Ion-channel currents of smooth muscle cells isolated from the prostate of guinea-pig


Dr Sung Joon Kim, Assistant Professor, Sungkyunkwan University School of Medicine, Suwon 440–746, Korea.



To characterize the voltage-activated ion-channel currents in guinea-pig prostate smooth muscle cells (GPSMCs).


GPSMCs were isolated using collagenase, and used in a whole-cell patch clamp study.


When GPSMCs were dialysed with a CsCl solution all the outward K+ currents were blocked and the step-like depolarization (holding voltage − 70 mV) of the cell membrane evoked inward currents that were completely blocked by nifedipine (1 µmol/L). With KCl solution, step depolarizations showed outward K+ currents composed of fast, transient outward current (Ito) and outward currents that did not inactivate. Ito was resistant to a high concentration of tetraethylammonium (TEA, 5 mmol/L) but was blocked by 4-aminopyridine (5 mmol/L). The half-activation and half-inactivation voltages of Ito were 6 mV and − 58 mV, respectively. With low Ca2+ buffer (0.1 mmol/L EGTA) in the solution, there were spontaneous transient outward currents (STOCs) at depolarized membrane voltages (0 mV). STOCs were blocked by TEA (1 mmol/L) or iberiotoxin (10 nmol/L) but were insensitive to apamin (100 nmol/L).


This voltage-clamp study showed that GPSMCs have l-type Ca2+ channels and more than two types of K+ channels. The voltage- and time-dependent changes of these ion channels and their interactions might be important in forming action potentials and regulating contractility.


(guinea-pig prostate) smooth muscle cell


voltage-activated Ca2+ channel


transient outward current




spontaneous transient outward current


sarcoplasmic reticulum.


The prostate, a major sex accessory gland of males, has a widely varying morphology depending on the species [1]. Microscopic studies of experimental animals and man commonly describe the ductal system of the prostate gland as being composed of exocrine epithelial cells overlying a fibromuscular stroma [2–5]. The main function of the prostate is to produce citrate-rich fluid that comprises the major portion of seminal plasma [6]. Its secretions also include proteolytic enzymes, zinc, fructose, acid phosphatase and prostaglandins. These factors, although not essential to fertilization, are thought to optimize the conditions for fertilization by increasing sperm motility and viability [1,7].

The human prostate is of considerable medical interest as it is the only male sex accessory gland that frequently undergoes abnormal growth during the later stages of life (e.g. BPH). As the prostate surrounds the urethra in a restricted space of the human lower pelvis it can cause severe urological symptoms, i.e. LUTS [2,4].

One of the obstacles hindering the progress of studies on BPH is the lack of adequate animal model. Although the prostate of the dog develops spontaneous prostate hyperplasia with age, the glandular nature of canine hyperplastic prostate differs from that of the ageing human prostate, which is characterized by an overall increase in stromal mass and fibrosis [2]. In this respect some researchers claim that the guinea-pig prostate has age-related changes in histological features similar to those of human BPH, i.e. an increase in perinuclear organelles and a progressive hypertrophy of the smooth muscle cells (SMCs) [5]. Morphological studies of cultured SMCs isolated from the guinea-pig prostate shows typical microscopic morphology of SMCs and they contract in response to the stimulation of α1 adrenoceptors [8].

The smooth muscle of the prostatic stroma contracts to propel its secretions into the prostatic urethra during ejaculation, and they can participate in the control of urine output from the bladder [9]. The contractile state of SMCs are regulated partly by entry of extracellular calcium through voltage-activated Ca2+ channels (VACCs), the ‘open’ probability of which is sensitive to changes in membrane potential [10]. Previous studies using conventional intracellular microelectrode techniques showed that the guinea-pig prostate has spontaneous electrical activity and concurrent contractile responses [11,12]. However, a direct investigation of ion-channel currents that underlie the formation of membrane potential and action potentials is lacking in the guinea-pig prostate.

In electrically excitable cells like neurones and muscle cells, the generation of action potentials arises from the complex interaction of various voltage-activated ion channels in the plasma membrane. The present study was undertaken to characterize the voltage-activated ion-channel currents in the SMCs freshly dissociated from the dorsolateral prostate of the guinea pig.


