A description of Ca2+ channels in human detrusor smooth muscle

Authors


C.H. Fry, Institute of Urology, 48 Riding House St, London W1W 7EY, UK.
e-mail: c.fry@ucl.ac.uk

Abstract

OBJECTIVE

To characterize the Ca2+ channels in human detrusor smooth muscle and to investigate their contribution to spontaneous electrical activity.

MATERIALS AND METHODS

Isolated human detrusor smooth muscle myocytes were used to measure ionic currents under voltage-clamp or membrane potential under current-clamp. Membrane potential oscillations were analysed in terms of oscillation frequency and amplitude using fast Fourier transforms.

RESULTS

Under voltage-clamp an inward current dependent on extracellular Ca2+ was recorded using Cs+-filled patch electrodes. The current could be separated into two components on the basis of their sensitivity to Ni2+, verapamil or nicardipine, and their dependence on holding and clamp potential. A Ni2+-sensitive component activated over a relatively negative range of potentials (−60 to −20 mV) comprised about a third of the total current and was designated a T-type Ca2+ current. A verapamil/nicardipine-sensitive component, activated at more positive potentials, was designated an l-type Ca2+ current. Using K+-based filling solutions spontaneous transient outward currents were recorded that had the characteristics of current flow through BK channels. Membrane potential oscillations, under current-clamp increased in frequency but not amplitude as the mean membrane potential was made less negative. The voltage-dependence of oscillation frequency was similar to that of the l-type, but not T-type, Ca2+ current activation curve. Furthermore oscillation frequency was slowed by verapamil but not Ni2+.

CONCLUSION

The study showed, for the first time, the presence of both T- and L-type Ca2+ channels in human detrusor smooth muscle; we propose a role for these channels in spontaneous activity. The results suggest that the L-type Ca2+ current can control membrane potential oscillation frequency. The significance of this finding for spontaneous contractions is discussed.

INTRODUCTION

In the human bladder contractile activation is mediated by the release of acetylcholine from motor nerves. Acetylcholine binds to muscarinic (M3) receptors on the smooth muscle membrane and subsequently generates the soluble second-messenger inositol trisphosphate which releases Ca2+ from intracellular Ca2+ stores [1]: the process is independent of any changes in membrane potential [2]. However, Ca2+ channels are involved in the overall regulation of contraction, as channel activity is required to fill the intracellular Ca-stores when the muscle is relaxed [3]. Moreover, in animal detrusor and human detrusor from some pathologies, Ca2+ channels have a more direct role via a second functional neurotransmitter, ATP [4]. ATP binds to ionotropic P2X1 receptors that depolarize the cell, and thus increase Ca2+ channel opening and Ca2+ influx [3].

Initial studies on human detrusor showed Ca2+ channel activity in cells obtained from both stable and unstable bladders. The current was considered to flow through L-type Ca2+ channels, as it was largely abolished by Ca-antagonists such as verapamil and nifedipine [5]. However, it was recently reported that a second class of channel is present in guinea-pig detrusor, that is activated at more negative potentials, i.e. a T-type Ca2+ channel [6]. The implications of this observation are profound; any attempt to reduce Ca2+ channel activity using Ca-antagonists, and thus influence contractile function, by using conventional Ca-antagonists may be only partly successful. Second, the presence of a Ca2+ influx route operating at more negative resting potentials changes the understanding of how detrusor contractile function is regulated. It was therefore considered important to determine if T-type Ca2+ channels are present in human detrusor and their influence if any on cellular electromechanical coupling.

MATERIALS AND METHODS

In the experiments we used freshly isolated cells from human detrusor biopsies. Detrusor samples were obtained either at cystectomy for bladder carcinoma or by cold-cup biopsy forceps from patients with no symptoms of bladder overactivity. Patients who had undergone radiotherapy were excluded. Samples were obtained after obtaining informed patient consent and the approval of the local Hospital Ethical Committee. Biopsies were placed in Ca2+-free HEPES-containing solution (below), the cells prepared by collagenase disruption of the biopsy specimen or from previous cell culture [7], and stored in Ca2+-containing HEPES-Tyrode's solution before experiments.

