INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Both pharmacological and molecular techniques have indicated a diversity of high-voltage-activated (HVA) Ca²⁺ channel types in neurons. This diversity has been associated with differences in biophysical properties, subcellular location, coupling to cellular events and to differential modulation by neurotransmitters (Hille 1994).

At least four HVA Ca²⁺ channel 1 subunits have been identified in CNS neurons (1A-D) (Birnbaumer et al. 1994; Snutch and Reiner 1992; Tsien et al. 1995). The 1B subunit has been identified with N-type channels, which are blocked by -conotoxin GVIA (Ctx GVIA) (McCleskey et al. 1987). 1C and 1D subunits correspond to L-type channels, which are sensitive to dihydropyridines (Bean 1989; Tsien et al. 1988). The 1A subunit, associated with P- and Q-type channels (Bourinet et al. 1996; Mermelstein et al. 1999; Piedras-Renteria and Tsien 1998; Stea et al. 1994), is sensitive to -agatoxin IVA and Ctx-MVIIC (Birnbaumer et al. 1994; Mintz et al. 1992a,b; Randall and Tsien 1995). Another calcium channel 1 subunit (1F) is only found in the retina and presumed to form L-type channels because of sequence homology to 1C and 1D (Bech-Hansen et al. 1998; Strom et al. 1998).

There has been some uncertainty, however, as to the type of currents generated by channels possessing the 1E subunit. The 1E subunit was originally suggested to be associated with low-voltage-activated, T-type currents (Bourinet et al. 1996; Soong et al. 1993), but may be associated with HVA, R-type currents (Piedras-Renteria and Tsien 1998; Schneider et al. 1994). To further confuse the issue, there are several reported similarities between R-type and T-type currents. By definition, R-type currents are insensitive to the aforementioned organic Ca²⁺ channel antagonists (Birnbaumer et al. 1994; Randall and Tsien 1995); however, T-type currents are also generally not blocked by these peptides or dihydropyridines (Huguenard 1996). Further, both currents are reported to be highly sensitive to Ni²⁺ block and, with strong depolarizations, to inactivate rapidly (Huguenard 1996; Randall and Tsien 1995).

Some reports suggest differences between R- and T-type currents. First, in most cell types, T-type currents activate ~20-30 mV more negative than R-type currents. Second, R-type currents deactivate with time constants of a few hundred microseconds (Randall and Tsien 1995, 1997); 10 times faster than T-type currents (Huguenard 1996; Matteson and Armstrong 1986). Third, the inactivation kinetics of T-type currents are steeply voltage dependent (Carbone and Lux 1984; Lee et al. 1999; Randall and Tsien 1997), whereas those of R-type currents are less so (Randall and Tsien 1997). Last, because of the pronounced voltage-dependent inactivation of T-type currents, they are largely inactivated at holding potentials more positive than 50 mV (Huguenard 1996).

While R-type Ca²⁺ currents have been observed in a variety of neuronal populations, reports on their biophysical properties have varied. This may be due to channel heterogeneity or to contamination by other voltage-dependent calcium currents that are not completely blocked under standard recording conditions. To unequivocally determine the properties of R-type currents in cortical and striatal neurons, we used three distinct pharmacological regimens as well as a biophysical strategy. To provide a complementary level of analysis, neurons also were subjected to single-cell RT-PCR analysis of class E 1 subunit expression.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Acute isolation

Four- to 6-wk-old Sprague-Dawley rats were anesthetized with methoxyfluorane and then decapitated. The brains were extracted and placed in an oxygenated high sucrose solution that contained (in mM) 250 sucrose, 2.5 KCl, 1 NaH₂PO₄, 11 glucose, 4 MgSO₄, 0.1 CaCl₂, 1 kynurenic acid, 1 pyruvic acid, 0.1 nitro-arginine, 0.05 glutathione, and 15 HEPES (pH 7.3 adjusted with 1N NaOH; 300 mOsm/l). The tissue was then sliced into 400-µM sections using a vibrating tissue slicer (Cambden Instruments). The slices were held for a minimum of 1 h in a carboxygen (95% O₂-5% CO₂)-bubbled artificial cerebral spinal fluid (ACSF) that contained (in mM) 125 NaCl, 3 KCl, 2 CaCl₂, 2 MgCl₂, 1.25 NaH₂PO₄, 26 NaHCO₃, 20 glucose, 1 kynurenic acid, 1 pyruvic acid, 0.1 nitroarginine, and 0.05 glutathione (pH 7.4 adjusted with 1 N NaOH; 310 mOsm/l). The primary motor and primary somatosensory cortex (hereafter referred to as sensorimotor cortex) or the dorsal striatum (posterior to anterior commisure) was then dissected from the slices with the aid of a stereo microscope.

