When the nerve terminals and the neuroeffector Ca²⁺ events were both visualized, it was found that the neuroeffector Ca²⁺ events occurred immediately beneath stimulated sympathetic varicosities. Similar, transient Ca²⁺ events evoked with field stimulation were found in small rat mesenteric arteries and were termed ‘junctional Ca²⁺ transients’ (Lamont & Wier, 2002; Lamont et al. 2003). Line scan images showed that junctional Ca²⁺ transients were evoked immediately following a Ca²⁺ increase in adjacent nerve fibres. Pharmacological studies showed that junctional Ca²⁺ transients were mediated through the activation of purinergic receptors (Lamont & Wier, 2002; Lamont et al. 2003).

In the present study, we used a combination of intracellular microelectrode recordings and a laser scanning confocal microscope to image fast Ca²⁺ events in UBSM. Two new Ca²⁺ events whose frequencies are elevated with nerve stimulation, were detected in rat UBSM: (1) large, rapid, global Ca²⁺ flashes that encompass the entire smooth muscle fibre, and are caused by Ca²⁺ influx through L-type voltage-dependent Ca²⁺ channels (VDCCs) during an action potential, and (2) much smaller, localized purinergic Ca²⁺ transients. Although similar in appearance to RyR-mediated Ca²⁺ sparks (Nelson et al. 1995; Herrera et al. 2001; Heppner et al. 2003), these purinergic Ca²⁺ transients are larger and longer than Ca²⁺ sparks and represent a novel elementary Ca²⁺ signal mediated by the activation of P2X receptors. Purinergic Ca²⁺ transients probably play a critical, initial role in nerve-evoked excitation–contraction coupling in UBSM.

Methods

Tissue preparation

Adult male rats were killed by pentobarbital sodium (130 mg (kg body weight)⁻¹), followed by thoracotomy in accordance with the guidelines for the use and care of laboratory animals (NIH publication 85-23, 1985) and as approved by the Institutional Animal Use and Care Committee of the University of Vermont. The urinary bladder was quickly removed and placed in cold Hepes-buffered saline. Strips of UBSM (approximately 0.5 mm thick × 1 mm wide × 5 mm long) were carefully removed from the serosal surface with sharp scissors and pinned to small sylgard blocks. The sylgard blocks with the attached UBSM strips were placed tissue side down, in a chamber specially designed to measure rapid Ca²⁺ responses. To allow fresh physiological saline solution (PSS) to flow under and around the tissue strip, thin spacers (0.2 mm) were placed on either side of the tissue. The tissue was superfused (1–2 ml min⁻¹) with a PSS at 37°C. For the experiments quantifying the small, localized Ca²⁺ transients, sucrose (12%) was added to the superfusing PSS to reduce tissue movement (Foster et al. 1989; Heppner et al. 1997). Sucrose was not added to the PSS in the simultaneous measurement of Ca²⁺ and membrane potential with microelectrodes, or in experiments involving the large, rapid global Ca²⁺ events. Drugs were dissolved in the superfusing salt solution and applied for at least 20 min prior to image acquisition unless otherwise indicated.

Ca²⁺ imaging and analysis

Strips of UBSM were imaged with a laser scanning confocal microscope (OZ; Noran Instruments) attached to a Nikon diaphot microscope. A Nikon ×60 water immersion objective (NA 1.2) was used to visualize the tissue. Images were acquired using an Intervision software package controlled by a Silicon graphics Workstation (O₂). To visualize Ca²⁺ events in UBSM, the tissue was placed in a Hepes solution containing the Ca²⁺ fluorescent dye, fluo-4 AM (10 μm) (Molecular Probes; Eugene, OR, USA) and pluronic acid (2.5 μg ml⁻¹; Molecular Probes). To facilitate loading of the dye, the tissue was kept in the dark for 60 min at 21–23°C followed by a wash in normal Hepes buffer (21–23°C) before placing the tissue in PSS. All experiments were conducted in PSS (37°C). A krypton–argon laser was used to excite fluo-4 and the emitted light was captured at wavelengths >500 nm. For most experiments, images were acquired at 30 images s⁻¹ (every 33.33 ms). The size of the field was 116 × 109 μm (512 × 480 pixels). To measure latency, images were acquired at 240 images s⁻¹, an acquisition rate that yields an interimage interval of 4.13 ms. Image analysis was conducted offline using customized software written in our laboratory (Dr Adrian Bonev). Evoked Ca²⁺ transients were detected by measuring an increase in the fractional fluorescence of Ca²⁺F/F_o= 1.3, that could be distinguished above the background noise. Baseline fluorescence was measured by averaging at least 10 images with no Ca²⁺ transients. Ca²⁺ transients were measured using a 1.5 × 1.5 μm box (7 × 7 pixels) and analysed as previously described (Perez et al. 1999; Jaggar & Nelson, 2000; Wellman et al. 2001) using customized software written in our laboratory.