All procedures on experimental animals followed the guidelines of the Institutional Animal Care & Use Committee of Sungkyunkwan University. In all, 20 guinea pigs were used for the study (male, 350–450 g); they were killed by 100% CO2 inhalation and bled by cutting the carotid artery. The partially fused dorsal and lateral lobes of the prostate were dissected from the posterior wall of the urethra. The tissue was cut with scissors into small pieces (1–2 mm3) in a phosphate-buffered Ca2+-free Tyrode's solution containing (mmol/L) 145 NaCl, 1.6 K2HPO4, 0.4 KH2PO4, 1 MgCl2, and 5 d-glucose, at pH 7.4. For the enzymatic digestion, collagenase (1.5 mg/mL, C. histolyticum, Wako Pure Chemicals, Tokyo, Japan), trypsin inhibitor (0.5 mg/mL, Sigma, St. Louis, MO, USA), and fatty-acid free BSA (1 mg/mL, Sigma) were added and incubated at 35 °C for 25 min. After digestion the tissue pieces were rinsed twice with fresh Ca2+-free Tyrode's solution and gently agitated using a fire-polished Pasteur pipette (2 mm tip diameter). Afterwards the cell suspension was kept at 4 °C for up to 8 h in the same solution. Spindle-shape cells (length 100–150 µm, width 5–7 µm) could be distinguished with an inverted microscope (IX-70, Olympus, Tokyo, Japan).


Isolated cells were transferred into a bath chamber mounted on the stage of the inverted microscope; the bath (volume ≈ 0.3 mL) was perfused with a HEPES-buffered Tyrode's solution containing (mmol/L) 140 NaCl, 5 KCl, 10 HEPES, 1 MgCl2, 1.3 CaCl2 and 5 d-glucose, at pH 7.4, at a rate of 10 mL/min at room temperature (22–25 °C). The tip resistance of the patch pipettes was 2.5–3 MΩ. The reference electrode was Ag/AgCl and the liquid junction voltage was nullified using the circuit of the amplifier (Axopatch 1-D, Axon Instruments, Foster City, USA). pCLAMP software v.7.0 and Digidata-1200 A (both from Axon Instruments) were used to acquire data and apply command pulses. Details of the voltage command protocols are given with the Figures. The voltage and current data were low-pass filtered (5 kHz) and displayed on a computer monitor. The data were stored in a computer and analysed using Origin vs. 6.1 (Microcal Software Inc, Northampton, USA).


The composition of the KCl pipette solution was (mmol/L) 135 KCl, 5 NaCl, 0.5 MgCl2, 3 MgATP, 10 HEPES at pH 7.3 titrated with KOH. Using the above KCl pipette solution, the cytosolic calcium-buffering conditions were controlled as follows. For some experiments the free Ca2+ activity of pipette solution was clamped to 100 nmol/L by including 1 mmol/L EGTA and 0.47 mmol/L CaCl2. For the other experiments 10 mmol/L EGTA with no CaCl2 was included in the pipette solution to minimize the calcium activity of dialysed cytosol. For further experiments 0.1 mmol/L EGTA with no CaCl2 was included.

To record the voltage-activated Ca2+ current all the outward K+ currents were blocked by dialysing the cells with CsCl pipette solution, containing (mmol/L) 135 CsCl, 5 NaCl, 0.5 MgCl2, 3 MgATP, 10 EGTA, 10 HEPES at pH 7.3 titrated with CsOH. All chemicals used were of the highest purity available and obtained from Sigma, except iberiotoxin, which was purchased from Tocris (Bristol, UK).


The membrane conductance (g) was determined as I/(V-Erev) where I is the peak amplitude of the whole-cell current, V is the test voltage and Erev is the reversal potential (−90 mV) for the K+ current. The peak amplitude of transient outward current (Ito) could be discriminated from the faster capacitative artefacts generated by step-like depolarizing pulses. To obtain the activation curve, the values of g/gmax were plotted and fitted to the Boltzmann equation;

1/(1 + exp (V − V1/2)/κ

where V1/2 and κ represent the half-activation voltage and slope factor, respectively.

