INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Low-voltage-activated T-type Ca²⁺ channels are suggested to play as physiological regulators of various neuronal properties including intrinsic neuronal oscillation and low-threshold spikes in the CNS (Huguenard 1996; Perez-Reyes 1998). These T-type Ca²⁺ channels have been also implicated in generation of pathophysiological conditions such as absence epilepsy (Huguenard and Prince 1994; Kim et al. 2001; Tsakiridou et al. 1995). In native tissues, presence of T-type Ca²⁺ currents has been recognized by common characteristics such as low threshold for activation, transient activation and inactivation kinetics, a "criss-crossing" pattern between current traces evoked by a voltage protocol, slow deactivation, and tiny unitary conductance. In terms of biophysical and pharmacological properties of T-type Ca²⁺ currents, however, there are tissue-to-tissue variations that may originate from three genes (1G, 1H, and 1I) encoding the 1 subunits of the T-type Ca²⁺ channels (Cribbs et al. 1998; Lee et al. 1999a; McRory et al. 2001; Perez-Reyes 1998) and their differential distribution in tissues (Talley et al. 1999). Thus the molecular identification of the isoform(s) might be prerequisite for studying tissue-specific functions and regulations of T-type Ca²⁺ channels.

The pelvic ganglia plexus innervates the pelvic viscera including the descending colon, the bladder, and the external genitalia (Dail et al. 1975; Langworthy 1965), being implicated in autonomic reflexes such as micturition and penile erection (Dail 1992). One of the unique features of the pelvic ganglia is that both sympathetic and parasympathetic neurons are co-localized within the same ganglion capsule (Dail 1992). More interestingly, the sympathetic neurons of rat major pelvic ganglia (MPG) are known to functionally express T-type Ca²⁺ channels, which have never been reported for other autonomic ganglia (Zhu et al. 1995). However, the physiological significance of the T-type Ca²⁺ channels has not been defined in MPG neurons. As an initial attempt toward understanding the functional roles of T-type Ca²⁺ channels in MPG neurons, we investigated the characteristics and molecular identity of the T-type Ca²⁺ channels using patch-clamp and RT-PCR techniques. Here, we show that the T-type Ca²⁺ currents in MPG neurons are primarily attributed to the Ni²⁺-sensitive 1H isoform. Some preliminary data have been published previously in abstract form (Jeong et al. 2001).

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Cell preparation

MPG neurons were enzymatically dissociated as described previously (Zhu et al. 1995). Briefly, adult (200-300 g) male Sprague-Dawley rats were anesthetized with carbon dioxide and decapitated using a laboratory guillotine. The major pelvic ganglia, located on the lateral surfaces of the prostate gland, were dissected out and placed in cold Hanks' balanced salt solution (HBSS). The ganglia were desheathed, cut into small pieces, and incubated with 0.7 mg/ml collagenase type D, 0.1 mg/ml trypsin (all from Boehringer Mannheim Biochemicals, Indianapolis, IN), and 0.1 mg/ml DNase type I (Sigma Chemical, St Louis, MO) in 10 ml of modified Earle's balanced salt solution (EBSS, pH 7.4) in a 25-cm² tissue culture flask. The EBSS was modified by adding 3.6 g/l glucose and 10 mM HEPES. The flask was then placed in a shaking water bath at 35°C for 1 h. After incubation, neurons were dissociated by vigorous shaking of the flask. After centrifugation at 50 × g for 5 min, the dissociated neurons were resuspended in minimum essential medium (MEM) containing 10% fetal bovine serum and 1% penicillin-streptomycin (all from Life Technologies, Grand Island, NY). For measurement of currents, neurons were then plated onto culture dishes (35 mm) coated with poly-L-lysine and maintained in a humidified 95% air-5% CO₂ incubator at 37°C. All neurons were used within 24 h after plating.

