| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1Department of Pharmacology and Medical Research Council Centre for Synaptic Plasticity, University of Bristol, Bristol BS8 1TD, United Kingdom; 2Department of Pharmacology, University of Tennessee, Memphis, Tennessee 38163; and 3Department of Pharmacology, University of Colorado Health Sciences Center, Denver, Colorado 80262
Submitted 21 July 2003; accepted in final form 29 March 2004
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
; McDonough et al. 1996). These channels arise from 2 separate genes, CACNA1C and CACNA1D, that encode the CaV1.2 (C class) and CaV1.3 (D class) subunits, respectively. Channel subunits are located within the soma of hippocampal pyramidal neurons, with CaV1.2 subunits being clustered at the base of the major dendrites and CaV1.3 being more diffuse (Bowden et al. 2001; Hell et al. 1993). Both channel subtypes possess a unique pharmacology in being sensitive to dihydropyridines, although some evidence suggests that CaV1.3 channels may be slightly less sensitive (Xu and Lipscombe 2001). Dihydropyridine antagonists inhibit whole cell L-type Ca2+ current, whereas agonists enhance the channel current (Hess et al. 1984). L-type current is also enhanced by the benzoylpyrrole FPL 64176 (Rampe and Lacerda 1991). Both BAY K 8644 and FPL 64176 induce a hyperpolarizing shift in whole cell current activation and inactivation by about 10 mV, but only FPL 64176 slows the rates of activation of L-type current in cardiac myocytes (Fan et al. 2000). Cell-attached patch recordings of single L-type channels in cardiac myocytes (Fan et al. 2000) and recombinant CaV1.2 channels expressed in Chinese hamster ovary (CHO) cells (Lauven et al. 1999) have shown that enhancement of whole cell L-type current results from increases of channel open time and probability (Fan et al. 2000; Lauven et al. 1999).
Despite the importance of L-type Ca2+ current to hippocampal physiology, the vast majority of studies that have assessed the activity of these channels have been performed using high concentrations of Ba2+ as charge carrier in the presence of agonists. It has been assumed that the effect of these channel agonists is independent of the identity of the conducting ion. However, application of BAY K 8644 to cell-attached patches from hippocampal CA1 pyramidal neurons displaying L-type channel activity did not produce the expected long-duration openings when using Ca2+ as the permeant ion (Marrion and Tavalin 1998). In contrast, L-type channel open times were obviously lengthened by BAY K 8644 when Ba2+ ions were the charge carrier (Cloues et al. 1997; Fisher et al. 1990; Kavalali and Plummer 1996). These data suggest that the effect of the dihydropyridine agonist depends on the identity of permeant ion. To determine whether the effects of BAY K 8644 and FPL 64176 are dependent on the charge carrier, we have used whole cell and single-channel measurements of pharmacologically isolated L-type channel currents from acutely dissociated rat hippocampal CA1 pyramidal neurons.
We have found that enhancement of L-type channel current by these compounds is from distinct mechanisms that both depend on the identity of the permeating ion. Augmentation of macroscopic L-type current by BAY K 8644 partly resulted from an increase in single-channel current, which was accompanied with an increase in channel open time only when the channel was conducting Ba2+ and not Ca2+ ions. In contrast, enhancement of whole cell current by FPL 64176 was not associated by a change in single-channel current. Augmentation was greatest when the channel conducted Ba2+ rather than Ca2+ ions. This resulted in part from a dramatic recruitment of very long duration openings, when only a modest increase in single-channel open time was observed when Ca2+ was the conducting ion. This gave rise to an obvious slowing of whole cell current deactivation seen when the channel conducted Ba2+ ions, compared with a slight slowing of deactivation when the current was carried by Ca2+ ions. The use of these channel agonists as diagnostic tools to identify L-type channel current has to be reconsidered with the identity of the conducting ion.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Acutely dissociated hippocampal CA1 neurons were obtained as previously described (Cloues et al. 1997). Briefly, Sprague–Dawley rats (9–14 days old) were killed and hippocampi rapidly dissected and cut into 300- to 400-µm-thick slices. Slices were incubated at 37°C in dissociation solution that was bubbled with O2 of composition (mM): Na2SO4, 82; K2SO4, 30; HEPES, 10; MgCl2, 5 (pH 7.4 with HCl), with added protease type XXIII (3 mg/ml) for 7–8 min. Tissue slices were then transferred to solution containing trypsin inhibitor (1 mg/ml) and bovine serum albumin (1 mg/ml) for 1 min and finally rinsed in dissociation solution containing no enzyme. The CA1 region was microdissected and triturated onto 35-mm tissue culture plates that were coated with poly-L-lysine as needed.