The dissected tissue was incubated for 20-30 min in an oxygenated ACSF-containing Pronase E (Sigma protease type XIV, 1.0 mg/ml at 32-35°C) (cf. Bargas et al. 1994; Lorenzon and Foehring 1995), 1 mM kynurenic acid, 1 mM pyruvic acid, 0.1 mM nitro-arginine, and 0.05 mM glutathione. Following the incubation period, the tissue was rinsed in a Na isethionate solution that contained (in mM) 140 Na isethionate, 2 KCl, 1 MgCl₂, 23 glucose, 15 HEPES, 1 kynurenic acid, 1 pyruvic acid, 0.1 nitro-arginine, and 0.05 glutathione (pH 7.3 adjusted with 1 N NaOH; 310 mOsm/l). The tissue was then triturated in the same solution using fire-polished Pasteur pipettes. The supernatant was collected and then poured into a plastic Petri dish (Lux) positioned on the stage of an inverted microscope (Nikon Diaphot 300). The cells were allowed several minutes to adhere to the Petri dish, and then a background flow of HEPES-buffered saline solution was initiated (~1 ml/min). This solution contained (in mM) 10 HEPES, 138 NaCl, 3 KCl, 1 MgCl₂, and 2 CaCl₂, pH 7.3 adjusted with 1 N NaOH, 300 mOsm/l.

Cultured cell preparation

For a few experiments, medium spiny cells from embryonic day 19 (E19) rat embryos were cultured for 2 wk according to the procedure outlined in Bargas et al. (1991). The cells were maintained in 5% CO₂ at 37°C.

Recording solutions and pharmacological agents

The external recording solution used to isolate Ca²⁺ channel currents consisted of (in mM) 125 NaCl, 20 CsCl, 1 MgCl₂, 10 HEPES, 5 BaCl₂, 0.001 TTX, and 10 glucose (pH 7.3 with TEA-OH; 300-305 mOsm/l). The internal recording solution included the following (in mM): 180 N-methyl-D-glucamine (NMG), 4 MgCl₂, 40 HEPES, 5-10 ethylene glycol-bis(-aminoethyl ether)-N,N,N',N'-tetraacetic acid (EGTA) or bis-(o-aminophenoxy)-N,N,N',N'-tetraacetic acid (BAPTA), 0.1 leupeptin, 0.4 guanosine triphosphate (GTP), 2 ATP, and 0.01 phosphocreatine, pH 7.2 (adjusted with 0.1 N H₂SO₄), 265-275 mOsm/l. The stock solutions of the calcium channel antagonists (with the exception of nifedipine) were dissolved in water. Stock solutions of the calcium channel antagonists -conotoxin-GVIA (Ctx-GVIA: 500 µM), -conotoxin-MVIIC (Ctx-MVIIC: 500 µM), and -agatoxin-IVA (AgTX: 100 µM; all from Bachem, Torrance, CA) were aliquoted and frozen. Each of the stocks was diluted to the appropriate concentrations in the external recording solution immediately prior to the experiment. Nifedipine (RBI, Nattick, MA) was first dissolved in 95% ethanol before being added to the external solution resulting in a final ethanol concentration of <0.05%. This concentration of ethanol had no effect on these cells (Bargas et al. 1994; Lorenzon and Foehring 1995). Nifedipine was protected from ambient light. Cytochrome C was combined with solutions containing AgTX to prevent nonspecific binding of AgTX to glass and plastic (Bargas et al. 1994; Lorenzon and Foehring 1995).

Single-cell RT-PCR

The methods utilized for the single-cell RT-PCR were similar to those described previously (Surmeier et al. 1996; Yan and Surmeier 1996). Electrodes contained ~5 µl of sterile recording solution (see above). Some cells were harvested without recording, with electrodes filled with water. The capillary glass used for making electrodes had previously been heated to 200°C for 4 h. Sterile gloves were worn during the procedure to minimize RNase contamination.

After aspiration, the electrode was broken and contents ejected into a presiliconized, 0.5-ml Eppendorf tube containing 5 µl diethyl pyrocarbonate (DEPC)-treated water, 0.5 µl RNAsin (28,000 U/ml), and 0.5 µl dithiothreitol (DTT; 0.1 M). One microliter of either oligo dT (0.5 mg/ml) or random hexanucleotides (50 ng/ml) was added and mixed before the mixture was heated at 70°C for 10 min and incubated on ice for more than 1 min. Single-strand cDNA was synthesized from the cellular mRNA by adding SuperScript II RT (1 µl, 200 U/ml), ×10 PCR buffer [2 µl, 200 mM TrisCl (pH 8.4)], 500 mM KCl, MgCl₂ (2 µl, 25 mM), RNAsin (0.5 ml, 28,000 U/ml), DTT (1.5 µl, 0.1 M), and mixed dNTPs (1 µl, 10 mM). The reaction mixture (20 µl) was incubated at 42°C for 50 min. The reaction was terminated by heating the mixture to 70°C for 15 min and then icing. The RNA strand in the RNA-DNA hybrid was then removed by adding 1 µl RNase H (2 U/ml) and incubating for 20 min at 37°C. All reagents except RNAsin (Promega, Madison, WI) were obtained from GIBCO BRL (Grand Island, NY). The cDNA from the reverse transcription (RT) of RNA in single cortical neurons was subjected to polymerase chain reactions (PCR) to detect the expression of mRNAs coding for Ca²⁺ channel 1 subunits.