RT-PCR analysis

Total RNA was isolated from dissociated MPG neurons using a guanidium thiocyanate-phenol-chloroform extraction method (Chomczynski and Sacchi 1987). The first strand cDNA was synthesized from 0.5 µg of total RNA using AMV reverse transcriptase (Boehringer Mannheim Biochemicals) by incubating first at room temperature for 10 min and then at 42°C for 50 min. The reaction was terminated by heating at 95°C for 5 min. PCR was performed using a pair of degenerate PCR primers corresponding to conserved portions (GVVVENF of domain III S6 and PINPTI of domain IV S3) of reported 1G, 1H, and 1I sequences (Lambert et al. 1998). The PCR primer sequences were as follows: forward primer, 5'GGCGT(G/C)GT(G/C)GT(G/C)GAGAACTT3'; reverse primer, 5'GATGATGGTGGG(A/G)TTGAT3'. The PCR reaction was initiated by a first denaturation at 94°C for 1 min, followed by 33 cycles consisting of 30 s at 94°C, 30 s at 58°C, and 30 s at 72°C. The resultant PCR product was separated on a 1% agarose gel and purified using a Qiaquick gel extraction column (Qiagen). The purified PCR products were ligated into pGEM-T vector (Promega, Madison, WI), and transformed into competent cells. After confirming the presence of PCR products in the vector by restriction enzyme digestion, 10 PCR products were sequenced. Their DNA sequences were translated into amino acid sequences that were compared with the reported T-type Ca²⁺ channel sequences for identification.

Electrophysiology

Ca²⁺ channel currents were recorded using the whole cell-ruptured configuration of the patch-clamp technique (Hamil et al. 1981) as described previously (Ikeda 1991; Jeong and Ikeda 1998). Patch electrodes were fabricated from a borosilicate glass capillary (1.65 mm OD, 1.2 mm ID, Corning 7052, Garner Glass, Claremont, CA). The patch electrodes were coated with silicone elastomer (Sylgard 184; Dow Corning, Midland, MI), fire polished on a microforge, and had resistances of 1.5-2.5 M when filled with the solution described below. An Ag/AgCl pellet connected via a 0.15 M NaCl/agar bridge was used to ground the bath. The cell membrane capacitance and series resistance were compensated (>80%) electronically using an Axopatch-1D amplifier (Axon Instruments, Foster City, CA). Voltage protocol generation and data acquisition were performed using S4 data acquisition software (written by Dr. Stephen R. Ikeda) on a Macintosh G4 computer equipped with a ITC16 data acquisition board (Instrutech, Port Washington, NY). Current traces were generally low-pass filtered at 5 kHz using the 4-pole Bessel filter in the clamp amplifier, digitized at 2 kHz, and stored on the computer hard drive for later analysis. For constructing activation curves, the tail currents were digitized at 20 kHz. Current-clamp recordings were performed under the gramicidin-perforated whole cell configuration of the patch-clamp technique using an EPC-9 amplifier and Pulse/Pulsefit(v8.50) software (Heka Electronik, Lambrecht, Germany). All experiments were performed at room temperature (~20-24°C).

Solutions and drugs

To isolate Ca²⁺ currents, patch pipettes were filled with an internal solution containing (in mM) 120 N-methyl-D-glucamine (NMG)-methanesulfonate (MS), 20 tetraethylammonium (TEA)-MS, 20 HCl, 11 EGTA, 1 CaCl₂, 10 HEPES, 4 Mg-ATP, 0.3 Na₂-GTP, and 14 creatine phosphate (pH 7.2). External recording solution contained (in mM) 145 TEA-MS, 10 HEPES, 10 CaCl₂, 15 glucose, and 0.0003 tetrodotoxin (TTX) (pH 7.4). As appropriate, Ca²⁺ was replaced with Ba²⁺ in the same concentration for testing permeation of divalent ions. For current-clamp recordings, patch pipettes were filled with a solution containing (in mM) 140 KCl, 5 EGTA, 10 HEPES, 0.5 CaCl₂, and 5 NaCl (pH 7.2). External recording solution contained 135 NaCl, 5.4 KCl, 1 MgCl₂, 1.8 CaCl₂, 5 HEPES, and 10 glucose (pH 7.4). Drugs were applied to single neurons via a gravity-fed fused silica capillary tube connected to an array of seven polyethylene tubes. The outlet of the perfusion system was located within 100 µm of the cell. The bath superfusion rate was approximately 1-2 ml/min. Mibefradil (a gift from Roche, Besel, Switzerland) and Ni²⁺ were freshly prepared before experiments. A stock solution of gramicidin (Sigma) was prepared at 50 mg/ml in dimethylsulfoxide (Sigma) and diluted in the pipette solution to a final concentration of 50 µg/ml before use.

