INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Sigma receptors are widely distributed in mammalian brain and peripheral systems and organs. These receptors have been pharmacologically defined into two subclasses of receptors, sigma-1 and sigma-2 (Hellewell and Bowen 1990; Quirion et al. 1992), with a major difference being the higher affinity of sigma-1 receptors for (+) pentazocine (Quirion et al. 1992) and the greater affinity of sigma-2 receptors for ibogaine (Bowen et al. 1995). While only the sigma-1 receptor has been cloned, studies using photolabeling techniques with sigma ligands on guinea pig brain and PC12 cell membranes suggest that distinct molecular entities exist that correspond to the two sigma receptor subtypes (Hellewell and Bowen 1990). The function of these receptors is not well understood; however, sigma receptors have been implicated in the modulation of various biochemical, behavioral, and physiological processes (Walker et al. 1990).

It has been suggested that sigma receptors may regulate the cardiovascular system (Ela et al. 1994). Sigma ligand binding sites have been detected in cardiac myocytes, and sigma ligands, including (+)-pentazocine and haloperidol, have been shown to alter contractility, Ca²⁺ influx, and contraction rate in cultured cardiac myocytes (Ela et al. 1994; Novakova et al. 1995). However, while direct effects of sigma ligands on cardiac muscle have been documented, very little is known about sigma receptors in autonomic neurons, and in particular sympathetic or parasympathetic neurons that innervate the heart. The presence of putative endogenous ligands of sigma receptors, including neuropeptide Y (Roman et al. 1989) and substance P (Larson and Sun 1993), in these ganglia (Hassall and Burnstock 1984; Karhula 1995; Kessler and Black 1982; Papka et al. 1981) suggests that sigma receptors may be activated under physiological conditions, affecting cell-to-cell signaling in the ganglia, and ultimately the regulation of cardiac function by the autonomic nervous system.

Some evidence does exist that suggests that sigma receptors may play an important role in the function of peripheral neurons. In the guinea pig ileum, for example, sigma receptors have been shown to block contractions of longitudinal muscle elicited by both electrical stimulus or by exogenous serotonin via inhibition of acetylcholine release from myenteric neurons (Campbell et al. 1989). Conversely, sigma receptors potentiate neurogenic twitch contraction in the mouse vas deferens by inhibiting K⁺ channels in sympathetic neurons of the hypogastric ganglion, which increases norepinephrine release from these cells (Campbell et al. 1987; Kennedy and Henderson 1990). In neurons of the CNS, sigma receptors have been shown to produce various cellular effects including inhibition of intracellular Ca²⁺ mobilization by N-methyl-D-aspartate (NMDA) in rat frontal cortical neurons (Hayashi et al. 1995) and depression of action potential firing in guinea pig hypoglossal neurons (Morin-Surun et al. 1999).

A process frequently targeted by sigma receptor modulation is intracellular calcium homeostasis. In the human neuroblastoma cell line, SK-NSH, sigma-2 receptors have been shown to evoke release of Ca²⁺ from intracellular stores (Vilner and Bowen 2000). Studies have also suggested that sigma ligands may block Ca²⁺ channels in hippocampal neurons and vascular smooth muscle (Church and Fletcher 1995; Flaim et al. 1985), although these effects were attributed to direct modulation of Ca²⁺ channels by the sigma ligands. The effect of sigma receptor activation on calcium channels, and in particular calcium channels of autonomic neurons, remains to be elucidated. Regulation of calcium channel function is a means by which various neurotransmitters exert their effects on autonomic neurons (Jeong et al. 1999). For example, both neuropeptide Y and norepinephrine depress calcium channel currents in rat intracardiac neurons (Jeong et al. 1999; Xu and Adams 1993). This inhibition of calcium channels is believed to be a mechanism by which the sympathetic nervous system modulates the activity of the parasympathetic nervous system. Similarly, acetylcholine, acting via M₄ muscarinic receptors, blocks calcium channel currents in intrinsic cardiac neurons (Cuevas and Adams 1997). This phenomenon is likely to represent a feedback mechanism in cholinergic parasympathetic neurons.

Experiments were undertaken to determine whether sigma receptors are present in autonomic neurons of the sympathetic superior cervical ganglion and the parasympathetic intracardiac ganglion, and whether activation of these receptors modulates the biophysical properties of calcium channels in these cells. Results indicate that sigma-1 receptor transcripts are expressed by individual autonomic neurons. Furthermore, sigma receptors were shown to depress peak Ca²⁺ channel currents, increase the rate of Ca²⁺ channel inactivation, and shift the voltage dependence of both steady-state inactivation and activation toward more negative potentials. Pharmacological experiments suggest that sigma-2 receptors modulate Ca²⁺ channels in these cells and that these receptors couple to Ca²⁺ channels via a signal transduction cascade that involves neither a diffusible cytosolic second messenger nor a G protein. A preliminary report of some of these results has been published (Zhang and Cuevas 2001).

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Preparation and electrical recording

Modulation of depolarization-activated Ca²⁺ channels by sigma receptor activation was studied in isolated neurons of neonatal rat intracardiac and superior cervical ganglia. The preparation of cultured neurons of neonatal rats (3-10 day old) and the electrophysiological recording methods used here have been previously described (Cuevas and Adams 1994; Cuevas et al. 2000). For the superior cervical ganglion preparation, neonatal rats were killed by inhalation of carbon dioxide; whereas for isolation of intracardiac neurons, rats were killed by decapitation. All procedures were done in accordance with the regulations of the Institutional Animal Care and Use Committee.