Conventional PCR was carried out with a thermal cycler (MJ Research, Watertown, MA) and thin-walled plastic tubes (Perkin Elmer, Norwalk, CT). PCR primers were developed from GenBank sequences with the commercially available software OLIGO (National Biosciences, Plymouth, MN) and have been described previously (Mermelstein et al. 1999). To detect individual mRNAs, 2.5 µl of the single-cell cDNA was used as a template for conventional PCR amplification. Reaction mixtures contained 2-2.5 mM MgCl₂, 0.5 mM of each of the dNTPs, 1-µM primers, 2.5 U Taq DNA polymerase and buffer (Promega). The thermal cycling program was 94°C for 1 min, 58°C for 1 min and 72°C for 1.5 min for 45 cycles.

PCR products were separated by electrophoresis in 1.5-2% agarose gels and visualized by staining with ethidium bromide. In representative cases, amplicons were purified from the gel (QIAquick Gel Extraction Kit, QIAGEN, Hilden, Germany) and sequenced by the University of Tennessee Molecular Resource Center or St. Jude Children's Research Hospital Molecular Resource Center. These sequences were found to match published sequences.

PCR reactions were carried out following procedures designed to minimize the chances of cross-contamination (Cimino et al. 1990). Negative controls for contamination from extraneous and genomic DNA were run for every batch of neurons. To ensure that genomic DNA did not contribute to the PCR products, neurons were aspirated and processed in the normal manner except that the reverse transcriptase was omitted. Contamination from extraneous sources was checked by replacing the cellular template with water. Both controls were consistently negative in these experiments.

Whole cell recordings

Whole cell recordings were acquired using a DAGAN 8900 or an Axopatch 200B electrometer. The recordings were monitored and controlled by pCLAMP6 (Axon Instruments) installed on a 486 computer. The electrodes were pulled from 7052 glass (Garner) and fire polished. Typically, series resistance compensation of 70-90% was employed. Cells were not included in the comparisons of biophysical properties if the estimated series resistance-related voltage error (peak I * uncompensated series resistance) was >5 mV. Voltage control was also assessed by observing tail currents after brief voltage steps (see Bargas et al. 1994; Lorenzon and Foehring 1995). Cells were discarded if tail currents were broad or unstable. The liquid junction potential was ~8 mV and was not subtracted in the presented data. A gravity-fed parallel array of glass tubes was used to apply the drugs to the cell being studied.

SYSTAT (SYSTAT, Evanston, IL) was used to carry out all statistical calculations. The population data are represented as median, mean ± SEM, or box plots (Tukey 1977). In the box plots, the internal line represents the median while the outer edges of the box represent the inner quartiles of the data set. The bars extending from the box depict the two outer quartiles of the data set. Data points greater than two times the inner quartiles were considered outliers and indicated as asterisks on the plots. Statistical differences were determined with the Kruskal-Wallace (comparisons of multiple populations) or Mann-Whitney U test (  0.05), unless otherwise noted.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

R-type currents

We used whole cell patch-clamp recordings from acutely dissociated pyramidal cells (sensorimotor cortex) or medium spiny neurons (dorsal striatum) to test whether there was a component of the current resistant to block by organic Ca²⁺ channel antagonists. Currents were evoked by repeated (every 5 s) voltage steps (from 80 to 0 mV) or voltage ramps (0.3 mV/ms) from 80 to +60 mV (Bargas et al. 1994; Lorenzon and Foehring 1995).

In pyramidal cells, we employed three different methods to isolate the R-type current. Two protocols involved sequential addition of specific blockers of L-type (nifedipine: 5-10 µM), N-type (Ctx-GVIA: 1 µM), and P/Q-type channel blockers (Fig. 1). In the first method, AgTx (100 nM) was added (in the presence of nifedipine and Ctx-GVIA) to block P- and some Q-type current. Subsequently, Ctx-MVIIC (1-3 µM) was added to nifedipine, GVIA, and AgTx to block Q-type currents (Fig. 1, A-C). The block of Q-type current by Ctx-MVIIC was slow ( ~ 106 s) (Mermelstein et al. 1999), requiring ~10 min to reach a quasi steady-state level in most neurons (Foehring and Armstrong 1996; McDonough et al. 1996; Mermelstein et al. 1999). With this combination of blockers, 21 ± 3% of the initial current remained (median: 16%; n = 18 cells; Fig. 1C). In most cells, this component inactivated rapidly (Fig. 1B). Also, it was completely blocked by 400 µM Cd²⁺ (Fig. 1A), indicating that it was carried through calcium channels.

View larger version (29K): [in this window] [in a new window]   Fig. 1. A component of the Ba2+ current is insensitive to organic Ca2+ channel blockers in acutely dissociated neocortical pyramidal neurons. A: peak current (absolute) plotted as a function of time. The toxins were present for the times indicated by the lines above the plot. AgTx (100 nM) blocked P-type and some Q-type channels. Nifedipine (Nif: 5 µM) blocked L-type channels. Ctx GVIA (GVIA: 1 µM) blocked N-type channels. Ctx MVIIC (MVIIC: 2 µM) blocked N-, P-, and Q-type channels. A component of the current was insensitive to all of these blockers (R-type). This resistive component was completely blocked by the inorganic Ca2+ channel blocker Cd2+ (400 µM). Currents were elicited by a 400-ms step from 90 to 10 mV, repeated every 5 s. B: representative current traces from the cell shown in A. Note rapid inactivation of toxin-insensitive current (top trace). C: box plot illustrating population data for the percent of the whole cell current insensitive to toxins (Ctx-MVIIC used to block the Q-type component). D: peak current (absolute) plotted as a function of time. In this experiment, the Q-type current was blocked with 1µM AgTx. The toxins were present for the times indicated by the lines above the plot. AgTx (25 nM) blocked P-type channels and 1 µM AgTx additionally blocked Q-type. All: 1 µM AgTx + 1 µM Ctx-GVIA + 10 µM nifedipine. An inactivating component of the current was insensitive to all of these blockers (R-type). This resistive component was completely blocked by the inorganic Ca2+ channel blocker Cd2+ (400 µM). E: representative current traces from the cell shown in D. F: box plot illustrating population data for the percent of the whole cell current insensitive to toxins (1 µM AgTx, 1 µM Ctx-GVIA, and 10 µM nifedipine used to isolate the R component).

In a separate group of cells (n = 16), L- and N- type currents were blocked with nifedipine and Ctx-GVIA, P-type currents were blocked with 25 nM AgTx (Randall and Tsien 1995) and Q-type currents with 1 µM AgTx (Randall and Tsien 1995) (Fig. 1, D-F). The block of Q-type current with µM AgTx had a  of ~45 s (Mermelstein et al. 1999). Under these conditions, 21 ± 3% of the initial current was unblocked (median: 20%; Fig. 1, C and D). This current also inactivated rapidly (Fig. 1D) and was blocked completely with 400 µM Cd²⁺ (Fig. 1D). We found no significant differences in percentage of current remaining or its biophysical properties (see Properties of R-type current) using these two methods, so we combined data for further analyses.

Dissociated medium spiny neurons were also tested with both isolation protocols, with no significant differences in the results, so the data were combined. Sequential application of the organic antagonists for >10 min left 18 ± 2% of the original current (median: 17%; n = 19; Fig. 2). In 8 of 10 cells tested with voltage steps, the resistant current inactivated rapidly. In the remaining two cells, the resistant current inactivated slowly (e.g., Fig. 2C).

View larger version (16K): [in this window] [in a new window]   Fig. 2. Current resistant to organic Ca2+ channel blockers in acutely dissociated striatal medium spiny neurons. Abbreviations as in Fig. 1. A: plot of peak current (absolute) in response to voltage ramp (shown in B) as a function of time. The blockers were present for the times indicated by the lines above the graph. These data suggest the presence of L-, P/Q-, and N-type currents as well as a current component resistant to all of the blockers. Inset: box plot illustrating the percent of the whole cell current insensitive to the blockers (R-type). B: voltage protocol and representative current traces from the same cell as in A. The numbers refer to the traces that correspond in time to the same numbers in A. C: current traces obtained in response to voltage steps (top) in a different cell. Note that the R-type current inactivates slowly in this cell (similar to other current subtypes in this particular cell). Slowly inactivating R-type current was observed in 2 of 10 cells tested.

In a few cortical cells, we isolated the residual current by preincubating the slices for at least 2 h in Ctx-MVIIC (2 µM) and then recorded in the presence of 10 µM nifedipine, 1 µM Ctx-GVIA, 1-3 µM Ctx-MVIIC, and 100 nM AgTx (cf., Fig. 4). The remaining current was small (median = 269 pA; n = 9 cells), but comparable in amplitude to the current isolated by sequential addition of blockers (median: 201 pA; n = 31 cells). In cultured medium spiny neurons (see METHODS), a component of the current was found to be insensitive to 12-h preincubation in Ctx-MVIIC (1 µM) and recording in the presence of Ctx-MVIIC plus nifedipine (5 µM; n = 9 cells; data not shown). These data suggest that all three protocols effectively isolated R-type currents and that this current represents a substantial fraction of the somatodendritic Ca²⁺ current in both pyramidal neurons and medium spiny neurons (see also Churchill and MacVicar 1998; Eliot and Johnston 1994; Foehring and Armstrong 1996; Magee and Johnston 1995; Magnelli et al. 1998; McDonough et al. 1996; Mermelstein et al. 1999; Pennington and Fox 1995; Randall and Tsien 1995; Yu and Shinnick-Gallagher 1997). We next characterized the biophysical and pharmacological properties of the residual (R-type) current in both cell types.