Field stimulation

The nerves in the strip were excited with platinum electrodes placed in the recording chamber, which were attached to the output of the stimulus isolation unit of a Grass S44 stimulator. Voltages used to evoke Ca²⁺ transients ranged from 30 to 100 V. A frequency of 1 Hz and a pulse duration of 0.2 ms was used throughout this study. Ca²⁺ transients were evoked with a train of stimuli applied for 10 s at a frequency of 1 Hz. This relatively low frequency allowed individual Ca²⁺ events to be identified, and facilitated quantification of these events during the stimulus train. Global Ca²⁺ events were evoked by increasing the stimulating voltage until threshold was reached. The smaller evoked Ca²⁺ transients were visualized by reducing the stimulating voltage just below threshold for the global Ca²⁺ events.

Microelectrode intracellular recording

To record voltage from UBSM cells, strips of tissue were prepared as above for Ca²⁺ imaging. Standard microelectrode techniques were used. Oxygenated PSS was superfused over the tissue strips at 1–2 ml min⁻¹ (37°C). Microelectrodes were pulled on a Flaming/Brown gas puller (Sutter Instruments, Novato, CA, USA) and had resistances of 60–80 MΩ when filled with 2.0 m KCl. The transmembrane potential was acquired using an Axoclamp 2A (Axon Instruments, Union City, CA, USA) and recorded using Axotape software (Axon Instruments). The membrane potential was measured as the difference between the reference electrode in the bath and the intracellular microelectrode.

Experimental design

Each experiment consisted of three to five 15 s recordings from the same field during which a 10 s stimulus train (1 Hz) was applied. One to three files were recorded under control conditions and after drug application. The time interval between each file was at least 5 min. The tissue was incubated for at least 20 min in drug-containing solutions before test stimuli were applied.

Drugs and solutions

UBSM strips were removed from the urinary bladder in Hepes solution with the following composition (mm): NaCl, 134; KCl, 6; MgCl₂, 1; CaCl₂, 2; Hepes, 10; glucose, 10; (pH 7.4). NaOH was used to adjust the pH of Hepes solutions. Evoked Ca²⁺ transients were measured in PSS with the following composition (mm): NaCl, 119; KCl, 4.7; NaHCO₃, 23.8; KH₂PO₄, 1.2; CaCl₂, 1.6; MgCl₂, 1.2, EDTA, 0.023; and glucose, 11.0 (pH 7.4). PSS was continuously bubbled with 95% O₂–5% CO₂. For experiments using nominal Ca²⁺-free solution, Ca²⁺ was omitted from the superfusing PSS. 2-Aminethoxydiphenyl-borate (2-APB) was obtained from Tocris (Ellisville, MO, USA), ryanodine was obtained from LC laboratories (Woburn, MA, USA) and diltiazem, α,β-methylene ATP, suramin, atropine, and tetrodotoxin (TTX) were obtained from Sigma (St Louis, MO, USA).

Statistics

Data were normalized to control, and expressed as mean ±s.e.m. At least three paired experiments were conducted for each paradigm using at least three rats. n refers to the number of evoked Ca²⁺ transients or the number of preparations as indicated in the text. To determine significance, Ca²⁺ events under control conditions were compared to Ca²⁺ events following drug treatment using a paired student's t test on the raw data. Significance was determined at P≤ 0.05.

Results

Rapid, global Ca²⁺ events in UBSM are caused by Ca²⁺ influx during an action potential

UBSM strips, loaded with fluo-4, appeared as dark grey, homogenous bundles when examined with a laser scanning confocal microscope. In a single field, several bundles of UBSM were often seen lying in parallel. Alongside and between the bundles of smooth muscle were thin, very bright fibres that usually ran in parallel with the smooth muscle bundles. Based on their location and morphology, these fibres were probably nerve processes innervating the smooth muscle (Gabella, 1995; Gabella & Davis, 1998; Drake et al. 2003).

In PSS, smooth muscle bundles displayed repetitive, large, very rapid increases in Ca²⁺ that spread quickly along the length of the bundle. These Ca²⁺ events occurred spontaneously or could be evoked with field stimulation. We termed these large, rapid, global Ca²⁺ events ‘global Ca²⁺ flashes’. These events were sensitive to P2X receptor inhibition. Treatment with α,β-methylene ATP (10 μm), a P2X receptor agonist, which desensitizes P2X receptors, and suramin (10 μm), an inhibitor of P2X receptors, significantly decreased the number of spontaneous global Ca²⁺ flashes to 20.0 ± 13.0% of control (n= 6 preparations; P < 0.05) and also significantly decreased the number of global Ca²⁺ flashes evoked by stimulation to 36.3 ± 12.0% of control (n= 8 preparations; P < 0.05). This suggests that P2X receptor activation contributes to spontaneous and evoked global Ca²⁺ transients.