Membrane currents in autonomic neurons, cultured for 24-72 h, were studied under voltage-clamp mode using the whole cell recording configuration of the patch-clamp technique (Hamill et al. 1981). Electrical access was achieved through the use of the amphotericin B perforated-patch method (Rae et al. 1991) to preserve the intracellular integrity of the neurons and prevent calcium current rundown (Xu and Adams 1992). For perforated-patch experiments, a stock solution of amphotericin B (60 mg/ml) in dimethylsulphoxide (DMSO) was prepared and diluted in pipette solution immediately prior to use to yield a final concentration of 198 µg/ml amphotericin B in 0.33% DMSO. Final patch pipette resistance was 1.0-1.3 M to permit maximal electrical access under the present recording configuration. Junction potentials generated by the ions in the pipette and bath solutions were compensated for via the Pipette Offset control of the Axopatch 200B. To test for amphotericin B incorporation into the membrane patch following gigaseal formation, the neurons were held at 60 mV, and 20-ms voltage pulses to 65 mV were applied at 1 Hz. In successful experiments there was an increase in a fast capacitive transient, the appearance of a slow capacitive transient, and a decrease in the series resistance (R_s). To minimize voltage error produced by R_s, R_s was monitored throughout the experiment, and only cells in which R_s was consistently 3 M following 50% R_s compensation were used. Also, to minimize space-clamp artifact, only cells with no large visible processes were selected for the experiments.

Depolarization-activated Ca²⁺ channel currents were evoked using voltage jumps from 90 mV to more positive potentials. Capacitive and leak currents were subtracted using the P/4 protocol, which assumes a linear relationship for these currents at voltages less than 60 mV (Xu and Adams 1992). Membrane currents were amplified using an Axopatch 200B patch-clamp amplifier (Axon Instruments, Union City, CA), filtered at 5 kHz (3 dB; 4-pole Bessel filter), and digitized at 20 kHz (Digidata 1200B).

Ca²⁺ currents elicited by long (2 s) depolarizations were fit using single or double exponential functions and the Clampfit 6.0.5 program (Axon Instruments). Activation and steady-state inactivation kinetics were described using Boltzmann distributions, and dose-response curves were fit using the Hill equation. Analysis of these data were conducted using the SigmaPlot 2000 program (SPSS Science, Chicago, IL). Data points represent means ± SE. Statistical difference was determined using paired t-test for within-group experiments, and unpaired t-test for between groups experiments, and was considered significant if P < 0.05.

RT-PCR

RT-PCR techniques, similar to those previously reported (Cuevas et al. 2000), were used for the detection of sigma-1 receptor expression in autonomic neurons. Total RNA was isolated from intracardiac ganglia and associated tissue and from superior cervical ganglia (SCG; RNeasy, Qiagen, Hilden, Germany). RNA was reverse-transcribed in a 20-µl reaction volume using the SuperScript First-Strand Synthesis System for RT-PCR (Invitrogen, San Diego, CA). As a negative control, a PCR reaction with only water was conducted to eliminate the possibility of false positives due to contaminating cDNA. Primers specific for sigma-1 receptor transcripts were designed to span an intron to discriminate between genomic DNA and cDNA. The sequences of the primers used were: sigma-1_(sense)-GTCTTTTGCACGCCTCGCTGTCTGAGTACG, sigma-1_(antisense)-ACCCTCTCTGGATGGAGGTGAGTGC, which yielded a product size of 639 base pairs. PCR reactions were conducted using the SuperScript System with Platinum Taq DNA polymerase (Invitrogen). The cycling parameters were one cycle of 94°C for 2 min; 30 cycles of 94°C for 30 s, 61°C for 45 s, and 72°C for 1 min; and 1 cycle of 72°C for 5 min.

For single-cell RT-PCR experiments, SCG and intracardiac neurons were dissociated, and cytoplasm was extracted from isolated neurons as previously described (Poth et al. 1997). Briefly, the cellular content of individual neurons was harvested using the dialyzing whole cell configuration of the patch-clamp technique. The patch pipettes were filled with 3 µl of 1× SuperScript One-Step RT-PCR Reaction Mix (Invitrogen) containing 1 U/µl RNAsin (Promega, Madison, WI). Following extraction of the cytoplasm, the content of the pipette was expelled into a microfuge tube and quickly frozen on dry ice. Single-cell RT-PCR experiments were conducted immediately following the extraction using SuperScript One-Step RT-PCR with Platinum Taq (Invitrogen). Negative controls for these experiments involved suctioning extracellular solution via a patch pipette located directly above the cells. These controls were carried through all subsequent reactions to rule out the possibility of contamination from cytoplasm from nearby cells or sigma receptor clones isolated in the laboratory. The cycling parameters were 1 cycle of 50°C for 30 min and 95°C for 2 min; 40 cycles of 94°C for 30 s, 61°C for 45 s, and 72°C for 1 min; and 1 cycle of 72°C for 5 min.

RT-PCR products were gel purified using a QIAEX II Gel Purification kit (QIAGEN) and sequenced by the Molecular Biology Core Facility at the H. Lee Moffitt Cancer Center and Research Institute.

Solutions and reagents

The bath solution used in these experiments was a physiological saline solution (PSS) composed of (in mM) 70 NaCl, 70 tetraethylammonium chloride (TEA), 5 BaCl₂, 1.2 MgCl₂, 7.7 glucose, 0.0005 tetrodotoxin (TTX), and 10 HEPES (pH to 7.2 with NaOH). Barium was used as the charge carrier to maximize Ca²⁺ channel current amplitude, and to minimize any intracellular Ca²⁺-dependent current rundown (Xu and Adams 1992). All drugs, including sigma ligands, were bath applied at room temperature at a rate of ~2 ml/min into a 0.3-ml recording chamber, which permitted rapid exchange of bath solution. The pipette solution used for perforated-patch experiments contained (in mM) 75 Cs₂SO₄, 55 CsCl, 5 MgSO₄, and 10 HEPES (pH to 7.2 with N-methyl-d-glucamine). Block of ionic current through Ca²⁺ channels was achieved via bath application of 100 µM CdCl₂ (Xu and Adams 1992). For studies using conventional (dialyzing) whole cell recording configuration, the pipette solution contained (in mM) 140 CsCl, 2 MgCl₂, 2 ethylene glycol-bis (-aminoethyl ether)-N,N,N',N'-tetraacetic acid (EGTA), 2 Mg₂ATP, 0.1 GTP lithium salt (GTP), and 10 HEPES-CsOH, pH to 7.2. In some experiments GTP was replaced with 100 µM guanosine 5'-O-(2-thiodiphosphate) trilithium salt (GDP--S) to inhibit G protein activation.

