HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) All Versions of this Article: 578/3/693    most recent jphysiol.2006.121905v1 Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Tringham, E. W. Articles by Usowicz, M. M. Search for Related Content PubMed PubMed Citation Articles by Tringham, E. W. Articles by Usowicz, M. M. Related Collections Neuroscience

¹ Department of Pharmacology, University of Bristol, University Walk, Bristol BS8 1TD, UK

	Abstract

Top Abstract Introduction Methods Results Discussion References

 
To determine if the properties of Ca²⁺ channels in cerebellar Purkinje cells change during postnatal development, we recorded Ca²⁺ channel currents from Purkinje cells in cerebellar slices of mature (postnatal days (P) 40–50) and immature (P13–20) rats. We found that at P40–50, the somatic Ca²⁺ channel current was inhibited by -agatoxin IVA at concentrations selective for P-type Ca²⁺ channels (85%; IC₅₀, <1 nM) and by the dihydropyridine (–)-(S)-Bay K8644 (70%; IC₅₀, 40 nM). (–)-(S)-Bay K8644 is known to activate L-type Ca²⁺ channels, but the decrease in current was not secondary to the activation of L-type channels because inhibition by (–)-(S)-Bay K8644 persisted in the presence of the L-type channel blocker (R,S)-nimodipine. By contrast, at P13–20, the current was inhibited by -agatoxin IVA (86%; IC₅₀, 1 nM) and a minor component was inhibited by (R,S)-nimodipine (8%). The dihydropyridine (–)-(S)-Bay K8644 had no clear effect when applied alone, but in the presence of (R,S)-nimodipine it reduced the current (40%), suggesting that activation of L-type channels by (–)-(S)-Bay K8644 masks its inhibition of non-L-type channels. Our findings indicate that Purkinje neurons express a previously unrecognized type of Ca²⁺ channel that is inhibited by -agatoxin IVA, like prototypical P-type channels, and by (–)-(S)-Bay K8644, unlike classical P-type or L-type channels. During maturation, there is a decrease in the size of the L-type current and an increase in the size of the atypical Ca²⁺ channel current. These changes may contribute to the maturation of the electrical properties of Purkinje cells.

(Received 27 September 2006; accepted after revision 22 November 2006; first published online 23 November 2006)
Corresponding author M. Usowicz: Department of Pharmacology, University of Bristol, University Walk, Bristol BS8 1TD, UK. Email: m.m.usowicz{at}bris.ac.uk

	Introduction

Top Abstract Introduction Methods Results Discussion References

 
Voltage-gated Ca²⁺ channels regulate electrical activity and many intracellular processes (Catterall et al. 2005). The classification of Ca²⁺ channels as T-, N-, L-, P-, Q- or R-type is based on the electrophysiological and pharmacological properties of Ca²⁺ channel currents in different cell types (Catterall et al. 2005). A finer subdivision of this classification follows from the identification of 10 classes of the pore-forming Ca_V (1) Ca²⁺ channel subunit and alternative pre-mRNA splicing of the gene encoding each Ca_V class, as well as different classes and splice variants of the accessory ₂, and subunits (Catterall et al. 2005).

Currents mediated by native L-type channels or by recombinant channels consisting of one of the four classes of the Ca_V1 family (Ca_V1.1, Ca_V1.2, Ca_V1.3 and Ca_V1.4) are inhibited or activated by dihydropyridines (DHPs). The effect (inhibition or activation) varies between DHPs and between the enantiomers of some DHPs, while the potency of inhibition varies between the different Ca_V1 classes (Triggle & Rampe, 1989; Bechem & Hoffmann, 1993; Striessnig et al. 1998; Xu & Lipscombe, 2001; Koschak et al. 2001). Currents mediated by native non-L-type channels and by recombinant channels containing a Ca_V2 (Ca_V2.1, Ca_V2.2, Ca_V2.3) or Ca_V3 (Ca_V3.1, Ca_V3.2, Ca_V3.3) subunit are considered to be insensitive to DHPs but can be distinguished on the basis of inhibition by various toxins (Llinás et al. 1989; Regan et al. 1991; Usowicz et al. 1992; Mintz et al. 1992a; Boland et al. 1994; Dupere et al. 1996; McDonough et al. 1996; Tottene et al. 2000). Prototypical P-type channels in the soma of cerebellar Purkinje cells dissociated from the cerebellum of immature rats (postnatal days (P) 5–21) (Mintz et al. 1992a; Mintz et al. 1992b) and channels containing a cloned Ca_V2.1 subunit (Catterall et al. 2005) are inhibited by -agatoxin IVA and are insensitive to DHPs. However, it has been suggested that the DHP Bay K8644 (racemic form), an activator of L-type channels, may inhibit P-type currents in Purkinje cells of mature cerebellum (Usowicz et al. 1992). This effect was not thoroughly studied and the enantiomers effecting the inhibition were not known. We have therefore examined the sensitivity of Ca²⁺ channel currents in the soma of mature (P40–50) and immature (P13–20) rat cerebellar Purkinje cells to the individual enantiomers of Bay K8644, the DHP L-type channel blocker nimodipine, and -agatoxin IVA.

We find that during maturation of the Purkinje cell, there is a decrease in the sizes of the Ca²⁺ channel currents carried in the soma by L-type channels or prototypical P-type channels, and a doubling of the size of an atypical current inhibited by both -agatoxin IVA and (–)-(S)-Bay K8644. The predominance of the atypical Ca²⁺ channel in mature but not in immature Purkinje cells may contribute to the different electrical properties of Purkinje cells at different stages of postnatal development (McKay & Turner, 2005). Furthermore, these data extend the known functional diversity of Ca²⁺ channels. They also suggest that the use of (–)-(S)-Bay K8644 to detect the expression and physiological roles of L-type channels may prevent the identification of some -agatoxin-IVA-sensitive channels.