The Ca2+-free HEPES-containing solution contained (mmol/L): NaCl, 105.4; NaHCO3, 22.3; KCl, 3.6; MgCl2.6H2O, 0.9; NaH2PO4.2H2O, 0.4; HEPES, 19.5; glucose, 5.4; Na pyruvate, 4.5; when required CaCl2.2H2O, (1.8 mmol/L) was added. The solution was buffered to pH 7.1 with NaOH. For experimental recording, cells were superfused with (mmol/L): NaCl, 118; NaHCO3, 24.0; KCl, 4.0; CaCl2.2H2O, 1.8; MgCl2.6H2O, 1.0; NaH2PO4.2H2O, 0.4; glucose, 6.1; Na pyruvate, 5.0, gassed with 95% O2−5% CO2, pH 7.35. Verapamil, nicardipine and NiCl2 were stored as aqueous stock solutions at concentrations 100–1000 times that used experimentally.

Membrane potentials and voltage-clamp recording were made with patch-type electrodes filled with a high-EGTA, Cs+-containing solution (mmol/L): CsCl, 20; aspartic acid, 110; MgCl2.6H2O, 5.45; Na2ATP, 5.0; Na4GTP.2H2O, 0.1; EGTA, 5.0; HEPES, 5.0; pH 7.2, adjusted with CsOH. To record membrane potentials and outward currents a high-K solution was used, KCl replacing CsCl, and the pH adjusted with KOH. All recordings were made at 37 °C, with reference to a 3 mol/L KCl liquid junction. An Axopatch 1-D was used as an analogue interface with the patch-electrode with a Digidata A-D converter (Axon Instruments, Foster City, CA, USA). Cell membrane capacitance, c m was measured under current-clamp mode.

In the resting state the membrane potential was maintained at − 60 mV under voltage-clamp. Ionic currents were activated with different depolarizing step pulses from holding potentials, V h, set previously for at least 2 s. Inward current magnitude, I, was measured as the difference between that at peak activation and at the end of the clamp step. Activation curves were generated from the chord conductance, g, of the current-voltage (i-v) relationship (= I/(V m – E r)) where V m is the voltage of the test step and E r the reversal potential of the current. The curve was fitted to equation 1:

d = g/g max= 1/[1 + exp((V0.5Vm)/k)]

V0.5 is the voltage required for half-maximum activation and k is a constant, the slope factor (mV) and g max the maximum slope conductance at positive potentials. The voltage-dependence of current availability (the inactivation curve) was obtained by preconditioning the cell to different potentials (see text). The curve was fitted to equation 2

f = l/Imax= 1/[1 + exp((VmV0.5)/k)]

where I max is the test pulse current evoked by preconditioning at the most negative potentials. d and f are dimensionless variables that vary between 0 and 1.

Action potentials were elicited by passing depolarizing pulses (5–40 pA) under current-clamp; the action potential threshold voltage, V th, was measured as described previously [7].

To measure membrane potential oscillation amplitude and frequency, digitized membrane potential records (sampling frequency 1 kHz) were subjected to fast Fourier transform analysis [8,9] using appropriate software. Power vs frequency spectra were used to calculate the mean geometric frequency and amplitude of oscillations with power = 10 mV2; this corresponds to finding the mean frequency of oscillations in a data record with amplitudes of > ≈ 3 mV.

All interventions were preceded and followed by exposure to normal Tyrode's solution and the mean value of the variable before and after intervention used as a control value. All values are expressed as the mean (sd). Changes to a variable in a test solution were compared with the control value using Student's t-test. The null hypothesis was rejected if P < 0.05.

RESULTS

SEPARATION OF L- AND T-TYPE CA2+ CHANNELS

Inward current in all detrusor cells was deemed to be carried by Ca2+ as it was abolished completely in zero-Ca solution. Figure 1A shows examples of inward current elicited in a human detrusor cell by depolarization from a holding potential, V h, of − 100 mV to either −30 or + 10 mV. The peak current was larger when the cell was depolarized to + 10 than −30 mV. In addition, when depolarized to − 30 mV the current was completely abolished by 0.1 mmol/L NiCl2, but was only slightly reduced at + 10 mV. NiCl2 at this concentration has been shown to block selectively T-type Ca2+ currents in guinea-pig detrusor [6] and in other cell types [10]. The large residual current when depolarized to + 10 mV was blocked completely also by adding verapamil (20 µmol/L) or nifedipine (5 µmol/L) to the superfusate. Therefore the Ni2+-sensitive current is regarded as a T-type Ca2+ current. Figure 1B shows a similar pair of current traces when the cell was depolarized to either − 30 or + 10 mV, when in this case 5 µmol/L nifedipine was added initially to the superfusate. The current on depolarization to − 30 mV was now little affected by nifedipine, but that evoked by depolarization to + 10 mV was significantly attenuated. Residual currents were blocked by 0.1 mmol/L NiCl2. The nifedipine-sensitive current is regarded as an l-type Ca2+ current.

