α₂-ADRENOCEPTORS COUPLE TO INHIBITION OF R-TYPE CALCIUM CURRENTS IN MYENTERIC NEURONS

Abstract

α₂-adrenoceptors inhibit Ca²⁺ influx through voltage-gated Ca²⁺ channels throughout the nervous system and Ca²⁺ channel function is modulated following activation of some G-protein coupled receptors. We studied the specific Ca²⁺ channel inhibited following α₂-adrenoceptor activation in guinea pig small intestinal myenteric neurons. Ca²⁺ currents (I_Ca²⁺) were studied using whole-cell patch clamp techniques. Changes in intracellular Ca²⁺ (Δ[Ca²⁺]_i) in nerve cell bodies and varicosities were studied using digital imaging where Ca²⁺ influx was evoked by KCl (60 mM) depolarization. The α₂-adrenoceptor agonist, UK 14,304 (0.01 − 1 μM) inhibited I_Ca²⁺ and Δ[Ca²⁺]_i; maximum inhibition of I_Ca²⁺ was 40%. UK 14,304 did not affect I_Ca²⁺ in the presence of SNX-482 or NiCl₂ (R-type Ca²⁺ channel antagonists). UK 14,304 inhibited I_Ca²⁺ in the presence of nifedipine, ω-agatoxin IVA or ω-conotoxin, inhibitors of L-, P/Q- and N-type Ca²⁺ channels. UK 14,304 induced Inhibition of I_Ca²⁺ was blocked by pertussis toxin pretreatment (1 μg/ml for 2 hr). α₂-Adrenoceptors couple to inhibition of R-type Ca²⁺ channels via a pertussis toxin-sensitive pathway in myenteric neurons. R-type channels may be a target for the inhibitory actions of norepinephrine released from sympathetic nerves on to myenteric neurons.

Gastrointestinal motility is controlled in part by the enteric nervous system (ENS) and by extrinsic sympathetic nerves. The ENS provides primary control over gut function while the sympathetic innervation modulates ENS activity (1). The myenteric plexus, a division of the ENS, is particularly important for control of gastrointestinal motility (2). Using electrophysiological criteria, myenteric neurons can be classified as S or AH neurons (3) and AH neurons may be intrinsic primary afferent neurons (4). In AH neurons the action potential has a Ca²⁺ shoulder that is dependent on activation of N-type Ca²⁺ channels (5). Ca²⁺ entering during the action potential triggers calcium release from intracellular stores (6,7). The rise in intracellular Ca²⁺ then activates a Ca²⁺-dependent K⁺ channel, causing a long-lasting afterhyperpolarization (AHP)(4,8,9). S neurons are interneurons and motorneurons (10). Although action potentials in S neurons are blocked completely by tetrodotoxin (sodium channel antagonist), they are also associated with increases in intracellular calcium that are partly dependent on activation of N-type Ca²⁺ channels (11).

Guinea pig myenteric neurons express multiple Ca2+ subtypes (12). The total Ca²⁺current (I_Ca²⁺) in myenteric neurons is composed of contributions of Ca²⁺ influx through L- N- P/Q- and R-type Ca²⁺ channels (13-18). The R-type channels make the largest contribution to the total I_Ca²⁺ in myenteric neurons of guinea-pig small intestine maintained in the primary culture (13). R-type Ca²⁺ channels are composed of the pore forming α_1E subunit (19,20). SNX 482 (21) or low concentrations of NiCl₂ (13,22) can block selectively currents passing through R-type channels.

Sympathetic nerves inhibit intestinal motility by releasing norepinephrine which activates inhibitory α₂-adrenoceptors expressed by myenteric neurons and their nerve endings (24,25). α₂-Adrenoceptors couple to inhibition of calcium influx through voltage-gated calcium channels via a G-protein dependent pathway in the nervous system (26-28). α₂-Adrenoceptors couple to the inhibitory G_o/G_i-protein to inhibit adenylyl cyclase or to interact directly with calcium channels (28). Coupling between α₂-adrenoceptor and specific subtypes of voltage-gated calcium channels has not been studied myenteric neurons. This study was designed to identify the type of calcium channel that is inhibited following activation of the α₂-adrenoceptor in myenteric neurons and in their varicosities. These studies were done using whole-cell patch clamp techniques and calcium imaging methods to study Ca²⁺ dynamics in guinea-pig small intestinal myenteric neurons maintained in primary culture.

METHODS AND MATERIALS

All animal use protocols were approved by the Institutional Animal Use and Care Committee at Michigan State University.