	Methods

Top Abstract Introduction Methods Results Discussion References

 
Cerebellar slices

Parasagittal slices of cerebellar vermis (250–300 µm) were prepared from male Wistar rats as previously described (Usowicz et al. 1992). Briefly, adult (selected by age, P40–50, or by weight, 150–200 g) or young rats (P13–20) were culled by cervical dislocation and decapitated, in accordance with the United Kingdom Animals (Scientific Procedures) Act 1986, and with the University of Bristol Ethical Review Committee. The cerebellum was exposed and cooled with ice-cold oxygenated Krebs-Henseleit solution (mM: 124 NaCl, 1.3 MgSO₄, 5 KCl, 2.4 CaCl₂, 1.2 KH₂PO₄, 26 NaHCO₃, 10 D-glucose, pH 7.4, bubbled with 95% O₂/5% CO₂). The vermis was excised and cut into slices on a Vibratome (Series 1000; Pelco, Redding, CA, USA) or a Leica VT1000S vibrating microtome (Leica Microsystems, Nussloch, Germany), under continuous oxygenation. Slices were kept at room temperature for 1–7 h before recording. For recording, individual slices were viewed on a FS Axioskop microscope (Carl Zeiss, Welwyn Garden City, UK) and superfused continuously (1–1.5 ml min^–1) with oxygenated (100% O₂) Hepes-buffered saline solution (mM: 133 NaCl, 2.5 KCl, 2.4 CaCl₂, 10 D-glucose, 20 Hepes, 1.3 MgCl₂, pH 7.4 with NaOH, plus 1 µM tetrodotoxin (TTX)) at 21–25°C. Prior to recording with an Axopatch 200A amplifier (Axon Instruments, Union City, CA, USA), the soma of a Purkinje cell was cleaned of any overlying debris by applying a fine stream of extracellular solution from a pipette (tip diameter, 5 µm).

Cell-attached recording

For cell-attached voltage-clamp recording, patch pipettes (thick-walled borosilicate glass capillaries with an inner filament, Harvard Apparatus, Kent, UK) were backfilled with a filtered (0.2 µm, cellulose acetate membrane) solution containing (mM): 5 BaCl₂, 10 CsCl, 10 Hepes, 150 TEA-Cl, 0.1 EGTA, pH 7.4 with TEA-OH, plus 1 µM TTX. The resistances of the pipettes, when filled with this solution, were between 4 and 10 M (90% were between 4.5 and 7.7 M). The pipette potential was set between +30 and +50 mV, in order to set the patch potential between 30 and 50 mV negative to the resting cell potential. Currents were evoked by applying a depolarizing voltage ramp to the pipette (0 (r) –170 mV, 0.53 mV ms^–1) followed by a repolarizing voltage ramp (–170 (r) 0 mV, 0.53 mV ms^–1) at 0.2 Hz for 2.5–7.5 min (Usowicz et al. 1992; Dupere et al. 1996; Kanumilli et al. 2006), or sometimes by applying voltage jumps (50 ms duration, four jumps to each potential, 0.2 Hz), using a Cambridge Electronic Design (CED) 1401 plus A/D interface and CED patch and voltage-clamp software (v.6.37; CED, Cambridge, UK). Evoked currents were low-pass filtered at 10 kHz (four-pole Bessel filter on the amplifier) and then at 2.04 kHz (ramps) or 5 kHz (jumps), eight-pole low-pass Bessel filter (Frequency Devices, Haverhill, MA, USA), acquired at 7 kHz (ramps) or 40 kHz (jumps) and analysed off-line by CED patch and voltage-clamp programs. Further analysis and subtraction of a linear leak current were performed with Origin v.6 (OriginLab Inc., Northampton, MA, USA). For currents evoked by voltage jumps, the leak current for each recording was estimated by averaging currents recorded during 30 mV jumps (10 or 20) that did not activate Ca²⁺ channels. This was scaled up and then subtracted from the average current evoked by four voltage jumps to the same potential, to yield the Ca²⁺ channel current.

At the end of each cell-attached recording, a whole-cell configuration was established in order to measure the resting membrane potential (TTX in the extracellular solution prevented spontaneous spiking). This was measured immediately, before it could be altered by diffusion of the Ba²⁺/TEA pipette solution into the cell. The reported patch potentials (e.g. –100 to +70 mV) were calculated as the resting cell potential (e.g. –60 mV) minus the pipette holding potential (e.g. +40 mV) and minus the voltage ramp (0 (r) –170 mV) or the voltage jump applied to the pipette. They were not corrected for the liquid junction potential between the bath solution and the pipette solution, which was only –1 mV. Any inaccuracy in the measurement of the cell resting potential does not affect the findings presented here, because the analysis of the cell-attached currents consisted of measurement of the size of the peak current, irrespective of the potential at which it occurred (see below). We found it was not possible to make cell-attached recordings from mature Purkinje cells after zeroing the membrane potential by applying an extracellular solution containing 135 mM KCl instead of NaCl, in the presence of 1 µM TTX, as is the case for many cell-attached recordings of Ca²⁺ channels in cell lines or dissociated neurons. Most Purkinje cells became swollen and transparent within minutes of the application of the high KCl solution. It was difficult to seal a pipette onto the few surviving Purkinje cells.

Concentration–inhibition curves

Cell-attached recordings were made with or without a drug in the pipette. Either different pipette holders were used to hold a pipette filled with a drug-containing or a drug-free solution, or the Ag/AgCl wire of a single pipette holder was cleaned with ethanol and water between recordings. Recordings in the presence or absence of a drug were interleaved throughout the day. In some cases, the experiments were performed ‘blind’, the identity of the pipette solutions not being revealed till the end of the day. For each cell-attached recording with a drug-free pipette solution, currents were evoked by 30 depolarizing voltage ramps. Similarly, for each recording with a drug-containing pipette solution, 30 ramps were applied to each patch, unless it was apparent during the recording that a drug-induced change in peak current had not reached steady state within 2.5 min. In such instances, a further 30 or 60 ramps were applied and only the later currents, which did not differ in size over 15–30 ramps, were averaged in order to get a measure of the peak Ca²⁺ channel current, I_pk, for that patch. Recordings in 100 nM -conotoxin MVIIC were kept for as long as possible, since it blocks P-type channels extremely slowly (McDonough et al. 1996). The rates of block by different concentrations of -conotoxin MVIIC were quantified by fitting single exponential functions to plots of peak current against time by non-linear least squares regression (Levenberg-Marquardt algorithm, Origin v.6). The sizes of I_pk in different patches in the absence of drug and in the presence of the same drug concentration are summarized as means ± S.E.M. The relationship between I_pk and log drug concentration was fitted by a logistic curve, of the form:

(1)

where max is the maximum I_pk in the absence of drug, min is the minimum current remaining, i.e. the current not blocked by the drug, IC₅₀ is the molar drug concentration that produces 50% block of the sensitive current, and n_H is the Hill slope. Statistical significance was calculated using a two-tailed unpaired Student's t test, or one-way ANOVA followed by Dunnett's test, with significance taken at the P < 0.05 level.