All chemical reagents used were of analytical grade. Ibogaine hydrochloride, (+)-pentazocine, haloperidol, 1,3-Di-O-tolylguanidine (DTG), metaphit, and tetrodotoxin were purchased from Sigma Chemical (St. Louis, MO).

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

To determine whether sympathetic and parasympathetic neurons may be the target of sigma receptor ligands, autonomic neurons were first screened for the expression of transcripts encoding the sigma-1 receptor using RT-PCR techniques. Oligonucleotide primers were designed to span introns 2 and 3 of the sigma-1 gene to differentiate between cDNA and genomic DNA. RT-PCR of total RNA extracts from SCG and intracardiac ganglia and associated tissue (e.g., cardiac myocytes, Schwann cells, and fibroblasts) showed that sigma-1 receptor transcripts are expressed in these cells (Fig. 1A). However, since sigma-1 receptors have been found in nonneuronal cells, it seemed prudent to test for the presence of sigma-receptor transcripts at the single-cell level. Using single-cell RT-PCR techniques, transcripts encoding the sigma-1 receptor were shown to be expressed in individual intracardiac and SCG neurons (Fig. 1B). Sigma-1 receptor transcripts were detected in 57% of intracardiac neurons (4 of 7) and 67% of SCG neurons (4 of 6). Sequencing of the products obtained from individual autonomic neurons indicated exact sequence homology to the known rat brain sigma-1 receptor (Seth et al. 1998). Splice variants of the sigma-1 receptor have been reported in the rat and mouse as submissions to GenBank (accession numbers AF087827 and AF226605, respectively). The oligonucleotide primers used here were specifically designed to detect conventional sigma-1 transcripts and both of these sequence variants, but no such isoforms were detected in these cells.

View larger version (31K): [in this window] [in a new window]   Fig. 1. Detection of sigma-1 receptor transcripts in rat intracardiac and superior cervical ganglia. A: RNA from rat intracardiac and superior cervical ganglia were reverse transcribed and amplified via PCR using oligonucleotide primers specific for the sigma-1 receptor (sense, GTCTTTTGCACGCCTCGCTGTCTGAGTACG; antisense, ACCCTCTCTGGATGGAGGTGAGT GC). B: RT-PCR reaction results for 2 isolated neurons from rat intracardiac (ICG) and superior cervical (SCG) ganglia using the sigma-1 primers. A and B: arrows indicate predicted size for the sigma-1 receptor product (639 bp), and standards are 100 bp ladder.

Sigma receptor-mediated attenuation of Ca²⁺ channel currents

Sigma receptors have been shown to modulate calcium homeostasis in various cell types. One of the mechanisms by which sigma receptors appear to modulate cellular calcium is through attenuation of calcium influx through the cell membrane. However, the effects of sigma receptor activation on calcium channel function have not been determined.

Ca²⁺ channel currents were isolated by inhibiting Na⁺ currents with extracellular TTX, and K⁺ channels with intracellular Cs⁺ and extracellular TEA and Ba²⁺. Ba²⁺ was used as the charge carrier through open calcium channels in most experiments for reasons discussed in METHODS. The effect of sigma ligands on the Ba²⁺ current-voltage (I-V) relationship was examined using brief (250 ms) step depolarizations of 10-mV increments (50 to +90 mV) from a holding potential of 90 mV. Figure 2A shows a family of depolarization-activated Ba²⁺ currents (I_Ba) recorded from a single SCG neuron in the absence (Control) and presence of the sigma receptor ligand, haloperidol (10 µM). Bath application of haloperidol depressed peak I_Ba amplitude at all potentials positive to 10 mV within 3 min of drug application. Figure 2B shows the average I-V relationship obtained for six neurons before (Control) and after bath application of 10 µM haloperidol. Under control conditions, I_Ba was activated at approximately 30 mV and the I-V relation was maximal at 0 mV, reversing at approximately +50 mV. In the presence of haloperidol, the I-V relationship exhibited a similar voltage dependence, but the peak I_Ba amplitude was reduced at all voltages. At 0 mV, I_Ba decreased from a control value of 1,565 ± 76 pA to 969 ± 222 pA in the presence of 10 µM haloperidol, (n = 6). Inhibition of Ca²⁺ channel currents occurred to a similar degree when Ca²⁺ was the charge carrier (data not shown).

View larger version (14K): [in this window] [in a new window]   Fig. 2. Inhibition of Ca2+ channel currents in sympathetic neurons by the sigma receptor agonist, haloperidol. A: family of depolarization-activated Ba2+ currents recorded from a single SCG neuron in the absence (Control) and presence of 10 µM haloperidol. B: whole cell current-voltage relation obtained in the absence () and presence of 10 µM haloperidol (). Data points represent means ± SE for 6 cells.

Bath application of 100 µM Cd²⁺ completely blocked the depolarization-activated Ba²⁺ current, both in the absence and the presence of haloperidol (data not shown). The observed Cd²⁺ block, coupled with the lack of shift in the reversal potential for the depolarization-activated currents (Fig. 2B) in the presence of sigma ligands, suggests that these drugs are not activating or inhibiting another membrane conductance.

Concentration-dependent inhibition of I_Ba by sigma ligands

To determine whether the effect of haloperidol on Ca²⁺ channels is mediated by sigma receptor activation, the ability of various sigma ligands to elicit a similar response was assessed. Figure 3A shows representative currents recorded from three different SCG neurons (1-3) in the absence (Control) and presence of haloperidol, (+)-pentazocine, and DTG. The peak inward I_Ba was measured before and after exposure to various concentrations of the sigma ligands haloperidol, (+)-pentazocine, DTG, and ibogaine. For these experiments, each cell was exposed to a minimum of three drug concentrations. A plot of the mean peak I_Ba as a function of drug concentration is shown in Fig. 3B. Haloperidol had the greatest potency of the ligands tested, and a fit of the data using the Hill equation gave a half-maximal inhibitory concentration (IC₅₀) value of 6 µM. Similarly, the IC₅₀ values for ibogaine, (+)-pentazocine and DTG were 31, 61, and 133 µM, respectively, and the Hill coefficient was 1.1 for all drugs. Maximum inhibition of I_Ba by all sigma ligands tested was 95%.

View larger version (17K): [in this window] [in a new window]   Fig. 3. Dose-dependent inhibition of depolarization-activated Ca2+ channels by sigma receptor ligands in rat sympathetic neurons. A: whole cell currents evoked from 3 SCG neurons (1-3) by step depolarizations to 0 mV from a holding potential of 90 mV in the absence (Control) and presence of haloperidol, (+)-pentazocine and 1,3-Di-O-tolylguanidin (DTG) at the indicated concentrations. B: peak whole cell IBa amplitude, evoked by depolarizing to 0 mV from 90 mV, normalized to control and plotted as a function of sigma ligand concentration. Data points represent means ± SE for 5-7 neurons. The curves represent best fit to the data using the Hill equation. Half-maximal inhibition was 6 µM for haloperidol (), 31 µM for ibogaine (), 61 µM for (+)-pentazocine (), and 133 µM for DTG (), and the Hill coefficient was 1.1 for all compounds.

The effect of sigma ligands on Ca²⁺ channel current amplitude was reversible on wash out. Figure 4A shows a family of Ba²⁺ currents evoked by step depolarizations from a single neuron in the absence (Control), presence (+DTG), and following wash out for the indicated time points of the sigma ligand, DTG. Following inhibition of I_Ba, the current recovered to near control levels within 5 min of wash out (Fig. 4B). Similar reversal of inhibition was observed for all sigma ligands tested here. No significant rundown of I_Ba was observed in recordings 45 min.

View larger version (15K): [in this window] [in a new window]   Fig. 4. Reversibility of DTG-induced block of depolarization-activated Ca2+ channels. A: family of depolarization-activated Ca2+ channel currents recorded from a single rat SCG neuron in the absence (Control) and presence of bath-applied 1 mM DTG (+DTG), and following wash out of drug for the indicated time periods. B: peak Ba2+ current amplitudes before application (), during application (), and after removal () of 1 mM DTG for the cell in A. Values are plotted as a function of time. Gap represents period during which lower concentrations of DTG were applied, with 1 mM DTG application commencing at t = 13 min.

Sigma receptor antagonist depresses the effect of DTG on Ca²⁺ channels

To confirm whether the effect of sigma ligands on Ca²⁺ channels was mediated by activation of a sigma receptor, the irreversible sigma receptor antagonist, metaphit, was used. Metaphit is known to rapidly and specifically acetylate sigma receptors, which results in a block of ligand binding (Bluth et al. 1989). Isolated SCG neurons were preincubated in 50 µM metaphit (in PSS) for 10 min at room temperature. Following wash out of drug, Ba²⁺ current amplitude was similar to that recorded in control experiments (no preincubation; Fig. 5A), suggesting that preincubation in metaphit alone had no effect on Ca²⁺ channel currents. On application of DTG, Ba²⁺ current amplitude was depressed under both conditions, but in cells preincubated in metaphit the response to DTG was obtunded. Figure 5B shows a bar graph of relative mean I_Ba amplitude recorded in the presence of 100 µM DTG in control neurons (DTG; n = 7) or neurons preincubated in metaphit (Metaphit + DTG; n = 6). DTG decreased mean I_Ba by 28 ± 4% in cells exposed to metaphit, whereas in control cells the decrease was 49 ± 4%. The difference in DTG attenuation of I_Ba under both conditions was statistically significant (P < 0.01).

View larger version (11K): [in this window] [in a new window]   Fig. 5. Attenuation of DTG-mediated inhibition of Ca2+ channels by the sigma receptor antagonist, metaphit. A: depolarization activated (90 to 0 mV) Ba2+ currents recorded from 2 SCG neurons in the absence (Control, Metaphit) and presence of 100 µM DTG (DTG, Metaphit + DTG). Bottom traces are from a neuron preincubated in metaphit [50 µM in physiological saline solution (PSS), 10 min]. B: bar graph of the relative mean peak IBa (±SE) obtained by step depolarizations (90 to 0 mV) in control cells (DTG) or cells preincubated in metaphit (50 µM, 10 min; Metaphit + DTG) following bath application of 100 µM DTG. Current amplitudes were normalized to their respective controls (absence of DTG). Data were collected from 7 neurons for each condition, and asterisk denotes significant difference between the groups (P < 0.01).

Sigma receptors inhibit Ca²⁺ channels in parasympathetic neurons

To determine whether similar modulation of Ca²⁺ channels also occurs in parasympathetic intracardiac neurons, the effect of sigma receptor ligands on these channels was studied. For these experiments haloperidol, ibogaine, (+)-pentazocine, and DTG were used at concentrations near the IC₅₀ value for I_Ba inhibition, as determined in SCG neuron. Figure 6A shows currents evoked from four different neurons in the absence (Control) and presence of the indicated sigma ligands. A plot of the mean inhibition of I_Ba evoked by these sigma ligands is shown in Fig. 6B and is consistent with the observations made in SCG neurons. As in the case of SCG neurons, sigma ligands maximally inhibited peak I_Ba in intrinsic cardiac neurons by 95% (data not shown).

View larger version (25K): [in this window] [in a new window]   Fig. 6. Inhibition of Ca2+ channel currents in rat intracardiac neurons by sigma receptor ligands. A: Ba2+ currents evoked by step depolarizations to 0 mV from 90 mV recorded from 4 parasympathetic intracardiac neurons in the absence (Control) and presence of ibogaine, DTG, (+)-pentazocine, and haloperidol at the indicated concentrations. B: peak whole cell Ba2+ current amplitude normalized to control recorded in the presence of haloperidol (HAL), ibogaine (IBO), (+)-pentazocine (PTZ), and DTG. Bars represent means ± SE for 3-5 intracardiac neurons.

Effects of haloperidol on Ca²⁺ channel inactivation

Some receptors that modulate Ca²⁺ channels, such as M₄-muscarinic and ₂-adrenergic receptors, have been shown to differentially affect the rapid and slow component of Ca²⁺ channel current decay (Cuevas and Adams 1997; Xu and Adams 1993). To determine whether sigma receptors have a similar effect on Ca²⁺ channel inactivation kinetics, Ca²⁺ channel currents were evoked by step depolarization (2 s) to 0 mV from a holding potential of 90 mV in the absence and presence of 10 µM haloperidol. Figure 7A shows representative responses recorded from a single SCG neuron. Under control conditions, the time-dependent inactivation of I_Ba was biphasic, and best fit by the sum of two exponential functions with time constants of 198 ms (₁) and 2.2 s (₂). In the presence of haloperidol, the inward current was also best fit by the sum of two exponential functions, but both ₁ and ₂ were decreased to 117 ms and 1.1 s, respectively. Following wash out of haloperidol, ₁ and ₂ returned to near control levels and were 171 ms and 1.5 s, respectively. In five similar experiments, the time course of I_Ba decay was best fit by the sum of two exponential functions with mean time constants of 100 ± 15 ms (₁) and 1.2 ± 0.1 s (₂). In all SCG neurons studied, haloperidol decreased both ₁ and ₂ in a statistically significant manner (P < 0.01), and mean time constants were 62 ± 11 ms (₁) and 711 ± 146 ms (₂; n = 5). The peak amplitude of each of the two components of the fit was also depressed, with the amplitude of the first component decreasing by 40 ± 11% and that of the second component by 35 ± 7%. In all cells tested, the effect of haloperidol on the time course of decay of I_Ba was reversible on wash out and mimicked by ibogaine (data not shown).

View larger version (16K): [in this window] [in a new window]   Fig. 7. Sigma receptor modulation of time-dependent and steady-state inactivation of Ca2+ channels in SCG neurons. A: Ba2+ currents evoked from a single neuron by 2-s depolarizations to 0 mV from a holding potential of 90 mV in the absence (Control), presence of 10 µM haloperidol (Haloperidol), and following wash out of drug (Wash). Solid lines represent a best fit of the time course of current decay with the sum of 2 exponential functions. B: relative Ba2+ current amplitude as a function of prepulse amplitude in the absence () and presence () of 10 µM haloperidol. Data points represent means ± SE for 6 neurons. Solid lines represent best fit to the data with a single (control) or 2-component (haloperidol) Boltzmann distribution.

The effect of haloperidol (10 µM) on steady-state inactivation of Ca²⁺ channels in rat SCG neurons was studied using a double pulse protocol. Neurons were initially held at 90 mV, and 10-s prepulses from 120 to +10 mV were applied in 10-mV increments prior to a voltage step to +20 mV (20 ms) to activate (open) the available Ca²⁺ channels. A plot of the relative peak current amplitude [I_Ba/I_Ba(max)] as a function of prepulse voltage is shown in Fig. 7B (n = 6). The steady-state inactivation of I_Ba under control conditions exhibited a sigmoidal dependence on voltage and was best fit with a single Boltzmann function according to the equation

(1)

A fit of the mean relative I-V relationship exhibited half-maximal steady-state inactivation (V_h) at 27 mV and had a slope parameter (k) of 12. In the presence of haloperidol, however, the voltage dependence of steady-state inactivation was best fit by a two-component Boltzmann distribution

(2)

where i₁ and i₂ represent the fraction contributed by each component to the final function. The values for i₁ and i₂ were 0.21 and 0.78, respectively. The first component was half-maximally activated (V_h1) at 13 mV and had a slope parameter (k₁) of 6.3, whereas the second component was half-maximally activated (V_h2) at 53 mV and had a slope parameter (k₂) of 14.9.

Effects of haloperidol on the voltage dependence of Ca²⁺ channel activation

The voltage dependence of activation was examined by measuring tail current amplitude. Neurons were held at 90 mV, and brief steps (20 ms) to various test potentials (50 to +100) were applied prior to repolarization to 90 mV. Figure 8A shows Ba²⁺ currents obtained in the absence and presence of haloperidol (10 µM) in response to voltage steps to the indicated potentials and the ensuing tail currents elicited on repolarization to 90 mV. The corresponding I-V relationship obtained for the peak tail current amplitudes of six neurons is shown in Fig. 8B. Haloperidol significantly reduced the peak tail current amplitude at all voltages from 10 to +100 mV in a reversible manner. However, at 0 mV haloperidol decreased peak I_Ba tail current amplitude by 57% but by 73% at +50 mV. This effect does not appear to be a use-dependent phenomenon since I_Ba tail current amplitude did not decrease during a train of brief depolarizations (5 pulses, 20 ms) to +80 mV or by reversing the voltage protocol (data not shown). Also, at voltages where Ca²⁺ channels were maximally activated, inhibition by haloperidol was comparable (+80 mV, 77%; +90 mV, 78%; +100 mV, 76%). The reason for greater depression of I_Ba tail current amplitude by haloperidol at higher depolarizations is a drug-induced shift in Ca²⁺ channel activation.

View larger version (19K): [in this window] [in a new window]   Fig. 8. Sigma receptor evoked shift in voltage dependence of Ca2+ channel activation. A: Ca2+ channel tail currents evoked from a single neuron by repolarization to 90 mV from the indicated potentials in the absence (Control) and presence of 10 µM haloperidol (Haloperidol). The time scale is increased 5-fold at the start of repolarization. Mean peak tail current amplitude (B) and relative peak Ba2+ tail current amplitude (C) evoked by repolarization to 90 mV following a brief depolarization to the indicated potentials in the absence () and presence () of 10 µM haloperidol. Data points represent means ± SE for 6 neurons. Solid lines in C represent best fit to the data using a 2-component Boltzmann distribution.

Figure 8C shows a plot of the mean peak I_Ba tail current amplitude normalized to maximum I_Ba tail current amplitude in the absence and presence of haloperidol. Ca²⁺ channels exhibit sigmoidal activation at potentials positive to 40 mV under both conditions. Data points were best fit using a two-component Boltzmann distribution (Eq. 2). For control, i₁ and i₂ were 0.48 and 0.52, respectively, whereas in the presence of haloperidol i₁ and i₂ were 0.70 and 0.20, respectively. Half-maximal activation of the first component (V_h1) shifted from 4 mV in the absence (control) to 13 mV in the presence of haloperidol, while the second component (V_h2) shifted from +38 mV (control) to +22 mV (+haloperidol). This sigma receptor-induced shift in the voltage dependence of activation results in tail currents of greater amplitude at more negative potentials and is thus responsible for the difference in percent inhibition of Ba²⁺ tail currents by haloperidol at low and high depolarizations. The effects of haloperidol on the voltage dependence of Ca²⁺ channel steady-state inactivation and activation were reversible on wash out and were mimicked by ibogaine (data not shown).

Effect of intracellular dialysis with GTP and GDP--S on sigma receptor inhibition of I_Ba

In some systems, sigma receptors have been shown to couple to effector targets via a signal transduction cascade involving a G protein. To determine whether a G protein is involved in the sigma receptor-mediated modulation of Ca²⁺ channels in autonomic neurons, intrinsic cardiac neurons were dialyzed with pipette solution containing either 100 µM GTP or 100 µM GDP--S. In neurons dialyzed with GTP, sigma receptor-induced inhibition of I_Ba was similar to that observed in neurons electrically accessed using the perforated-patch method. Figure 9A shows representative currents in response to step depolarizations from 90 to 0 mV recorded from two neurons dialyzed with either GTP (top traces) or GDP--S (bottom traces) in the absence and presence of 10 µM haloperidol. A summary of the peak I_Ba amplitudes elicited on depolarization to 0 mV, normalized to their respective control values, under the different experimental conditions is presented in Fig. 9B. Haloperidol decreased I_Ba by 68 ± 8% (n = 5) in neurons dialyzed with pipette solution containing GTP and by 62 ± 3% (n = 6) in neurons dialyzed with GDP--S. The difference between these two experimental groups was not statistically significant. DTG inhibition of I_Ba was reversible on wash out of drug when cells were dialyzed with either GTP or GDP--S (data not shown).

View larger version (12K): [in this window] [in a new window]   Fig. 9. Sigma receptor inhibition of Ca2+ channels is not blocked by intracellular GDP--S. A: depolarization activate (90 to 0 mV) Ba2+ currents recorded from neurons dialyzed with pipette solutions containing either 100 µM GTP (top traces) or GDP--S (bottom traces) in the absence (Control) or presence of 10 µM haloperidol (Haloperidol). B: peak IBa recorded from neurons dialyzed with either GTP or GDP--S in the presence of 10 µM haloperidol or 100 µM ACh. Currents are normalized to their respective controls (absence of drug). Asterisk denotes significant difference between groups of cells exposed to ACh.

To determine whether the dialysis with GDP--S was sufficient to block G protein-mediated events, these cells were also exposed to 100 µM ACh to elicit muscarinic receptor-evoked inhibition of Ca²⁺ channels. Muscarinic receptors have been shown to couple to Ca²⁺ channels via a pertussis toxin-sensitive G protein in intrinsic cardiac neurons (Cuevas and Adams 1997). Following cell dialysis with GTP containing pipette solution, ACh depressed I_Ba by 40 ± 6% (n = 3). This ACh-mediated inhibition was reduced to 17 ± 4 in cells dialyzed with GDP--S (n = 3), which was significantly different from the reduction observed when GTP was used.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The results presented here provide the first evidence of sigma receptors being expressed in mammalian parasympathetic and sympathetic neurons. Transcripts encoding sigma-1 receptors were detected in individual neurons from intracardiac and superior cervical ganglia in neonatal rats. Sigma receptors were shown to inhibit all Ca²⁺ channel subtypes present in neurons of both autonomic ganglia with high efficacy. Furthermore, sigma receptors were demonstrated to differ from other modulators of Ca²⁺ in these cells on the basis of their effects on the biophysical properties of the channels and the signal transduction cascade coupling them to these Ca²⁺ channels.

Two pharmacologically distinct subtypes of sigma receptors have been identified: sigma-1 and sigma-2 receptors, respectively. Photoaffinity labeling experiments with sigma receptor-specific ligands have also revealed the presence of a 25-kDa polypeptide in guinea pig brain corresponding to the sigma-1 receptor and a polypeptide doublet of 18 and 21 kDa that is believed to represent the sigma-2 receptor (Hellewell and Bowen 1990). The mammalian sigma-1 receptor has been cloned in several species including guinea pig, human, and rat (Hanner et al. 1996; Kekuda et al. 1996; Seth et al. 1998). These receptors are expressed in the brain and in several peripheral tissues and organs (Walker et al. 1990). Because of the presence of sigma-1 receptors in cardiac muscle (Ela et al. 1994), and possibly other tissues associated with intracardiac ganglia or in support cells in the SCG, we used single-cell RT-PCR techniques to demonstrate that transcripts for this receptor are present specifically in autonomic neurons. The sequences of the sigma-1 receptor transcripts cloned from both intracardiac and SCG neurons were identical to that reported for the rat brain sigma-1 receptor (Seth et al. 1998); no splice variations of the sigma-1 receptor were detected in these neurons. Sequences for two isoforms of the sigma-1 receptor have been submitted previously to GenBank (Mei and Pasternak 1998; Wang et al. 2000). One of these isoforms, sigma-1 receptor, was cloned from mouse and is reported to have sigma-2-like binding activity (Wang et al. 2000). However, transcripts encoding these truncated forms of the sigma-1 receptor were not observed, indicating that they do not mediate the cellular responses to sigma ligands reported in this study.

In the present study, the use of sigma receptor agonists suggests that sigma receptors inhibit Ca²⁺ channels in neurons from parasympathetic intracardiac ganglia and sympathetic superior cervical ganglia. These agonists were shown to inhibit Ca²⁺ channels in over 90% of the cells tested (n > 100). In autonomic neurons, various cell membrane receptors have been shown to be coupled to Ca²⁺ channels. In mammalian intracardiac neurons, M₄ muscarinic, -adrenergic, neuropeptide Y, and µ-opioid receptors have all been shown to depress Ca²⁺ channel currents (Adams and Trequattrini 1998; Cuevas and Adams 1997; Jeong and Wurster 1997a; Kennedy et al. 1998; Xu and Adams 1993). However, maximum inhibition of peak Ca²⁺ by activation of these receptors is 75%, and their primary target is N-type calcium channels, which account for ~70% of the whole cell calcium current in these cells (Cuevas and Adams 1997; Jeong and Wurster 1997a; Xu and Adams 1993). In contrast, activation of sigma receptors in these neurons inhibits 95% of the peak current, indicating that all Ca²⁺ channel types are affected. It has been reported that these cells express N-, L-, P/Q-, and R-type calcium channels (Cuevas and Adams 1997; Jeong and Wurster 1997b; Xu and Adams 1993).

The sigma receptor inhibition of heterogeneous populations of Ca²⁺ may have significant physiological implications. Attenuation of N-type Ca²⁺ channels by -conotoxin GVIA fails to block synaptic transmission in parasympathetic ganglia, but broad-spectrum Ca²⁺ channel inhibitors, such as cadmium, eliminate excitatory postsynaptic potentials (Seabrook and Adams 1989). Inhibition of multiple classes of Ca²⁺ channels may contribute to the reported sigma receptor-mediated decrease in guinea pig ileum longitudinal muscle contraction (Campbell et al. 1989; Kinney et al. 1995). It has been proposed that a decrease in ACh release is responsible for this attenuation in muscle contraction (Campbell et al. 1989) and block of presynaptic Ca²⁺ would depress transmitter release. Therefore activation of sigma receptors may block signaling through autonomic ganglia and inhibit modulation of effector targets by peripheral neurons. In the cardiovascular system, inhibition of parasympathetic input to the heart may account for the increased heart rate, arrhythmias, and sudden cardiac death observed in some patients during therapy with haloperidol (Mehta et al. 1979; Settle and Ayd 1983; Turbott and Cairns 1984). Furthermore, haloperidol evokes a prolongation of the QT interval in the electrocardiogram (Kriwisky et al. 1990), which is similar to that induced by atropine-mediated vagal block (Annila et al. 1993).

In addition to affecting a broader population of Ca²⁺ channel types than other endogenous modulators of autonomic Ca²⁺ channels, activators of sigma receptors have profoundly different effects on Ca²⁺ channel biophysics. Whereas ACh and norepinephrine (NE), for example, have no effects on the steady-state inactivation of Ca²⁺ channels (Cuevas and Adams 1997; Xu and Adams 1993), sigma ligands shift the steady-state inactivation curve to more negative potentials. The fact that the voltage dependence of inactivation was best fit by a one-component Boltzmann distribution in the absence of sigma receptor activation but by a two-component Boltzmann distribution in the presence of haloperidol suggests that sigma receptors may not equally modulate steady-state inactivation in all Ca²⁺ channel subtypes.

Activation of sigma receptors also altered the voltage dependence of activation of Ca²⁺ channels in a manner distinct from other known Ca²⁺ channel inhibitors. In the presence of sigma receptor agonists, the Ca²⁺ channel activation curve was shifted toward more negative potentials. Other inhibitors of Ca²⁺ channels, such as µ-opioid, muscarinic and -adrenergic agonists, shift the activation curve to more positive potentials (Adams and Trequattrini 1998; Cuevas and Adams 1997; Xu and Adams 1993). Thus stronger depolarizations are required in the presence of these agents to activate the same number of Ca²⁺ channels. The shift in the Ca²⁺ channel activation curve toward more positive potentials has been explained by a "willing-reluctant" model first proposed by Bean (Bean 1989). According to this model, Ca²⁺ channels are converted in the presence of a modulator from a "willing" state to a "reluctant" state that requires stronger depolarization to open the channel. Such a shift in the voltage dependence of activation is not observed here. Further evidence for the lack in willing to reluctant shift in the presence of sigma ligands is provided by experiments in which prolonged depolarizations were applied to activate Ca²⁺ channels. The fast component of the inward Ca²⁺ current (₁), represents channels in the willing state, and this component was not preferentially inhibited. In contrast, ACh and NE primarily depress the fast inactivating component (₁) in long depolarizations and have little effect on the amplitude of ₂ (Cuevas and Adams 1997; Xu and Adams 1993). However, ACh also activates "voltage-independent" mechanisms of Ca²⁺ channel inhibition that result in depression of I_Ba at all voltages tested (Cuevas and Adams 1997; Mathie et al. 1992), as is observed here. Similarly, NE inhibition of Ca²⁺ channel activation in rat intracardiac neurons exhibits voltage-dependent and -independent components (Xu and Adams 1993). One of the outcomes of sigma receptor-mediated increase in the rate of Ca²⁺ channel inactivation and attenuation of the amplitude of both ₁ and ₂ is a greater decrease in net Ca²⁺ entry through the channels compared with other Ca²⁺ channel inhibitors.

Previous studies have suggested a possible relationship between sigma receptors and calcium channels. Dextromethorphan, a nonselective sigma receptor agonist, decreased K⁺ depolarization-evoked Ca²⁺ uptake into brain synaptosomes and PC12 cells (Carpenter et al. 1988). The half-maximal inhibition of Ca²⁺ uptake by dextromethorphan was consistent with the effect being mediated by a sigma-2 site, and sigma-2 receptors have been reported in brain and PC12 cells (Hellewell and Bowen 1990; Reid et al. 1990). It has also been suggested that some sigma ligands, including dextromethorphan, inhibit Ca²⁺ currents by directly interacting with Ca²⁺ channels (Church and Fletcher 1995; Flaim et al. 1985). Conversely, in frog melanotrophs, micromolar concentrations of (+)-pentazocine have been shown to enhance calcium conductances through activation of sigma receptors (Soriani et al. 1999). The present study shows that structurally dissimilar sigma ligands are able to modulate the biophysical properties of Ca²⁺ channels. Since these ligands have similar effects on the biophysical properties of the channels, it is unlikely that they would be acting on different sites of the channel. Although the exact binding site for these drugs on the cloned sigma-1 receptor has not been identified, any single site that permits binding to such a broad array of drugs is likely to be quite complex (see Walker et al. 1990) and conserved. Thus the lack of any significant homology between Ca²⁺ channels and the cloned sigma-1 receptor suggests that the effects of sigma ligands are likely mediated by a sigma receptor and not a direct effect on the Ca²⁺ channel. The argument against a direct effect of sigma ligands on Ca²⁺ channels is significantly strengthened by the observation that the sigma receptor antagonist, metaphit, blocks the DTG-mediated attenuation of Ca²⁺ channels.

Consistent with the effects of sigma ligands on Ca²⁺ channels being mediated by specific binding of the drugs to a sigma receptor is the finding that the rank order potency and IC₅₀ values for the various sigma ligands tested here are in agreement with those reported previously for sigma-2 receptors. Sigma-2 receptors have been shown to modulate Ca²⁺ release from intracellular stores in human SK-N-SH neuroblastoma cells (Vilner and Bowen 2000). The rank order potency reported in that study, haloperidol > ibogaine > (+)-pentazocine  DTG, and micromolar EC₅₀ values are in agreement with our findings. Sigma-2 receptors have also been reported to mediate the inhibition of guinea pig ileum longitudinal muscle contraction (Kinney et al. 1995). In that study, haloperidol was shown to depress electrically evoked contractions with an IC₅₀ nearly identical to that reported here (~6 µM). In primary cultures of rat frontal cortical neurons, sigma-2 receptor activation blocked N-methyl-D-aspartate (NMDA) mobilization of intracellular free Ca²⁺ (Hayashi et al. 1995). The IC₅₀ for haloperidol and (+)-pentazocine inhibition of peak free Ca²⁺ were ~6 and 40 µM, respectively, also in agreement with the values reported here for these drugs. One of the strongest lines of evidence for sigma-2 receptors mediating the inhibition of Ca²⁺ channels in autonomic neurons is our observation that ibogaine, a sigma-2-selective agonist (Bowen et al. 1995), exhibits greater potency than (+)-pentazocine. The affinity of sigma-1 receptors for (+)-pentazocine is ~2,000-fold greater than for ibogaine, whereas the affinity of sigma-2 receptors for ibogaine is ~6-fold higher than for (+)-pentazocine (Vilner and Bowen 2000). The IC₅₀ for ibogaine inhibition of I_Ba in these autonomic neurons is twofold greater than that determined for (+)-pentazocine.

The primary mechanism by which other known modulators of Ca²⁺ in autonomic neurons depress Ca²⁺ currents is via the activation of pertussis toxin-sensitive G proteins (Adams and Trequattrini 1998; Cuevas and Adams 1997; Xu and Adams 1993). However, no such G protein appears to be implicated in the signal transduction cascade coupling sigma receptors and Ca²⁺ channels in these cells, since intracellular dialysis with GDP--S failed to inhibit the effects of sigma ligands. Furthermore, the inability of cell dialysis to block the effects of sigma receptors suggests that a diffusable cytosolic second messenger is likely not involved. Similarly, sigma receptors have been shown to modulate K⁺ channels in rat pituitary cells through a membrane-delimited signaling pathway that does not incorporate a G protein (Lupardus et al. 2000). (+)-Pentazocine exhibited an IC₅₀ value of ~50 µM for the inhibition of K⁺ channels in neurohypophysial terminals (Lupardus et al. 2000), suggesting that, like Ca²⁺ channel inhibition in autonomic neurons, it is mediated specifically by sigma-2 receptors.

Taken together, molecular biology studies and pharmacological studies conducted here suggest that both sigma-1 and sigma-2 receptors are expressed in intracardiac and superior cervical ganglion neurons. However, our experiments indicate that only the sigma-2 receptor, which remains to be cloned, couples to Ca²⁺ channels in these cells. Given that the sigma-2 receptor has been shown to be a distinct molecular entity (Hellewell and Bowen 1990), it is doubtful that the sigma receptor shown to modulate Ca²⁺ channels here is a modified form of the sigma-1 receptor gene product. Thus the cellular function of the sigma-1 receptors found in these autonomic neurons remains to be determined. One possibility is that sigma-1 receptors are responsible for the changes in action potential firing evoked by sigma receptor activation in these cells (unpublished observation).

In conclusion, rat intracardiac and SCG neurons express sigma-1 and sigma-2 receptors, and activation of these receptors alters the biophysical properties of Ca²⁺ channels and attenuates whole cell Ca²⁺ channel currents. Pharmacological experiments suggest that the modulation of Ca²⁺ channels is mediated by sigma-2 receptors. Because of the importance of Ca²⁺ channels in the function and regulation of the autonomic nervous system, sigma receptors are likely to have a significant role in the modulation of autonomic nerve activity and thus on regulation of the cardiovascular system and other effector targets.