To identify the origin of the global Ca²⁺ flashes, a single muscle bundle was impaled with a microelectrode to record changes in the membrane potential. Simultaneous recordings of Ca²⁺-activated fluorescence changes and membrane potential showed that each global Ca²⁺ flash was evoked by a single action potential (Fig. 1). Since the upstroke of the action potential is dependent on Ca²⁺ influx through VDCCs, the global Ca²⁺ flash should be reduced when these channels are blocked. Diltiazem, a reversible inhibitor of VDCCs, markedly decreased these events (Fig. 2). For example, in one preparation, 68 global Ca²⁺ flashes were recorded in PSS during a 15 s recording using 1 Hz stimulation. Diltiazem (100 μm; 20 min) eliminated these events. Following a 35 min rinse in PSS, 70 global Ca²⁺ flashes were recorded. Simultaneous recordings of Ca²⁺ fluorescence and membrane potential from bundles of UBSM, along with the sensitivity of the global Ca²⁺ flashes to an inhibitor of VDCCs indicate that each global Ca²⁺ flash results from Ca²⁺ influx during an action potential. Immediately following the global Ca²⁺ flashes, the UBSM would contract suggesting that the VDCC-mediated Ca²⁺ influx is causally related to UBSM contraction.

Figure 1. Simultaneous recording of Ca²⁺ and voltage from a single bundle of UBSM.

A, images recorded from UBSM in normal PSS (37°C) loaded with fluo-4. The muscle bundles are orientated horizontally. Images were acquired at 30 images s⁻¹. The letters above each image correspond to changes in Ca²⁺-activated fluorescence in B (upper trace) and were recorded before (a), during (b) and after (c) a spontaneous action potential in the UBSM bundle impaled with the microelectrode. B, simultaneous recordings of changes in Ca²⁺-activated fluorescence (upper trace) and voltage (lower trace) from a single bundle of UBSM. Changes in Ca²⁺-activated fluorescence were measured from the red box in A located on the impaled muscle bundle. The microelectrode impalement site is indicated by the green rectangle. Notice that each of the three action potentials induced a simultaneous increase in the Ca²⁺-activated fluorescence. The large increase in Ca²⁺-activated fluorescence from a single muscle bundle during an action potential is also shown above in Ab.

Figure 2. Global Ca²⁺ flashes in UBSM recorded during field stimulation.

A, average images (acquired at 30 images s⁻¹) showing bundles of smooth muscle before (a), approximately 100 ms following a single stimulus pulse (b), and approximately 1 s following field stimulation (c). Several bundles demonstrated an increase in Ca²⁺-activated fluorescence changes following stimulation. B, the changes in Ca²⁺-activated fluorescence were recorded from the regions of interest (coloured boxes) in A showing a rapid increase in Ca²⁺-activated fluorescence and a slow return to resting Ca²⁺ levels following field stimulation. Letters a, b and c correspond to the images in A. C, the relative number of global Ca²⁺ flashes from paired experiments recorded during field stimulation in control and with diltiazem (100 μm). Inhibition of VDCCs significantly decreased the number of these global Ca²⁺ flashes. *P < 0.05, n= 3 preparations.

Nerve stimulation elevates the frequency of local Ca²⁺ transients

In addition to the global Ca²⁺ flashes, smaller, more localized Ca²⁺ transients were also evident in UBSM. These localized Ca²⁺ transients became more apparent when the global Ca²⁺ flashes were inhibited with either VDCC blockers or when sucrose (12%) was added to the superfusing PSS to restrict tissue movement. The frequency of these small Ca²⁺ transients could be elevated by low-frequency field stimulation (1 Hz) (Fig. 3) (movie file provided as supplemental data). To evoke these small Ca²⁺ transients with field stimulation, the stimulating voltage was decreased to a level just below the threshold needed to induce global Ca²⁺ flashes. To measure the effect of field stimulation on the frequency of these events, Ca²⁺ transients were recorded from the same field before and after stimulation (10 s at 1 Hz). Nerve stimulation significantly elevated the frequency of the Ca²⁺ transients about 7-fold, from 23 ± 8.4 to 156 ± 35.9 events (P < 0.05; n= 4 preparations).

Figure 3. Local Ca²⁺ responses in UBSM evoked with field stimulation.

A, images of UBSM show smooth muscle (acquired at 30 images s⁻¹) bundles running diagonally from the lower left to the upper right of each image before (left panel) and after (right panel) a single stimulus pulse. Coloured boxes enclose regions of interest where evoked Ca²⁺ responses occur, and correspond to the coloured traces below in B. The white arrows in the right panel show Ca²⁺ responses occurring in response to a single stimulus and correspond to the second stimulus pulse from the left in B. B, changes in Ca²⁺-activated fluorescence in response to seven stimulus pulses (1 Hz). Note that with each pulse a Ca²⁺ transient occurs in some of the selected regions of interest (movie file provided as supplemental data).

Purinergic receptor, but not muscarinic receptor, activation mediates the localized Ca²⁺ transients

Parasympathetic nerve fibres innervating UBSM corelease ACh and ATP in response to nerve stimulation (Kasakov & Burnstock, 1982; Theobald & de Groat, 1989). Activation of muscarinic receptors induces IP₃-mediated release of Ca²⁺ from internal stores. ATP activates P2X receptor channels which permit the influx of Na⁺ and Ca²⁺ ions into UBSM cells. To study the link between nerve transmission and evoked Ca²⁺ transients, we used selective pharmacological inhibitors of muscarinic and purinergic receptors. Atropine (10 μm), which inhibits muscarinic receptors, did not significantly affect the frequency of evoked local Ca²⁺ transients (Fig. 4). To assess the contribution of P2X receptors to the evoked Ca²⁺ transients, we used compounds that inhibit P2X receptor function. Suramin (10 μm) and α,β-methylene ATP (10 μm) nearly abolished the local Ca²⁺ transients evoked with field stimulation (Fig. 4). In addition, P2X receptor inhibition significantly decreased the number of spontaneous Ca²⁺ transients to 19.0 ± 7.0% of control (20 s file; α,β-methylene ATP, 10 μm and suramin, 10 μm) (P < 0.05; n= 4 preparations). These results indicate that the majority of spontaneous Ca²⁺ transients and evoked Ca²⁺ transients are purinergic in nature and probably reflect the influx of Ca²⁺ into UBSM through P2X receptors. We refer to these events as ‘purinergic Ca²⁺ transients’ hereafter.

Figure 4. Inhibition of P2X receptors, but not muscarinic receptors, significantly decreased the frequency of evoked Ca²⁺ transients.

Column graph showing no change in evoked Ca²⁺ transients with muscarinic receptor inhibition by atropine (10 μm), but nearly complete abolition of these events when P2X receptors are blocked with suramin (10 μm) and α,β-methylene ATP (10 μm). To account for the variability in the number of events between different preparations, the data were normalized to control. *P < 0.05, n= 3 preparations.

Latency of purinergic Ca²⁺ transients from onset of nerve stimulation

To explore the relationship between nerve stimulation and purinergic Ca²⁺ transients, the time between nerve stimulation and the appearance of the local Ca²⁺ event (latency) was determined by acquiring images at the rate of 240 images s⁻¹. At this acquisition rate, the time interval between images was 4.13 ms. Figure 5A and B shows the onset and decay of two Ca²⁺ events following a single stimulus. The purinergic Ca²⁺ transients occurred as early as 8 ms after a stimulus, with most Ca²⁺ events occurring 12–16 ms following a stimulus (Fig. 5C). These purinergic Ca²⁺ transients often appeared to originate from nerve fibres, and occurred repeatedly from the same site.

Figure 5. Field stimulation (1 Hz) evokes purinergic Ca²⁺ transients in UBSM.

A, four panels showing the stimulus flash followed by the onset of purinergic Ca²⁺ transients. The beginning of the grey region (upper left corner of grey area) of the upper left panel shows the precise time the stimulus was delivered to the tissue. About 12 ms following the stimulus (upper right panel), a purinergic Ca²⁺ transient appears in the light blue region of interest, followed 25 ms after the stimulus by a second purinergic Ca²⁺ transient in the red region of interest (lower left panel). The lower right panel shows the purinergic Ca²⁺ transients near their peak. B, the changes in Ca²⁺-activated fluorescence from the regions of interest in A before and after a stimulus pulse. Notice the rapid onset and slow decay of these purinergic Ca²⁺ transients. C, the latency of purinergic Ca²⁺ transients is displayed graphically. The earliest purinergic Ca²⁺ transients were recorded about 8 ms after the stimulus and reached a peak between 12 and 16 ms after stimulation.

TTX inhibits purinergic Ca²⁺ transients evoked with field stimulation

Since both UBSM and nerve fibres could potentially be activated by field stimulation, a pulse duration of 0.2 ms was used to selectively activate nerve fibres (Hashitani et al. 2000; Herrera et al. 2000). However, to ensure that the purinergic Ca²⁺ transients were not caused by direct activation of UBSM, tetrodotoxin (TTX) 2 μm was added to the superfusing solution to inhibit voltage-gated Na⁺ channels located on nerve processes. In paired experiments, the number of spontaneous purinergic Ca²⁺ transients recorded during 20 s files was not significantly affected by TTX (2 μm) (TTX 109.0 ± 24.0% of control, n= 4 preparations). In contrast, TTX (2 μm), significantly decreased the frequency of evoked purinergic Ca²⁺ transients (TTX 17.0 ± 8.0% of control, P < 0.05, n= 4 preparations), indicating that evoked purinergic Ca²⁺ transients were dependent on the activation of nerve fibres.

Purinergic Ca²⁺ transients do not depend on Ca²⁺ influx through VDCCs

The opening of VDCCs allows Ca²⁺ influx into UBSM. It is possible that localized Ca²⁺ entry through VDCCs into UBSM contributes to purinergic Ca²⁺ transients. To examine this possibility, we applied the VDCC inhibitor, diltiazem (100 μm), before field stimulation (1 Hz). Inhibiting VDCCs did not significantly reduce the frequency (diltiazem, 87.0 ± 7.0% of control, n= 3 preparations) or amplitude (control, 2.0 ± 0.04 F/F₀, n= 246 Ca²⁺ events; diltiazem, 2.0 ± 0.03 F/F₀, n= 310 Ca²⁺ events) of the purinergic Ca²⁺ transients, indicating that Ca²⁺ influx through VDCCs does not contribute to purinergic Ca²⁺ transients.