Primary Culture of Myenteric Neurons

Newborn (1−2 days old) guinea pigs were killed by severing the major neck blood vessels after deep halothane anesthesia. The entire length of small intestine was placed in cold (4°C) sterile-filtered Krebs’ solution of the following composition (millimolar): 120 NaCl, 5 KCl, 2.5 CaCl₂, 1.2 MgCl₂, 1.3 NaH₂PO₄, 25 NaHCO₃, and 11 glucose. The longitudinal muscle myenteric plexus was stripped free using a cotton swab and cut into 5-mm pieces. Tissues were divided into 4 aliquots, and each aliquot was transferred to 1 ml of Krebs’ solution containing 1 mg/ml trypsin for 15−20 min at 37 °C. Tissues were then triturated 30 times and centrifuged at 900 g for 5 min using a bench-top centrifuge. The supernatant was discarded, and the pellet was resuspended and incubated (30 min, 37°C) in Krebs’ solution containing 1 mg/ml collagenase. The suspension was triturated again 30 times and centrifuged at 900 g for 5 min. The pellet was suspended in minimum essential medium (MEM) containing 10% fetal bovine serum, gentamicin (10 μg/ml), penicillin (100 U/ml) and streptomycin (50 μg/ml). Cells were plated on glass cover slips coated with poly-l-lysine (50 μg/ml for 2 h) and maintained in an incubator at 37°C with 5% CO₂ for up to 2 weeks. After 2 days in culture, cytosine arabinoside (10 μM) was added to the MEM to limit smooth muscle and fibroblast proliferation. The medium was changed twice weekly.

Electrophysiological methods

Whole cell patch-clamp recordings were carried out at room temperature. Fire-polished patch pipettes with tip resistances of 3−6 MΩ were used for whole cell recordings. Seal resistances for all recordings were ≥ 5GΩ. The extracellular solution contained (millimolar): 97 NaCl, 20 tetraethylammonium, 4.7 CsCl, 5 CaCl₂, 1.2 MgCl₂, 1.2 NaH₂PO₄, 25 NaHCO₃, 11 glucose, and 0.0003 TTX. The pipette solution contained (millimolar): 160 CsCl, 2 MgCl₂, 10 EGTA, 10 HEPES, 1 ATP, and 0.25 GTP. The pH and osmolality for the extracellular solutions were adjusted to 7.2−7.4 (using CsOH or KOH) and 310−320 mosmol/kg H₂O (using CsCl). I_Ca²⁺ was recorded by depolarizing the membrane potential to −10 mV from a holding potential of −70 mV. All recordings were made using an Axopatch 200A amplifier (Axon Instruments, Foster City, CA). Data were acquired using pClamp 6.0 (Axon Instruments) and currents were sampled at 5 kHz and were filtered at 2 kHz (4-pole Bessel filter).

Calcium imaging

The methods for measuring intracellular Ca²⁺ in myenteric neurons maintained in primary culture were similar to those published by others (29-32). Myenteric neurons grown on glass coverslips were loaded for 45 min at 37°C in Opti-MEM (Invitrogen, Carlsbad, CA) solution containing the calcium indicator dye, Fluo-4 AM (2 μM)(Molecular Probes, Inc., Eugene, OR) and Pluronic F-127 (1% v/v)(Molecular Probes). After Fluo-4 loading, cover slips containing neurons were transferred to a cover glass-based chamber mounted on a microscope (Olympus, IX 70). The chamber was perfused continuously with oxygenated (95 %O₂, 5% CO₂) Krebs’ solution. The Fluo-4 was excited at 488 nm using a high speed multi-wavelength illuminator (DeltaRam, Photon Technology, Inc., Birmingham, NJ) and fluorescence emission was collected in the 500 − 560 nm range using a CCD camera (8 bit acquisition, IC-100, Photon Technology) connected to the microscope. Ca²⁺ influx through voltage-gated calcium channels was evoked using elevated extracellular KCl (60 mM). Δ[Ca²⁺]_i was calculated by subtracting the baseline fluorescence from the peak fluorescence measured during KCl depolarization; these values are expressed as arbitrary fluorescence units (AFU). Each cover slip of neurons was exposed to three successive applications of elevated KCl applied at 10 minute intervals. The first application was the control response while the second two applications occurred in the presence of drugs used to block the calcium response. Pilot studies revealed that the KCl-induced increases in calcium fluorescence declined significantly after 30 minutes. The acquired fluo-4 fluorescence signals were stored on a computer hard drive for subsequence analysis using ImageMaster software (Version 5.0, Photon Technology).

Drug application

In whole cell patch-clamp recordings, drugs were applied using an array of quartz, gravity-fed flow tubes (320 μm ID and 450 μm OD; Polymicron Technologies, Phoenix, AZ). The distance from the mouth of the tubes to the cell examined was about 200 μm, and the position of the tubes was controlled manually using a micromanipulator. For pertussis toxin treatment, myenteric neurons were incubated in MEM containing pertussis toxin (1 μM) for 2 hr before the patch-clamp experiments. For calcium imaging experiments, drugs were applied via addition to the bath perfusion solution.