Properties of R-type current

NEOCORTICAL PYRAMIDAL NEURONS. We first measured the kinetics (time-to-peak; _activation) and steady-state voltage dependence of current activation and the kinetics of deactivation (at 70 mV; Fig. 3). The average time-to-peak (TTP) at 0 mV was 5.5 ± 0.3 ms (median: 5.1 ms; n = 40 cells; Fig. 3, A and B), significantly shorter than the TTP for N-, L-, or P-type currents in these cells (Lorenzon and Foehring 1995) (Table 1). The _activation was best fit by assuming third-order kinetics and fitting the initial current trajectory with an exponential function: I = I_o * (exp  t/)³ + b, where I is current, I_o is the initial baseline, t is time, and b is the maximum current. Mean _activation was 3.6 ± 0.4 ms at 0 mV (median: 3.2 ms; n = 15). Deactivation time constants were determined by fitting an exponential to the tail current generated by stepping from 0 to 70 mV (Fig. 3A). The average time constant was 221 ± 66 µs (median: 241 µs; n = 24; Fig. 4B). This is similar to other HVA subtypes in these cells (e.g., L-type: 0.4 ± 0.2 ms; N-type: 0.4 ± 0.2 ms) (Lorenzon and Foehring 1995). This feature clearly distinguished this resistant current from a T-type current (Randall and Tsien 1995, 1997).

View larger version (23K): [in this window] [in a new window]   Fig. 3. Activation and deactivation properties of R-type current in neocortical pyramidal neurons. A, left: the R-type current reached its peak quickly (5.5 ms in this cell) and deactivation was rapid at 70 mV. Inset: voltage protocol. Right: tail current following step shown at increased gain. B: box plots summarizing R-type current time-to-peak (TTP) and deactivation data. C1: the voltage dependence of activation was studied with slow (0.3 mV/mV) voltage ramps. The Goldman-Hodgkin-Katz equation was used to generate a term for relative permeability (see text). C2: this term was plotted as a function of voltage, and the data were fit with a Boltzmann function (see text). D: the box plots summarize data from 33 cells for half-activation voltage (Vhalf) and slope factor.

View larger version (31K): [in this window] [in a new window]   Fig. 4. Pyramidal cell R-type current activation properties determined with voltage steps. A-D are from a neocortical pyramidal cell preincubated in Ctx-MVIIC (1 µM) and then recorded from in the presence of nifedipine (10 µM), AgTx (100 nM), Ctx-GVIA (1 µM), and Ctx-MVIIC (1 µM). A: representative current traces in response to voltage steps from 80 mV to various potentials. Traces shown were taken at 75 (no current), 60, 55, 45, 40, 35, 30, 25, 20, and 15 mV (largest current). Note lack of crossing of current traces (see text). Inset: voltage protocol (every other step indicated). B: peak current-voltage (I-V) relationships for data in A. C: expanded view of tail currents (at 70 mV) following voltage steps in A. Dotted line indicates where tail amplitude was measured (300 ms after step). D: plot of tail current amplitudes (from C) vs. voltage. The data were well fit by a single Boltzmann relationship (equation in text).

1

1

1

View larger version (25K): [in this window] [in a new window]   Fig. 5. Inactivation properties of R-type current in neocortical pyramidal neurons. A: the R-type current was rapidly inactivating. In this cell, the R-type current inactivated by 67% following a 400-ms step to 0 mV. Inactivation was characterized by 2 time constants: 45 ms (35% of inactivation) and 771 ms (65% of inactivation). B: box plot summaries of data from 36 cells for percent (%) inactivation (at 400 ms), the fast (1) and slow (2) inactivation time constants, and the percent of the total inactivation accounted for by each of the 2 time constants. C: representative traces of steady-state inactivation for R-type current [prepulses were for 2 s; protocol below (not to scale)]. D: Boltzmann fit of steady-state inactivation for cell in C. Similar data were obtained for 6 of 7 cells (see text).

Neostriatal medium spiny neurons (acutely dissociated)

As with pyramidal neurons, the current time-to-peak (at 0 mV) was short (6.9 ± 0.5 ms; n = 6; Fig. 6A). The _activation (fit as in cortical neurons) was 5.1 ± 0.5 ms (n = 6) and _deactivation was 275 ± 47 µs (n = 4; Fig. 6B), suggesting that the residual current is an HVA current. The steady-state voltage dependence of activation was tested with voltage ramps (see above; Fig. 6, C and D). The mean V_h was 15 ± 1.2 mV (n = 10), and V_c was 7 ± 0.2 ms (n = 10).