P2X receptor channels are permeable to both Ca²⁺ and Na⁺ ions. Therefore, purinergic Ca²⁺ transients should depend on external Ca²⁺. Removing external Ca²⁺ significantly decreased the frequency of purinergic Ca²⁺ transients. In the absence of stimulation, the frequency of spontaneous purinergic Ca²⁺ transients was significantly decreased in Ca²⁺-free external solution to 14.0 ± 9.0% of control (n= 4 preparations; P < 0.05). Returning Ca²⁺ to the external solution increased the frequency of purinergic Ca²⁺ transients to control levels (106.0 ± 17.0% of control, n= 4 preparations). The frequency of evoked purinergic Ca²⁺ transients was also inhibted significantly in Ca²⁺-free solution to 15.0 ± 5.0% of control (n= 4 preparations). When Ca²⁺ was returned to the external solution, the frequency of evoked purinergic Ca²⁺ transients returned to 77.0 ± 10.0% of control. Although it is likely that the release of transmitter from nerve terminals would also be blocked, these findings indicate that purinergic Ca²⁺ transients are dependent on external Ca²⁺ and suggest that internal Ca²⁺ release does not underlie these events.

Ca²⁺ release through IP₃ receptors and RyRs does not contribute to the purinergic Ca²⁺ transients

In smooth muscle, IP₃-mediated Ca²⁺ release produces Ca²⁺ waves (Iino et al. 1993), and Ca²⁺ release through RyRs can produce Ca²⁺ sparks which activate large-conductance potassium channels and lead to membrane hyperpolarization (Nelson et al. 1995; Jaggar et al. 2000; Herrera et al. 2001; Heppner et al. 2003). To study the contribution of Ca²⁺ release from internal stores to purinergic Ca²⁺ transients, we examined purinergic Ca²⁺ transients following inhibition of IP₃ receptors with 2-APB (100 μm), and inhibition of RyRs with ryanodine (10 μm). The inhibition of IP₃ receptors did not significantly alter the frequency (2-APB, 81.0 ± 10.0% of control, n= 4 preparations) or amplitude (control, 2.0 ± 0.02 F/F₀, n= 711 purinergic Ca²⁺ transients; 2-APB, 2.0 ± 0.02 F/F₀, n= 541 purinergic Ca²⁺ transients) of purinergic Ca²⁺ transients. Similarly, the inhibition of RyRs did not significantly alter the frequency (ryanodine 106.0 ± 10.0% normalized to control, n= 4 preparations) or amplitude of purinergic Ca²⁺ transients (control, 1.9 ± 0.18 F/F₀, n= 339 purinergic Ca²⁺ transients; ryanodine, 1.9 ± 0.18 F/F₀, n= 639 purinergic Ca²⁺ transients). These findings suggest that Ca²⁺ release from internal stores does not significantly contribute to purinergic Ca²⁺ transients.

Ca²⁺ sparks and purinergic Ca²⁺ transients represent two discrete elementary Ca²⁺ events in UBSM

Ca²⁺ sparks are localized, transient events mediated by the release of sarcoplasmic reticulum Ca²⁺ through RyRs in smooth muscle (Nelson et al. 1995; Jaggar et al. 2000), including UBSM (Imaizumi et al. 1999; Kotlikoff et al. 1999; Herrera et al. 2001; Ohi et al. 2001). Purinergic Ca²⁺ transients are also brief, localized Ca²⁺ events, but represent Ca²⁺ influx through purinergic receptors. In the presence of ryanodine (10 μm), which blocks Ca²⁺ sparks, a population of larger events remained that were evoked by field stimulation. These events were sensitive to the purinerigic receptor inhibitors, α,β-methylene ATP (10 μm) and suramin (10 μm). Inhibition of purinergic receptors left a population of Ca²⁺ events that were smaller in amplitude and frequency (Ca²⁺ sparks). When the kinetics of each type of elementary Ca²⁺ transient were measured, we found that the purinergic Ca²⁺ transients were larger in amplitude, duration and spread compared to Ca²⁺ sparks (Fig. 6).

Figure 6. Ca²⁺-activated fluorescence for purinergic Ca²⁺ transients and Ca²⁺ sparks in UBSM.

The upper trace represents a purinergic Ca²⁺ transient (*) through P2X receptors recorded in the presence of ryanodine (10 μm) to inhibit Ca²⁺ sparks. The lower trace represents a Ca²⁺ spark (*) recorded in the presence of suramin (10 μm) and α,β-methylene ATP (10 μm) to block P2X receptors. Kinetic data from these traces show that purinergic Ca²⁺ transients mediated through P2X receptors are larger in amplitude and duration than Ca²⁺ sparks. In the table * indicates a significant difference (P < 0.05) between purinergic Ca²⁺ transients and RyR Ca²⁺ sparks.