To measure the steady-state inactivation of the Ito, a double-pulse protocol was used. With a holding potential of − 80 mV, a pre-pulse potential from −110 mV to − 40 mV (using 10 mV increments) was applied for 1050 ms and a test potential of + 50 mV (160 ms) was given to the voltage-clamped cells. The inter-pulse interval was 10 s. The normalized current values (I/Imax) were plotted against the pre-pulse voltage and fitted to the Boltzmann equation to obtain the kinetics of voltage-dependent inactivation. The data are presented as original recordings, current-voltage (I-V) curves and the mean (sem).



The spindle-shape cells after enzymatic digestion were regarded as putative guinea-pig prostate SMCs (GPSMCs). Using CsCl pipette solution the membrane voltage was held at − 70 mV and step-like depolarizing pulses (400 ms) applied with incremental increases of 10 mV. Figure 1a shows typical records of currents measured in response to a voltage step to 0 mV. The step-depolarization evoked an inward current which subsequently showed a slow inactivation. In the same cell, a bath application of nifedipine (1 µmol/L), an l-type Ca2+-channel blocker, completely abolished the inward current. For five cells the mean peak current was −220 (48.2) pA at 0 mV. Current responses of GPSMCs to various levels of depolarizing pulses were normalized to the peak current at 0 mV and plotted according to test voltages (I-V curves, Fig. 1b). The I-V curves had a typical ‘inverted bell-shape’ with a peak at 0 mV (open circle, Fig. 1b). By treating with nifedipine the inward currents were completely abolished throughout the tested voltages (closed squares, Fig. 1b), suggesting that the voltage-activated inward currents of GPSMCs were exclusively a result of the activation of l-type Ca2+ channels.

Figure 1.

Voltage-activated inward current of GPSMCs. a, an original record of membrane current elicited by a 400 ms voltage step from −70 mV to 0 mV, which was then perfused with 1 µmol/L nifedipine. b, The relation between peak inward current and voltage (I/V curves). The symbols show mean (sem) values from five cells of the control (open green circle), and 1 µmol/L nifedipine (closed red square).


The GPSMCs were dialysed with KCl pipette solution containing 100 nmol/L free Ca2+ buffered with 1 mmol/L EGTA. Various levels of step-like depolarizing pulses (1200 ms) were applied from two different holding voltages, − 90 mV and −50 mV (Fig. 2a,b). The current responses from the same cell had different shapes depending on the holding voltages applied. When held at −90 mV, the step-pulse evoked fast, transient activation of Ito that inactivated to a quasi-steady-state level (Fig. 2a). In contrast, the Ito elicited by step pulses from a −50 mV holding voltage did not show such inactivation, which implied that Ito had been already inactivated at this holding voltage. In Fig. 2b, there was a small, transient inward current at the step-pulse to −10 mV, which was interpreted as the activation of the Ca2+ channels. Fig. 2a shows current records from the same cell as in Fig. 2b, where the voltage-activated inward current might have been masked by the concomitant activation of Ito.

Figure 2.

Outward currents activated by membrane depolarization in GPSMCs. a, b, show representative current traces obtained by step-pulse protocols as indicated above each panel. a, holding voltage, − 90 mV; b, holding voltage, − 50 mV. The pulse duration was 1.2 s and the inter-pulse interval 10 s. c, d, show I/V curves of the initial peak Ito (10 ms, closed red square) and steady-state (1200 ms, open green circle). c, holding voltage, − 90 mV; d holding voltage, − 50 mV.

The mean amplitudes of membrane currents measured at ≈ 5 ms (initial peak Ito) and 1200 ms (quasi-steady state current) after the start of step pulses were separately summarized as I-V curves. Figure 2c,d presents the I-V curves obtained under the holding voltages of −90 mV and −50 mV, respectively. From the I-V curves of Fig. 2c it was evident that the activation of Ito started at > −50 mV.

SMCs express abundant Ca2+-activated K+ channels with large single-channel conductances (> 200 pS), termed BKCa or maxi-K+ channels [10,13]. When GPSMCs were dialysed with KCl pipette solution containing 10 mmol/L EGTA with no CaCl2 (Ca2+-free KCl solution), the initial peak of Ito was still prominent, indicating that the Ito channels are Ca2+-independent (Fig. 3a). From the peak amplitude of Ito, the slope conductance at each voltage was calculated, which reflects the voltage-dependent activation of Ito channels. In Fig. 3c, the means of slope conductance from five GPSMCs were plotted against the test voltages and fitted by the Boltzmann equation; the half-activation voltage was 6 (2.3) mV.