Data analysis

Current traces and current-voltage (I-V) relationships were corrected for linear leakage current as determined from hyperpolarizing pulses. Activation and inactivation data, and time courses of tail current deactivation were fitted by nonlinear regression assuming a Boltzmann distribution and a single exponential relaxation, respectively. All curve fitting was performed with the IGOR data analysis package (Wave-Metrics, Lake Oswego, OR). Data were presented as means ± SE. Student's t-test, as appropriate, was applied to the data to determine statistical significance. P < 0.05 was considered significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Electrophysiological characteristics of T-type Ca²⁺ currents in MPG neurons

To evoke the voltage-dependent inward Ca²⁺ currents, ramp and test pulses were applied from a holding potential of 100 mV under the whole cell configuration of the patch-clamp techniques. As reported previously (Zhu et al. 1995), we could recognize the sympathetic neurons among dissociated rat MPG neurons by presence of T-type Ca²⁺ currents identified as a prominent hump at low-voltage range (50 to 20 mV) of the I-V relationship (Fig. 1, A and B, left). Consistent with the previous findings (Zhu and Yakel 1997; Zhu et al. 1995), the capacitance (C_m) of the sympathetic neurons was larger than that of the parasympathetic neurons lacking of the hump. On average, C_m of neurons expressing T-type Ca²⁺ channels was 77 ± 3 pF (n = 27), while that of neurons lacking of T-type Ca²⁺ channels was 32 ± 2 pF (n = 24, P < 0.01). In an I-V curve, T-type Ca²⁺ currents were first activated around 50 mV and peaked between 30 and 20 mV when 10 mM Ca²⁺ was used as a charge carrier (Fig. 1B). In contrast, the high-voltage-activated (HVA) Ca²⁺ currents (mostly -conotoxin GVIA-sensitive N-type) began to activate at 20 mV and peaked at +10 mV in sympathetic MPG neurons held at 100 mV. Similar results were obtained with parasympathetic MPG neurons (data not shown). Accordingly, there might be a small overlap between T- and N-type Ca²⁺ currents around 20 mV in sympathetic MPG neurons. In the representative current traces recorded from a sympathetic MPG neuron, both activation and inactivation were slow near threshold potential but became faster at more depolarized potentials producing "criss-crossing pattern" by the voltage protocol (Fig. 1B, right). To assess voltage dependence of T-type Ca²⁺ channel activation, we measured slowly deactivating tail-currents after test pulses of varying amplitude (Fig. 2, A and C). Since the time-to-peak of the T-type Ca²⁺ currents varied at different test voltages, we applied pulses of the varying duration to prevent underestimation of tail-current activation, especially at threshold potentials (Montiel et al. 2000) (see Fig. 2A). The curve fitting using a Boltzmann function showed that the midpoint (V_0.5) and the slope factor (k) were 38.3 ± 0.4 mV and 5.8 ± 0.4 mV (n = 5), respectively (Fig. 2C). As shown in a previous study (Serrano et al. 1999), inactivation of T-type Ca²⁺ currents reached a steady state within 1 s in MPG neurons (data not shown). Thus the steady-state inactivation parameters of T-type Ca²⁺ currents were also evaluated after applying a voltage protocol consisting of 1.4-s conditioning prepulses between 80 and 25 mV followed by test pulses to 30 mV (Fig. 2B). On average, the half-inactivation voltage (V_0.5) was 57.4 ± 0.8 mV with the slope factor of 4.3 ± 0.2 mV (n = 5; Fig. 2C). Overlap of both activation and inactivation curves revealed a prominent window current that might allow Ca²⁺ influx at a resting condition. In addition, we assessed voltage dependence of slow T-type Ca²⁺ channel deactivation by measuring time constants for slow decay of tail currents (Fig. 3). The slow deactivation was more prominent in a high deactivation voltage range.