Discussion

In this study, we identified three different Ca²⁺ signals in UBSM: (1) global Ca²⁺ flashes, which are large, rapid Ca²⁺ signals that arise from Ca²⁺ influx through VDCCs during an action potential, last for several seconds, and occupy the entire smooth muscle cell; (2) purinergic Ca²⁺ transients, which are smaller, local Ca²⁺ transients in UBSM mediated by Ca²⁺ influx through purinergic receptor channels; (3) Ca²⁺ sparks, which are localized Ca²⁺ events mediated by Ca²⁺ release through RyRs.

Global Ca²⁺ flashes represent Ca²⁺ influx during action potentials

Global Ca²⁺ flashes in UBSM occurred spontaneously, were evoked with field stimulation, and were sensitive to the inhibition of P2X receptors. These events travelled along the length of the UBSM bundle and were immediately followed by tissue contraction. These events were dependent on the influx of Ca²⁺ through VDCCs, since they were substantially reduced by the VDCC blocker, diltiazem (see Figs 1 and 2). This view is supported by the finding that the upstroke of the action potential is mediated by Ca²⁺ influx through VDCCs (Klockner & Isenberg, 1985; Mostwin, 1986; Heppner et al. 1997; Hashitani et al. 2000; Hashitani & Brading, 2003a,b) and is consistent with the observation that action potentials occur spontaneously in UBSM (Creed et al. 1983; Heppner et al. 1997; Hashitani et al. 2000, 2001; Hashitani & Brading, 2003a,b). Simultaneous recordings of voltage and Ca²⁺ in the guinea-pig UBSM revealed a Ca²⁺ transient associated with each action potential similar to the findings of the present study (Hashitani et al. 2001, 2004a,b). The block of Ca²⁺ global flashes by purinergic receptor inhibitors indicates that the activation of purinergic receptors is involved in the generation of VDCC-dependent action potentials and global Ca²⁺ flashes.

Nerve stimulation evokes purinergic Ca²⁺ transients

The frequency of purinergic Ca²⁺ transients increased with field stimulation when the pulse duration was selective for nerve fibres. When conduction along nerve fibres was blocked by inhibiting voltage-gated Na⁺ channels with TTX, evoked purinergic Ca²⁺ transients were not detected. These results indicate that the activation of nerve fibres in the UBSM significantly elevates the frequency of purinergic Ca²⁺ transients.

A property of synaptic transmission is latency, or the brief delay between presynaptic cell stimulation and the response in the postsynaptic cell. The latency of purinergic Ca²⁺ transients recorded with line scan confocal imaging was <3 ms in mesenteric arteries (Lamont & Wier, 2002), and about 6 ms in the vas deferens (Brain et al. 2002). Using high-speed confocal imaging (240 images s⁻¹), events could be detected as soon as 8 ms following the stimulus, with the peak latency of purinergic Ca²⁺ transients in UBSM at 12–16 ms. There was approximately 4.1 ms between images at an acquisition rate of 240 images s⁻¹, and therefore, the maximum error in latency would be 4.1 ms. Therefore, the latency of purinergic Ca²⁺ transients in UBSM was consistent with the delay found in other smooth muscle tissues. This result, in conjunction with the inhibitory effect of TTX, indicates that purinergic Ca²⁺ transients result from nerve fibre stimulation and synaptic transmission, and not direct stimulation of UBSM.

External Ca²⁺, but not VDCCs, is necessary to evoke purinergic Ca²⁺ transients

The purinergic Ca²⁺ transients should be dependent on external Ca²⁺, since P2X receptors are highly permeable to Ca²⁺. Removing Ca²⁺ from the external solution significantly decreased the frequency of purinergic Ca²⁺ transients. This effect was largely reversible, since the frequency of purinergic Ca²⁺ transients increased with the re-introduction of Ca²⁺ to the external medium. The removal of external Ca²⁺ could also inhibit the release of ATP from nerve terminals and thereby decrease purinergic Ca²⁺ transients. Our experiments do not differentiate between these two possibilities. However, if the removal of external Ca²⁺ had no effect on purinergic Ca²⁺ transients, then this would argue against Ca²⁺ influx through P2X receptor channels.

Another possibility is that Ca²⁺ influx through VDCCs contributes to the purinergic Ca²⁺ transients. VDCCs in UBSM (Creed et al. 1983; Heppner et al. 1997; Herrera et al. 2001; Herrera & Nelson, 2002; Hashitani & Brading, 2003a,b) are voltage sensitive and play a key role in mediating Ca²⁺ entry into UBSM cells, such as would occur during an action potential (Klockner & Isenberg, 1985; Mostwin, 1986; Heppner et al. 1997; Hashitani & Brading, 2003b). However, Ca²⁺ influx through VDCCs does not appear to contribute to purinergic Ca²⁺ transients for several reasons: (1) the duration of our stimulating pulse (0.2 ms) is selective for nerve fibres, and not likely to depolarize UBSM; (2) inhibition of VDCCs with diltiazem did not significantly decrease the frequency of purinergic Ca²⁺ transients; (3) the spatial spread of Ca²⁺ was local, and did not include the entire cell as would be expected with VDCC activation. Therefore, in this study, diltiazem-sensitive VDCCs do not appear to contribute to the purinergic Ca²⁺ transients.

