The effects of extracellular and intracellular pH on intracellular Ca²⁺ regulation in guinea-pig detrusor smooth muscle

Abstract

1. Intracellular pH (pH_i) and intracellular [Ca²⁺] ([Ca²⁺]_i) were measured during changes to superfusate P_CO2 and/or [NaHCO₃]. Changes to superfusate P_CO2 produced sustained changes to pH_i and [Ca²⁺]_i, while changes to [NaHCO₃] altered only extracellular pH (pH_o).

2. Carbachol or caffeine induced a transient rise of [Ca²⁺]_i due to Ca²⁺ release from an intracellular store. This Ca²⁺ transient was reduced by extracellular acidosis, but increased by intracellular acidosis. Alkalosis in either compartment produced opposite effects to acidosis. Changes to the Ca²⁺ transient mirrored those to phasic tension previously reported in this preparation.

3. A raised superfusate [K⁺] also induced a Ca²⁺ transient, due to transmembrane influx of Ca²⁺. This transient was depressed by extracellular acidosis, but unaffected by changes to pH_i. The L-type Ca²⁺ current was similarly affected by changes to pH_o, but not by alteration of pH_i.

4. The results suggest that extracellular acidosis depresses the Ca²⁺ transient by reducing transmembrane influx through the L-type Ca²⁺ channel. The increase in the carbachol- and caffeine-induced Ca²⁺ transients by intracellular acidosis is due to enhancement of Ca²⁺ uptake into intracellular stores as a result of a raised resting [Ca²⁺]_i.

Detrusor smooth muscle can experience large variations of extracellular pH (pH_o) and intracellular pH (pH_i) under various conditions. For example, CO₂ can diffuse readily across the urothelium and urinary P_CO2 can rise to values as high as 20 kPa (Pitts, Ayers & Scheiss, 1948), the urothelium can be damaged and rendered permeable to the acid urine (Eldrup, Therup, Nielsen, Hald & Hainau, 1983), and H⁺ can be accumulated in the bladder wall during ischaemia caused by conditions such as systemic hypotension or during both bladder filling and spontaneous contractions, particularly in the obstructed bladder (Bellringer, Ward & Fry, 1994; Azadzoi, Pontari, Vlachiotis & Siroky, 1996; Batista, Wagner, Azadzoi, Krane & Siroky, 1996).

Changes to the extracellular and intracellular pH have profound and unusual effects on the contractile function of detrusor smooth muscle. Extracellular acidosis, from a normal pH value of about 7.4, depresses electrically elicited phasic contractions; however, intracellular acidosis, from a normal pH value of about 7.1, increases the magnitude of such contractions (Liston, Palfrey, Raimbach & Fry, 1991). Alkalosis in these two compartments exerts opposite effects to those of acidosis. Such effects have been observed in both human detrusor and tissue from several animal preparations, including the guinea-pig. In the case of extracellular pH changes, altered contractility is mediated by depression of the L-type Ca²⁺ channel (Fry, Gallegos & Montgomery, 1994a). However, the mechanisms whereby intracellular acid-base changes exert their effects on the force of contraction are unknown and form the subject of this paper.

In striated muscle, several steps in excitation-contraction coupling are altered by changes to intracellular pH; in particular, changes to the myofilament sensitivity for Ca²⁺ contribute significantly to pH-dependent inotropic effects (Orchard & Kentish, 1990). However, we have recently shown that changes to myofilament Ca²⁺ sensitivity in detrusor muscle cannot explain fully the inotropic effects of changes to intracellular pH (Wu, Kentish & Fry, 1995), which suggests that modulation of the availability of intracellular Ca²⁺ could be a factor. Phasic detrusor contractions are initiated predominantly by cholinergic motor nerves embedded in the muscle mass, mediated by a transient rise in intracellular [Ca²⁺] ([Ca²⁺]_i), but independent of changes to the membrane potential (Wu & Fry, 1997). However, in many small animals and in human detrusor from some abnormally functioning bladders there is an additional, probably purinergic component which raises [Ca²⁺]_i via membrane depolarization and activation of the L-type Ca²⁺ current, I_Ca (Sibley, 1984; Inoue & Brading, 1991; Wu, Wallis & Fry, 1995; Bayliss, Wu & Fry, 1997). The present study has investigated whether changes to extracellular and intracellular pH could modulate such Ca²⁺ transients and where the sites of action of H⁺ are located in the cell.

The results show that intracellular acidosis increases the magnitude of the intracellular Ca²⁺ transient evoked by the cholinergic agonist carbachol, and that such an action is most probably mediated via an enhancement of the Ca²⁺ sequestration rate by the sarcoplasmic reticulum. Intracellular pH changes had no effect on the rate of Ca²⁺ movement across the surface membrane.

METHODS

Preparation

Single detrusor smooth muscle cells were isolated from guinea-pig urinary bladders. Guinea-pigs (500–900 g) were killed by cervical dislocation, and small detrusor strips were cut from the dome of the bladder and placed in Ca²⁺-free solution containing (mm): NaCl, 105.4; NaHCO₃, 20.0; KCl, 3.6; MgCl₂.6H₂O, 0.9; NaH₂PO₄.2H₂O, 0.4; glucose, 5.5; sodium pyruvate, 4.5; and Hepes, 4.9, pH 7.1. Isolated detrusor myocytes were prepared with a collagenase-based mixture as described previously (Montgomery & Fry, 1991).

Solutions

Detrusor myocytes were superfused at 37°C with Tyrode solution gassed with 95 % O₂- 5 % CO₂ (pH 7.38 ± 0.01, n= 56) and of the following composition (mm): NaCl, 118; NaHCO₃, 24.0; KCl, 4.0; MgCl₂.6H₂O, 1.0; NaH₂PO₄.2H₂O, 0.4; CaCl₂, 1.8; glucose, 6.1; sodium pyruvate, 5.0. Ca²⁺-free superfusates contained no added CaCl₂ and 0.1 mm EGTA. To change separately intracellular pH (pH_i) and extracellular pH (pH_o), the Tyrode solution was modified by altering [NaHCO₃] and the percentage of CO₂ in the gas mixture. pH_o was altered by varying [NaHCO₃] between 9.6 and 48 mm at constant P_CO2 (5 % CO₂, 4.6 kPa). pH_i was varied by changing, in proportion, [NaHCO₃] and P_CO2; NaHCO₃ was varied between 9.6 and 48 mm and CO₂ was varied between 2 % (1.8 kPa) and 10 % (9.2 kPa). Values of pH_o and pH_i in these solutions are shown in Table 1. Because Ca²⁺ and HCO₃⁻ form ion pairs in solution, changes to [NaHCO₃] necessitated changes to [CaCl₂] in these superfusates to maintain constant the ionized calcium concentration (Fry & Poole-Wilson, 1981). Solutions containing 9.6 and 48 mm NaHCO₃ contained 1.55 and 2.34 mm CaCl₂, respectively, to maintain the same ionized calcium concentration as was present in Tyrode solution (24 mm NaHCO₃, 1.8 mm CaCl₂), and were checked with a Ca²⁺-selective electrode.

Table 1.

Values of pH_o and pH_i in guinea-pig detrusor myocytes by variation of superfusate [NaHCO₃] (between 9.6 and 48 mm) and P_CO2 (between 1.8 and 9.2 kPa)

	24 mm, 4.6 kPa	48 mm, 9.2 kPa	9.6 mm, 1.8 kPa	48 mm, 4.6 kPa	9.6 mm, 4.6 kPa
pHo	7.38 ± 0.01 (56)	7.39 ± 0.01 (17)	7.39 ± 0.01 (19)	7.70 ± 0.02 (13)*	7.02 ± 0.01 (9)*
pHi	7.10 ± 0.03 (14)	6.98 ± 0.03 (7)*	7.24 ± 0.07 (7)*	7.06 ± 0.04 (6)	See text

The particular values of the variables are given at the head of each column.

^*pH_o or pH_i significantly different (P < 0.01) from the value in normal Tyrode solution (24 mm HCO₃⁻, 4.6 kPa CO₂). Numbers of myocytes (pH_i) or number of measurements (pH_o) are shown in parentheses.

Measurement of pH_i and [Ca²⁺]_i

pH_i and [Ca²⁺]_i were measured by epifluorescence microscopy using BCECF (2′,7′-bis(carboxyethyl)-5(6′)-carboxyfluorescein) and fura-2 (Calbiochem). Cells were loaded with acetoxymethyl (AM) esters of the indicators (4–5 μm, 30 min at room temperature, 20–23°C) then superfused at 37°C with Tyrode solution in a chamber on the stage of an inverted microscope. In voltage-clamp experiments, acid forms of the indicators (100 μm) were dialysed into cells via patch pipettes. Cells were illuminated alternately at 510/430 nm or 340/380 nm and emitted light was collected between 530 and 580 nm for BCECF or between 400 and 510 nm for fura-2. BCECF signals were calibrated to pH_i values with 10 μm nigericin (Sigma) in a high-K⁺ medium between pH 4.0 and 9.0 (Eisner, Nichols, O'Neill, Smith & Valdeolmillos, 1989). Similarly fura-2 signals were converted to [Ca²⁺]_i values by the equation (Grynkiewicz, Poenie & Tsien, 1985):

R is the fluorescence ratio at 340/380 nm excitation, ‘min’ and ‘max’ subscripts refer to values at zero and saturating [Ca²⁺], respectively; b is the fluorescence ratio at zero and saturating [Ca²⁺] at 380 nm alone; K_d (224 nm) is the dissociation constant for Ca²⁺ with fura-2. It is assumed that K_d is independent of pH_i over the range of measured values (see Discussion).

The peak values of intracellular Ca²⁺ transients were measured when cells were stimulated by brief application of solutions containing 10 μm carbachol, 44 mm KCl or 10 mm caffeine. Preliminary experiments showed that exposures to carbachol and caffeine for 20–30 s and to KCl for 40 s were sufficient to elicit a maximal intracellular Ca²⁺ transient.

Electrophysiolgical recordings

Ca²⁺ currents (I_Ca) through L-type Ca²⁺ channels were measured using patch-type electrodes in whole-cell configuration (Axopatch-1D; Axon Instruments). An A/D converter (Digidata 1200; Axon Instruments) applied voltage steps and recorded membrane currents with a sampling frequency of 4 kHz and cut-off frequency of 2 kHz; voltage-clamp protocols were supported by pCLAMP software (Axon Instruments). [Ca²⁺]_i was measured simultaneously with the indicator fura-2. Patch electrodes (3–5 MΩ) were filled with a solution containing (mm): CsCl, 20; aspartic acid, 110; MgCl₂, 5.45; Na₂ATP, 5.0; and Hepes, 5.0, pH 7.1 adjusted with CsOH. Preliminary experiments showed that 100 μm K₅-fura-2 was required to achieve a sufficient signal-to-noise ratio and 50 μm EGTA to prevent run-down of the Ca²⁺ current, while not attenuating significantly the Ca²⁺ transient.

Statistics

Numerical data are quoted as means ±s.e.m., since experimental observations were generally repeated on the same cell. Data sets from single observations per cell are quoted as means ±s.d., and are identified specifically in the text. Student's two-tailed paired or unpaired t tests were used to test for significance between sets of data and the null hypothesis was rejected when P < 0.05.

RESULTS

pH_i and resting [Ca²⁺]_i of guinea-pig detrusor myocytes

pH_i in myocytes superfused with normal Tyrode solution was 7.10 ± 0.03 (n= 14; Table 1, column 1). The table shows that pH_i could be altered independently of pH_o, when superfusate [NaHCO₃] and P_CO2 were varied in proportion, and also that changes to [NaHCO₃] alone had no effect on pH_i. Note that a previous study has shown that pH_i was unaffected when superfusate [NaHCO₃] was reduced from 24 to 6 mm at constant P_CO2 (Fry, Gallegos & Montgomery, 1994b). The changes to pH_i during alteration of superfusate P_CO2 were rapid and sustained for the period of the intervention and recovered completely on return to control superfusate.

Resting [Ca²⁺]_i in undialysed myocytes in Tyrode solution was 77 ± 41 nm (n= 254), which is consistent with a threshold value of about 100 nm for activation of detrusor myofibrils (Wu et al. 1995). The values of the resting [Ca²⁺]_i were not significantly different from a normal distribution (Kolmogorov-Smirnov test, d= 0.078, P > 0.05). A decrease in pH_i by, on average, 0.12 units (Table 1) raised the mean value of [Ca²⁺]_i by 12 ± 1 nm (P < 0.01, n= 50) and an increase in pH_i by, on average, 0.14 units decreased [Ca²⁺]_i by 23 ± 6 nm (P < 0.01, n= 33). Extracellular pH changes had no significant effects on resting [Ca²⁺]_i values.

Carbachol-induced intracellular Ca²⁺ transients - the effects of changes to pH_o and pH_i

Acetylcholine, and its analogue carbachol induce contraction in detrusor smooth muscle (Sibley, 1984). Figure 1A shows that in single detrusor myocytes repeated brief (20–40 s) applications of carbachol (10 μm) induced transient increases of [Ca²⁺]_i - Ca²⁺ transients. The transients reached a maximum value within 2–3 s before relaxing, despite the continuous presence of carbachol. In the majority of cells (75 %, n= 150), the relaxation phase showed an undershoot before return to the baseline value. The magnitude of the Ca²⁺ transient was dependent on the carbachol concentration, with half-maximal activation at 1.2 ± 0.6 μm (mean ±s.d., n= 5) and maximum activation at 10 μm.

Figure 1. Intracellular Ca²⁺ transients induced by carbachol.

Properties of the carbachol-induced Ca²⁺ transient. A, repeated brief applications (filled bars) of 10 μm carbachol (carb) gave consistent Ca²⁺ transients. B, prior application of 1 μm atropine (Atr) blocked the Ca²⁺ transient. C, prior application of 10 mm caffeine (caff) blocked the Ca²⁺ transient. D, prior application of 5 μm nicardipine (nicard) had no effect on the Ca²⁺ transient. E, a Ca²⁺ transient during superfusion with a zero-Ca²⁺ solution.

Several other features of the carbachol-induced Ca²⁺ transients are also illustrated in Fig. 1. Panel B shows that the Ca²⁺ transient was abolished by prior application of the muscarinic receptor antagonist atropine (1 μm). Panels C and D show that the Ca²⁺ transient was also abolished by prior treatment with caffeine (10 mm) but that 30 or 60 s pre-treatment with the L-type Ca²⁺ channel antagonist nicardipine (5 μm) had no effect. Finally, panel E shows that such transients could still be evoked in a nominally Ca²⁺-free superfusate, although the magnitude was reduced - note also a reduction in resting [Ca²⁺] in the Ca²⁺-free solution. These observations suggest that the source of Ca²⁺ for the Ca²⁺ transient is an intracellular store, probably the sarcoplasmic reticulum.

Figure 2 shows the effect of altering pH_o and pH_i on the magnitude of the carbachol-induced Ca²⁺ transient. Panels A and B show the effect of acidosis. Panel A shows that intracellular acidosis at constant pH_o, achieved by increasing the superfusate P_CO2 and [NaHCO₃], increased the magnitude of the Ca²⁺ transient to 121 ± 4 % of control (mean ±s.e.m., P < 0.01, n= 50). However, extracellular acidosis (B) had the converse effect, namely it attenuated the Ca²⁺ transient magnitude to 83 ± 5 % of control (P < 0.01, n= 11).

Figure 2. The effects of changes to pH_o and pH_i on the carbachol-induced Ca²⁺ transient.

Cells were briefly exposed to 10 μm carbachol and either pH_o or pH_i was changed by altering the superfusate P_CO2 and/or [HCO₃⁻]. A, intracellular acidosis (48 mm HCO₃⁻, 9.2 kPa CO₂). B, extracellular acidosis (9.6 mm HCO₃⁻, 4.6 kPa CO₂). C, intracellular alkalosis (9.6 mm HCO₃⁻, 1.8 kPa CO₂). D, extracellular alkalosis (48 mm HCO₃⁻, 4.6 kPa CO₂).

Intracellular and extracellular alkalosis also had opposite effects. An intracellular alkalosis (C) reduced the magnitude of the Ca²⁺ transient (83 ± 4 % of control, P < 0.01, n= 33), whereas an extracellular acidosis (D) augmented the transient (123 ± 6 % of control, P < 0.01, n= 32).

The relationship between the carbachol-induced Ca²⁺ transient and contraction

A previous study (Liston et al. 1991) established the relationship between the magnitude of phasic isometric detrusor contractions and the value of either pH_o or pH_i. It was found that intracellular acidosis increases, while extracellular acidosis decreases, the force of contraction. The dependence of contractile force on the magnitude of the Ca²⁺ transient, when either pH_o or pH_i is altered, can be gauged by plotting the two variables against each other. Figure 3 shows the percentage change to Ca²⁺ transient magnitude as a function of contractile strength when pH_o and pH_i are altered. The interdependence of the two variables, regardless of whether they were altered by changing pH_o or pH_i, suggests that pH-induced variations of phasic contractions are significantly influenced by changes to the magnitude of the cholinergic Ca²⁺ transient.

Figure 3. The relationship between the magnitudes of the carbachol-induced Ca²⁺ transient and phasic tension.

Values of the Ca²⁺ transient and phasic tension are plotted as a function of those recorded in control Tyrode solution (100 %, ▴). Ca²⁺ transient values are plotted as means ±s.d.; mean tension values are taken from Liston et al. (1991). Data points were obtained during extracellular alkalosis (▪) and acidosis (•) and during intracellular alkalosis (^) and acidosis (□).

High-K⁺-induced intracellular Ca²⁺ transients - the effects of changes to pH_o and pH_i

In several muscle types, pH-dependent inotropic effects are in part mediated by their actions on transmembrane Ca²⁺ fluxes, in particular on voltage-dependent Ca²⁺ channels. In detrusor a rise in intracellular Ca²⁺ can be elicited by activation of the L-type Ca²⁺ channel (Ganitkevich & Isenburg, 1991). We therefore investigated the effect of altered pH on Ca²⁺ transients evoked by membrane depolarization induced by either raising extracellular [K⁺] or under voltage clamp.

Figure 4 shows Ca²⁺ transients elicited by increasing the extracellular [K⁺] from 4 to 40 mm for 40 s. The KCl-induced Ca²⁺ transients had longer time courses than the carbachol-evoked transient, were totally abolished by 5 μm nicardipine (n= 5) and were partially inhibited by 10 μm ryanodine (72 ± 14 % of control, mean ±s.d., n= 4) - not shown.

Figure 4. The effects of changes to pH_o and pH_i on the high-K⁺-induced Ca²⁺ transient.

Cells were briefly exposed to Tyrode containing 44 mm KCl and either pH_o or pH_i was changed by altering the superfusate P_CO2 and/or [HCO₃⁻]. A, extracellular acidosis (9.6 mm HCO₃⁻, 4.6 kPa CO₂). B, extracellular alkalosis (48 mm HCO₃⁻, 4.6 kPa CO₂). C, intracellular acidosis (48 mm HCO₃⁻, 9.2 kPa CO₂).

The three panels in Fig. 4 show that changes to extracellular pH affected the magnitude of the KCl-induced Ca²⁺ transient but that changes to intracellular pH were without effect. Panel A shows that extracellular acidosis decreased the Ca²⁺ transient magnitude - to 76 ± 5 % of control (P < 0.01, n= 12). Panel B shows that extracellular alkalosis increased the Ca²⁺ transient - to 127 ± 5 % of control (P < 0.01, n= 10). However, panel C shows that intracellular acidosis had no significant effect on the Ca²⁺ transient (104 ± 5 % of control, P > 0.05, n= 16). Similarly, but not illustrated, intracellular alkalosis had no significant effect (103 ± 3 % of control, P > 0.05, n= 22).

Ca²⁺ current (I_Ca) and depolarization-induced Ca²⁺ transients - the effects of changes to pH_o and pH_i

The above data show that changes to pH_o exert effects on high-K⁺-induced Ca²⁺ transients whereas changes to pH_i do not. To obtain more direct evidence that these effects are mediated via the L-type Ca²⁺ channel, Ca²⁺ transients and the L-type Ca²⁺ current, I_Ca, were measured simultaneously during changes to pH_o and pH_i.

Ca²⁺ transients and I_Ca were evoked by steps to between -40 and +60 mV from a holding potential of −60 mV. The examples given in Fig. 5A show that the decay of I_Ca is accelerated in the final stages when the Ca²⁺ transient begins, demonstrating the Ca²⁺ dependence of I_Ca. The voltage dependencies of both signals were bell shaped and a mirror of each other, with maxima at +10 mV (see Fig. 6 below). At potentials between +60 and +80 mV I_Ca reversed while the Ca²⁺ transient magnitude approached, but did not reach, a zero value. The reversal of I_Ca and a maintained but small Ca²⁺ transient may be explained by outward permeation of other ions such as Cs⁺ and Na⁺ which counterbalance a small residual Ca²⁺ influx (Hess & Tsien, 1984). Both I_Ca and the Ca²⁺ transient were abolished by verapamil (10 μm). These observations demonstrate a causal relationship between I_Ca and the Ca²⁺ transient and also minimize the possibility of an involvement by Na⁺-Ca²⁺ exchange to the Ca²⁺ transient. The Ca²⁺ transient was also attenuated by ryanodine (20–50 μm), suggesting the involvement of Ca²⁺-induced Ca²⁺ release in generating the final Ca²⁺ transient. To examine the effects of pH changes on these phenomena, a voltage step of 1 s duration was used, which gave the maximum increase in the intracellular [Ca²⁺], and a 50 s pulse interval to allow complete recovery.

Figure 5. The effect of changes to pH_o to simultaneously recorded I_Ca and Ca²⁺ transient.

A, I_Ca (upper traces) and Ca²⁺ transients (lower traces) during superfusion with normal Tyrode solution (left and right traces, 24 mm HCO₃⁻) and during extracellular acidosis (middle traces, 9.6 mm HCO₃⁻). Traces evoked by a 1 s voltage-clamp step from -60 to +10 mV. B, rising phases of Ca²⁺ transients during control (24 mm HCO₃⁻), extracellular acidosis (9.6 mm HCO₃⁻) and extracellular alkalosis (48 mm HCO₃⁻).

Figure 6. Voltage dependence of I_Ca and the Ca²⁺ transient.

A, voltage dependence of I_Ca and the Ca²⁺ transient in normal Tyrode solution (□) and during extracellular acidosis (•) or extracellular alkalosis (▪). B, control experiment showing changes to intracellular pH during superfusion with solutions of raised (left) and lowered (right) P_CO2, at constant pH_o in a patch pipette dialysed cell.

Figure 5A shows three pairs of traces, each consisting of a current trace (upper panel) and a Ca²⁺ transient (lower panel). The first and third pair of traces were obtained under control conditions, the centre pair during an extracellular acidosis. The magnitude of both I_Ca and the Ca²⁺ transient were reversibly attenuated in the low-pH_o solution. Combined data showed that an extracellular acidosis from pH 7.38 to pH 7.02 reduced I_Ca and the Ca²⁺ transient to 87 ± 2 and 83 ± 4 %, respectively, of control (P < 0.01, n= 15). Extracellular alkalosis to pH 7.70 increased I_Ca and the Ca²⁺ transient to 116 ± 3 and 132 ± 3 %, respectively, of control (P < 0.01, n= 12).

Figure 5B shows that the rate of rise in the Ca²⁺ transient was similarly affected; acidosis decreased the rate from a control of 4.3 ± 0.5 to 3.9 ± 0.45 μmol l⁻¹ s⁻¹ (P < 0.01, n= 15) and alkalosis increased the rate from a control of 3.9 ± 0.8 to 5.0 ± 1.0 μmol l⁻¹ s⁻¹ (P < 0.01, n= 12). Because the rate of rise in the Ca²⁺ transient corresponds to the instantaneous amplitude of I_Ca (Becker, Singer, Walsh & Fay, 1989), this suggests that extracellular H⁺ exerts a primary effect on the L-type Ca²⁺ channels.

Changes to intracellular pH, however, had no significant effect on either I_Ca or the Ca²⁺ transient. The magnitudes of the two variables were 107 ± 5 and 102 ± 5 %, respectively, of control when pH_i was reduced by 0.13 units (P > 0.05, n= 17); they were 101 ± 4 and 99 ± 4 %, respectively, of control when pH_i was increased by 0.15 units (P > 0.05, n= 16).

These effects of pH were also examined over the entire voltage range from -50 to +30 mV. Figure 6 shows the voltage dependence of both I_Ca and the Ca²⁺ transient and demonstrates that, at all voltages, extracellular acidosis decreased and extracellular alkalosis augmented both variables. No effects of altering pH_i were found at any voltage in this range.

The lack of effect of interventions designed to alter pH_i could be due to the fact that no such change actually occurred in these cells dialysed with a patch pipette. To exclude this possibility, pH_i was measured in patch-dialsysed cells exposed to the superfusates with altered P_CO2 and [NaHCO₃]. pH_i was measured by including the free-acid form of BCECF (30 μm) in the patch pipette. Figure 6B shows that, when cells were exposed to solutions in which proportional changes to P_CO2 and [NaHCO₃] were made, pH_i changed in the expected direction and by a similar magnitude to that measured in undialysed cells (see Table 1). Therefore the lack of effect of changes to pH_i on I_Ca or the Ca²⁺ transient cannot be attributed to an inability to alter pH_i under these experimental conditions.

Caffeine-induced intracellular Ca²⁺ transients - the effects of changes to pH_o and pH_i

The effects of intracellular pH changes on the magnitude of the carbachol-induced Ca²⁺ transient cannot be explained by changes to transmembrane Ca²⁺ fluxes, at least through the L-type Ca²⁺ channel. This suggests that the actions of intracellular H⁺ are more likely to be on an intracellular site of Ca²⁺ regulation. One possibility is that the ability of the sarcoplasmic reticulum (SR) to store and/or release Ca²⁺ is altered by changes to pH_i. To examine this, the influence of changes to pH_i on the caffeine-induced Ca²⁺ transient was measured.

Figure 7 shows Ca²⁺ transients elicited by brief exposure to caffeine (10 mm). These transients had a slightly longer time course than the carbachol-induced Ca²⁺ transients, but like the latter, often exhibited a pronounced transient undershoot following relaxation. Panel A shows that these Ca²⁺ transients could be attenuated by prior application of 10 μm carbachol. This observation, together with the obverse experiment, whereby the carbachol-dependent Ca²⁺ transient is diminished by pre-treatment with caffeine (Fig. 1C), indicates that there is an overlap of the carbachol-sensitive and caffeine-sensitive Ca²⁺ stores in the muscle cell.

Figure 7. The effects of changes to pH_o and pH_i on the caffeine-induced Ca²⁺ transient.

A, caffeine- and carbachol-induced Ca²⁺ transients. Brief application of carbachol (10 μm) a short time before application of caffeine (10 mm) reduced the caffeine-induced Ca²⁺ transient. B, effect of intracellular acidosis (left trace, 48 mm HCO₃⁻, 9.2 kPa CO₂) and intracellular alkalosis (right trace, 9.6 mm HCO₃⁻, 1.8 kPa CO₂) on the magnitude of the Ca²⁺ transient.

Figure 7B shows that intracellular acidosis increased the caffeine-induced Ca²⁺ transient, the mean value being 121 ± 5 % of control (P < 0.01, n= 41). Conversely, Fig. 7B also shows that an intracellular alkalosis attenuated the Ca²⁺ transient, the mean value being 77 ± 5 % of control (P < 0.01, n= 24). The magnitudes of these changes were quantitatively similar to the pH_i-mediated effects on the carbachol-induced Ca²⁺ transient. This suggests that the SR is the site at which intracellular H⁺ augments the Ca²⁺ transient and hence the force of contraction.

Possible mechanisms for the actions of intracellular H⁺ on SR Ca²⁺ regulation

Changes to the magnitude of the carbachol-induced or caffeine-induced Ca²⁺ transient by intracellular H⁺ could be mediated either by effects on the SR Ca²⁺ release channels or by influencing the accumulation of Ca²⁺ via the SR Ca²⁺ pump. Measurement of the rate of rise in the carbachol-induced Ca²⁺ transient showed that this variable was unaffected at different values of pH_i. This suggests that an effect on the Ca²⁺ release channels is unlikely.

The second possibility was investigated by measuring the magnitude of the caffeine-induced Ca²⁺ transient as an estimate of the amount of Ca²⁺ within the intracellular stores. Intracellular acidosis increased not only the magnitude of the Ca²⁺ transient but also the resting [Ca²⁺]_i. It was therefore possible that the the raised [Ca²⁺]_i was responsible for greater Ca²⁺ accumulation and hence a larger transient. To test this hypothesis the relationship between the caffeine-induced transient and [Ca²⁺]_i was plotted as shown in Fig. 8. The changes are plotted as percentage variation from the control situation in normal Tyrode solution (^, during intracellular acidosis; •, during intracellular alkalosis). A significant and positive relationship between the two variables (r= 0.55, where r is the correlation coefficient; P < 0.01, n= 54) supported the hypothesis.

Figure 8. The relationship between resting [Ca²⁺]_i and the caffeine-induced Ca²⁺ transient.

The magnitude of the caffeine-induced Ca²⁺ transient is plotted as a function of the resting [Ca²⁺]_i. Values are expressed as a percentage of those obtained in Tyrode solution: ^, with intracellular acidosis; •, with intracellular alkalosis. The straight line was obtained by least-squares regression.

The reason for the raised [Ca²⁺]_i is illustrated in Fig. 9A, which shows an experiment where a cell was exposed to a Ca²⁺-free superfusate (containing 0.1 mm EGTA) and [Ca²⁺]_i monitored. The resting [Ca²⁺]_i was low in this solution (about 20 nm), but when the pH_i was lowered by changing the superfusate a small, reversible increase in [Ca²⁺]_i was still recorded, suggestive of an intracellular source for this phenomenon.

Figure 9. Source of the rise of [Ca²⁺]_i during intracellular acidosis.

A, trace of the resting [Ca²⁺]_i during superfusion with a zero-Ca²⁺ superfusate. During the time indicated an intracellular acidosis was induced by raising, in proportion, the superfusate P_CO2 and [HCO₃⁻]. B, the increase in [Ca²⁺]_i when the superfusate was switched from a zero-Ca²⁺ to normal (1.8 mm) Ca²⁺-containing superfusate. The experiment was carried out at control pH_o and pH_i (24 mm HCO₃⁻, 4.6 kPa CO₂) and during intracellular acidosis (48 mm HCO₃⁻, 9.2 kPa CO₂). The recovery phases were best fitted to a single exponential process: the time constants are displayed next to the line fits.

Figure 9B shows an experiment to examine if a transmembrane source of Ca²⁺ might, however, also contribute to the increase in [Ca²⁺]_i in acidosis. The traces show the recovery of [Ca²⁺]_i when a Ca²⁺-containing superfusate was returned after a period of exposure to a low-Ca²⁺ solution. The rate of recovery was similar in the superfusate with normal P_CO2 and [NaHCO₃] (τ= 65 s, where τ is recovery time constant) compared with that (τ= 64 s) in the solution which would generate an intracellular acidosis. This suggests that the rate of transmembrane Ca²⁺ influx was not altered by intracellular acidosis and therefore was unlikely to account for the raised value of [Ca²⁺]_i recorded when pH_i was reduced.

DISCUSSION

Actions of H⁺ on detrusor smooth muscle

A pH_i value of 7.10 in guinea-pig detrusor myocytes (Table 1) is similar to that reported for human and ferret detrusor myocytes (Liston et al. 1991; Fry et al. 1994b). The sustained nature of the pH_i change when superfusate P_CO2 was altered and the lack of effect upon changes to superfusate [NaHCO₃] at constant P_CO2 measured in this study is in contrast to observations in some other smooth muscles (Aickin & Brading, 1984; Aalkjær & Cragoe, 1988; Austin & Wray, 1993) and may reflect different pH-regulating mechanisms or surface membrane permeability to H⁺ between detrusor and these other muscles.

Changes to extracellular or intracellular pH have been shown to have opposite effects on phasic contractions in detrusor muscle evoked either by stimulation of the attendent motor nerves embedded in the tissue or by direct electrical stimulation of the muscle (Liston et al. 1991). In this study, extracellular acidosis was negatively inotropic, a finding similar to that in cardiac and several other smooth muscles (Wray, 1988; Orchard & Kentish, 1990); however, intracellular acidification increased the force of phasic contractions.

These observations suggested that pH changes exerted direct effects on the smooth muscle cell, and a previous study showed that extracellular effects of H⁺ could be attributed to an action on the L-type Ca²⁺ channel (Fry et al. 1994a). The intracellular actions of H⁺ are more equivocal but they could act by altering the myofilament sensitivity to Ca²⁺, or by influencing the intracellular pathways which link surface membrane activation to myofilament interaction. A previous study which investigated the actions of H⁺ on detrusor myofilament Ca²⁺ sensitivity (Wu et al. 1995) found that an increase in pH_i, from a normal value of 7.1, decreased Ca²⁺ sensitivity and hence could explain in part the negative inotropic effect of intracellular alkalosis. However, a decrease in pH_i was without effect so that other actions of H⁺ were required to explain the augmentation of contraction.

Involvement of membrane transporters

One possible action of intracellular acidosis is to increase the intracellular [Na⁺] via Na⁺-H⁺ exchange or Na⁺-HCO₃⁻ co-transport, which in turn could increase intracellular Ca²⁺ via Na⁺-Ca²⁺ exchange; such a mechanism has been proposed in cardiac muscle as a means whereby intracellular H⁺ can, under some conditions, increase force production (Bountra & Vaughan-Jones, 1989). However, several lines of evidence indicate that this is unlikely: (i) the increase in force by decreased pH_i is unaffected by amiloride or DIDS (4,4′-diisothiocyanatostilbene-2,2′-disulphonic acid; Liston et al. 1991), (ii) the importance of Na⁺-Ca²⁺ exchange in guinea-pig detrusor is questionable (Ganitkevich & Isenberg, 1991), and (iii) intracellular acidosis did not modulate the rate of recovery of [Ca²⁺]_i on return from a zero-Ca²⁺ superfusate (Fig. 9).

Carbachol-, high-K⁺- and caffeine-induced Ca²⁺ transients: actions of intracellular and extracellular H⁺

Extracellular application of carbachol, caffeine or a high-K⁺ solution evoked an intracellular Ca²⁺ transient. However, the source of Ca²⁺ differed; the high-K⁺ solution generated a transmembrane flux, since the associated Ca²⁺ transient was rapidly abolished by zero-Ca²⁺ superfusates or application of the L-type Ca²⁺ channel blocker nicardipine, whereas carbachol and caffeine probably released Ca²⁺ from an intracellular store. The morphology of the carbachol- and caffeine-evoked Ca²⁺ transients was different from the high-K⁺-induced transient, showing an undershoot on recovery (see also Baró, O'Neill & Eisner. 1993). Furthermore, prior application of caffeine or carbachol attenuated the Ca²⁺ transient evoked by the second agonist, indicating that the Ca²⁺ source was similar. In addition, they were less affected by zero-Ca²⁺ superfusates and nicardpine, and the carbachol-induced Ca²⁺ transient, at least, was independent of changes to membrane potential (Wu & Fry, 1997).

Acidosis confined to the extracellular space attenuated all three types of Ca²⁺ transient (alkalosis increased all transients), but there were differences in their response to intracellular pH changes. The high-K⁺-induced transient was unaffected by changes to pH_i, whereas carbachol- and caffeine-evoked transients were increased by acidosis. It has been shown previously that extracellular H⁺ reduces current flow through the L-type Ca²⁺ channel (Fry et al. 1994a) and this would explain the attenuation of the high-K⁺-induced transient. However, the lack of effect of intracellular H⁺ on the Ca²⁺ transient evoked by high-K⁺ suggested that the L-type channel was insensitive to pH changes in this compartment, and this was corroborated by direct measurement of I_Ca during variation of pH_i. This insensitivity of the L-type Ca²⁺ channel to pH_i changes is in contrast to the situation in other muscles where intracellular acidosis is effective in reducing the magnitude of I_Ca (Kaibara & Kameyama, 1988).

The attenuation of the carbachol- and caffeine-induced Ca²⁺ transients by extracellular acidosis is less straightforward. The Ca²⁺ transient evoked by carbachol is not dependent on a change of the membrane potential (Wu & Fry, 1997) so that extracellular H⁺ must exert a secondary effect. One possibility is that H⁺ may have reduced the filling of intracellular Ca²⁺ stores via the surface membrane. The importance of such fluxes in refilling acetylcholine-sensitive intracellular Ca²⁺ stores has been observed in other smooth muscles (Bourreau, Kwan & Daniel, 1993). Such refilling may involve transmembrane Ca²⁺ influx, as it can be diminished in detrusor by L-type Ca²⁺ channel blockers (Wu & Fry, 1997) and thus may be attenuated by extracellular H⁺.

The augmentation of the carbachol-induced Ca²⁺ transient by intracellular acidosis mirrored in a proportional manner the increase in phasic contractile force under similar conditions. Several observations suggest that the site of action might be at the sarcoplasmic reticulum: in particular, caffeine and carbachol released Ca²⁺ from the same intracellular site, and Ca²⁺ transients evoked by both agonists were altered by a similar amount with changes to pH_i.

Fura-2 signals and pH_i changes

It is possible that the altered magnitude of the Ca²⁺ transients with changes to pH_i might be an artefact due to pH-dependent changes of Ca²⁺-binding to fura-2, even though the fluorochrome has been considered relatively insensitive to pH changes (Grynkiewicz et al. 1985). The calculated values of [Ca²⁺] quoted in the Results ignore the influence of pH-dependent Ca²⁺ binding to fura-2 itself. An in vitro control experiment was carried out whereby the binding constant, K_d, of Ca²⁺ to fura-2 (pentapotassium salt, 10 μm) was measured between pH 6.8 and pH 7.3. Over this range acidification decreased the pK_d (-log₁₀K_d) value from 6.73 to 6.48 - i.e. there was a decrease in Ca²⁺ binding. The effect of this would be to diminish the magnitude of the intracellular Ca²⁺ signals with acidosis by a maximum of 15 %, an effect opposite to that which was actually observed. Therefore if any pH-dependent effects of Ca²⁺ binding to fura-2 were taken into account the result would be to accentuate the observed effects of changes to pH_i on the intracellular Ca²⁺ transients.

Mode of action of pH_i changes on intracellular Ca²⁺ stores

An increase in the carbachol- and caffeine-induced Ca²⁺ transient during intracellular acidosis could arise from (i) an increased uptake into the sarcoplasmic reticulum during the quiescent phase or (ii) a greater release during activation. The rate of rise in the carbachol transient was independent of pH_i, which is not consistent with an augmentation of Ca²⁺ release by intracellular H⁺. This is also consistent with other observations which show that, if anything, H⁺ depresses function of the ryanodine and IP₃ receptor (Meissner & Henderson, 1987; Tsukioka, Iino & Endo, 1994) and so is unlikely to enhance Ca²⁺ release.

It is, however, possible that a decrease in pH_i can enhance Ca²⁺ uptake into the SR. Intracellular acidosis increased the resting [Ca²⁺]_i and the positive relationship between this [Ca²⁺] and the caffeine-induced Ca²⁺ transient suggested that this rise in [Ca²⁺]_i was indeed responsible for greater accumulation of Ca²⁺ by the SR. This is supported by the observation that SR Ca²⁺ uptake is increased as [Ca²⁺]_i is raised, with a half-maximal value of about 0.75 μm in smooth muscle (Stout, 1991).

The source of the raised resting [Ca²⁺]_i during acidosis is unknown but could arise from an intracellular source such as displacement of Ca²⁺ from soluble proteins or from mitochondria (Fry & Harding, 1989). An extracellular source of this Ca²⁺ is thought unlikely, as the increase still occurred in a zero-Ca²⁺ superfusate and variation of pH_i did not alter the recovery of resting [Ca²⁺]_i when returning from a zero-Ca²⁺ to a Ca²⁺-containing superfusate. A second possibility is that H⁺ directly augments Ca²⁺ uptake by the detrusor SR (Xu, Narayanan, Samson & Grover, 1993; Elmoselhi, Blennerhasset, Samson & Grover, 1995), although this is disputed by others (Okabe, Kohno, Kato, Odajima & Ito, 1985).

Diversity of responses in smooth muscle to pH changes

The above experiments, when compared with data from other studies, emphasize the diversity of responses to changes of extracellular and intracellular pH in various smooth muscles. The surface membrane of detrusor smooth muscle is very impermeable to polar acid-base moieties such that when the extracellular [NaHCO₃] is altered at constant P_CO2, there is little alteration of pH_i. This has enabled the separate effects of intracellular and extracellular H⁺ on contractile force to be distinguished, and their different modes of action explored. This relative impermeability has also been seen in vas deferens (Aickin, 1984) and myometrium (Parratt, Sheader, Taggart & Wray, 1994), as well as in cardiac muscle (Ellis & Thomas, 1976). In contrast some other smooth muscles transmit rapidly and more completely changes of extracellular pH to the intracellular space (Austin & Wray, 1993; Ramsey, Austin & Wray, 1994) so that alterations to the pH of either compartment are more likely to have comparable contractile effects.

In addition, changes to pH_i can have variable effects on contractile function. The increase in force by acidosis in detrusor phasic tension is mirrored acutely, for example, in some vascular smooth muscle preparations (Matthews, Graves & Poston, 1992; Apkon & Boron, 1995), although more prolonged exposure depresses force. Acidification could reduce force by, for example, reducing myofilament Ca²⁺ sensitivity, reducing I_Ca or depressing IP₃-dependent SR release. On the other hand, by increasing the resting [Ca²⁺]_i and enhancing SR Ca²⁺ uptake, force could be increased. In detrusor, the balance seems to favour those factors which tend to increase force. The enhanced contractile response of detrusor smooth muscle to intracellular acidification can be considered as an advantage, as this situation is most likely to occur in normal filling of the bladder when the wall is stretched and blood flow is reduced. The greater contractility will thus facilitate the efficient expulsion of urine.