Figure 1.

Separation of T- and L-type Ca2+ currents in human detrusor smooth muscle cells. A, A myocyte was voltage-clamped from a holding potential of − 100 mV to either − 0 mV (left) or + 10 mV (right). In each case two records are superimposed, one in the absence and the other in the presence of 0.1 mmol/L NiCl2. B, A similar protocol with a cell clamped to either − 30 or + 10 mV in the absence or presence of 5 µmol/L nifedipine.

KINETICS OF L- AND T-TYPE CA2+ CURRENT

In the following experiments cells were held at a potential of − 100 mV and experiments carried out in the presence of either 0.1 mmol/L NiCl2, to elicit L-type Ca2+ current, or 5 µmol/L nifedipine, to elicit T-type Ca2+ current. Figure 2A shows the current-voltage (i-v) relationships for the L- and T-type Ca2+ currents in human detrusor myocytes. The T-type current was activated over a more negative range of potentials and was smaller in peak amplitude than the L-type current. For the T-type current a maximum value was recorded at −10 mV, with a value of 1.0 (0.6) pA/pF (n = 11); for the L-type current a maximum value was recorded at + 10 mV, of 3.7 (2.1) pA/pF (n = 18). Thus the maximum T-type current is ≈ 21% of total inward Ca2+ current; the unit of current (pA/pF) is effectively normalized to current per unit area of membrane, as specific membrane capacitance (C m= 1 µF/cm2[11]) is assumed to be constant; actual cell membrane capacitance, c m, was measured when the whole-cell clamp was initially established before experimental recording.

Figure 2.

Figure 2.

A Current-voltage relationships of the T- and L-type Ca2+ currents in human detrusor smooth muscle. The current axis is normalized to unit cell capacitance (area). B,C Activation (closed red circles) and inactivation (open green circles) curves for the T-type Ca2+ current (B) and the L-type Ca2+ current (C). The lines were fitted to equations 1 and 2 in the Methods. The points are the mean (sd). The area of overlap of the two curves is denoted as the ‘window’ current.

Figure 2.

Figure 2.

A Current-voltage relationships of the T- and L-type Ca2+ currents in human detrusor smooth muscle. The current axis is normalized to unit cell capacitance (area). B,C Activation (closed red circles) and inactivation (open green circles) curves for the T-type Ca2+ current (B) and the L-type Ca2+ current (C). The lines were fitted to equations 1 and 2 in the Methods. The points are the mean (sd). The area of overlap of the two curves is denoted as the ‘window’ current.

Figure 2.

Figure 2.

A Current-voltage relationships of the T- and L-type Ca2+ currents in human detrusor smooth muscle. The current axis is normalized to unit cell capacitance (area). B,C Activation (closed red circles) and inactivation (open green circles) curves for the T-type Ca2+ current (B) and the L-type Ca2+ current (C). The lines were fitted to equations 1 and 2 in the Methods. The points are the mean (sd). The area of overlap of the two curves is denoted as the ‘window’ current.

Figure 2 shows the voltage-dependence of activation (B) and inactivation (C) for the T- and L-type Ca2+ currents. The activation curve is a measure of the voltage range over which the Ca2+ current is activated upon depolarization; the inactivation curve expresses the proportion of current elicited upon depolarization from different resting potentials. For the T-type current both curves lie at more negative potentials than the corresponding ones for the L-type Ca2+ current. Quantitatively, the V0.5 for the T-type activation and inactivation curves were − 36.9 (7.4) and − 64.8 (10.9) mV, respectively, and for the L-type were −3.4 (5.2) and −24.8 (5.2) mV, respectively. The two curves for both the T- and L-type currents show a region of overlap, the so-called ‘window’ current. Over the potentials encompassed by the window current, channels will be partly activated and not fully inactivated, so that a small but constant current can flow. For the T-type current this window range of potentials is near to the average resting potential (see below) and so provides a possible mechanism for Ca2+ influx at such potentials.