View larger version (22K): [in this window] [in a new window]   Fig. 6. Activation and deactivation properties of R-type current in acutely dissociated medium spiny neurons. A: activation was rapid. The time-to-peak was ~7 ms in this cell [protocol above (not to scale)]. B: deactivation kinetics (different cell from that in A). Exponential fits were obtained from tail currents following protocol as in A. In this cell, deactivation was 268 µs (at 70 mV). The box plot summarizes deactivation data for 4 cells. C: the voltage dependence of R-type current was studied with a voltage ramp (0.3 mV/ms). D: the resulting current was converted to an estimate of permeability with the Goldman-Hodgkin-Katz equation (see text). The relative permeability was then plotted as a function of voltage and fit with a Boltzmann function. The box plots summarize the half activation voltage and slope factor data for 10 cells.

Most medium spiny neurons tested (8/10) had both rapid and slow components to inactivation at 0 mV (Fig. 7A). In these eight cells, the fast component had an inactivation time constant that averaged 66 ± 7 ms (53 ± 5% of inactivation); the slower component had a time constant that averaged 1,404 ± 176 ms (47 ± 5%). During a 400-ms step to 0 mV, the R-type current declined to 53 ± 4% (n = 8) of its original amplitude. In a fraction of the medium spiny neurons (2/10), the R current decayed with only a single time constant of intermediate rate (677, 899 ms).

View larger version (17K): [in this window] [in a new window]   Fig. 7. Inactivation properties of R-type current in acutely dissociated medium spiny neurons. A: inactivation was rapid. In most cells, the current decay was best fit with 2 exponentials (protocol above). B: box plot summary of data (n is number of cells tested). The median fast  was 64 ms. The median slow  was ~1.3 s. Percent inactivation was determined as [(peak I)  (I at 400 ms)/peak I]. Median percent inactivation was 58%. C: voltage dependence of inactivation. Peak current was elicited by a 30-ms step to 0 mV, following 2,000-ms prepulses to various potentials (protocol in inset). The circles represent means ± SE for 6 cells tested. Mean half-inactivation voltage was 33 mV (slope factor = 17 mV).

We tested the voltage dependence of inactivation in medium spiny neurons using a protocol similar to that described above for pyramidal cells (Fig. 7C). The half-inactivation voltage averaged 33 ± 4 mV, with a slope factor of 17 ± 1 mV (n = 6).

Ni²+ sensitivity

The R-type current in neurons (Churchill and MacVicar 1998; Eliot and Johnston 1994; Forti et al. 1994; Magee and Johnston 1995; Randall and Tsien 1995), as well as 1E-associated current in expression systems (Soong et al. 1993; Williams et al. 1994; Zamponi et al. 1996) have been reported to be more sensitive to block by Ni²⁺ than other HVA Ca²⁺ channels. To test this hypothesis, Ni²⁺ dose-response relationships were constructed for neocortical pyramidal and striatal medium spiny neurons.

We first asked whether the current blocked by a low dose of Ni²⁺ (10-20 µM) inactivated rapidly, like R-type current. Of five neocortical pyramidal cells tested, the percent inactivation of the Ni²⁺-sensitive current at 400 ms was 23, 70, 0, 0, and 43%; these results suggest that Ni²⁺ did not selectively block the R current (data not shown). Similarly, in medium spiny neurons, Ni²⁺ (10 µM) blocked a current that was indistinguishable in inactivation rate from the total current (n = 5; data not shown).

We then tested the response of the whole Ba²⁺ current to doses of Ni²⁺ between 100 nM and 2 mM. The data were obtained with either steps to 10 mV (30 ms) or with voltage ramps. We plotted relative peak current (I/I_max) as a function of log dose of Ni²⁺. For pyramidal neurons, the resulting curve was well fit by a single Langmuir isotherm of the form
where the EC₅₀ is the concentration at which 50% of the current was blocked. The median EC₅₀ for the whole current was ~85 µM (n = 3). In medium spiny neurons, the EC₅₀ for Ni²⁺ block of the whole cell current was 96 µm (n = 6; data not shown). Neither dose-response plot had features indicative of more than one Ni²⁺ binding site.

To test the Ni²⁺ sensitivity of the R-type current directly, we preincubated neocortical slices in 2 µM Ctx-MVIIC for at least 2 h and then recorded in the presence of 100 nM AgTx, 1 µM Ctx-GVIA, 10 µM nifedipine, and 2 µM Ctx-MVIIC (see METHODS). These data were also well fit by a single Langmuir isotherm, with an EC₅₀ of ~48 µM (not shown).

Single-cell RT-PCR

Single-cell RT-PCR revealed that most neocortical pyramidal neurons express detectable levels of mRNAs for all five HVA Ca²⁺ channel 1 subunits tested (1A-E; data not shown). Detectable levels of class E 1 subunit mRNA were found in six of seven identified pyramidal neurons (Fig. 8A). In addition, nearly every cell expressed detectable levels of 1A (4/5 cells) (see Mermelstein et al. 1999) and 1B mRNA (4/5). All cells expressed detectable levels of either 1C (5/5) or 1D (3/5) mRNA (data not shown). Both 1C and 1D mRNA were co-expressed in three of five of cells tested.