Contaminating N- and P/Q-type Ca2+ channel currents were eliminated by preincubation of cells in the dissociation solution supplemented with 
-conotoxins GVIA (1 µM) and MVIIC (5 µM) for 30 min before recording (Cloues et al. 1997; Marrion and Tavalin 1998; McCleskey et al. 1987; McDonough et al., 1996). Cells were used for either whole cell macroscopic current or single-channel measurements for only 1 h after this preincubation, to prevent the opening of contaminating N- and P/Q-type channels following relief of block. Hippocampal neurons possess R-type current that is resistant to blocking by these toxins and contributes about 17–35% of the macroscopic current (McDonough et al. 1996; Sochivo et al. 2003). A holding potential of –60 mV was chosen to study both single and macroscopic L-type channel activity in toxin-pretreated neurons. This holding potential inactivates over 80% of R-type current (Sochivko et al. 2003), eliminating these channels as a source of contamination.
Whole cell recording
Cells were superfused with a modified Krebs' solution of composition (mM): NaCl, 125; KCl, 5.85; tetraethylammonium chloride, 22.5; MgCl2, 1.2; HEPES(Na), 10; and D-glucose, 11 (pH 7.4 with HCl) at room temperature. The external solution was supplemented with tetrodotoxin (1 µM) to block contaminating Na+ currents. Whole cell L-type Ca2+ current was carried by either 10 mM Ca2+ or Ba2+ using Cl– salts. Whole cell electrodes were fabricated from KG-33 (borosilicate) glass (Friedrich & Dimmock, Millville, NJ) and filled with a solution of composition (mM): CsMeSO4, 120; tetraethylammonium chloride, 30; BAPTA, 10; MgCl2, 5; Na2ATP, 5; and HEPES, 10 (pH 7.2). Currents were recorded with an Axopatch 200A (Axon Instruments, Union City, CA), using electrodes of resistance 1–3 M
. Capacitance and series resistance compensation (>90%) were used throughout. Whole cell currents were evoked by 50 ms voltage steps (–50 to +30 mV) from a holding potential of –60 mV, low-pass filtered at 1 kHz through an 8-pole Bessel filter (Frequency Devices, Haverhill, MA), and acquired at 100 µs intervals using PULSE (HEKA, distributed by Instrutech, New York, NY). Currents were leak-subtracted with a P/4 protocol from a holding potential of –100 mV. Whole cell macroscopic L-type channel current was analyzed using PULSEFIT (HEKA; Instrutech). The time-course of deactivation currents was approximated by the fitting of a single exponential, with the fit initiated from 0.1 ms after repolarization and terminated 20 ms after the end of the activation prepulse.
Single-channel recording
After incubation, cells were superfused with a solution containing (mM): K aspartate, 125; KCl, 35; MgCl2, 5; HEPES(Na), 10; EGTA, 10; and CaCl2, 5.64 (to give an estimated free concentration of 60 nM) (pH 7.4 with HCl). Cells in this solution had an approximately 0 mV membrane potential. Cell-attached patch recordings were made using thick-walled (1.5 mm OD, 0.5 mm ID) quartz electrodes (7–10 M) filled with (mM): tetraethylammonium chloride, 135; HEPES, 10 (pH 7.4). Single-channel currents were carried by either Ca2+ (10, 60, or 160 mM) or Ba2+ (10, 30, or 110) using Cl– salts. Concentrations of divalent ion in excess of 30 mM were compensated for by equimolar substitution of tetraethylammonium chloride. The pipette solution was supplemented with -conotoxin GVIA (5 µM) to counteract antagonism of block by the high concentration of divalent ion used in the electrode. L-type channel activity was evoked by 200 ms voltage steps to 0 mV from a holding potential of –60 mV. All potentials are expressed as the negative of the potential imposed on the pipette, without correction for a –17 mV (for 160 mM Ca2+) or a +16 mV (for 10 mM Ca2+) liquid junction potential. Channel currents were recorded with an Axopatch 200A amplifier, filtered at 1 kHz with an 8-pole Bessel filter and acquired at 100 µs intervals using PULSE. Single channels were analyzed using TAC (Bruxton, Instrutech), which uses a cubic spline interpolation procedure. The "50% threshold" technique was used to estimate event amplitudes and durations. The threshold was adjusted for each opening, and each transition was inspected visually before being accepted.