MEMBRANE POTENTIAL OSCILLATIONS AND THE ROLE OF CA2+ CURRENTS

Oscillations of membrane potential have been recorded in detrusor from several animal species [12] and implicated in the basis of spontaneous contractile activity. The origin of such oscillations is unclear but one possibility is an interaction between inward Ca2+ currents and Ca2+-activated outward currents. By this mechanism an inward Ca2+ current depolarizes the cell, raises the intracellular [Ca2+] and activates outward K+ currents, thus repolarizing the cell. Repolarization reduces intracellular [Ca2+] and thus the cycle begins afresh. Figure 3 shows examples of spontaneous transient outward currents (STOCs), elicited after generating inward Ca2+ currents. In the upper panel, inward current was through T-type channels as the level of depolarization would be insufficient to elicit L-type Ca2+ currents. In the lower panel the larger inward current was contributed by L- and T-type currents; the larger the inward current, the greater was STOC frequency.

Figure 3.

STOCs evoked by depolarization of a cell under voltage-clamp from −80 to either −50 mV (top) or 0 mV (bottom). The STOCs are evident as upward deflections on the current trace. The initial inward current is arrowed in each trace.

To examine this hypothesis further the membrane potential oscillations were analysed quantitatively in the absence and presence of Ca2+-channel blockers. Data were obtained from a total of 17 cells with a mean resting membrane potential of −60.7 (4.7) mV. Figure 4A shows an example of membrane potential recordings in an isolated cell in which the membrane potential was artificially adjusted to different values by passing current through the recording patch electrode into the cell; the amplitude and frequency of oscillations was recorded over a period of 60–200 s. Figure 4B shows that the amplitude of oscillations was not significantly dependent on the membrane potential, but that the frequency increased as the cell was depolarized, especially at values less negative than −40 mV. The data points from the traces shown in Fig. 4A are shown by the large squares, rather than circles. The solid line through the data points is the activation curve for the L-type Ca2+ current (Fig. 2B) and corresponds closely to the data points, and the dotted line is the corresponding curve for the T-type Ca2+ current. The line has been vertically scaled but the slope and position on the voltage axis has not been altered. The significance of this correspondence is considered later.

Figure 4.

A, Recording of membrane potential in a human detrusor smooth muscle. Current was passed into the cell during the periods shown. B, The relationship between membrane oscillation frequency and mean membrane potential. The two lines are the activation curves for the T-type Ca2+ current (green dashed) and the l-type Ca2+ current (red solid) as shown in Fig. 2 and vertically scaled to the highest frequency value. The four red closed squares are the values from the experiment shown in A.

To gain insight into the role of T- and L-type Ca2+ currents in contributing to these oscillations, experiments were also carried out in the presence of 0.2 mmol/L NiCl2 or 20 µmol/L verapamil; the former was included especially when the membrane potential was in the range − 60 to − 20 mV, as this encompassed the activation range of the T-type Ca2+ current. NiCl2 had no significant effect on the frequency of the membrane potential oscillations, with values of 0.279 (0.022) without and 0.366 (0.085) Hz with NiCl2, and neither did it affect the amplitude of oscillations. The effect of verapamil was assessed at potentials of − 20 to − 5 mV, when the L-type Ca2+ current would be activated. In this case the frequency of oscillations was significantly lower, at 0.309 (0.012) with and 1.013 (0.022) Hz without verapamil (P < 0.05), and similar to that obtained at more negative potentials. These results show that depolarization to ≥− 40 mV increases the chance of generating larger membrane potential fluctuations, and that Ca2+ entry through L-type channels is crucial in this process.

In two experiments it was possible to switch from current- to voltage-clamp to record membrane potential oscillations, and then STOCs. The STOC frequency was ≈ 10 times greater than membrane potential oscillation frequency, indicating that STOCs were not always effective in generating a membrane potential oscillation.

DISCUSSION

These experiments show two types of Ca2+ current in human detrusor, one activated at relatively positive membrane potentials, the L-type current, and one at more negative membrane potentials, near the resting membrane potential of detrusor myocytes. The significance of this finding is two-fold: first, any attempt to block Ca2+ current in this tissue must recognize that specific channel inhibitors will be only partly effective; second, the presence of a route for cellular Ca2+ influx near the resting membrane potential will have implications for understanding how detrusor contractile function is regulated.