Purinergic Ca²⁺ transients do not require Ca²⁺release from internal stores

Ca²⁺ stores in the sarcoplasmic reticulum are critical to many cellular functions and underlie Ca²⁺ transient events such as Ca²⁺ sparks and Ca²⁺ waves. Ca²⁺ sparks are generated by the release of Ca²⁺ through RyRs located on the sarcoplasmic reticulum (Nelson et al. 1995; Herrera et al. 2001; Heppner et al. 2003). Ca²⁺ waves are generated by the activation of IP₃ receptors (Iino et al. 1993). Activation of muscarinic receptors leads to the production of IP₃. Both M2 and M3 receptor subtypes are found in the urinary bladder. The M3 receptor is coupled to the Gq family of proteins and generates the production of IP₃ and internal Ca²⁺ release (Caulfield, 1993). The M3 subtype mediates contraction in the urinary bladder (Hegde & Eglen, 1999) and could trigger the release of Ca²⁺ from the sarcoplasmic reticulum. However, the frequency of purinergic Ca²⁺ transients was not reduced when muscarinic receptors were blocked with atropine, suggesting that muscarinic receptor activation does not contribute to the initiation of the purinergic Ca²⁺ transients in UBSM.

Inhibition of Ca²⁺ release through RyRs (ryanodine, 10 μm) or through IP₃ receptors (2-APB, 100 μm) and block of the production of IP₃ by inhibition of the muscarinic receptor (atropine, 10 μm) did not alter the frequency of purinergic Ca²⁺ transients in UBSM. Similar results were also observed in arteries. Gitterman & Evans (2001) found that Ca²⁺-induced Ca²⁺ release was not involved in P2X-receptor-mediated contractions in rat mesenteric arteries, but that all the Ca²⁺ for contraction enters through P2X receptors. Therefore, in UBSM the release of Ca²⁺ from internal stores is not crucial to the generation of the purinergic Ca²⁺ transients.

Purinergic Ca²⁺ transients are mediated through purinergic receptors

P2X receptors belong to a family of ATP-sensitive ligand-gated cation channels (P2X₁–P2X₇) formed by the assembly of 3–6 individual subunits (Khakh et al. 2001). In the cat UBSM, the P2X₂ receptor is the predominant subtype followed by the P2X₁ receptor (Birder et al. 2004). However, in rodents the homomeric P2X₁ receptor subtype is clearly associated with the membranes of UBSM (Lee et al. 2000; Elneil et al. 2001; Vial & Evans, 2000) and underlies the P2X receptor activity in UBSM as demonstrated in P2X₁ receptor-deficient mice (Vial & Evans, 2000). Although some studies found no evidence of P2X₁ clustering in rat or human urinary bladders (Elneil et al. 2001), other studies employing rat urinary bladder demonstrated P2X₁ receptors distributed in clusters on smooth muscle membranes adjacent to nerve varicosities (Hansen et al. 1998; Dutton et al. 1999; Yunaev et al. 2000). These studies provide anatomical evidence of nerve fibre varicosities and P2X₁ receptors, possibly in clusters, adjacent to one another.

Activation of P2X receptors allows the influx of cations, including Ca²⁺ and Na⁺ ions, into the cell. Recent Ca²⁺ imaging studies identified ATP-evoked Ca²⁺ events in the vas deferens (Brain et al. 2002, 2003) and in mesenteric arterial smooth muscle (Lamont & Wier, 2002; Lamont et al. 2003). In the vas deferens, stimulation of sympathetic fibres released ATP that activated P2X receptors. The Ca²⁺ influx through the P2X receptor was termed ‘neuroeffector Ca²⁺ transients’ (Brain et al. 2002, 2003). Similarly, in mesenteric artery smooth muscle, Ca²⁺ transients, termed ‘junctional Ca²⁺ transients’ were evoked by the activation of P2X receptors located on the smooth muscle, by ATP released from sympathetic fibres. This influx of Ca²⁺ through the P2X receptors in mesenteric arteries comprises the junctional Ca²⁺ transient (Lamont & Wier, 2002; Lamont et al. 2003). The kinetics of these Ca²⁺ events are similar to the evoked purinergic Ca²⁺ transients presently described in UBSM. Since evoked Ca²⁺ transients are blocked by inhibitors of P2X receptors, it is likely that the activation of a cluster of P2X₁ receptors by nerve-evoked release of ATP underlies the evoked Ca²⁺ transients observed in the present study (Fig. 7). In UBSM, voltage recordings revealed excitatory junction potentials (EJPs), which are brief membrane potential depolarizations mediated by P2X receptors. EJPs initiate action potentials in UBSM, transient increases in Ca²⁺ that are followed by contractions (Bramich & Brading, 1996; Hashitani et al. 2000). Since the activation of P2X receptor channels permits the influx of both Ca²⁺ and Na⁺ ions it is likely that the purinergic Ca²⁺ transients detected in the present study represent the Ca²⁺ component of the EJP. Consistent with this hypothesis is the finding that global Ca²⁺ transients, which represent Ca²⁺ influx during an action potential, are significantly inhibited when purinergic Ca²⁺ transients are blocked. This suggests that the activation of purinergic receptors through EJPs is involved in the initiation of VDCC-dependent action potentials and global Ca²⁺ transients.