Whole-cell recording

For whole-cell voltage-clamp recording from Purkinje cells, slices were superfused with Hepes-buffered saline. For recordings from P13–20 animals, the solution also contained 10 µM SR 95531 hydrobromide and 20 µM Na₂CNQX in order to inhibit spontaneous synaptic currents, which were more frequent than in mature cells. Patch pipettes (thin-walled borosilicate glass capillaries with an inner filament; Harvard Apparatus, Kent, UK) were backfilled with a filtered (0.2 µm, cellulose acetate membrane) solution containing (mM): 100 CsCl, 10 Hepes, 4.5 MgCl₂, 4 MgATP, 14 phosphocreatine-di Tris, 0.3 GTP-Tris, 10 EGTA, pH 7.2 with TEA-OH. The pipette shanks were coated with Sylgard resin and polished down to resistances of 1.25 M. Once a whole-cell configuration was established, the first component of the two-component capacitance transients evoked by 5 mV voltage jumps, which probably represent the somatic capacitance and the dendritic capacitance, was cancelled (28.1 ± 0.4 pF for mature Purkinje cells, n = 145; 26.4 ± 0.4 pF for immmature Purkinje cells, n = 88). The series resistance compensation and correction were set at 85–95% with a 10 µs lag, using controls on the Axopatch 200A amplifier. A depolarizing voltage ramp (–100 mV (r) –70 mV, 0.55 mV ms^–1) followed immediately by a repolarizing voltage ramp (+70 mV (r) –100 mV, 0.55 mV ms^–1) was applied at 0.05 Hz. The cell was then superfused with an extracellular Hepes-based solution supplemented with 1 µM TTX, in which 2.4 mM CaCl₂ was replaced by 2 mM BaCl₂, from a needle (made of polyimide, 360 µm in diameter; Digitimer Ltd, Welwyn Garden City, UK) placed under the water immersion objective 0.5 cm away from the soma. During the first 6–10 min of recording, the outward current was reduced in size and the inward Ca²⁺ channel current carried by 2 mM Ba²⁺ was revealed, as the pipette solution dialysed in to the cells and the extracellular solution around the cell was exchanged for the Ba²⁺-containing solution. The main extracellular cation was Na⁺ and not TEA⁺ as in the cell-attached recordings, because pilot whole-cell recordings revealed that the condition of the slice deteriorated in the presence of 150 mM TEA too rapidly for us to examine the effects of certain drugs. The deterioration also occurred in the presence of 20 mM TEACl. Under these conditions, Ca²⁺ channel currents were recorded in the presence of an outward current, which was revealed as the inward currents were inhibited, and which has previously been described as a current through K⁺ channels insensitive to block by intracellular Cs⁺ (Bushell et al. 2002). This limits the accuracy of our measurements of pharmacological blockade since the outward current may have caused us to underestimate the inward current in drug-free solution, and may have masked a small drug-resistant inward current. The currents were low-pass filtered at 2 kHz (10 kHz, four-pole Bessel filter on the amplifier, followed by 2.04 kHz, eight-pole low-pass Bessel filter; Frequency Devices, Haverhill, MA, USA), acquired at 6.2 kHz and analysed off-line by CED patch and voltage-clamp programs. Further analysis was performed with Origin v.6.

In some experiments, the effects of drugs on the whole-cell Ca²⁺ channel currents were examined by switching the solution flowing out of the 360 µm needle from a drug-free to a drug-containing ‘2 mM Ba²⁺ extracellular solution’, which also contained 1 µM TTX. The needle was connected via a multibarrel manifold (Digitimer Ltd) and Teflon tubing and valves to solutions stored in glass syringes. In other experiments, once the whole-cell Ca²⁺ channel currents had become established, a drug-containing pipette (10 µm in diameter) was lowered into the bath and placed 20–60 µm away from the soma (see Fig. 4A). The bath perfusion was switched off, the flow of the 2 mM Ba²⁺ solution out of the needle was stopped, and drug-containing, 2 mM Ba² ± TTX solution, was forced out of the pipette on to the soma by applying mouth pressure to the pipette via a piece of tubing. The pressure was maintained by closing a tap at the mouth-piece. The series resistance (mean value before 85–95% compensation: 4.6 ± 0.3 M for mature Purkinje cells, n = 66; 4.9 ± 0.2 M for immature Purkinje cells, n = 88) was monitored with 5 mV jumps throughout each recording. Recordings were terminated if the series resistance increased above 10 M (before compensation). The holding current was monitored throughout each recording so as to ensure that any drug-induced decreases in Ca²⁺ channel current were not due to a decrease in the input resistance of the cell or a decrease in the accuracy of the voltage-clamp.

View larger version (21K): [in this window] [in a new window]   Figure 4.  Whole-cell Ca2+ channel currents in mature cerebellar Purkinje cells run down slowly and are almost completely blocked by P-type-selective concentrations of -agatoxin IVA A, example of a current evoked by a depolarizing voltage ramp (–96 to +74 mV) applied to a Purkinje cell in the whole-cell configuration, approximately 16 min after establishing the whole-cell recording configuration and about 14 min after commencement of superfusion of the cell with an extracellular solution containing 2 mM Ba2+ as the main divalent cation, and 1 µM TTX to block Na+ channels. The horizontal dashed line (left) denotes the zero current level. A linear leak current was estimated by fitting a straight line (dotted line) to the current recorded during the negative potentials of the voltage ramp that did not evoke an inward current (–96 to –65 mV). It was extrapolated over the whole length of the recording, and the peak inward Ca2+ channel current, Ipk, was measured from the dotted line. B, scatter diagram and box plot of whole-cell Ipk measurements in many mature Purkinje cells, made about 10 min after establishing each whole-cell recording. C, plots of whole-cell Ipk, normalized by size at time zero, against time in 13 cells (grey circles) showing that the size of the current runs down slowly (6% decrease in 10 min, measured from the mean Ipk, empty circles). D, time course of block and unblock of the inward current recorded in a single cell by 100 µM Cd2+. The superimposed traces to the right show the effect of Cd2+ on the currents. Note that the decrease in inward current was not accompanied by a change in holding current (horizontal dashed line indicates zero current level for all three traces). E, comparison of the time course of block of whole-cell Ipk by superfusion of 100 nM -agatoxin IVA on to five cells (filled circles, mean; vertical bars, ±S.E.M.), with a plot of the mean whole-cell Ipk recorded in 13 cells in the absence of drug (open circles, as in C). The superimposed current traces to the right show the effect of 100 nM -agatoxin IVA on currents evoked by depolarizing voltage ramps in a single cell. The inhibition of the inward current by -agatoxin IVA was not accompanied by a change in holding current (horizontal dashed line denotes zero current level for all three traces). It revealed an outward current (trace above the dotted line), as did the inhibition by Cd2+ (see D and Methods).