Figure 3.

Voltage-activated Ito in GPSMCs; a, current traces obtained by step-pulses from − 70 to + 90 mV (using 20 mV increments, holding voltage, − 90 mV). The KCl pipette solution containing 10 mmol/L EGTA was used to dialyse the cell. b, a double-pulse protocol was applied to obtain the steady-state voltage-dependent inactivation. Inset: time-expanded traces. The current responses to the test pulse (+ 90 mV) are shown with greater time resolution, which shows a clear discrimination between capacitative artefacts and Ito. c, d, The normalized activation (c) and inactivation (d) of Ito were plotted and fitted using the Boltzmann equation.

The results shown in Fig. 2 also suggested that the Ito was sensitively abolished by a relatively depolarized holding voltage. To investigate the voltage dependence of inactivation more precisely, cells were held at different levels of conditioning pre-pulses and then a common test pulse (+ 50 mV) applied. The amplitude of peak Ito at the test pulse decreased as the level of conditioning pre-pulse was depolarized (Fig. 3b). The amplitudes of Ito at the common test voltage were normalized to the maximum current recorded after the conditioning pre-pulse of −110 mV, and the mean values of eight GPSMCs plotted against the conditioning pre-pulses (Fig. 3d). When the data obtained with conditioning potentials between −110 and 0 mV were fitted by a conventional Boltzmann distribution, the half-inactivation potential and slope factor were −58 (1.9) mV and 5.5 (0.19), respectively.


Figure 4 shows representative effects of K+-channel blockers on voltage-activated Ito in the same GPSMC. In this experiment, KCl pipette solution containing 1 mmol/L EGTA (100 nmol/L free Ca2+) was used. Step-pulses of −50 to + 70 mV (20 mV increment, 1200 ms) were applied from the holding voltage of −90 mV. After confirming the current responses under control conditions, tetraethylammonium (TEA) was applied. With increasing concentrations of TEA, the steady-state outward currents were suppressed while the amplitude of initial peak, i.e. Ito, was unaffected. The Ito was blocked only by applying another type of K+-channel blocker, 4-aminopyridine (Fig. 4e,f). However, as shown in Fig. 4f, even the combined application of TEA and 4-aminopyridine failed to block the Ito completely. The Ito resistant to TEA and 4-aminopyridine showed a delayed activation and inactivation, slower than the initial Ito (Fig. 4f).

Figure 4.

Effects of K+ channel blockers on voltage-activated Ito in GPSMCs. In each panel, the membrane currents were induced by step-pulses from − 50 mV to + 70 mV (using 20 mV increments, holding voltage: − 90 mV). All the panels of current were obtained from the same GPSMC and the sequential effects of TEA (a, control, b, TEA 0.1 mmol/L, c, TEA 1 mmol/L, d, TEA 5 mmol/L) and 4-aminopyridine (e, TEA 1 mmol/L, 4-aminopyridine 1 mmol/L, f, TEA 1 mmol/L, 4-aminopyridine 5 mmol/L) are shown.

The summarized effects of TEA and 4-aminopyridine are shown as I-V curves in Fig. 5. The currents were measured at ≈ 5 ms and 1200 ms of the step depolarization and normalized to the amplitude of the initial peak (at 5 ms) of Ito evoked by + 10 mV of step pulse in each GPSMC. The upper panels (Fig. 5a,b) and the lower panels (Fig. 5c,d) show the initial peak Ito and the steady-state amplitudes of outward currents, respectively.

Figure 5.

A summary of the effects of K+-channel blockers. Amplitudes of the initial peak Ito (a, b) or the steady-state current (c, d) were plotted against the test voltages. a and c show the effects of TEA (control, green open circle; TEA 1 mmol/L, light red open square; TEA 5 mmol/L, red closed square) and b and d the effects of 4-aminopyridine (1 mmol/L, red open triangle; 5 mmol/L light red closed triangle) and 1 mmol/L TEA.