Open duration histograms, constructed from openings to level –1, were logarithmically binned and a square-root transformation of the ordinate (number of events/bin) was used, with the distribution being fitted by a sum of exponential probability density functions using the maximum-likelihood method. Statistical significance of multiple components was determined using the ratio of maximum likelihoods method (Colquhoun and Sigworth 1995; Horn and Lange 1983). Therefore, all histograms show data fit with the minimum number of exponential distributions that were statistically required to best describe the data. Missed events were not corrected for. The rise time of the filter was about 400 µs, so that any events briefer than 166 µs were missed because they never reached the 50% threshold after filtering. Only those events that exceeded 1 ms in duration were used for estimation of amplitudes when filtered at 1 kHz to ensure that only those channel openings where the full amplitude was achieved were included. In addition, channel opening of events carried by Ca2+ ions under control conditions displayed an open time that was about twice the filter dead time. Therefore, the values of open time from exponential fits may be subject to some error. Increasing the divalent concentration from 10 to 110 mM (or 160 mM in the case of Ca2+ ions) will affect the screening of surface charges, causing a depolarizing shift in current activation. This did not affect the comparison of open times of channels conducting different concentration of divalent ion because hippocampal L-type channel open times are insensitive to voltage (Cloues et al. 1997). Only those events that closed before the end of the voltage step were used for estimation of channel amplitude and open time. Channel open probability could not be quantified for 2 reasons. First, superimpositions of channel opening would not be observed when channel open times were extremely brief. This prevented any estimate of the number of channels in the patch for correct calculation of channel open probability. Second, FPL 64176 caused a dramatic increase in channel open time, which caused openings to exceed the pulse duration. This enhanced behavior could not be quantified and resulted in an underestimate of channel open probability, making comparison meaningless.
(±) BAY K 8644 and FPL 64176 were dissolved in ethanol to a stock concentration of 10 mM. Either compound was added to the superfusing external solution to give a final concentration of 1 µM. For whole cell recordings, compounds were added after current amplitude had stabilized and current–voltage relationships had been established. In some experiments, whole cell macroscopic currents were recorded in alternating Ca2+- and Ba2+-containing external solution before addition of the agonists. In single-channel studies, either compound was added after formation of the cell-attached patch. Additional cell-attached patches were recorded from the same culture dish, in the continued presence of compound. Data were pooled from patches acutely treated with agonists and those subjected to the continued presence of the agonist. The results indicated that neither compound produced long-duration openings when Ca2+ ions were used as the charge carrier. The charge carrier was alternated (between Ca2+ and Ba2+ ions) in successive patches to confirm that the compound was active. All control data were obtained on experimental equipment that had not been in contact with either compound, because neither BAY K 8644 nor FPL 64176 could be completely removed by washing. Values are given as means ± SE.
All compounds and reagents were from Sigma (Poole, Dorset, UK), except (±) BAY K 8644 and HEPES (Na and free acid) were purchased from Calbiochem (San Diego, CA). -Conotoxins GVIA and MVIIC were obtained from Bachem California (Torrance, CA).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Whole cell recording from acutely dissociated rat CA1 hippocampal neurons preincubated in -conotoxins MVIIC and GVIA revealed the L-type channel current (see METHODS). Macroscopic currents showed a modest decay (
20–25%) during the voltage step when Ca2+ ions were the permeant ion (Fig. 1, A and C). In contrast, the whole cell L-type current was sustained throughout a step depolarization from a holding potential of –60to 0 mV when Ba2+ ions were the charge carrier (Figs. 1, B and D). The macroscopic L-type current recorded using either 10 mM Ca2+ or Ba2+ ions as the charge carrier activated from about –40 mV, with the peak current observed at 0 to +10 mV (Fig. 1). Deactivation of L-type current was rapid regardless of the identity of the charge carrier (Fig. 1 and Table 1). The decay time course of deactivation currents was voltage-dependent, increasing e-fold in about 30 mV with 
increasing from 1.28 ± 0.08 ms at –60 mV to 3.37 ± 0.32 ms at –30 mV (n = 11, data not shown).
|
|
40% less) than that carried by Ba2+ (Fig. 1, A and B). BAY K 8644 did not affect the activation time course of whole cell currents carried by either ion (Fig. 1, A and B). The effect on the macroscopic L-type channel current of FPL 64176 was substantially different from that of BAY K 8644. The enhancement of the peak current by FPL 64176 was significantly greater than that seen with BAY K 8644 (Fig. 1) and was associated with an almost 20 mV hyperpolarizing shift in current activation. In addition, FPL 64176 slowed the activation of the whole cell current, regardless of the identity of the charge carrier (Fig. 1, C and D).
The time course of deactivation of macroscopic currents is determined, at least in part, by the channel open time. Both agonists have been shown to prolong L-type channel open time (e.g., Fan et al. 2001; Hess et al. 1984), suggesting that the time course of deactivation of macroscopic currents should be slowed. In contrast to this prediction, BAY K 8644 did not change the decay rate of deactivation current carried by either permeant ion (Fig. 1, A and B and Table 1). In accord with these predictions, FPL 64176 produced a modest slowing of the deactivation tail current time course when Ca2+ ions were the charge carrier, but produced a dramatically slow deactivation tail current when Ba2+ was the permeant ion (Fig. 1, C and D and Table 1). These effects suggested that the mechanism(s) underlying enhancement of the macroscopic current by the dihydropyridine were very different from that produced by the benzoylpyrrole. Single-channel studies were carried out to determine the effects on channel gating that underlie enhancement of the macroscopic current.