Inward Ca2+ currents have no direct role in contractile activation by acetylcholine, as applying muscarinic agonists increases the intracellular [Ca2+] with no simultaneous alteration to membrane potential [2]. However, Ca2+ influx through voltage-activated channels refills intracellular stores at rest and thus replenishes them for subsequent activity [3]. The average resting membrane potential of guinea-pig detrusor myocytes is ≈ −60 mV [7], but there was considerable spontaneous variation of − 70 to − 40 mV, similar to the observations here using human detrusor myocytes from stable bladders. This range of membrane potentials is too negative to permit significant steady-state Ca2+ influx through L-type channels, as the window current range on the i-v relationship (Fig. 2) is more positive. However, the T-type current window current is within this range and provides a mechanism for Ca2+ influx. The fraction of total maximum inward current (T-type/(L-type + T-type)) is about 0.21, similar to that reported in guinea-pig cells (0.26) [6] and suggests that this is a significant route of transmembrane Ca2+ flux in this tissue. The dual nature of Ca2+ currents in detrusor also partly explain the failure of L-type Ca2+-channel antagonists to abolish completely contractile activity [13].

Membrane potential recordings showed considerable spontaneous fluctuations that increased in frequency, but not mean amplitude, as the mean resting potential became less negative. Such fluctuations occasionally elicited larger depolarizing spikes but could not be called action potentials, as peak amplitudes were not constant and only rarely exceeded 0 mV. Interest in such oscillations arises as they could give rise to spontaneous muscle contractions and thus may contribute to resting detrusor tone, or uncontrolled bladder contractions if they were particularly large.

One hypothesis of potential oscillations is an interaction between inward Ca2+ current and outward current activated by a rise of the intracellular [Ca2+]; the latter is generally through Ca2+-activated K+ channels (e.g. BK channels). BK channel activation manifests as STOCs and these were recorded in the present experiments (Fig. 3) after T-type and L-/T-type Ca2+ channel activation. The detection of STOCs in human detrusor corroborates their description in other species [7,14,15]. The hypothesis is strengthened by quantitative calculations, the input resistance, R inp, of resting isolated human detrusor cells is ≈ 2 GΩ[7], so that a current of 10 pA would cause a membrane potential fluctuation of up to 20 mV. Although this is an upper estimate, as R inp would be less during ion current activation, these values are certainly in the range of STOC amplitude and E m fluctuations. These observations show that all elements of the oscillator are sufficient in human detrusor myocytes.

The question arises whether T- and/or L-type Ca2+ channel activity is significant in providing the Ca2+ influx arm of the oscillator. At more negative potentials, when T-type current is activated, oscillation frequency and STOC frequency were finite, albeit low; at less negative potentials both frequencies were increased. Several lines of evidence in these results suggest that L-type channel Ca2+ influx is important. Oscillation frequency increased at a range of potentials similar to those when L-type channels are activated. This is especially evident in the line through the data of Fig. 4B, which is the activation curve of the L-type Ca2+ current derived in Fig. 2B, scaled vertically but not altered along the voltage axis. Thus the oscillation frequency increased over the same range of voltages as L-type Ca2+ channel activation. The importance of these channels is strengthened because verapamil reduced the frequency of oscillations to the baseline (negative membrane potential) value when the membrane was depolarized to −20 to −5 mV. Conversely, NiCl2 had no significant effect on the baseline frequency over the range of membrane potentials of − 65 to −25 mV, when T-type Ca2+ channels would be activated, and thus suggests a less significant role for them. It is therefore unclear if oscillatory electrical activity forms the basis of spontaneous contractions, as the latter are effectively attenuated by both L- [16] and T-type channel blockers [17].

However, spontaneous contractions are especially evident in detrusor under conditions when the intracellular [Ca2+] might be raised, and thus may be a feature of tissue exposed to relatively abnormal conditions. Hence they are exaggerated under conditions such as hypoxia, acidosis and application of cardiac glycosides [18,19], and are a feature of detrusor from rodents, e.g. rats and ferrets, in which the intracellular [Na+] (and hence via Na+-Ca2+ exchange) the intracellular [Ca2+] is raised [20,21]. It remains to be determined if and how spontaneous contractions contribute to increased mechanical activity in the overactive bladder.

ACKNOWLEDGEMENTS

We are grateful to St Peter's Trust for financial assistance

Abbreviations
STOC

spontaneous transient outward current.

Ancillary