Stock solutions of -agatoxin IVA (100 µM; Peptides Institute, Japan or Scientific Marketing Associates, Barnet, UK) and -conotoxin MVIIC and -conotoxin GVIA (1 mM; Peninsula Laboratories Europe Ltd, UK) were prepared, respectively, in sterile degassed water or in filtered Milli-Q water, and stored as 10–15 µl aliquots at –80°C (water was degassed by vacuum for approximately 30 min and then perfused with nitrogen for a further 30 min). Stock solutions of Na₂CNQX (Tocris Bioscience, UK) and SR 95531 hydrobromide (Tocris Bioscience) were prepared in water. Dilutions to the final concentrations were made on the day of the experiment. The (–)-(S) and (+)-(R) enantiomers of Bay K8644 (Research Biochemicals Inc., MA, USA; or Tocris Bioscience, UK; or Sigma), racemic Bay K8644 (Research Biochemicals Inc.), nimodipine (Bayer PLC, UK), and FPL 64176 (Research Biochemicals Inc. or Sigma) were prepared as 3 or 10 mM stock solutions in 100% dimethylsulfoxide (DMSO) and stored at 4°C or as aliquots at –20°C. Because of the photosensitivity of Bay K8644, nimodipine and FPL 64176, solutions were prepared under subdued light and stored in brown glass vials. Furthermore, the microscope light was turned down during recording of currents in the presence of these drugs. In addition, since these drugs absorb to plastic, glass capillaries were used to make dilutions to the final concentration on the day of recording, and glass Pasteur pipettes pulled to a fine tip were used to introduce the drug solutions into pipettes. Cell-attached recordings with 0.1% DMSO, a concentration equivalent to the highest concentration present in the solutions containing Bay K8644, did not potentiate or inhibit the Ca²⁺ channel currents (P42–43, n = 9; P16–20; n = 10). The purity of the enantiomers of Bay K8644 (purchased from Research Biochemicals Inc.) was checked by chiral thin-layer chromatography (MN-ChiralPlate, Camlab, UK), using methanol:water:acetonitrile (1:1:4) as the elution system. This showed that each stereoisomer was enantiomerically pure, as they both eluted as a single spot. This observation also strongly supports the assumption that each enantiomer was free from contamination from degradation products resulting from storage of stock Bay K8644 solutions.

	Results

Top Abstract Introduction Methods Results Discussion References

 
Pharmacological identification of P-type Ca²⁺ channel currents in cell-attached patches of the soma of mature Purkinje cells

The pharmacology of Ca²⁺ channels in the soma of mature Purkinje cells was investigated by recording macroscopic currents carried by 5 mM Ba²⁺ in cell-attached patches of cells in cerebellar slices, with drug-free and drug-containing pipette solutions (Usowicz et al. 1992; Dupere et al. 1996; Kanumilli et al. 2006). Spontaneous firing of the Purkinje cells was prevented by the inclusion of 1 µM TTX in the bathing and pipette solutions, and K⁺ currents were inhibited by Ba²⁺, TEACl and CsCl in the pipette solution. We initially used cell-attached recording because this recording mode does not perturb the intracellular environment and therefore circumvents potential rundown of Ca²⁺ channel currents. It obviates concerns about inadequate space-clamp that hinder accurate whole-cell recording of voltage-gated currents from most fully differentiated neurons. Inclusion of the drugs in the recording pipette avoids uncertainty about the effective drug concentration at the channels of interest, which can arise during bath application of peptide toxins and lipophilic drugs as a result of binding to the perfusion system and to the slices. Also, recording from cells in slices avoids the potential for alteration of channel properties during enzymatic dissociation of neurons.

The currents were evoked by depolarizing voltage ramps in somatic patches of cells in cerebellar slices (Fig. 1A). We found that the seal between the pipette and the adult Purkinje neurons could be weakened by voltage jumps applied to the cell-attached pipette but was more able to withstand the small increments in voltage that occur during a voltage ramp. Hence, voltage ramps allowed for recordings of longer duration over a wider range of voltages. Since the patch potential in the cell-attached configuration depends on the voltage applied to the pipette and the cell membrane potential, any change in the cell membrane potential or in the resistance of the pipette-membrane seal can alter the size of the recorded currents. Therefore, the stability of each recording was assessed by applying a minimum of 30 voltage ramps to each patch and comparing the relative position of the currents (Usowicz et al. 1992; Dupere et al. 1996; Kanumilli et al. 2006). During a stable recording, there was no change in the shape of the currents, as depicted by the superimposed traces in Fig. 1A. Multiple traces obtained during stable recordings were averaged to give the mean current for each patch (black line, Fig. 1A). A leak current (dashed line, Fig. 1A) was then estimated by fitting a straight line to the linear part of the mean current (a and b, Fig. 1A). It was subtracted from the mean current, to give the mean Ca²⁺ channel current carried by 5 mM Ba²⁺ for each patch (Fig. 1B), which was measured at the peak (I_pk). Unstable recordings were identified by leftward shifts in the currents, as shown in Fig. 1C, and were excluded. Finally, currents in the presence or absence of drugs were obtained from cells in the same slices by alternatively recording with drug-free and drug-containing pipettes, so as to ensure that any drug-induced differences were not due to differences in current size between different slices or animals.

View larger version (17K): [in this window] [in a new window]   Figure 1.  Cell-attached recording of Ca2+ channel currents carried by 5 mM Ba2+ in the soma of Purkinje cells in cerebellar slices A, superimposed current traces (grey) recorded from a single patch (P43 animal), showing constancy in the shape of currents evoked by 30 voltage ramps (top) applied at 0.2 Hz. The average of these currents is shown by the black trace. A linear leak current (dashed line) was estimated by fitting a straight line to the linear part of the mean trace (a–b) and extrapolating it over the full range of the ramp. B, the mean Ca2+ channel current obtained by subtracting the leak current from the mean total current. The size of the Ca2+ channel current was measured as Ipk from the zero current level (dotted line) as indicated. C, example of an unstable cell-attached recording. The current traces were evoked by the 1st, 6th, 12th, 18th and 24th voltage ramp, and illustrate a characteristic leftward shift accompanied by a decrease in size.

View larger version (18K): [in this window] [in a new window]   Figure 2.  Pharmacological confirmation that inward currents evoked by depolarizing voltage ramps in somatic patches of mature cerebellar Purkinje cells (P40–50) are mediated mostly by P-type Ca2+ channels A, superimposed Ca2+ channel currents recorded with (bottom, 13 patches) and without 100 µM Cd2+ (top, 20 patches) in the patch-pipette, from cells within the same slices. In this and subsequent figures, each current trace is the mean of currents evoked by 15–35 voltage ramps in a single patch, from which a leak current has been subtracted, as in Fig. 1. B, concentration dependence of inhibition of Ipk by Cd2+. Symbols and vertical bars plot means ± S.E.M. of measurements of Ipk in the numbers of patches given in parentheses. The continuous curve is the best fit with the logistic equation, for which IC50 is 40 µM, Hill slope is 1.6, max – min is 28.6 pA (100%) (max is the maximum Ipk in the absence of drug, and min is the minimum current remaining). C, superimposed Ca2+ channel currents recorded in 51 patches with control pipette solution, and in 41 patches with 30 nM -agatoxin-IVA in the pipette, from the same slices. D, concentration dependence of the decrease of Ipk by -agatoxin-IVA. The continuous line is the best fit of the logistic equation, with IC50 0.4 nM, Hill slope 0.59, max – min 26 pA (85%). E, superimposed Ca2+ channel currents recorded with drug-free pipette solution (eight patches) and with 10 µM -conotoxin MVIIC in the pipette (six patches) showing 99% inhibition. F, concentration dependence of the rate of Ca2+ channel inhibition by -conotoxin MVIIC. Plots illustrate the time course of inhibition by three different concentrations in three different patches. Each is fitted with a single exponential function (continuous line), characterized by the time constants shown. The zero time point of each exponential was defined as the time of seal formation. Recordings started 60 s (1 µM) or 80 s (100 nM, 10 µM) after seal formation.

(–)-(S)-Bay K8644 inhibits somatic P-type Ca²⁺ channels in mature cerebellar Purkinje cells

The (–)-(S)-enantiomer of the DHP Bay K8644, which is a potent and selective activator of L-type Ca²⁺ channels (EC₅₀ = 30 nM) (Triggle & Rampe, 1989; Bechem & Hoffmann, 1993; Striessnig et al. 1998), reduced the size of the Ca²⁺ channel currents in mature cerebellar Purkinje cells (Fig. 3A). Inclusion of (–)-(S)-Bay K8644 in the pipette decreased the size of the inward currents recorded in cell-attached somatic patches, in a concentration-dependent manner with an IC₅₀ of 40 nM (Fig. 3B). The maximal decrease was 70%, which is less than the 85% inhibition caused by -agatoxin IVA or the 99% inhibition effected by -conotoxin MVIIC. There are two possible explanations for the effect of (–)-(S)-Bay K8644. One is that the drug affects the activity of a proportion of the P-type channels that are sensitive to block by -agatoxin IVA and -conotoxin MVIIC. Another is that as the recording pipette is lowered towards the soma (–)-(S)-Bay K8644 leaks out of the pipette, activates L-type Ca²⁺ channels, and the resultant conformational change or Ca²⁺ influx through L-type channels diminishes P-type channel activity. We thought the latter scenario unlikely because the bath was constantly perfused, and any solution leaking out would have been washed away. Also, in situ hybridization shows a lack of Ca_V1.2 and Ca_V1.3 mRNA in mature Purkinje cells (Tanaka et al. 1995; Ludwig et al. 1997). However, immunohistochemistry suggests the presence of Ca_V1.2 and Ca_V1.3 protein in the soma of adult Purkinje cells, with Ca_V1.2 more abundant than Ca_V1.3 (Hell et al. 1993; Chung et al. 2000; Mize et al. 2002; Kim et al. 2004). Therefore, we looked for evidence for the expression of L-type channels by including in the recording pipette (R,S)-nimodipine, which blocks L-type channels containing Ca_V1.2 by 90% and Ca_V1.3 by 50% (Xu & Lipscombe, 2001), or FPL 64176, which is a non-DHP activator of L-type channels (Baxter et al. 1993). We found that currents recorded in the presence of (R,S)-nimodipine were not smaller than control currents recorded in the same slices (Fig. 3C) and currents recorded in the presence of FPL 64176 were not greater than control currents recorded in the same slices (Fig. 3D). This suggests that if L-type channels are expressed, they carry only a minor component of the current. However, it was still possible that activation by (–)-(S)-Bay K8644 of a population of L-type channels that was too small to be detected, because of variation in current size between patches, might indirectly contribute to the decrease in the total current caused by (–)-(S)-Bay K8644. Therefore, we investigated if nimodipine would reduce the effect of (–)-(S)-Bay K8644. We found that inclusion of 1 µM (R,S)-nimodipine together with 300 nM (–)-(S)-Bay K8644 in the pipette did not affect the block by 300 nM (–)-(S)-Bay K8644 (Fig. 2E). Therefore, a primary effect on the P-type channels by (–)-(S)-Bay K8644 is the most likely explanation for the current decrease.

View larger version (22K): [in this window] [in a new window]   Figure 3.  Ca2+ channel currents in somatic patches of mature cerebellar Purkinje cells are reduced by (–)-(S)-Bay K8644 This effect does not involve the activation of L-type channels. A, Ca2+ channel currents evoked in cell-attached patches by depolarizing voltage ramps in the absence (40 patches) and presence of 300 nM (–)-(S)-Bay K8644 (27 patches) in the patch-pipette, recorded from the same slices. B, concentration dependence of the decrease of Ipk by (–)-(S)-Bay K8644. Symbols and vertical bars plot means ± S.E.M. Numbers of patches for each concentration are given in parentheses. The continuous line is the best fit of the logistic equation, for which IC50 is 40 nM, Hill slope is 2.9, max – min is 21.1 pA (70%). C, D and E, bar charts showing that mean Ipk in cell-attached patches is not altered by the inclusion of 1 µM (R,S)-nimodipine (P = 0.55, Student's t test) or 30 nM or 3 µM FPL 64176 (P = 0.24, ANOVA) in the pipette, whereas 300 nM (–)-(S)-Bay K8644 reduced Ipk by 64% when applied in the presence of 1 µM (R,S)-nimodipine (*P = 0.011, Student's t test). F, superimposed mean (black line) ± S.E.M. (grey) of currents evoked in multiple patches (numbers in parentheses) by voltage jumps, demonstrating inhibition of the currents by (–)-(S)-Bay K8644 in the presence or absence of (R,S)-nimodipine. G, the mean currents in F normalized to the maximum current, showing that only a minor change in the activation kinetics (arrow) accompanies the inhibition of the currents. H, plot of means ± S.E.M. of Ipk against different concentrations of (+)-(R)-Bay K8644. I, bar chart illustrating reduction in Ipk by racemic Bay K8644 (*P = 0.003, Student's t test).

The activation of L-type channels by (–)-(S)-Bay K8644 is accompanied by alterations in the kinetics of channel activation, inactivation and deactivation (Bechem & Hoffmann, 1993; Grabner et al. 1996; Sinnegger et al. 1997; Xu & Lipscombe, 2001; Koschak et al. 2001). From our recordings of currents evoked with voltage ramps, it was not possible to determine if (–)-(S)-Bay K8644 affected channel kinetics. Indeed, it was possible that the observed decrease in I_pk was not due to a true inhibition, but was due to a pronounced slowing of channel activation that prevented the currents from reaching their true final amplitude at the voltages transiently traversed during the voltage ramp. Therefore, we made recordings in which currents were evoked by voltage jumps to a potential ( –10 mV) similar to the median voltage ( –5 mV) at which the peaks of the currents evoked by depolarizing voltage ramps occurred. Near the peak of the current–voltage relationship, the (–)-(S)-Bay K8644-induced slowing of the activation of an L-type component should be apparent for channels containing a Ca_V1.2 (Grabner et al. 1996) or a Ca_V1.3 subunit (Xu & Lipscombe, 2001; Koschak et al. 2001). Figure 3F shows that inclusion of (–)-(S)-Bay K8644 in the pipette, in the presence or absence of (R,S)-nimodipine, reduced the size of the currents evoked by voltage jumps. Therefore, the decrease caused by (–)-(S)-Bay K8644 in the peaks of the currents evoked by depolarizing voltage ramps does represent a decrease in the activity of P-type Ca²⁺ channels. Figure 3F shows that there was no clear lengthening of the tail current evoked by voltage jumps, further demonstrating a paucity of somatic L-type Ca²⁺ channel current. Also, although superposition of the currents evoked by voltage jumps suggests a slight slowing of the activation kinetics in (–)-(S)-Bay K8644 (arrow, Fig. 3G), this also occurred in the presence of (R,S)-nimodipine. It cannot therefore be interpreted as evidence for the presence of L-type currents.

The effects of Bay K8644 on L-type Ca²⁺ channels are known to be stereoselective, that is, the (–)-(S)-enantiomer potently activates the channels whereas the (+)-(R)-enantiomer has no effect or weakly inhibits L-type channels (Triggle & Rampe, 1989; Bechem & Hoffmann, 1993; Striessnig et al. 1998). Consequently, the activity of racemic Bay K8644 is dominated by its constituent (–)-(S)-enantiomer, and (R,S)-Bay K8644 rather than (–)-(S)-Bay K8644 is frequently used to activate L-type Ca²⁺ channels when probing the identity of Ca²⁺ channels. Therefore, we sought to determine the effect of (+)-(R)-Bay K8644 on P-type Ca²⁺ channels in the soma of mature Purkinje cells. We found that (+)-(R)-Bay K8644 did not clearly inhibit or potentiate I_pk over a range of concentrations (10 nM – 3 µM, Fig. 3H). However, it was possible that a small effect of (+)-(R)-Bay K8644 might have been overlooked because of variation in current size between patches. Therefore, we examined the effect of 2 µM racemic Bay K8644 and compared it with that of 1 µM (–)-(S)-Bay K8644 and with that of 1 µM (+)-(R)-Bay K8644. We found that 2 µM (R,S)-Bay K8644 inhibited the current by 66% (Fig. 3I). This is in good agreement with the 70% inhibition caused by 1 µM (–)-(S)-Bay K8644 (Fig. 3B). These findings suggest that the effects of Bay K8644 at somatic P-type channels of adult Purkinje cells are stereoselective, but the stereoselectivity is different than that at L-type channels, and the activity of racemic Bay K8644 on P-type Ca²⁺ channels in the soma of mature Purkinje cells is dominated by the inhibitory actions of the (–)-(S)-enantiomer.

The experiments described so far suggest that 85% of the somatic Ca²⁺ channel current is inhibited by -agatoxin IVA and 70% is inhibited by (–)-(S)-Bay K8644. To determine if the currents not blocked by -agatoxin IVA were also not blocked by (–)-(S)-Bay K8644, a mixture of -agatoxin IVA (30 nM) and (–)-(S)-Bay K8644 (3 µM) was included in the recording pipette. This mixture caused an inhibition of 85% (control I_pk, –40.8 ± 4.8 pA, n = 48; 3 µM Bay K8644 + -agatoxin IVA I_pk, –5.9 ± 1.6, n = 5), which is no more than the inhibition caused by -agatoxin IVA alone. These data suggest that a small fraction of the current is insensitive to both drugs (15%), a small fraction (15%) can be inhibited by -agatoxin IVA but not by (–)-(S)-Bay K8644, like prototypical P-type channels, and the majority of the current (70%) can be blocked by (–)-(S)-Bay K8644 or -agatoxin IVA.

Whole-cell recording of Ca²⁺ channel currents in mature Purkinje cells

Disadvantages of exploring the pharmacology of Ca²⁺ channels by comparing currents recorded from cell-attached patches with drug-free and drug-containing pipettes are that there is no pre-drug baseline against which a drug effect in an individual recording can be measured, and it is not possible to wash off the drug to determine the reversibility of the effect. Also, variation in current size between patches necessitates recordings from many cells for a drug-induced difference to be detected, it may obscure a small effect on the currents, and it limits the accuracy of the IC₅₀ values. For these reasons, it was necessary to confirm the inhibitory effect of (–)-(S)-Bay K8644 detected by cell-attached recording by a second recording method. We decided to record whole-cell Ca²⁺ channel currents and measure their sensitivity to different Ca²⁺ channel blockers. By keeping the effective series resistance to less than 1 M, we expected to be able to voltage clamp the soma (Llano et al. 1994). Figure 4A shows an example of a whole-cell current evoked by a depolarizing voltage ramp applied to the soma, which was superfused with an extracellular solution containing 2 mM Ba²⁺ instead of Ca²⁺, and 1 µM TTX. Voltage ramps were used to evoke the currents because, unlike voltage jumps, these gave inward currents with little sign of a regenerative component that might be expected from the activation of dendritic spikes (Llano et al. 1994). The charge carrier was not 5 mM Ba²⁺ as in cell-attached recordings because whole-cell currents carried by 5 mM Ba²⁺ were too big for the amplifier to record.

The sizes of the whole-cell peak inward current, I_pk, carried by 2 mM Ba²⁺, in many cells approximately 10 min after establishing the whole-cell configuration, are summarized in Fig. 4B. The median and mean values were –7.8 nA and –8.3 ± 0.2 nA (n = 145). One of our concerns was that the currents would rundown too quickly for us to examine their sensitivity to drugs, as suggested by the fast rundown (70% in 10 min) of channels containing a cloned Ca_V2.1 subunit (Wu et al. 2002). However, we found that the inward currents showed relatively slow rundown (6% in 10 min, Fig. 4C). The currents were carried by Ca²⁺ channels, since they were inhibited by 100 µM Cd²⁺, as exemplified by the recording from a single cell in Fig. 4D. The majority of the current was inhibited by 100 nM -agatoxin IVA (Fig. 4E). The insets in Fig. 4D and E show perfect superposition of the current traces (not leak-subtracted) at voltages below the threshold for the inward currents, in the absence and presence of the drugs. This demonstrates that there was no change in holding current and hence that the decrease in inward current was not due to a drug-induced decrease in input resistance or accuracy of voltage clamp. By comparing the amplitude of the currents remaining in -agatoxin IVA with currents recorded in drug-free solution at an equivalent time point in Fig. 4E, the mean inhibition is estimated as 93%. It is noteworthy that the inhibition took 30 min (Fig. 4E), even though -agatoxin IVA was applied from a needle placed only 0.5 cm away from the cell under investigation.

In contrast with the clear blockade of the currents by Cd²⁺ or -agatoxin IVA, the application of (–)-(S)-Bay K8644 (300 nM and 3 µM) had variable effects on the inward current (not shown), contrary to the inhibition detected by cell-attached recording. It was possible that this might reflect restricted access of this lipophilic drug to the cell, because of adsorption of the drug to the perfusion system or to the slice. We decided to limit this by applying the drug from a glass pipette placed close to the cell body. During development of this method of local application, we tried to apply the 2 mM Ba²⁺ solution on to the cell body by pressure ejection from pipettes with tip diameters similar to or larger than the diameter of the Purkinje cell soma, so as to ensure that the concentration at the soma was the same as that in the pipette. However, we found that this was not possible. The force of the solution either damaged the cells or increased the series resistance, which decreased the size of the currents recorded (not shown). We found that to avoid such effects, it was necessary to use pipettes with smaller tips, 10 µm in diameter, to position them no closer than 20 µm from the soma (Fig. 5A), and to eject the solution by gentle mouth pressure, at a maximum rate of 0.75 µl min^–1. Experiments with drug-free Ba²⁺ solution indicated that this mode of local application did not accelerate current rundown (Fig. 5B), but more forceful ejection of solution tended to damage the cell (not shown). We also found that withdrawal of the drug-containing pipette to increasing distances from the soma reduced the effect on the currents (not shown), suggesting that diffusion of drug to the dendrites and inhibition of dendritic channels did not contribute to the effects described in this study. Therefore, by using this method of application, we explored the effects of (–)-(S)-Bay K8644 on channels in the soma and not in the dendrites, just as the cell-attached experiments examined drug effects on somatic channels. In order to distinguish the pipettes used to apply drugs during whole-cell recording from the drug-free and drug-containing pipettes used in the cell-attached experiments, they will be referred to as micropipettes and ‘pipette application’ of drug to the soma will be referred to as ‘local application’.

View larger version (41K): [in this window] [in a new window]   Figure 5.  Whole-cell Ca2+ channel currents in mature Purkinje cells are inhibited by local application of -agatoxin IVA or (S)-Bay K8644 to the soma from a micropipette A, digitized image illustrating the application of drug from a micropipette (right) to the soma of a Purkinje cell, that had been previously cleaned of overlying tissue by applying a stream of extracellular solution from a different pipette (not in view), and to which a recording-pipette (left) is attached in a whole-cell configuration. B, plot of whole-cell Ipk normalized to Ipk at time zero (circles and vertical bars: means ± S.E.M.), against time for recordings in which a micropipette (pipette symbol) was used to apply drug-free extracellular solution to the soma (open circles, eight cells) or solution containing 300 nM -agatoxin IVA (filled circles, six cells). Time zero denotes the start of the application. In this and subsequent figures, the concentration in the micropipette is denoted by the subscript ‘pip’, and diagonal lines to the right of the pipette symbol signify maintained ejection of solution from the micropipette. The current traces in the inset show the inhibition of the inward current by -agatoxin IVA in a single cell. C, plots of normalized whole-cell Ipk (mean ± S.E.M.) against time showing that, when compared with currents recorded in drug-free solution (grey circles, same data as in B) (R,S)-nimodipine did not reduce Ipk(1 µMpip, red circles, seven cells; 10 µMpip, open circles, three cells), whereas (–)-(S)-Bay K8644 inhibited the peak current (3 µMpip, open triangles, nine cells) and this inhibition was not diminished by inclusion of 1 µM (R,S)-nimodipine in the micropipette (green triangles, five cells). D, plots of whole-cell Ipk(mean ± S.E.M.) against time for recordings in which the cell was superfused for 7–10 min with DMSO (0.04%) or (R,S)-nimodipine 1 or 10 µM, before the local application of DMSO (open circles, seven cells) or (R,S)-nimodipine (purple circles, nine cells) or a mixture of (–)-(S)-Bay K8644 plus (R,S)-nimodipine at two concentrations (green triangles, four cells; open triangles, three cells). The concentration of DMSO is equivalent to that in the solution containing 3 µM (–)-(S)-Bay K8644 and 1 µM (R,S)-nimodipine. E, time courses of whole-cell Ipk in two experiments showing reversibility of inhibition by (–)-(S)-Bay K8644. Each cell was superfused with (R,S)-nimodipine 1 or 10 µM, the perfusion was stopped (–)-(S)-Bay K8644 (3 or 30 µMpip) in the presence of (R,S)-nimodipine 1 or 10 µMpip was applied to the soma for 15 min, the micropipette was removed and superfusion of the cell by (R,S)-nimodipine was restarted. F, current traces showing the effect of (–)-(S)-Bay K8644 on currents recorded in a single cell, in the presence of (R,S)-nimodipine.

Figure 5C shows that the local application of (–)-(S)-Bay K8644 (3 µM in micropipette) to the soma caused a 26% decrease in the size of the current (relative to the currents recorded in drug-free solution, grey symbols). This is the same as the degree of inhibition caused by a 1000x lower concentration (30 nM) in cell-attached mode (Fig. 3B). Figure 5C also shows that local application of nimodipine alone (1 or 10 µM in micropipette) did not inhibit the current and the time course of inhibition was the same when (–)-(S)-Bay K8644 (3 µM in micropipette) was applied in the presence of (R,S)-nimodipine (1 µM in micropipette). The absolute current inhibited by (–)-(S)-Bay K8644 was 2.03 ± 0.21 nA (n = 14 cells, which corresponds to a mean current density of 72 ± 6 pA pF^–1; currents normalized by the soma capacitance, pooling recordings in the absence and presence of (R,S)-nimodipine, corrected for rundown). These findings suggest that inhibition by (–)-(S)-Bay K8644 was not secondary to the activation of L-type channels, but to further confirm this, we carried out recordings in which the cell soma was superfused with 1 µM (R,S)-nimodipine prior to the local application of (–)-(S)-Bay K8644 (3 µM in micropipette) in the presence of (R,S)-nimodipine (1 µM in micropipette). This did not alter the percentage inhibition by (–)-(S)-Bay K8644 (28%, Fig. 5D) or the absolute current inhibited (1.79 ± 0.24 nA, n = 4, P = 0.58, Student's t test; current density, 70 ± 4 pA pF^–1, P = 0.71, Student's t test). The inhibition was reversible (Fig. 5E and F).

To determine if a larger fraction of the whole-cell current was sensitive to (–)-(S)-Bay K8644, we applied a 10-fold higher concentration of (–)-(S)-Bay K8644 (30 µM in micropipette), in the presence of a 10-fold higher concentration of (R,S)-nimodipine (10 µM in micropipette). This blocked 70% of the current (Fig. 5D) in a reversible manner (Fig. 5E). The absolute current blocked (4.75 ± 1.1 nA or 174 ± 35 pA pF^–1, n = 3 cells) was more than that blocked by the lower concentration of (–)-(S)-Bay K8644 (3 µM in micropipette, P = 0.00013 for absolute current, P = 0.0001 for current density, Student's t test). It is worth noting that the close agreement of the fraction of whole-cell current inhibited by (–)-(S)-Bay K8644 applied to the soma (70%), with the (–)-(S)-Bay K8644-sensitive fraction (70%) of somatic current measured by including (–)-(S)-Bay K8644 in cell-attached recording pipettes (Fig. 3B), demonstrates that dendritic Ca²⁺ channels make little or no contribution to the whole-cell currents recorded. If they had, the percentage inhibition of the somatodendritic current by the local application of drug to the soma would be smaller than the percentage inhibition of somatic current in cell-attached somatic patches by (–)-(S)-Bay K8644 in the pipette. The lack of dendritic contribution to the whole-cell currents may reflect the selective activation of channels in the somatic component of Purkinje cells by the relatively slow voltage ramp.

Different pharmacological properties of somatic Ca²⁺ channels in young Purkinje cells

Our finding that (–)-(S)-Bay K8644 or (R,S)-Bay K8644 inhibited 85% of the -agatoxin-IVA-sensitive Ca²⁺ channel current in mature Purkinje cells is in marked contrast to previous reports that (R,S)-Bay K8644 does not inhibit -agatoxin-IVA-sensitive Ca²⁺ channels in cerebellar Purkinje cells (Regan et al. 1991; Mintz et al. 1992a). To determine if this discrepancy reflects the younger ages of the Purkinje cells examined in the previous studies (P7–21), we investigated the effects of -agatoxin IVA and (–)-(S)-Bay K8644 on Ca²⁺ channels in Purkinje cells in cerebellar slices obtained from P13–20 rats.

The concentration-dependent inhibition relationship for currents evoked by depolarizing voltage ramps in the absence and presence of -agatoxin IVA, in cell-attached patches of the soma of young Purkinje cells (Fig. 6A), shows that the majority of the inward current carried by 5 mM Ba²⁺ was inhibited by P-type-selective concentrations of -agatoxin IVA. The parameters characterizing the block (86%; IC₅₀, 1.3 nM) agree with values previously obtained by whole-cell recording from dissociated Purkinje cells and bath application of -agatoxin IVA (82–90%, 2–3 nM, Mintz et al. 1992a,b; Lorenzon et al. 1998). Inclusion of 300 nM (–)-(S)-Bay K8644 in the cell-attached recording pipette did not inhibit the currents (control I_pk, –27.2 ± 7.8 pA, n = 11; 300 nM (–)-(S)-Bay K8644 I_pk, –40.3 ± 6.7 pA, n = (14), in marked contrast to the inhibition of somatic currents in mature cerebellar Purkinje cells (Fig. 3). Rather, the mean values for I_pk, in the absence and presence of (–)-(S)-Bay K8644, –27 and –40 pA, suggest an increase in current size, although this was not statistically significant (P = 0.21, Student's t test) possibly due to the variation in current size between patches. There was also a non-significant leftward shift and an increase in the size of the mean currents (Fig. 6B, formed by averaging currents across patches). These findings are consistent with the expression of a small component of L-type Ca²⁺ channel current (Triggle & Rampe, 1989; Bechem & Hoffmann, 1993; Striessnig et al. 1998), as deduced previously for P7–21 cells from the effects of (–)-(S)-Bay K8644 on whole-cell currents (Regan, 1991) and for P4–7 Purkinje cells in culture from the inhibition or enhancement of Ca²⁺ oscillations by nimodipine and (–)-(S)-Bay K8644, respectively (Liljelund et al. 2000). However, we did not detect a slowing of the activation or deactivation kinetics of currents evoked by voltage jumps in the presence of 300 nM (–)-(S)-Bay K8644 (Fig. 6C), that typifies currents mediated by L-type channels (Bechem & Hoffmann, 1993; Grabner et al. 1996; Sinnegger et al. 1997; Xu & Lipscombe, 2001; Koschak et al. 2001), but there was a slight acceleration of inactivation (Fig. 6C). In summary, our findings hint at the presence of a small L-type component in P13–20 cells that is too small to be clearly identified by the use of (–)-(S)-Bay K8644 under our cell-attached recording conditions.

View larger version (17K): [in this window] [in a new window]   Figure 6.  Ca2+ channel currents in somatic patches of young cerebellar Purkinje cells (P13–20) are inhibited by -agatoxin IVA but not by (–)-(S)-Bay K8644 A, concentration dependence of inhibition of inward currents (5 mM Ba2+) evoked by depolarizing voltage ramps in cell-attached patches in the soma by -agatoxin IVA included in the recording pipette. Symbols and vertical bars plot mean ± S.E.M. of Ipk measured in the numbers of patches given in parentheses. The continuous curve is the best fit of the logistic equation to points at 0–100 nM, for which IC50 is 1.3 nM, Hill slope is 0.6, max – min is 35.3 pA (86%). The dotted line indicates that the points at 1 and 3 µM, which are concentrations not selective for P-type channels, were excluded from the fit. B, mean Ca2+ channel currents (black) ±S.E.M. (grey) generated by averaging leak-subtracted currents evoked by depolarizing voltage ramps in 11 somatic cell-attached patches in the absence of drug, and in 14 patches in the presence of 300 nM (–)-(S)-Bay K8644. C, average currents (black) ±S.E.M. (grey) evoked by voltage jumps in cell-attached patches in the absence (nine cells) and presence of 300 nM (–)-(S)-Bay K8644 in the pipette (five cells).

View larger version (28K): [in this window] [in a new window]   Figure 7.  Whole-cell Ca2+ channel currents in immature Purkinje cells are inhibited by -agatoxin IVA, or by (–)-(S)-Bay K8644 when applied in the presence of (R,S)-nimodipine A, example of a whole-cell current evoked by a depolarizing voltage ramp (–97 to +73 mV) applied to a P16 Purkinje cell in the whole-cell configuration, superfused with an extracellular solution containing 2 mM Ba2+ and 1 µM TTX. The horizontal dashed line (left) denotes the zero current level. The dotted line denotes the estimated linear leak current, from which inward current (Ipk) was measured. B, scatter diagram and box plot of whole-cell Ipk measurements in m