A widely known electrophysiological phenomenon in SMCs under voltage-clamp conditions is the generation of STOCs, caused by bursts of opening of Ca2+-activated K+ channels in single voltage-clamped SMCs, and thought to reflect the sporadic release of calcium ions from the sarcoplasmic reticulum (SR) [13,14]. STOCs from GPSMCs occurred when the cell membrane was held at a depolarized voltage (e.g. 0 mV) and dialysed with KCl pipette solution containing 0.1 mmol/L EGTA (Figs 6 and 7). In Fig. 6a, the amplitudes of STOCs at 0 mV varied but were < 600 pA, indicating that the amount or ‘packet’ of calcium ions spontaneously released from SR might have a ‘quantal’ limit.

Figure 6.

STOCs recorded in GPSMCs, from original traces of membrane currents. Cells were dialysed with KCl solution (0.1 mmol/L EGTA) and clamped at 0 mV. The bath application of caffeine (10 mmol/L) evoked a large outward current (b).

Figure 7.

Effects of K+-channel blockers on STOCs, from representative original traces of membrane currents showing the effects of TEA (a), apamin (b) and iberiotoxin (c), respectively. Cells were dialysed with KCl solution (0.1 mmol/L EGTA) and were clamped at 0 mV.

The Ca2+-release pathway generating STOCs is reportedly the ryanodine receptor Ca2+ channel (RyRs) in the SR membrane of SMCs [13,14]. Caffeine is widely used to investigate ryanodine receptors. The threshold of RyRs for cytosolic calcium concentration ([Ca2+]c) is lowered by caffeine and therefore RyRs are activated even in the resting state of [Ca2+]c[15]. Figure 6b shows the effects of caffeine on STOCs in GPSMCs. Bath-applied caffeine (10 mmol/L) induced a large Ito which was considered to reflect the increase of [Ca2+]c from the simultaneous release of Ca2+ from the SR [13,14]. The generation of STOCs was transiently inhibited after washing out the caffeine, indicating that a refilling of the SR was required for the subsequent generation of STOCs. Also, the transient inhibition of STOCs after caffeine removal might have been partly a result of a transient decrease of [Ca2+]c below the resting level, as has been reported in guinea-pig detrusor muscle [16].

Using pharmacological blockers of Ca2+-activated K+ channels, the subtype of Ca2+-activated K+ channels underlying STOCs of GPSMCs was investigated. After confirming the generation of STOCs in GPSMCs under voltage-clamp conditions, various K+-channel blockers were tested (Fig. 7). A bath application of TEA (1 mmol/L), or iberiotoxin (10 nmol/L) markedly decreased the amplitudes of STOCs. The blocking effects of these three pharmacological agents commonly indicate that BKCa channels are involved in generating STOCs [17]. However, apamin (100 nmol/L), a well-known blocker of small conductance Ca2+-activated K+ channels had no effect on STOCs of GPSMCs. This pharmacological sensitivity of STOCs indicated that BKCa channels are responsible for STOCs in GPSMCs.


We tried to characterize the voltage-activated ion-channel currents in the GPSMCs; as with other types of SMC, the l-type Ca2+ current and the TEA-sensitive STOCs were consistently present in GPSMCs. An interesting feature of GPSMCs was the presence of the fast Ito.

In SMCs both influx via VACCs and release from the SR provide the Ca2+ for contractile responses. Although nifedipine-sensitive l-type Ca2+ channels are present in all kinds of SMCs [10] there is a report suggesting the presence of T-type Ca2+ channels in detrusor SMCs of guinea-pig [18]. However, in that study, no inward current was activated at <  −40 mV and the application of nifedipine completely abolished the inward current, which indicates that the l-type Ca2+ channels with a high threshold for activation comprise the voltage-activated Ca2+ influx pathway in GPSMCs. There is a question about the holding voltage of test pulses to record VACCs in this study. As T-type Ca2+ channels have a lower threshold of voltage for their inactivation and activation, the holding voltage of − 70 mV might have partly inactivated the T-type Ca2+ channels before their activation by step pulses. However, there was no detectable residual inward current at ≤  −40 mV. Moreover, in a previous report using conventional intracellular microelectrode techniques [11], the most negative resting membrane potential measured in the guinea-pig prostate was − 50 to − 60 mV. Therefore, the physiological responses of GPSMCs may be better reflected when held at − 70 mV rather than at − 90 mV, or more negative holding voltages.