Effect of L-type channel agonists on single-channel properties
Increased single-channel current induced by bay k 8644.
BAY K 8644 or the related dihydropyridine agonist (+)-(S)-202-791 increased single L-type channel conductance in hippocampal neurons and GH3 pituitary cells when Ba2+ was the permeating ion (Cloues et al. 1997; Mantegazza et al. 1995). In contrast, no effect of BAY K 8644 on channel conductance was observed in dorsal root ganglion cells (Fox et al. 1987), pancreatic 
cells (Smith et al. 1993), or cardiac myocytes (Hess et al. 1986). Therefore single L-type channels were resolved using 10, 30, and 110 mM Ba2+ and 10, 60, and 160 mM Ca2+ in the electrode solution to determine whether this effect was common to both divalent ion charge carriers. Figure 2A shows examples of single-channel currents recorded at 0 mV in the presence of 10, 30, and 110 mM Ba2+. The mean values of single-channel current (i) at 0 mV are plotted as a function of divalent ion concentration in Fig. 2D (Hess et al., 1986). Increasing the concentration of the permeant ion increased the amplitude of single-channel currents recorded at 0 mV, with the amplitude displaying saturation at high divalent ion concentration. The relationship was fitted with a hyperbolic Langmuir isotherm, giving an apparent dissociation constant (Kd) of 13.5 mM (Fig. 2D). Single-channel open times and probability were obviously enhanced (see following text) in the presence of BAY K 8644 (Fig. 2B). In agreement with a previous report (Cloues et al. 1997), single-channel amplitude was increased in the presence of the dihydropyridine. Amplitudes were well described by a Langmuir isotherm, giving a Kd of 18.6 mM (Fig. 2D).
|
|
|
Effect of l-type channel agonists on channel open times with ba2+ as the permeant ion.
The open times of L-type channels in the absence of channel agonists are extremely brief (Aoyama et al. 2003; Cloues et al. 1997; Guia et al. 2001; Marrion and Tavalin 1998). Representative openings with 10 and 110 mM Ba2+ as the charge carrier are shown in Figs. 5 (control) and 6 (control), respectively. These openings were best described by the sum of 2 exponentials of time constants () 0.4 and 2.8 ms [Figs. 5 and 6 (control)]. Channel open times were obviously prolonged in the presence of BAY K 8644 (Fig. 5, BAY K 8644; Fig. 6, BAY K 8644). Openings were best described by the sum of 3 exponentials of 1.0, 3.1, and 13.8 ms when 10 mM Ba2+ was the permeant ion and 0.6, 2.2, and 6.5 ms when 110 mM Ba2+ ions were the charge carrier. Therefore the dihydropyridine evoked an additional open state with an open time constant at 0 mV of 6.5–14 ms (Figs. 5 and 6).
|
|
Effect of l-type channel agonists on channel open time with ca2+ as the permeant ion.
In contrast to expectations, BAY K 8644 and FPL 64176 had only a modest effect on L-type single-channel open times carried by Ca2+ ions (Figs. 7 and 8). This was observed when either 10 mM (Fig. 7) or 160 mM (Fig. 8) Ca2+ was the permeant ion. The amplitude of single-channel currents recorded with 160 mM Ca2+ at 0 mV was identical to amplitudes recorded previously that were sensitive to the dihydropyridine antagonist nimodipine (Marrion and Tavalin 1998). In addition, recordings using Ba2+ or Ca2+ as the charge carrier were alternated to confirm that each channel agonist was functional. As presented above, each channel agonist evoked long open time activity when Ba2+ was the charge carrier. Therefore the lack of a substantial effect of agonist on channels conducting Ca2+ ions is a property of the agonist and not merely a problem of keeping the compound active. Single L-type channels conducting Ba2+ ions displayed variable open times under control conditions, with some patches that displayed a monoexponential open duration distribution ( 0.4 ms), whereas others exhibited a biexponential distribution that included an additional set of openings with longer durations ( 2.3 ms) (see Figs. 7 and 8 and Cloues et al. 1997). Single-channel open time was extremely brief when channels conducted Ca2+ ions under control conditions, being best described by a single exponential of time constant 0.34–0.54 ms (Figs. 7 and 8, Control). It appeared that channels conducting Ca2+ ions did not display the slower open time constant ( 2.5 ms) observed when Ba2+ was the permeating ion (Figs. 5 and 6, control). Addition of BAY K 8644 promoted a minority of openings to display the slower open time component ( 2.5 ms) (Figs. 7 and 8, BAY K 8644). In contrast, single channels showed no change in open-state kinetics when recorded with 10 mM Ca2+ as the charge carrier and in the presence of FPL 64176 (Fig. 7, FPL 64176). FPL 64176 may have produced a small effect on channels recorded with 160 mM Ca2+, evoking the slower open time component seen in the presence of BAY K 8644 ( 2.3 ms, Fig. 8, FPL 64176). Therefore neither agonist affected channel open time beyond that already observed by channels conducting Ba2+ ions under control conditions.
|
|
A slowing of current activation evoked by FPL 64176 has been observed in cardiac myocytes, an effect that was independent of the identity of the charge carrier (Fan et al. 2000). In many single-channel sweeps, FPL 64176–modified long-duration openings appeared clustered at the end of the voltage step, with channels remaining open after repolarization (Figs. 4B, 5, 9, A and C). In most sweeps, long-duration openings were preceded by short-duration openings (Figs. 4B, 5, and 9A). This is illustrated in Fig. 9A, where a membrane patch containing a single channel is shown in the presence of FPL 64176. Step depolarization to 0 mV from a holding potential of –60 mV evoked channel opening. Long-duration openings characteristic of being modified by the benzoylpyrrole were observed after a marked delay and subsequent to short-duration openings (marked *) (Fig. 9A). Open time analysis of short-duration events that immediately preceded an opening in excess of 10 ms duration were best described by a single exponential with time constant of 0.73 ms (Fig. 9C). This open time distribution was reminiscent of channel openings in the absence of FPL 64176 (compare Fig. 8C to Figs. 5–8). The combination of short-duration openings followed by long-duration activity produced a kinetic slowing of the ensemble current (Fig. 9B).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
0.4 ms) (Cloues et al. 1997; Kavalali et al. 1997; Figs. 2, 5, and 6), which is prolonged by either BAY K 8644 or FPL 64176 when channel currents are carried by Ba2+ ions (Figs. 3, 6, and 7). It has been assumed that the effect of channel agonists is independent of the identity of the permeant ion. However, this study has shown that the mode of action of both agonists is dependent on the divalent ion charge carrier.
L-type channel agonists enhanced the macroscopic L-type current by different effects on single-channel gating. For example, BAY K 8644 enhanced the whole cell L-type channel Ba2+ current partly by an increase in single-channel conductance (Fig. 2) and a recruitment of a longer-duration open state (Figs. 5 and 6). A similar BAY K 8644–induced increase in single-channel conductance was observed when Ca2+ ions were the charge carrier (Fig. 3), contributing to enhancement of the macroscopic current. This increased single-channel current amplitude for either charge carrier in the presence of BAY K 8644 may result from the lowering of affinity for the permeant ion during conduction. Whole cell Ba2+ currents were augmented to a greater extent than Ca2+ currents by BAY K 8644, an effect that can be as least partially explained by a lack of effect of the dihydropyridine on channel open times when Ca2+ was the permeant ion (Figs. 7 and 8). Similar effects of BAY K 8644 have been reported in hair cells, where agonist-induced long-duration openings were observed only when channels conducted Ba2+ and not Ca2+ ions (Rodríguez-Contreras and Yamoah 2003). FPL 64176 strongly enhanced the whole cell L-type channel Ba2+ current, an effect achieved in part by a dramatic increase in channel open time (Figs. 5 and 6) that explained the dramatic slowing in the time course of current deactivation (Figs. 1D and 9). FPL 64176-enhancement of whole cell Ca2+ current did not result from an increase of single-channel current (Fig. 4) or channel open time (Figs. 7 and 8), thus suggesting that it was mediated primarily by an increase in the frequency of opening. Collectively, these data suggest that although the two classes of L-channel agonists have different effects on conductance, both have qualitatively similar effects on open time behavior that are dependent on the permeant ion.
L-type channels exhibit Ca2+-dependent inactivation, resulting in a decline in whole cell current during a maintained depolarization (Eckert and Chad 1984). Macroscopic current from hippocampal neurons displayed clear Ca2+-dependent inactivation, a loss of current that was not observed when using Ba2+ ions as the charge carrier (Fig. 1). It is possible that the action of channel agonist may be state-dependent, with a greater effect being observed on channels not inactivated by a Ca2+-dependent process. This is consistent with the larger effect of BAY K 8644 on macroscopic current carried by Ba2+ ions (Fig. 1). It has been reported that BAY K 8644 increases the rate of inactivation of Ca2+-carried macroscopic current in both cardiac L-type (Sanguinetti et al. 1986) and recombinant CaV1.2 subunits expressed in Xenopus oocytes (Noceti et al. 1998). Stationarity plots of channel open probability at 0 mV showed that BAY K 8644 and FPL 64176 increased the rate of decay of channel open probability conducting Ca2+ ions, an effect that was more prominent at higher concentrations of exter-nal Ca2+ ions (data not shown; Noceti et al. 1998). This was not observed when channels conducted Ba2+ ions. These data demonstrate that Ca2+-dependent inactivation is sensitive to open probability (Noceti et al. 1998). However, it is unclear whether this suggests channel agonists favor channels that are not inactivated or that they promote inactivation by increasing channel open probability at the start of the voltage step. The open time of channels conducting Ca2+ ions within the first 50 ms of the voltage step was longer ( 0.69 ms) than in the remaining 150 ms ( 0.47 ms). It is possible that hippocampal L-type channels display the change in gating resulting from Ca2+-dependent inactivation reported in cardiac myocytes (Imredy and Yue 1992). However, channels in the presence of BAY K 8644 showed a similar small reduction of open time during the pulse (openings <50 ms: values of 0.5 and 2.8 ms; openings >50 ms: values of about 0.41 and 1.9 ms; data not shown). Therefore BAY K 8644 did not promote a large prolongation of channel open time at the start of the pulse. This suggests that the action of BAY K 8644 is not state-dependent.
L-type channel gating has been modeled with a modal scheme that can be illustrated as follows 
The channel is proposed to reside within mode 1 during normal gating, with occasional sojourns to mode 2 (Hess et al. 1984). BAY K 8644 is proposed to increase the transition rate into mode 2 (marked *) (Hess et al. 1984). A minor change to this scheme can satisfy the data obtained in this study. Assuming that gating is modal (Hess et al. 1984; we have no data supporting this assumption), the channel can reside in any of the following states 
The channel resides in mode 1 during normal gating, giving rise to open times best described by a dominant exponential distribution of time constant 0.4 ms (from O1), with a very minor additional component of time constant 2.3 ms (from O2) (Figs. 4 and 5). Sojourns into mode 2 are not permissible under control conditions. Addition of agonist increases the transition rates marked , but the transition rate between C1 and C4 into mode 2 can occur only when Ba2+ binds within the channel. This would give rise to an open time distribution consisting of 3 components derived from mode 1 ( values of 0.4 and 2.3 ms) and mode 2 activity ( = 10 ms). The transition rate between C2 and C3 that resides within mode 1 activity is about 10-fold slower when the channel is conducting Ca2+ in contrast to Ba2+ ions. This would provide a monoexponential open time distribution ( = 0.4 ms) during control activity (Figs. 6 and 7). Addition of agonist to the channel conducting Ca2+ ions would increase the transition rate between C1 and C2 (marked ), producing a dominant exponential distribution of time constant 0.4 ms with a very minor additional component of time constant 2.3 ms (Figs. 6 and 7). The transition rate between C1 and C4 into mode 2 is not permitted while the channel conducts Ca2+ ions, with mode 2 observed only when the channel conducts Ba2+ ions. This scheme suggests that gating is dependent on the divalent ion, with channel states accessible only if a nonphysiological charge carrier is conducted through the channel. Finally, it supports the experimental observation that agonist-modified openings are very unlikely to be seen spontaneously.
Slowing of the deactivation current by BAY K 8644 was not observed in this study with either cation as the charge carrier, despite the recruitment of long-duration openings at 0 mV when using Ba2+ ions as the charge carrier. It is worth noting that channel openings were not observed to exceed the pulse duration, supporting the lack of a slowing of macroscopic current deactivation. BAY K 8644 slows the macroscopic deactivation tail current in cardiac myocytes, an effect thought to arise from the slow closing rate associated with mode 2 openings (Bechem and Hoffmann 1993; Fan et al. 2000; Tsien et al. 1986). Slowing of macroscopic deactivation by the related dihydropyridine agonist (+)-202-791 has been reported in a variety of neurons when using Ba2+ ions as the charge carrier (Kammermeier and Jones 1997; Plummer et al. 1989; Regan et al. 1991). Contrary results have been reported from hippocampal neurons, where BAY K 8644 slowed current deactivation only when the current was carried by Ba2+ and not Ca2+ ions (Docherty and Brown 1986; Toselli and Taglietti 1992), whereas slowing was observed by others when using Ca2+ ions as the charge carrier (Ishibashi et al. 1998). In contrast, a lack of effect of BAY K 8644 on current deactivation has been reported in pancreatic cells (Smith et al. 1993). BAY K 8644 slowed tail currents when recombinant CaV1.2 channels were expressed in dysgenic myotubes or in CHO cells (Aoyama et al. 2003; Wilkens et al. 2001), but not when coexpressed with the cardiac 2a subunit in Xenopus oocytes (Noceti et al. 1998). Conflicting results have also been observed for the effect of BAY K 8644 on CaV1.3 channels. For example, BAY K 8644 slowed whole cell deactivation tail currents from native CaV1.3 channels in pinealocytes (Chik et al. 1997), whereas those in dorsal root ganglion neurons were not affected (Wyatt et al. 1997). This dichotomy is also apparent with recombinant CaV1.3 channels, as slowing of CaV1.3-mediated current by BAY K 8644 has been observed by some (Xu and Lipscombe 2001) but not others (Bell et al. 2001; Koschak et al. 2001). Recent evidence indicates that the related dihydropyridine agonist (+)-202-791 produces only a modest slowing of deactivation tail currents in both PC12 cells and rat superior cervical ganglion cells whose L-type current is thought to be predominantly mediated by CaV1.2 and CaV1.3, respectively (Liu et al. 2003). Thus the deactivation rate in the presence of dihydropyridine agonists does not appear to distinguish between either class of channel.
FPL 64176 evoked an obvious slowing of the deactivation tail current carried by Ca2+ ions in cardiac myocytes (Fan et al. 2000). In the present study, FPL 64176 also dramatically slowed the whole cell deactivation current when Ba2+ was the charge carrier (Fig. 1D) but produced only a modest prolongation of deactivation time course when Ca2+ was the conducting ion (Fig. 1C). In addition, the agonist did produce a kinetic slowing of macroscopic current activation, regardless of the identity of the charge carrier (Fig. 1, C and D). This resulted from a delay of FPL 64176–induced long-duration channel openings, which occurred subsequent to brief openings and caused most openings to occur toward the end of the voltage step (Figs. 4, 5, 6, and 9A). Similar behavior has been observed for recombinant CaV1.2 channels treated with FPL 64176 (Lauven et al. 1999). These data suggest that the binding site for FPL 64176 is primarily accessible from the open state. Similar FPL 64176–induced slowing of whole cell activation and deactivation in Ba2+-containing solutions has also been recently observed in other neuronal cell types, which like hippocampal neurons may contain a combination of CaV1.2 and CaV1.3 (Liu et al. 2003). This suggests that the slowing of activation reflects a property of the agonist that is likely to be common to both classes of L-type channels.
The amplitude of single hippocampal L-type channels conducting either Ba2+ (110 mM) or Ca2+ (160 mM) was much smaller than that recorded in cardiac tissue. For example, single L-type channels in the absence of channel agonists conducting Ca2+ ions (160 mM) at 0 mV was about 191 fA in hippocampal neurons (Marrion and Tavalin 1998; Figs. 2 and 7), whereas those conducting Ba2+ ions (110 mM) was about 950 fA (Figs. 2 and 6). In contrast, L-type channel amplitude in saturating Ca2+ and Ba2+ ion concentrations was about 410 fA and 1.4 pA, respectively, in the absence of channel agonists, in cardiac myocytes (Guia et al. 2001; Hess et al., 1986). These differences may be partly attributable to the apparent affinity of the ion for the channel being higher in hippocampus (Kd 4.2 mM for Ca2+ and 13.5 mM for Ba2+) compared with cardiac tissue (Kd 14 mM for Ca2+ and 28 mM for Ba2+; Hess et al. 1986).
The effect of these L-type channel agonists on single-channel conductance has been controversial. For example, BAY K 8644 increased single-channel conductance in hippocampal neurons (Figs. 2 and 3; Cloues et al. 1997), GH3 pituitary cells (Mantegazza et al. 1995), cardiac myocytes (Kokubun and Reuter 1984; Lacerda and Brown 1989), and smooth muscle cells (Caffrey et al. 1986). In contrast, no effect of the dihydropyridine agonist on channel conductance was observed in dorsal root ganglion cells (Fox et al. 1987), pancreatic cells (Smith et al. 1993), or cardiac myocytes (Hess et al. 1986). In contrast, FPL 64176 increases single L-type channel conductance in cardiac myocytes (Fan et al. 2001), but not in hippocampal neurons (Fig. 4). It is tempting to suggest that CaV1.3 channels dominated the hippocampal L-type current, with little arising from CaV1.2 channels, because hippocampal neurons express both CaV1.2 and CaV1.3 (Bowden et al. 2001), whereas cardiac myocytes exclusively express CaV1.2 subunits. Certainly, the larger conductance expected from CaV1.2 channels was not observed in this study. Therefore it is possible that, even though the CaV1.2 channel is present, it does not provide a significant component of Ca2+ entry into hippocampal neurons. However, several splice variants of the pore-forming 
subunit exist for both CaV1.2 and CaV1.3 and may contribute to the conflicting behavior of L-type Ca2+ channels in various tissues (Safa et al. 2001; Welling et al. 1997). In addition, auxiliary subunits may also contribute to the aforementioned differences between hippocampal neurons and other tissues. Therefore it remains important to systematically determine the effects of channel composition on L-channel activity and pharmacology.
Regardless of the precise molecular identity of the L-type channels recorded in the present study, it is clear that use of these compounds with nonphysiological charge carriers may not be simply extrapolated to physiological conditions. Therefore the identity of the permeant ion must be taken into consideration when using agonists to characterize L-type channels.
|
GRANTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Present address for S. E. Bowden: Ionix Pharmaceuticals Ltd., Cambridge, UK.
|
FOOTNOTES |
|---|
Address for reprint requests and other correspondence: N. V. Marrion, Department of Pharmacology and MRC Centre for Synaptic Plasticity, School of Medical Sciences, University of Bristol, University Walk, Bristol BS8 1TD, UK (E-mail: N.V.Marrion{at}bris.ac.uk).
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Bechem M and Hoffmann H. The molecular mode of action of the Ca agonist (–)BAY K 8644 on the cardiac Ca channel. Pfluegers Arch 424: 343–353, 1993.[CrossRef][ISI][Medline]
Bell DC, Butcher AJ, Berrow NS, Page KM, Brust PF, Nesterova A, Stauderman KA, Seabrook GR, Nürnberg B, and Dolphin AC. Biophysical properties, pharmacology, and modulation of human, neuronal L-type (1D, CaV1.3) voltage-dependent calcium currents. J Neurophysiol 85: 816–827, 2001.
Bowden SEH, Fletcher S, Loane DL, and Marrion NV. Somatic co-localization of rat SK1 and D class (Cav1.3) L-type calcium channels in rat CA1 hippocampal pyramidal neurons. J Neurosci 21: RC175, 1–6, 2001.
Caffrey JM, Josephson IR, and Brown AM. Calcium channels of amphibian stomach and mammalian aorta smooth muscle cells. Biophys J 49: 1237–1242.
Chik CL, Liu QY, Li B, Klein DC, Zylka M, Kim DS, Chin H, Karpinski E, and Ho AK. Alpha 1D L-type Ca(2+)-channel currents: inhibition by a beta-adrenergic agonist and pituitary adenylate cyclase-activating polypeptide (PACAP) in rat pinealocytes. J Neurochem 68: 1078–1087, 1997.[ISI][Medline]
Cloues RK, Tavalin SJ, and Marrion NV. -Adrenergic stimulation selectively inhibits long-lasting L-type calcium channel facilitation in hippocampal neurons. J Neurosci 17: 6493–6503, 1997.
Colquhoun D and Sigworth FJ. Fitting and statistical analysis of single-channel records. In: Single-Channel Recording (2nd ed.), edited by Sakmann B and Neher E. New York: Plenum Press, 1995, p. 438–588.
Docherty RJ and Brown DA. Interaction of 1.4-dihydropyridines with somatic Ca currents in hippocampal CA1 neurones of the guinea pig in vitro. Neurosci Lett 70: 110–115, 1986.[CrossRef][ISI][Medline]
Eckert R and Chad J. Inactivation of Ca channels. Prog Biophys Mol Biol 44: 261–267, 1984.
Elliott EM, Malouf AT, and Catterall WA. Role of calcium channel subtypes in calcium channel subtypes in calcium transients in hippocampal CA3 neurons. J Neurosci 15: 6433–6444, 1995.
Fan JS, Yuan Y, and Palade P. Kinetic effects of FPL 64176 on L-type Ca2+ channels in cardiac myocytes. Naunyn Schmiedebergs Arch Pharmacol 361: 465–476, 2000.[CrossRef][ISI][Medline]
Fan JS, Yuan Y, and Palade P. FPL-64176 modifies pore properties of L-type Ca2+ channels. Am J Physiol Cell Physiol 280: C565–C572, 2001.
Fisher RE, Gray R, and Johnston D. Properties and distribution of single voltage-gated calcium channels in adult hippocampal neurons. J Neurophysiol 64: 91–104, 1990.
Fox AP, Nowycky MC, and Tsien RW. Single-channel recordings of three types of calcium channels in chick sensory neurones. J Physiol 394: 173–200, 1987.
Guia A, Stern MD, Lakatta EG, and Josephson IR. Ion concentration-dependence of rat cardiac unitary L-type calcium channel conductance. Biophys J 80: 2742–2750, 2001.
Hell JW, Westernbroek RE, Warner C, Ahlijanian MK, Prystay W, Gilbert MM, Snutch TP, and Catterall WA. Identification and differential subcellular localization of the neuronal class C and class D L-type calcium channel 1 subunits. J Cell Biol 123: 949–962, 1993.
Hess P, Lansman JB, and Tsien RW. Different modes of Ca channel gating behaviour favoured by dihydropyridine Ca agonists and antagonists. Nature 311: 538–544, 1984.[CrossRef][Medline]
Hess P, Lansman JB, and Tsien RW. Calcium channel selectivity for divalent and monovalent cations. Voltage and concentration dependence of single channel current in ventricular heart cells. J Gen Physiol 88: 293–319, 1986.
Horn R and Lange K. Estimating kinetic constants from single channel data. Biophys J 43: 207–223, 1983.