View larger version (25K): [in this window] [in a new window]   Fig. 1. Whole cell inward Ca2+ currents evoked by ramp and step voltage protocols in major pelvic ganglia (MPG) neurons. A: phenotype-specific expression of T-type Ca2+ currents identified with a hump at a low voltage range (50 mV to approximately 20 mV). Ca2+ currents were evoked by a ramp from 100 to +80 mV for 160 ms. Note that T-type Ca2+ currents could be observed in large-sized MPG neurons (see text). B: current-voltage (I-V) relationship and representative traces of T-type Ca2+ currents in sympathetic MPG neurons. Ca2+ currents were evoked by test pulses in a range between 100 and +80 mV. The hump by T-type Ca2+ currents was highlighted with closed circles on the I-V curve.

View larger version (24K): [in this window] [in a new window]   Fig. 2. Voltage dependence of activation and steady-state inactivation of T-type Ca2+ channels in MPG neurons. A: representative traces of tail currents evoked by deactivation to 80 mV following test pulses of varying duration based on the time-to-peak at different test voltages (7, 10, 15, 20, 25, and 30 ms for 25, 30, 35, 40, 45, and 50 mV, respectively). B: representative traces of T-type Ca2+ currents induced by test pulses to 30 mV for 100 ms after holding neurons at different potentials between 100 and 20 mV (as indicated) for 1.4 s. Traces during the prepulses are not shown for clarity. C: activation and steady-state inactivation curves. For constructing activation curve, amplitudes of the tail currents were measured 1 ms after the end of the test pulses, normalized to the maximal tail current obtained at 25 mV, and plotted as a function of test pulses (closed circles). All tail currents were digitized at 20 kHz. The activation curve was fitted by the Boltzmann function: m = [1 + exp(V0.5  V)/k]1 where V0.5, the half-activation voltage, was 38.3 ± 0.4 mV (n = 5) and k, a slope factor, was 5.8 ± 0.4 mV per e-fold change in conductance. For constructing inactivation curve, the amplitudes of currents evoked by the test pulses were normalized to the current evoked from a holding potential of 100 mV and plotted as a function of holding potentials (open cicles). The steady-state inactivation curve was fitted by the Boltzmann function: h = [1 + exp(V  V0.5)/k]1, where V0.5 and k were 57.4 ± 0.8 mV and 4.3 ± 0.2 mV (n = 5), respectively.

View larger version (14K): [in this window] [in a new window]   Fig. 3. Voltage dependence of deactivation kinetics. Tail currents were evoked by test pulses to 30 mV from a holding potential of 100 mV followed by deactivation to different potentials between 100 and 40 mV. Deactivation time constants were determined by fitting tail currents with a single exponential and plotted as a function of deactivation potential (n = 6). Data are presented as means ± SE.

Recently, the rat 1H has been shown to be more permeable to Ba²⁺ than Ca²⁺ while 1I and 1G to be either equally permeable or less permeable to Ba²⁺ than Ca²⁺, respectively (McRory et al. 2001). In our experiments, replacing 10 mM Ca²⁺ with 10 mM Ba²⁺ produced almost twofold increase of the HVA Ca²⁺ currents. Conversely, Ba²⁺ carried the same amount of currents as Ca²⁺ did through the T-type Ca²⁺ channels (data not shown).

Pharmacological characterization of T-type Ca²⁺ currents in MPG neurons

Mibefradil, an antihypertensive drug, has been known to selectively block T-type Ca²⁺ currents (Merke et al. 1994; Mishra and Hermsmeyer 1994). Thus effects of mibefradil on T-type Ca²⁺ currents were examined in MPG neurons. Noncumulative application of mibefradil blocked the T-type Ca²⁺ currents in a concentration-dependent manner with an IC₅₀ of 3 µM (n = 5; Fig. 4). A recent study has shown that Ni²⁺, an inorganic Ca²⁺ channel antagonist, blocked the 1H (IC₅₀ = 13 µM) more potently than the 1G (IC₅₀ = 259 µM) and the 1I (IC₅₀ = 216 µM) (Lee et al. 1999b). As illustrated in Fig. 5, Ni²⁺ produced a concentration-dependent block of T-type Ca²⁺ currents with an IC₅₀ of 10 µM (n = 6). The high nickel sensitivity of T-type Ca²⁺ currents led us to make a tentative conclusion that the 1H is mainly expressed to form the functional T-type Ca²⁺ channels in MPG neurons.

View larger version (14K): [in this window] [in a new window]   Fig. 4. Concentration-dependent block of T-type Ca2+ currents by mibefradil in MPG neurons. A: representative currents evoked by test pulses to 30 mV for 200 ms from a holding potential of 100 mV in the absence and presence of increasing concentrations of mibefradil. B: concentration-response relationship for the mibefradil. Normalized current block was plotted as a function of mibefradil concentration. The curve was fitted with the Hill equation B = (1 + IC50/[mibefradil]n)1, where B, IC50, and n are normalized block, the concentration of mibefradil for half-maximal block, and Hill factor, respectively. Data are presented as means ± SE.

View larger version (14K): [in this window] [in a new window]   Fig. 5. Effects of Ni2+ on T-type Ca2+ currents in MPG neurons. A: representative currents evoked by test pulses to 30 mV for 200 ms from a holding potential of 100 mV in the absence and presence of increasing concentrations of Ni2+. B: concentration-response relationship for the Ni2+. Normalized current block was plotted as a function of Ni2+. The curve was best fitted with the Hill equation Block = (1 + IC50/[Ni2+]n)1, where IC50, and n are the concentration of Ni2+ for half-maximal block and Hill factor, respectively. Data are presented as means ± SE.

Molecular identification of T-type Ca²⁺ channels in MPG neurons

The molecular nature of T-type Ca²⁺ channels was identified using RT-PCR. A pair of degenerate PCR primers, which were previously demonstrated to amplify any of 1G, 1H, and 1I (Lambert et al. 1998), were used to amplify the T-type Ca²⁺ channel sequences nonselectively in MPG neurons. The PCR product shown in the lane 2 of Fig. 6A appeared to be a single band of which size was about 470 base pairs. When 10 PCR products were sequenced and compared with those of the reported T-type Ca²⁺ channels (McRory et al. 2001), all the PCR products were identified as rat 1H (Fig. 6B). Interestingly, 7 of the 10 PCR products were identical to the reported rat 1H sequence, but the remaining ones were found to be a long form of 1H that has six more amino acids in a linker between domain III and IV. Taken together, these findings strongly support the fact that the T-type Ca²⁺ channels expressed in rat MPG neurons are primarily attributable to the 1H among three cloned isoforms (1G, 1H, and 1I).

View larger version (49K): [in this window] [in a new window]   Fig. 6. T-type Ca2+ channel subtypes in rat MPG neurons identified by RT-PCR. A: PCR products were separated on an agarose gel with size markers. Abbreviations in the picture are as follows: M, DNA size marker; lane 1, a negative control of RT-PCR in which AMV reverse transcriptase was not included; lane 2, PCR products from total RNA of MPG neurons. B: deduced amino acid sequences of PCR products were compared with those from Domain IIIS6 to Domain IVS3 of the reported T-type Ca2+ channel sequences. Based on the sequence similarity, pelvic Ta is identical to the sequence of the rat 1H and pelvic Tb is a splice variant of the rat 1H containing additional 6 amino acids, which are underlined. The amino acid residues of the pelvic 1H sequence distinguished from the rat 1G and 1I are marked with asterisks (*) below the alignment. The sources of T-type channel sequences are as follows: rCavT3.1 is from the rat 1G (GenBank access no. AF027984); rCavT3.2 is from the rat 1H (AF290213); rCavT3.3 is from the rat 1I (AF086827).

Involvement of T-type Ca²⁺ channels in generation of rebound spikes in MPG neurons

Several studies have shown that T-type Ca²⁺ channels are responsible for generation of low-threshold spikes (LTS) in the CNS. Accordingly, we tested whether the LTS could be generated in sympathetic MPG neurons expressing T-type Ca²⁺ channels. Since ~80% of T-type Ca²⁺ channels resides in inactivation states at resting membrane potentials (around 50 mV; Fig. 2C), hyperpolarizing currents were injected into MPG neurons to recruit sufficient T-type Ca²⁺ channels under a current-clamp mode. As illustrated in Fig. 7A, sympathetic MPG neurons (n = 9) produced rebound spikes after anodal-break, whereas parasympathetic neurons were silent (n = 11). When the current-clamp mode was switched to the voltage-clamp mode, T-type Ca²⁺ currents were detected only in neurons producing the rebound spikes (Fig. 7A). More interestingly, the number of the rebound spikes was reduced by 10 µM Ni²⁺ that blocked 50% of T-type Ca²⁺ currents and had little effect on HVA Ca²⁺ currents in MPG neurons (n = 9; Fig. 7B). Furthermore, generation of the rebound spikes was completely prevented by 100 µM Ni²⁺ that blocked most of the T-type Ca²⁺ currents. In contrast, the rebound spikes were not affected by 10 µM of Cd²⁺ that was able to block 26 ± 6% (n = 6) and 85 ± 3% (n = 6) of T- and N-type Ca²⁺ currents in MPG neurons, respectively. However, the number of rebound spikes was significantly reduced by 100 µM of Cd²⁺ that blocked >80% of T-type Ca²⁺ currents (Fig. 7B). In some experiments, Ni²⁺ (10 and 100 µM) did not affect high-threshold spikes generated by depolarizing current injection, although 100 µM Cd²⁺ slightly augmented the excitation, which appears to be due to inhibition of Ca²⁺-activated K⁺ currents (data not shown). Taken together, these data suggest that T-type Ca²⁺ channels may play a role in generation of LTS in MPG neurons.

View larger version (27K): [in this window] [in a new window]   Fig. 7. Involvement of T-type Ca2+ channels in generation of low-threshold spikes in MPG neurons. A: rebound spikes were induced in sympathetic but not in parasympathetic neurons. In a gramicidin-perforated whole cell configuration of the current clamp, neurons were hyperpolarized to 110 mV to approximately 115 mV by current injection (100 and 20 pA for sympathetic and parasympathetic MPG neurons, respectively). After recording the voltages, the current-clamp mode was switched to the voltage-clamp mode for recording ionic currents evoked by test pulses to 40 mV from a holding potential of 100 mV. T-type Ca2+ currents were isolated by subtracting Ni2+-insensitive ones from the whole cell currents. In these voltage-clamp experiments, 1.8 mM Ca2+ was used as a charge carrier. B: effects of divalent ions on generation of rebound spikes. Either Ni2+ or Cd2+ at 10 and 100 µM was applied.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

T-type Ca²⁺ channels had never been revealed in autonomic ganglia until described in rat MPG neurons (Zhu et al. 1995). Interestingly, the expression of T-type Ca²⁺ channels is phenotype specific within MPG, providing a reliable electrophysiological marker for distinguishing the sympathetic one from the parasympathetic population (Zhu and Yakel 1997; Zhu et al. 1995). Thus T-type Ca²⁺ channels are believed to play important roles in differential functions assigned to the sympathetic MPG neurons. To date, however, characteristics of T-type Ca²⁺ channels have been poorly defined in MPG neurons. In this regard, we first described general biophysical and pharmacological properties of T-type Ca²⁺ channels. More importantly, we demonstrated that 1H underlies the T-type Ca²⁺ currents in MPG neurons based on two lines of evidence. First, the peak T-type Ca²⁺ currents in MPG neurons were blocked by low micromolar concentrations of Ni²⁺. The low potency (IC₅₀ = 10 µM) for the nickel blockade is reminiscent of human 1H currents that have much higher sensitivity to Ni²⁺ than the 1G and 1I currents (Lee et al. 1999b; Williams et al. 1999). Second, the RT-PCR experiment disclosed that rat MPG neurons predominantly expressed mRNAs encoding 1H, which is consistent with the pharmacological identification of T-type Ca²⁺ currents in MPG neurons and other tissues such as zona glomerulosa of adrenal glands (Schrier et al. 2001), and skeletal muscle myoblasts (Bijlenga et al. 2000). A recent study in HEK293 cells has suggested that T-type 1H Ca²⁺ channels were more permeable to Ba²⁺ than Ca²⁺ (McRory et al. 2001). In our experiments, however, T-type Ca²⁺ channels in MPG neurons were equally permeable to Ca²⁺ and Ba²⁺. This permeation pattern (Ca²⁺ = Ba²⁺) is rather similar to that for human 1H (Williams et al. 1999) and is consistent with those for T-type Ca²⁺ currents in many other tissues (Huguenard 1996). Since the monoclonal antibodies or blockers specific to T-type Ca²⁺ channel isoforms are not available, the high Ni²⁺ sensitivity appears to be the only reliable hallmark discriminating 1H from 1G and 1I Ca²⁺ channels in native tissues.

Overall properties of T-type Ca²⁺ currents in MPG neurons were qualitatively similar to the reported ones for native and recombinant currents (Huguenard 1996; Perez-Reyes 1998). Nevertheless, it might be worth to note some prominent differences present between T-type Ca²⁺ currents in MPG neurons and recombinant 1H currents expressed in HEK 293 cells (McRory et al. 2001). First, the window current, generated by the overlap of activation and inactivation voltage range appears to be larger for the T-type Ca²⁺ currents in MPG neurons. Second, the voltage dependence of deactivation in MPG neurons was steeper at high-voltage range than those of the recombinant 1H currents. Sequence analysis of the RT-PCR products indicates that the T-type Ca²⁺ currents in MPG neurons arise from two splicing variants of 1H. Thus the biophysical differences may originate from heterogenous expression of short and long forms of 1H (pelvic Ta and Tb). Alternatively, it is possible for unidentified auxiliary subunits present in native tissues to modify the current kinetics in comparison to the expression system.

We presume that the large window currents of T-type Ca²⁺ channels may regulate intracellular Ca²⁺ concentration contributing to activation of Ca²⁺-dependent ion channels or other signaling under resting conditions in MPG neurons. The Ca²⁺ influx via T-type Ca²⁺ channels may also contribute the amplitude and frequency of action potentials. Furthermore, the large Ca²⁺ influx during the slow channel deactivation (particularly during the early repolarizing phase of action potentials) may affect the shape of spike and the amplitude of afterpotential influencing cell excitability (Bijlenga et al. 2000). Similar to other neurons in the CNS (Huguenard 1996; McRory et al. 2001; Perez-Reyes 1998), it seems likely that MPG neurons are capable of generating LTS if there are hyperpolarizing inputs that increase the availability of T-type Ca²⁺ channels (Fig. 7). Recently, we found that GABA_A receptors were functionally expressed in sympathetic but not in parasympathetic MPG neurons (I. D. Kong, S. K. Cha, J. W. Lee, and S. W. Jeong, unpublished observation). Thus it might be interesting to examine any functional correlation between T-type Ca²⁺ channels and GABA receptors.

In summary, we identified the molecular nature of T-type Ca²⁺ channels selectively expressed in the sympathetic population of MPG neurons. Combining pharmacological and molecular data led us to conclude that T-type Ca²⁺ currents in MPG neurons arise primarily from 1H among three isoforms. Thus we suggest that MPG neurons may provide a good model system for comparative studies between native and recombinant 1H on regulatory and functional aspects of the T-type Ca²⁺ channels.