View larger version (45K): [in this window] [in a new window]   Fig. 8. Class E 1 subunit mRNA is expressed in neocortical pyramidal neurons and striatal medium spiny neurons. A: gel indicating the expression of class E mRNA in a neocortical pyramidal cell. Calcium calmodulin kinase type II (CaMKII) mRNA expression indicates that this was a pyramidal neuronCaMKII is expressed in pyramidal neurons but not nonpyramidal neurons in cortex (Jones et al. 1993). Similar data were obtained in 6/7 cells. B: class E mRNA is expressed in medium spiny neurons. Similar data were observed in 14/15 cells tested. This particular cell expressed substance P mRNA but not enkephalin mRNA, indicating that it was a projection neuron (medium spiny) rather than a local circuit interneuron.

Detectable levels of mRNA for 1E were also found in nearly every striatal medium spiny neuron tested (14/15; Fig. 8B). The majority of medium spiny neurons express detectable levels of mRNA for 1A, 1B, and 1C, or 1D (Bargas et al. 1994; Mermelstein et al. 1999; Song and Surmeier 1996).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

We have previously provided pharmacological evidence for the expression of L-, N-, P-, and Q-type channels in neocortical pyramidal and striatal medium spiny cells (Bargas et al. 1994; Lorenzon and Foehring 1995; Mermelstein et al. 1999). Here we demonstrate that a component of the HVA Ca²⁺ current in both cell types is resistant to block by organic antagonists (R-type). We characterized the biophysical properties and Ni²⁺ sensitivity of these R-type currents. Single-cell RT-PCR was used to test for the expression of 1E subunits in these cells.

The data presented here and our previous work (Bargas et al. 1994; Mermelstein et al. 1999) demonstrate that neocortical pyramidal neurons and striatal medium spiny neurons express mRNAs for at least five different 1 subunits: 1A, B, C, D, and E. The 1E mRNA was ubiquitously expressed in both pyramidal and medium spiny neurons, in agreement with previous in situ hybridization studies reporting the presence of 1E mRNA in cortex and dorsal striatum (Ludwig et al. 1997; Plant et al. 1998; Soong et al. 1993; Tanaka et al. 1995; Williams et al. 1994; Yokayama et al. 1995). Considering that 1A-D subunits are associated with P-, Q-, N-, and L-type currents (Mermelstein et al. 1999; Piedras-Renteria and Tsien 1998; Stea et al. 1994), these data are also consistent with 1E subunits being associated with R-type. The association between 1E subunits and R-type currents has been suggested on the basis of sequence homology between the 1E subunit and other HVA subunits (Williams et al. 1994). Also, macroscopic current properties are similar in 1E-associated channels expressed in oocytes (Williams et al. 1994) or cell lines (Page et al. 1997; Stephens et al. 1997; Williams et al. 1994) to R-type currents in cerebellar granule cells (Randall and Tsien 1995). More recently, antisense experiments have strengthened the link between R-type currents and 1E subunits in cerebellar granule cells (Piedras-Renteria and Tsien 1998).

R-type versus T-type current

R-type current was first described in cerebellar granule neurons, where relative to other HVA Ca²⁺ currents, R-type currents inactivate more rapidly and at more negative potentials (Randall and Tsien 1995, 1997). In these respects, R-type currents are similar to T-type currents. Granule cell R-type currents (Randall and Tsien 1995, 1997) typically inactivate more slowly and activate and inactivate at more positive potentials than T-type currents. Typical T-type currents have half-activation voltages of approximately 50 to 60 mV and half-inactivation voltages of approximately 80 mV, although a wide range of values has been reported (Huguenard 1996; Lee et al. 1999).

R- and T-type channels can most clearly be distinguished by their deactivation kinetics. Whereas R-type currents in granule cells deactivate with time constants of a few hundred microseconds at negative potentials (Randall and Tsien 1995) (~240 µs at 80 mV), T-type currents deactivate with time constants of a few milliseconds (Huguenard 1996; Matteson and Armstrong 1986; Randall and Tsien 1997). T-type currents (but not R-type) also typically show "criss-crossing" traces in I-V protocols (Randall and Tsien 1997). T-type and 1E channels can also be distinguished by their single channel conductance (Schneider et al. 1994; Wakamori et al. 1994; but see Meir and Dolphin 1998). T-type channels are also generally more permeable to Ca²⁺ than Ba²⁺ (Huguenard 1996). There is some controversy as to whether R-type and 1E channels are more permeable to Ca²⁺ than Ba²⁺ (e.g., Bourinet et al. 1996) or vice versa (e.g., Magnelli et al. 1998; Williams et al. 1994).