Data Analysis

Effects of different treatments were compared using Student's t-test or repeated measures analysis of variance followed by Student Newman Keul's post hoc test. Data are expressed as mean ± s.e.m. and n values indicate the number of neurons from which data were obtained. P < 0.05 was considered statistically significant.

Drugs

ATP (Disodium salt, Sigma Chemical Co., St. Louis MO, A-3377); CdCl₂ (Sigma, C-3141); collagenase (EMD Biosciences, Inc. San Diego, CA) 234200); cytosine β-d-Arabinoside (Ara-C, Sigma, C-6645); fetal bovine serum (Sigma, F-2442); Fluo-4 AM (Molecular Probes, F-14201); Gentamicin (Sigma, G-1272); GTP (Sodium salt, Sigma, G-8752); minimum essential medium (MEM, Sigma, M-0268); NiCl₂ (Sigma, N-5756); nifedipine (Sigma, N-7634); Opti-MEM (GIBCO, 51985−034); penicillin/streptomycin (Sigma, P-0781); pertussis toxin (PTX, Sigma, P-7208); Pluronic F-127 (Molecular Probes, P-6866); poly-L-lysine (Sigma, P-2636); SNX 482 (Alomone, Inc. Jerusalem, Israel, S-500); tetraethylammonium chloride (TEA, Sigma, T-2265); tetrodotoxin (TTX, Sigma, T-5651); trypsin (Sigma T-5266); UK 14,304 (Sigma, U-104); ω-Agatoxin IVA (ATX, Alomone, A-500); ω-Conotoxin GVIA (CTX, Alomone, C-300);

RESULTS

UK 14,304 inhibits I_Ca2+

UK 14,304 is a selective α2 adrenoceptor agonist (33) that has been used in a number of studies of α₂-adrenoceptors in the enteric nervous system (27,28). Therefore, we used this agonist to study the effects of α₂-adrenoceptor activation on I_Ca²⁺ in myenteric neurons. I_Ca²⁺ was activated by depolarizing the membrane potential from a holding potential of −70 mV to a test potential of −10 mV. I_Ca²⁺ did not decline during repeated depolarizations applied over 60 minutes as described previously (13). Direct measurements of soma I_Ca²⁺ using whole cell recording techniques showed that I_Ca²⁺ was inhibited by UK 14,304, in a concentration-dependent (0.0001 − 1 μM) and reversible manner (Fig. 1). At the maximum concentration (1 μM), UK 14,304 inhibited I_Ca²⁺ by 45.0 ± 6.4 % from −1.8 ± 0.3 nA to −1.0 ± 0.2 nA (P < 0.05, n = 7; Fig. 1).

Figure 1.

Inhibitory effect of UK 14,304 on I_Ca²⁺. A, Activation of α₂-adrenoceptors with UK 14,304 inhibited I_Ca²⁺. I_Ca²⁺ was activated by depolarizing the membrane potential from −70 mV to −10 mV. B, Concentration-response relationship for UK 14,304 induced inhibition of I_Ca²⁺. Inhibition of I_Ca²⁺ is plotted as amplitude in the presence of each concentration of UK 14,304 normalized to the control I_Ca²⁺. Points represent the mean ± s.e.m. of data from 4−7 cells.

UK 14,304 selectively inhibits R-type calcium channels

Myenteric neurons express functional N-, P/Q, L- and R-type Ca²⁺ channels (12,13). In the present study, it was found that the relative contribution of each Ca²⁺ channel subtype to the total current varied somewhat from cell-cell. In order to estimate the relative contributions of each channel type we determined the percent inhibition of the total Ca²⁺ current caused by ω-agatoxin (ATX, P/Q –type channel antagonist, 0.1 μM), (ω-conotoxin (CTX, N-type channel antagonist, 0.1 μM), nifedipine (L-type channel antagonist, 1 μM) and NiCl₂ (R-type channel antagonist, 50 μM) in 10 neurons. Based on these studies it was found that P/Q-type channels contributed 17 ± 2 % (range = 9 − 32%), L-type channels contributed 22 ± 2% (range = 15 − 40%), N-type channels contributed 29 ± 4% (range = 13 − 48%) and R-type channels contributed 45 ± 3% (range = 29 − 57%).

We next investigate the actions of UK 14,304 on I_Ca²⁺ carried by specific Ca²⁺ channel subtypes. In these neurons, nifedipine (1 μM) reduced I_Ca²⁺ by 19 ± 1% from −1.6 ± 0.2 to −1.3 ± 0.2 nA (P < 0.05, n = 5; Fig. 2A,B). UK 14,304 (1 μM) inhibited nifedipine-resistant I_Ca²⁺ by 54 ± 2.0 % to −0.7 ± 0.09 nA (P < 0.001, n = 5). I_Ca²⁺ recovered after washing with nifedipine-containing solution (n = 4; Fig. 2B).

Figure 2.

Inhibitory effect of UK 14,304 on I_Ca²⁺ in the presence of the L-type channel antagonist, nifedipine. A, