Ca2+-activated Cl− currents (I Cl,Ca) were examined using fluorescence confocal microscopy to monitor intracellular Ca2+ liberation evoked by flash photolysis of caged inositol 1,4,5-trisphosphate (InsP 3) in voltage-clamped Xenopus oocytes. Currents at +40 mV exhibited a steep dependence on InsP 3 concentration ([InsP 3]), whereas currents at −140 mV exhibited a higher threshold and more graded relationship with [InsP 3]. Ca2+ levels required to half-maximally activate I Cl,Ca were about 50% larger at −140 mV than at +40 mV, and currents evoked by small Ca2+ elevations were reduced >25-fold. The half-decay time of Ca2+ signals shortened at increasingly positive potentials, whereas the decay of I Cl,Calengthened. The steady-state current-voltage (I-V) relationship for I Cl,Ca exhibited outward rectification with weak photolysis flashes but became more linear with stronger stimuli. Instantaneous I-V relationships were linear with both strong and weak stimuli. Current relaxations following voltage steps during activation of I Cl,Ca decayed with half-times that shortened from about 100 ms at +10 mV to 20 ms at −160 mV. We conclude that InsP 3-mediated Ca2+liberation activates a single population of Cl−channels, which exhibit voltage-dependent Ca2+ activation and voltage-independent instantaneous conductance.
- calcium-activated chloride current
- voltage dependence
- inositol 1,4,5-trisphosphate
numerous Ca2+-activated Cl− channels are present in the plasma membrane ofXenopus oocytes (13) and serve to generate the fertilization potential that provides a fast electrical block to polyspermy in the egg (7). Voltage-clamp recordings of Cl−current provide a convenient reporter of subplasmalemmal Ca2+ concentration ([Ca2+]) and are widely used to study both endogenous Ca2+ signaling pathways in the oocyte and to monitor the expression of exogenous Ca2+-mobilizing receptors (8, 14). Ca2+-activated Cl− currents (I Cl,Ca), however, show complex time- and dose-dependent characteristics that complicate interpretation of the underlying Ca2+ signals (11, 21). In particular, it remains unclear whether they are generated through a single population of Cl− channels or through two or more channel types that have different sensitivities to Ca2+ (3, 11).
This problem is particularly vexing in the case of responses mediated by the inositol 1,4,5-trisphosphate (InsP 3) messenger pathway. Ca2+ signals evoked by InsP 3 involve a rapid, transient liberation of Ca2+ from endoplasmic reticulum stores, followed by influx of extracellular Ca2+ via a plasma membrane pathway activated by store depletion (2, 23). These two sources of Ca2+ can be discriminated by recordingI Cl,Ca evoked by hyperpolarizing voltage steps because entry of extracellular Ca2+ depends on the electrochemical gradient across the cell membrane, whereas intracellular Ca2+ mobilization is largely independent of membrane potential (11, 19, 28). We had previously interpreted the resulting Cl− currents as arising through a single class of Cl− channels (28), but a more detailed electrophysiological analysis led Hartzell (11) to propose the involvement of two channel types: one generating an outwardly rectifying current activated by both intracellular Ca2+liberation and Ca2+ influx and a second channel with lower affinity for Ca2+ that is activated selectively by the high subplasmalemmal Ca2+ levels resulting from Ca2+influx during hyperpolarizing voltage steps. An alternative hypothesis, however, is that only a single class of I Cl,Cachannels is present in the oocyte, but that the sensitivity of these channels for Ca2+ is voltage dependent (12).
To study the influence of membrane potential on the Ca2+sensitivity of Cl− channels, we recorded Cl− currents and cytosolic Ca2+ signals generated by Ca2+ arising from transient liberation of Ca2+ from intracellular stores induced by flash photolysis of caged InsP 3. Currents recorded at positive holding potentials exhibited a lower threshold and steeper dose dependence on InsP 3 concentration ([InsP 3]) than corresponding currents at hyperpolarized potentials, even though intracellular Ca2+ signals monitored by confocal microfluorimetry were similar at both voltages. Furthermore, the steady-state current-voltage (I-V) relationship of the Cl− current changed progressively from strongly outward rectifying to linear with increasing intracellular [Ca2+]. These results are consistent with the presence of a single class of Cl− channels that show increasing sensitivity to Ca2+ at more positive potentials.
Immature (stage V and VI) oocytes of Xenopus laevis were obtained as previously described (25). Frogs were anesthetized by immersion in a 0.17% aqueous solution of MS-222 (3-aminobenzoic acid ethyl ester) for 15 min, and a small portion of the ovary was removed surgically through an abdominal incision, after which the wound was sutured and animals were allowed to recover. After manual removal of epithelial layers, oocytes were loaded 30–60 min before recording with caged InsP 3 [myo-inositol 1,4,5-trisphosphate, P4(5)-1-(2-nitrophenyl) ethyl ester] together with the low-affinity Ca2+ indicator Oregon Green 488 1,2-bis(2-aminophenoxy)ethane-N,N,N′,N′-tetraacetic acid-5N (OG-5N) to final intracellular concentrations of about 5 and 50 μM, respectively (26).
Recordings were made at room temperature, with oocytes bathed in normal Ringer solution (in mM, 120 NaCl, 2 KCl, 1.8 CaCl2, 5 HEPES, at pH about 7.2). Ca2+-free solution contained no added Ca2+, and additionally 1 mM EGTA and 5 mM Mg2+. Measurements of membrane currents were obtained using a two-electrode voltage clamp (GeneClamp 500; Axon Instruments; Foster City, CA), with KCl-filled microelectrodes broken to resistances of 1–3 MΩ. Simultaneously, measurements were made of intracellular Ca2+ signals (monitored by OG-5N fluorescence) evoked by photostimulation, using a linescan confocal microscope and flash photolysis system as described previously (4, 17). In brief, a custom-built confocal scanner assembly was interfaced to an Olympus IX 70 inverted microscope fitted with a ×40 oil-immersion objective (NA 1.35). Fluorescence excited in the oocyte by a diffraction-limited spot of 488 nm light from an argon ion laser was monitored through a confocal aperture at λ > 510 nm while the spot was repeatedly scanned along a 50-μm line. The microscope was focused at the depth of the pigment granules within the oocyte, which lie about 2 μm inward from the cell surface. Measurements are presented as a ratio relative to the resting fluorescence before stimulation (ΔF/F) and represent an average across the scan line. Scans were obtained at a rate of 125 Hz, and current records were low-pass filtered at 50 Hz. The low-affinity indicator OG-5N was used to avoid saturation during strong InsP 3-evoked Ca2+ signals and provide a fluorescence signal linearly proportional to [Ca2+] over the range of interest. The dissociation constant of OG-5N is about 31 μM [measured using Molecular Probes buffer kit 3 in the presence of 1 mM Mg2+(24)], and we observed a maximal fluorescence change of 12.6 following injections of saturating amounts of Ca2+ into oocytes (n = 3) through a micropipette filled with 100 mM CaCl2. Thus for small Ca2+ signals, the dye provides a fluorescence change of ∼0.20 ΔF/F per micromole of free Ca2+.
Flashes of ultraviolet (UV; 340–400 nm) light, derived from a continuous mercury arc lamp, were used to photorelease InsP 3 from the caged precursor loaded into the oocyte. The relative amount of InsP 3 formed was varied using an electronic shutter to regulate the duration of light flashes and neutral density filters to control intensity. The photolysis light was introduced through the epifluorescence port of the microscope and was focused uniformly as a 200-μm diameter spot on the cell, centered around and parafocal with the laser scan line. For experiments with Ca2+ injections, oocytes were not loaded with caged InsP 3 or indicator, and injections were made by applying pneumatic pressure pulses to a third micropipette filled with 0.5 M CaCl2 while the cell was voltage clamped (13, 18). All experiments were done in the animal hemisphere of oocytes, because this contains a higher density of Ca2+-activated channels (10, 13) and InsP 3-sensitive Ca2+ release sites (6).
Fluorescence signals and currents were quantified by measuring the amplitudes of their respective peaks. We did not attempt to correlate simultaneous measurements of current and fluorescence, because of uncertainties in their kinetic relationship (e.g., delays due to Ca2+ diffusion and Cl− channel activation). However, the peak current was usually attained shortly (ca 100 ms) following the peak fluorescence, at a time when the fluorescence had declined only slightly (<10%). To compare responses at different membrane potentials, measurements of currents were expressed as membrane conductance, so as to normalize for ohmic changes in current and better reveal the voltage dependence of activation ofI Cl,Ca. Conductances were calculated using measured values of reversal potentials for I Cl,Ca, which were close to −20 mV, corresponding to the Cl−equilibrium potential in the oocyte (13). Outward currents at voltages positive to this potential are thus carried by an influx of Cl−, and inward currents at more negative potentials by efflux of Cl−.
In addition to Cl− currents evoked by InsP 3-mediated Ca2+ liberation, oocytes from some donor frogs display Cl− currents on depolarization due to activation of voltage-dependent membrane Ca2+ channels (13). These latter currents were absent or negligible (<50 nA) in the oocytes used for the present experiments. Furthermore, they would not have complicated measurements of InsP 3-evoked signals, because the Cl− currents generated by Ca2+ entry through endogenous voltage-gated channels decay within about 2 s (13), whereas photolysis flashes were delivered after the oocytes had been held at a given potential for >10 s.
OG-5N and caged InsP 3 were obtained from Molecular Probes (Eugene, OR). All other reagents were from Sigma Chemical (St. Louis, MO).
Cl− currents and Ca2+ fluorescence signals evoked by photoreleased InsP3. Figure1 shows records of Ca2+fluorescence signals (top traces) and currents (bottom traces) obtained at clamp potentials of −160 mV (Fig.1 A) and +40 mV (Fig. 1 B) in a single oocyte in response to photorelease of InsP 3 by UV light flashes of increasing durations. These data are representative of results obtained in a total of 12 oocytes. As we have described previously, Ca2+ waves (22) and Ca2+-activated currents (18) were evoked only by flashes exceeding a certain threshold duration. With suprathreshold stimuli, the amplitudes of the Ca2+ signals then increased progressively with increasing flash duration and were similar at the two holding potentials. In contrast, the corresponding membrane currents differed markedly at negative and positive potentials. Inward currents at −160 mV were brief and varied progressively in magnitude with the size of the underlying Ca2+ transient. Outward currents at +40 mV, on the other hand, were slower and exhibited a steeper dose-response characteristic.
Measurements of peak fluorescence signals and currents are shown plotted as a function of flash duration in Figs. 1, C andD. At −160 mV the current amplitude increased in a graded manner as the flashes were progressively lengthened. In contrast, currents at +40 mV exhibited a lower threshold and increased more steeply with increasing flash duration so as to more rapidly attain a maximum, saturating value. Maximal currents at +40 mV were smaller than at −160 mV, but this could be attributed simply to the different electrical driving force for Cl− flux through the channels. To eliminate the ohmic voltage dependence of the current, the data are expressed as membrane conductance in Fig. 1 E. This displays more clearly the differing dose dependence of activation ofI Cl,Ca at positive and negative potentials and shows that the maximal conductances at high [InsP 3] were similar at positive and negative potentials. The different dose dependence was not due to differences in [Ca2+] transients, because the corresponding fluorescence signals showed closely similar peak amplitudes at positive and negative holding potentials (Fig.1 C). Instead, the differences in dose dependence likely arise through voltage-dependent changes in Ca2+ activation of the Cl− channels.
A possible concern, however, was whether Ca2+ influx across the plasma membrane may have locally activated Cl−channels without producing appreciable fluorescence signals. This is unlikely because similar data (not shown) were obtained in four oocytes bathed in Ca2+-free Ringer solution. Furthermore, InsP 3-mediated activation of store-operated Ca2+ influx develops much more slowly than the transient responses evoked here (28) and, in any case, is not expected to contribute to the increased sensitivity of I Cl,Cato Ca2+ at positive potentials because the electrochemical gradient for Ca2+ entry is reduced at these voltages.
Density of Ca2+-activated Cl− channels. Figure2, inset, shows the mean values of currents evoked by supramaximal photolysis flashes (intensities >50times threshold) in oocytes from five different donor frogs. Measurements were made at a holding potential of +40 mV, using a 200-μm-diameter photolysis spot positioned near the animal pole. The maximal currents were relatively consistent within oocytes obtained from a given donor frog but varied more widely between oocytes from different frogs. Among the donors examined, mean currents ranged from ∼300–1,200 nA. The overall current was 843 ± 66 (SE) nA (35 oocytes), which corresponds to a conductance of 14.05 μS (assuming a Cl− equilibrium potential of −20 mV). Given the restricted spatial photorelease of InsP 3(200-μm-diam spot), the maximal conductance was ∼600 pS per square micrometer of cell surface (neglecting membrane infoldings) and the maximal whole cell conductance from the entire oocyte is predicted to be about 1,200 μS, assuming a diameter of 1.2 mm and a twofold lower density of Cl− channels in the vegetal hemisphere (6). The single channel conductance of Ca2+-activated Cl− channels in the oocyte membrane measured by patch-clamp studies is about 3 pS (27), indicating that the channel density in the animal hemisphere is roughly 200/μm2considering the oocyte as a smooth sphere, a value that may be reduced as much as 10-fold in terms of density per square micrometer of actual membrane area because of the numerous microvilli (8).
Relationship between ICl,Ca and cytosolic [Ca2+]. The relationship between the cytosolic Ca2+ signal and activation of I Cl,Ca was determined from experiments such as that in Fig. 1, using a range of flash strengths to evoke responses of varying magnitudes while fluorescence transients were imaged with the confocal scan line focused within 2 μm of the cell surface. Figure 2 A shows pooled measurements from 12 oocytes, plotting the peak conductance change at holding potentials of +40 and −150 mV as a function of OG-5N fluorescence ratio. The maximal conductance at high [Ca2+] was similar at both voltages, but the relationship at −150 mV was shifted to the right compared with that at +40 mV so that intermediate Ca2+ levels resulted in smaller conductance changes at the hyperpolarized potential. The fluorescence signal associated with half-maximal activation of I Cl,Ca increased by about 46% on polarization from +40 to −150 mV, but a much more pronounced effect was evident at low Ca2+ levels. For example, the conductance change associated with a fluorescence signal of 0.25 ΔF/F was nearly 20-fold greater at +40 mV than at −150 mV.
Measurements of fluorescence signals corresponding to half-maximal Cl− currents at various voltages are shown in Fig.2 B and provide an estimate of how the apparent affinity for Ca2+ activation of the Cl− conductance varies with voltage. Little difference was apparent between +40 mV and −60 mV, but the Ca2+ levels associated with half-maximal activation then increased progressively at more negative potentials and were about 57% greater at −150 mV compared with +40 mV. From the calibration factor given in methods, the fluorescence ratio values in Fig. 2 B correspond to free Ca2+ concentrations of about 2.1 μM at +40 and −60 mV, 3 μM at −120 mV, and 3.5 μM at −150 mV.
Voltage dependence of InsP3-evoked currents and Ca2+ signals. To explore further the interaction between voltage and [Ca2+] in the activation ofI Cl,Ca, we recorded fluorescence signals and membrane currents evoked by photorelease of InsP 3while the membrane potential was clamped over a range of holding voltages. Figure 3 shows representative records from an oocyte stimulated by photolysis flashes that were relatively strong (135 ms; Fig. 3 A) or weak (35 ms; Fig.3 B) in relation to the threshold flash duration required to evoke I Cl,Ca at +40 mV (about 25 ms). The InsP 3-evoked Ca2+ signals changed relatively little with voltage, other than a slight acceleration of their decay at more positive potentials. In contrast, the amplitudes and kinetics of I Cl,Ca changed markedly with voltage. Currents evoked by both strong and weak flashes reversed direction at about −18 mV in this oocyte, and at increasingly positive potentials, the currents became slower and exhibited a prolonged plateau. The peak amplitudes of outward currents evoked by both weak and strong flashes increased progressively as the membrane potential was clamped at voltages increasingly positive to the reversal potential. At more negative potentials, however, responses to strong flashes differed markedly from those evoked by weak flashes. Inward currents evoked by the 35-ms flashes increased only slightly or even declined with hyperpolarization beyond about −60 mV, despite the increased electrical driving force for Cl− efflux (Fig. 3 B). In contrast, the peak amplitudes of currents evoked by 135-ms flashes continued to increase progressively with hyperpolarization to at least −140 mV (Fig. 3 A).
I-V relationships (Fig. 3 C) derived from these data showed a nearly linear voltage dependence forI Cl,Ca evoked by the 135-ms flash, whereas currents evoked by the 35-ms flash showed a marked outward rectification. The ratio of the current amplitudes evoked by the strong and weak flashes varied from about 1.5 at a potential of +50 mV to about 4 at −180 mV. This shift from an outwardly rectifying to a linear I-Vrelationship with increasing photorelease of InsP 3was consistently observed in all oocytes examined (12 cells; 5 donor frogs).
The outward rectification of currents evoked by weak flashes was not due to decreased Ca2+ liberation at more negative potentials, because the corresponding fluorescence signals changed little or not at all at voltages between −180 and +60 mV (Fig.3 D). Instead, we interpret the differing shapes of theI-V relationships with weak and strong stimuli to arise through the voltage-dependent decrease in sensitivity of the Cl− conductance to Ca2+ at more negative potentials. Thus intracellular free [Ca2+] levels resulting from the strong flash were sufficient to maximally activate the conductance at all voltages between approximately +60 and −140 mV, whereas currents evoked by the weak flashes diminished at increasingly hyperpolarized potentials as the sensitivity of the Cl− conductance became progressively lower in relation to the smaller cytosolic [Ca2+] transient.
In accordance with this interpretation, the shape of the I-Vrelationship depended on the current density (current per unit area of membrane) rather than on the absolute whole cell current. For example, when the photolysis light was focused sharply as a 200-μm diameter spot, a flash of 60-ms duration evoked currents with a linear voltage dependence. After deliberately defocusing the light to illuminate a broader area of the cell, currents evoked by flashes of the same duration then exhibited a pronounced outward rectification, so that the current at −130 mV was only 55% of that with the sharply focused spot, even though the currents at +30 mV were similar (2 oocytes examined).
Voltage-dependent kinetics of Ca2+transients and I Cl,Ca. As illustrated in Fig.3, A and B, the kinetics of I Cl,Cavaried strongly with holding potential, although the time course of the corresponding Ca2+ signals exhibited only slight voltage dependence. To quantify these effects, we measured the time to fall to one-half the peak value (t 1/2) of currents and fluorescence signals evoked by supramaximal photolysis flashes over a range of holding potentials (Fig.4 A). The decay of the currents shortened markedly at increasingly negative potentials, witht 1/2 decreasing as an exponential function of voltage from about 1.6 s at +60 mV to 350 ms at −175 mV. In contrast, the voltage dependence of decay of the fluorescence signals was more shallow and in the opposite direction; thet 1/2 lengthened from about 1.5 s at +60 mV to about 1.75 s at strongly negative potentials.
The increasing discrepancy in time course between the Cl− currents and Ca2+ signals at more negative potentials was not due simply to a reduction in Ca2+ sensitivity, such that the current terminated more rapidly because the Ca2+ level, as indicated by the fluorescence signal, declined below the higher threshold level needed for activation of I Cl,Ca. This is illustrated in Fig. 4 B, which compares fluorescence and current signals evoked at −160 mV by photolysis flashes with durations of 22 and 50 ms. The current evoked by the stronger flash terminated rapidly, and no current was evident during an appreciable time for which the fluorescence signal remained higher than the peak level evoked by the weaker flash, to which there was a clear current response.
Graded rectification of ICl,Ca as a function of stimulus strength. Figure5 A plots pooled data from five oocytes (3 donor frogs), showing that the form of the I-Vrelationship of I Cl,Ca varied progressively from almost complete outward rectification with just suprathreshold photolysis flashes to near perfect linearity with very strong flashes. Normalized Cl− conductance changes derived from these data are shown in Fig. 5 B, to remove the ohmic voltage dependence of the current and display more clearly the effect of voltage on activation of I Cl,Ca. The strongest flash was >100 times stronger than the threshold required to evoke a current at +40 mV and produced a conductance that remained constant between +40 and −160 mV, suggesting that cytosolic [Ca2+] was sufficiently high to maximally activate the channels even at strongly hyperpolarized potentials. In contrast, the conductance change evoked by the weakest flash (∼1.2 times threshold) declined rapidly at potentials negative to +40 mV and decreased to ∼5% at −120 mV. Flashes of intermediate strengths gave intermediate conductance-voltage relationships.
Instantaneous and steady-state I-V relationships. The results presented above are consistent with I Cl,Ca arising through channels, the gating of which is determined by [Ca2+] and modulated by voltage and which exhibit an ohmic open channel I-V relationship (as demonstrated by the linear voltage dependence with strong stimuli). To confirm this latter point, we measured instantaneous I-V relationships (reflecting current flow through open channels) as well as steady-state relationships (reflecting both probability of channel opening and open channel current). Figure 6 shows data from an oocyte stimulated by relatively weak (1.5 times threshold at +20 mV) and stronger (7 times threshold) photolysis flashes. Steady-state current measurements were obtained, as in Fig. 2, by delivering photolysis flashes after clamping at a given potential for several seconds. To determine instantaneous currents, flashes were delivered with the oocyte clamped at +20 mV and the potential was stepped to more negative levels at about the time of the peak current (800 ms following the flash). Measurements of instantaneous currents were made about 8 ms after the voltage step, to allow time for capacitative currents to settle, and are presented after subtraction of passive currents evoked by equivalent voltage steps in the absence of photolysis flashes. The steady-state I-V relationship for the weak flash exhibited characteristic outward rectification, whereas the corresponding instantaneous I-V relationship was close to linear. With the strong flash, on the other hand, both the instantaneous and steady-state relationships were linear and closely similar.
Kinetics of Ca2+-sensitivity changes. To examine the rapidity with which the Ca2+sensitivity of the Cl− conductance changed with potential, we measured the relaxation times of tail currents following voltage steps. These measurements were not readily obtained from responses evoked by photoreleased InsP 3, due to their rapid decay (e.g., Fig. 1). Instead, we applied voltage pulses during submaximal currents evoked by microinjecting Ca2+into the oocyte through a pipette filled with 0.5 M CaCl2, which persist for a few seconds (13). Hyperpolarizing steps applied from a holding potential of −50 mV evoked large instantaneous inward currents that decayed rapidly, whereas positive-going steps evoked smaller instantaneous currents followed by a more slowly increasing outward current (Fig.7 A). Similar to the I-Vrelationships obtained with InsP 3-evoked responses, the steady-state currents measured following Ca2+injections exhibited outward rectification (Fig. 7 B), whereas instantaneous currents were more closely linear. The half-times of the current relaxations lengthened from ∼20 ms at −160 mV to nearly 100 ms at +10 mV (Fig. 7 C).
We found that Cl− currents in Xenopus oocytes activated by InsP 3-mediated liberation of intracellular Ca2+ exhibited a marked increase in sensitivity to [Ca2+] at increasingly positive membrane potentials. Consequently, the shape of the I-Vrelationship varied with [Ca2+]. Currents evoked by small Ca2+ elevations exhibited outward rectification, with little or no response at strongly negative potentials, whereas a linear I-V relationship was obtained with Ca2+ elevations sufficiently large to cause maximal activation over a wide voltage range. These findings are most easily explained by the presence of a single class of Ca2+-activated Cl− channels that exhibit a voltage-dependent activation by cytosolic Ca2+ and a voltage-independent open-channel conductance. The idea that there are two or more classes of Cl− channels with differing Ca2+ affinities (11) appears less plausible, because our data indicate that this would require both outwardly rectifying channels with high Ca2+ affinity and inwardly rectifying channels with low affinity, present at an appropriate relative density in the oocyte membrane so as to result in an overall linear I-Vrelationship with strong Ca2+ activation. Our results and conclusion are similar, however, to a recent study by Kuruma and Hartzell (12), who employed Ca2+ injections and Ca2+ influx through heterologously expressed Ca2+ channels to show that outward Cl−currents in the oocyte are more sensitive to Ca2+ than inward currents. In that study they also described apparent differences in anion selectivity and instantaneous I-V relationships between Cl− currents activated by different voltage-clamp protocols but concluded that these may arise through experimental limitations and that the hypothesis that a single type of Cl− channel remained valid (12).
In agreement with Kuruma and Hartzell (12), we believe a likely explanation for the voltage-dependent sensitivity of the Cl− current to cytosolic Ca2+ is that the apparent affinity of the channel for Ca2+ is voltage dependent, a property that is common among Ca2+-activated membrane channels. For example, Ca2+-activated Cl− channels in rat parotid acinar cells (1) and in pulmonary artery endothelial cells (15) exhibit increased activation at more positive potentials, as do large-conductance Ca2+-activated K+ channels (9). The increasing sensitivity of the Cl− current at more positive potentials may be due to Ca2+ remaining bound to the channel for a longer time, and in agreement relaxations ofI Cl,Ca following voltage steps occur with a half-time that lengthens from about 20 ms at −160 mV to about 100 ms at +10 mV. These values are closely similar to those obtained for half-times of deactivation of I Cl,Ca in excised patches of oocyte membrane following removal of Ca2+ (10), suggesting that the voltage-dependent changes in affinity of the Cl− channel arise largely through changes in off-rate for Ca2+ dissociation rather than in the rate of Ca2+ binding. Our measurements, however, represent population behavior of the underlying channels and do not allow a precise molecular interpretation. It therefore remains possible that processes other than changes in affinity may account for the voltage-dependent sensitivity of the Cl− current (e.g., changes in channel gating subsequent to Ca2+binding; modulation of local cytosolic [Ca2+] gradients near the plasma membrane). Detailed kinetic studies employing patch-clamp recording are required to unambiguously resolve these issues.