PYRAMIDAL AND MEDIUM SPINY NEURONS. T-type 1 subunits (primarily 1G) (Talley et al. 1999) and low-voltage-activated (LVA = T-type) currents are expressed in neocortical pyramidal neurons (Sayer et al. 1990; Tarasenko et al. 1998; Ye and Akaike 1993). The LVA current inactivated with a  of ~24 ms (Tarasenko et al. 1998) and the half-inactivation voltage was approximately 90 mV (Sayer et al. 1990; Tarasenko et al. 1998; Ye and Akaike 1993). T-type, but not R-type currents are completely inactivated at holding potentials of approximately 50 mV in cortical pyramidal cells (Sayer et al. 1990; Ye and Akaike 1993) and in other cell types (Huguenard 1996). When recording with Ba²⁺ as the charge carrier, we find T-type to be nearly undetectable in most acutely dissociated cells. When present, T-type currents are typically <50 pA in amplitude (Lorenzon and Foehring 1995; unpublished observations). This was also true for medium spiny cells (Bargas et al. 1994), although T-type currents are more prominent in embryonic or cultured medium spiny neurons (Bargas et al. 1991; Hoehn et al. 1993). The resistant current studied here (both cell types) is clearly identified as HVA on the basis of deactivation kinetics, lack of "criss-cross" in the I-V curve, and relative insensitivity to depolarized holding potentials (50 mV). In addition, no current was observed to activate negative to approximately 50 mV. The half-activation and inactivation voltages for R-type current reported here (both cell types) were also more positive than published values for T-type currents in cortical and striatal neurons (Bargas et al. 1991; Hoehn et al. 1993; Sayer et al. 1990; Tarasenko et al. 1998; Ye and Akaike 1993).

BIOPHYSICS OF R-TYPE CURRENTS. The R-type current comprised ~18% of the whole cell Ba²⁺ current in pyramidal neurons and 17% in medium spiny neurons. This compares to ~15-20% in cerebellar granule neurons (Randall and Tsien 1995), 32% in N. accumbens (Churchill and MacVicar 1998), and ~40% in embryonic motoneurons (Magnelli et al. 1998). We isolated the R-type current with combinations of organic blockers for L- (nifedipine), N- (Ctx-GVIA), and P/Q-type channels (AgTx and Ctx-MVIIC). We used three different isolation procedures, including preincubation for several hours in MVIIC, to rule out the possibility that the residual current reflects insufficient time for toxins to exert their effects. The deactivation time constant was rapid in pyramidal neurons (~221 µs at 0 mV) and medium spiny neurons (~270 µs), similar to values reported for R-type current in cerebellar granule cells (deactivation  was ~210 µs) (Randall and Tsien 1997), spinal motoneurons (330 µs) (Magnelli et al. 1998), and CA1 pyramidal cell dendrites (~200 µs) (Kavalali et al. 1997).

PHARMACOLOGY OF R-TYPE CURRENTS. There was no clear evidence for a high affinity Ni²⁺ block of the whole cell Ba²⁺ current in either pyramidal cells or medium spiny cells. The current sensitive to low doses (10-50 µM) of Ni²⁺ was not rapidly inactivating in these cells. In pyramidal cells, the small difference in Ni²⁺ affinity for the isolated R-type current versus the whole current (EC₅₀s 48 vs. 85 µM) is of limited diagnostic utility. Our data are consistent with studies on antagonist-resistant currents in cerebellar granule cells (R: 66 µM: Zhang et al. 1993), dentate granule cells (Eliot and Johnston 1994), raphe neurons (~50 µM: Pennington and Fox 1995) and expressed 1E channels (27 µM: Williams et al. 1994; 28 µM: Soong et al. 1993) or doe-1 channels (33 µM: Elinor et al. 1993). In spinal motoneurons, the R-type current was not especially Ni²⁺-sensitive (Magnelli et al. 1998). Zamponi et al. (1996) reported that in addition to channel block, Ni²⁺ causes a shift in the voltage dependence of gating that is greater in 1E channels than other channel types. In oocytes, this leads to an apparent K_i at 10 mV of 21 µM, which shifted to 144 µM at +10 mV. Our data were obtained at 10 mV.

DIVERSITY OF R-TYPE CURRENTS. We have discussed several examples of variability in the biophysical properties of R-type currents in various cell types. The R-type currents in neocortical pyramidal neurons and striatal medium spiny neurons were rapidly activating, deactivating, and inactivating. In both cell types, activation occurred at relatively hyperpolarized potentials. The primary difference between these two cell types was in the voltage dependence of inactivation, which occurred at much more hyperpolarized potentials in pyramidal cells than in medium spiny neurons. Variability was evident in a single cell type as well, especially in the kinetics of inactivation. In medium spiny neurons, some cells only expressed a slowly inactivating current, whereas most cells had a prominent rapidly inactivating component as well. What could be the basis for the biophysical differences between resistant currents in different cell types?

Functional consequences

The hyperpolarized activation voltage range of R-type current in neocortical pyramidal cells and medium spiny neurons may facilitate a role in synaptic i