Figure 7. Illustration showing the origin of the purinergic Ca²⁺ transients mediated by P2X receptors.

ATP released from parasympathetic nerve varicosities activates P2X receptors located on UBSM. Ca²⁺ influx through the P2X receptors underlies the purinergic Ca²⁺ transients. The membrane potential depolarization arising from Ca²⁺ and Na⁺ influx through P2X receptors would open VDCCs allowing significant Ca²⁺ influx into the cell.

Two elementary localized Ca²⁺ events in UBSM: purinergic Ca²⁺ transients and Ca²⁺ sparks

UBSM exhibits two distinct, localized, transient, elementary Ca²⁺ events: purinergic Ca²⁺ transients and Ca²⁺ sparks. Although similar in appearance, the origin, kinetics, and function of these events are quite different. Purinergic Ca²⁺ transients, mediated by external Ca²⁺ influx through purinergic receptors, have significantly larger amplitudes, longer decay times and greater spatial spread than Ca²⁺ sparks mediated by Ca²⁺ efflux through RyRs located on the sarcoplasmic reticulum (Fig. 6). Ca²⁺ sparks are known to have a ‘braking’ effect on cell excitability by activating large-conductance Ca²⁺-sensitive K⁺ channels in arterial smooth muscle (for review see Jaggar et al. 2000), and have a similar role in UBSM (Imaizumi et al. 1999; Herrera et al. 2001; Heppner et al. 2003). Purinergic Ca²⁺ transients may have just the opposite effect. P2X receptors are non-selective cation channels that exhibit nearly equal permeability to Na⁺ and Ca²⁺ ions (Schneider et al. 1991). Under physiological conditions, Ca²⁺ ions comprise about 6–8% of the cation influx, with the balance comprised mainly of Na⁺ (Schneider et al. 1991). The activation of a significant number of P2X clusters could trigger sufficient Na⁺ and Ca²⁺ entry to depolarize the membrane potential and activate VDCCs. Subsequent activation of VDCCs would contribute the large amounts of Ca²⁺ that underlie the rising phase of action potentials in UBSM and cause contraction (Creed et al. 1983; Klockner & Isenberg, 1985; Heppner et al. 1997; Hashitani et al. 2000, 2001; Herrera et al. 2001).

Clinical conditions increase purinergic activity in the urinary bladder

During certain clinical conditions the contribution of purinergic pathways to excitation–contraction coupling in UBSM is altered. During pregnancy in rats, the composition of purinergic receptors beneath varicosities in the urinary bladder is changed. There is a decrease in clusters of P2X₁, P2X₂, P2X₃, and P2X₅, and an increase in subtypes P2X₄, and P2X₆ (Yunaev et al. 2000). In normal human UBSM, purinergic pathways appear to play a minor role. Normally, the density of P2X receptors in human urinary bladders is low compared to rodent urinary bladders (Bo & Burnstock, 1995). However, in unstable, symptomatically obstructed urinary bladders, P2X₁ receptor subtype expression is significantly increased (for review see Boselli et al. 2001), and patients with idiopathic detrusor instability have a significant purinergic component of nerve-mediated contractions that is absent in normal human bladders (O'Reilly et al. 2002). This suggests a plasticity of P2X receptor expression that alters urinary bladder function. Our approach provides the means to examine changes in purinergic signalling at the elementary level in UBSM following outlet obstruction.

This study identifies two distinct types of Ca²⁺ signals exhibited in UBSM using confocal miscroscopy combined with fast Ca²⁺ imaging techniques: global Ca²⁺ flashes and novel, smaller, localized purinergic Ca²⁺ transients. The global Ca²⁺ flashes represent Ca²⁺ influx during action potentials. The smaller, localized purinergic Ca²⁺ transients are not RyR-mediated Ca²⁺ sparks, but instead represent Ca²⁺ influx through P2X receptors located on UBSM cells. These localized purinergic Ca²⁺ transients may represent the initial, crucial steps in the nerve-evoked cascade of events that leads to increases in intracellular Ca²⁺ and contraction of UBSM.

Supplementary material

The online version of this paper can be accessed at: DOI: 10.1113/jphysiol.2004.077826

http://jp.physoc.org/cgi/content/full/jphysiol.2004.077826/DC1 and contains supplemental material consisting of a movie clip from Fig. 3 showing purinergic Ca²⁺ transients in urinary bladder smooth muscle evoked by field stimulation (1 Hz). Notice that numerous purinergic Ca²⁺ transients occurr in response to each stimulus.

This material can also be found at:

http://www.blackwellpublishing.com/products/journals/suppmat/tjp/tjp740/tjp740sm.htm