A recent study showed that the guinea-pig prostate has spontaneous depolarizing transient membrane potentials superimposed by a single or multiple spike potentials [11]. The l-type Ca2+ channel currents noted in the present study would be responsible for generating spike potentials, as applying nifedipine greatly suppressed the spike potentials [11]. However, in that study the basal spontaneous depolarizing transient was resistant to nifedipine, suggesting that further investigation is required to understand the pace-making mechanism of guinea-pig prostate.


The electrophysiological and pharmacological profile of Ito (Figs 2 and 3) indicated that so-called A-type K+ channels are present in GPSMCs [19]. In addition, as shown in Figs 4f and 5, a substantial Ito remained after the combined application of TEA and 4-aminopyridine. The delayed activation and slow inactivation of this current suggests that a delayed-rectifier type of K+ current is present [19].

The threshold for activating Ito was very close to its half-inactivation voltage (− 58 mV), indicating that there might be a ‘window’ range of membrane voltages where there is a steady-state activation of voltage-gated K+ channels. In the intact guinea-pig prostate smooth muscle showing spontaneous potential changes, the mean value of the resting membrane potential is reportedly −50 mV [11]. Based on the present results, it is intriguing that the window-current of Ito might contribute to forming the resting membrane voltage in GPSMCs. In neuronal cells where diverse A-type K+ channels are expressed, A-type K+ channels operate transiently below the threshold range of action potential and serve as a damper in the inter-spike interval, to separate successive action potentials [19]. Because of its rapid inactivation, a moderate depolarization by intrinsic or extrinsic signals may reduce the available population of Ito channels and regulate the excitability of GPSMCs.


The STOCs in GPSMCs are sensitive to a low concentration of TEA or iberiotoxin, indicating the functional expression of BKCa channels; the latter are activated by both cytosolic Ca2+ and by membrane depolarization. Therefore, the BKCa channels of GPSMCs would be crucial in the repolarizing phase of spontaneous voltage changes. The role of KCa channels in determining resting membrane voltages has also been emphasized, especially in relation to the physiological role of STOCs.

In single GPSMCs held at depolarized potentials, STOCs were generally, although not invariably, present. It has been suggested that STOCs represent the release of stored calcium in relation to the internal surface of the membrane, so causing a temporary and localized rise in [Ca2+] and the simultaneous opening of up to 100 KCa channels [13]. The short duration of an individual STOC (< 100 ms) presumably reflects that calcium ions are released only transiently from the store and soon diffuse away or are removed by various mechanisms [13,14]. Under conditions of recording STOCs, a high concentration of caffeine induced a large Ito, which presumably indicates the massive release of calcium from stores in relation to KCa channels.

STOCs have been used as an assay of localized Ca2+ release in SMCs [13,14], and quantitative analysis has shown a close correlation between Ca2+ spark amplitude and amplitudes of STOCs [20]. In vascular SMCs the selective inhibition of STOCs induces membrane depolarization and subsequent contractile responses [21]. A rare patch-clamp study using prostate SMCs emphasized the role of KCa channels; Kurokawa et al.[22] suggested that, on stimulating with phenylephrine, a concomitant activation of KCa channels and release of stored Ca2+ could attenuate the contractile response. Moreover, the protein kinase A-dependent phosphorylation positively regulated the activity of BK channels in human prostate smooth muscle [23].

In summary, the present study is the first to report the voltage-activated ion-channel currents of GPSMCs; it is likely that activating l-type Ca2+ channels and the Ito, the delayed-rectifier type of K+ current and IK,Ca are responsible for the upstroke depolarization and re-polarization of spike potentials in the guinea pig prostate. As the fluctuation of membrane voltage and the superimposed action potentials are fundamental events regulating the contractility of prostatic stroma, precise knowledge of ion channels underlying such electrical events would provide a better understanding of the physiology of the prostate.


This research was supported by a grant (No. PF002108-03) from Plant Diversity Research Center of 21st Century Frontier Research Program funded by Ministry of Science and Technology of Korean government. Also, we greatly appreciate So-Young Lee for her technical assistance.