Fluorescence measurements of the intracellular free Ca2+transients underlying the Cl− currents indicated that half-maximal activation of the Cl− conductance required ∼3.5 μM Ca2+ at −150 mV and about 2 μM Ca2+ at +40 mV. These values are within the range of cytosolic Ca2+ concentrations attained during InsP 3-mediated signaling (22), and even at strongly negative potentials sufficient Ca2+ could be liberated in most oocytes by strong photorelease of InsP 3 to maximally activate the conductance. However, the voltage-dependent shift in the Ca2+ activation of I Cl,Ca(Fig. 2 B) results in much more prominent effects at low [Ca2+] than suggested by the relatively modest change in concentrations required for half-maximal activation. For example, conductances evoked by just suprathreshold photorelease of InsP 3 decreased by almost 95% with polarization from +40 to −140 mV. This large change may account for observations that intracellular Ca2+ elevations preferentially activate outward Cl− currents (12) and that Ca2+ influx (which is expected to produce a high, local [Ca2+] near the plasma membrane) readily activates inward currents at hyperpolarized potentials, whereas Ca2+ liberation from more distant intracellular stores is relatively ineffective (11, 12).
Gomez-Hernandez et al. (10) reported that a concentration of ∼27 μM Ca2+ was required for 40% activation of Cl− current in excised patches of Xenopusoocyte membrane. One explanation for the apparent discrepancy between this value and our results may be that the sensitivity of the channels is higher in the intact oocyte compared with an excised patch. Alternatively, high Ca2+ levels may be sensed by the Cl− channels if they are located close to InsP 3-sensitive Ca2+ release sites but would not be accurately reflected in the fluorescence measurements. If this is the case, such local Ca2+ gradients must be very narrowly delineated, as our fluorescence measurements were obtained from a confocal section <1 μm thick (17) focused within about 2 μm of the plasma membrane, and we did not observe larger signals closer to the plasma membrane (5).
A further unresolved question concerns the kinetic relationship between Ca2+ signals and I Cl,Ca. The decay of InsP 3-evoked fluorescence signals exhibited a slight slowing with polarization to more positive potentials, possibly because the mechanisms extruding Ca2+ across the plasma membrane functioned more effectively because of the reduced electrochemical gradient for Ca2+ across the membrane. In contrast, the decay of Cl− currents lengthened considerably at increasingly positive potentials, following an exponential relationship with an e-fold change int 1/2 per 210 mV. The increasing discrepancy between time courses of the fluorescence signal and I Cl,Cawith hyperpolarization is not readily explained as a simple consequence of the reduced sensitivity to Ca2+ but suggests that the kinetics of termination of the Cl− conductance are voltage sensitive. The reason why I Cl,Ca declines more rapidly than the Ca2+ signal at negative voltages, however, remains unclear. We had previously proposed that the Cl− channels exhibit an adaptive behavior to Ca2+ (21), but subsequent findings of sustained Cl− currents evoked by Ca2+ application to excised membrane patches (10) and by photolysis of caged Ca2+ in intact oocytes (unpublished data) make this idea less tenable. Instead, a close coupling between Ca2+release sites and Cl− channels may help explain the dissociation in time courses of the confocal fluorescence signal andI Cl,Ca.
The high sensitivity of the Cl− conductance to [Ca2+] at positive membrane potentials may assist in experiments where voltage-clamp recording of Cl− current is used as a convenient, endogenous monitor of intracellular Ca2+. Furthermore, it is likely to be of physiological importance in the generation of the fertilization potential that results from an InsP 3-dependent wave of intracellular Ca2+ spreading from the sperm entry site (16). The Cl− equilibrium potential for eggs in pond water is positive, so that opening of Cl− channels results in a depolarization. This is likely to be reinforced robustly by the regenerative characteristic imparted by the increasing sensitivity to intracellular Ca2+ at increasingly positive potentials.
We thank Dr. Jennifer Kahle for editorial assistance. This work was supported by the National Institute of General Medical Sciences Grant GM-48071.
Address for reprint requests and other correspondence: I. Parker, Laboratory of Cellular and Molecular Neurobiology, Dept. of Neurobiology and Behavior, Univ. of California, Irvine, CA 92697 (E-mail:).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.
- Copyright © 2000 the American Physiological Society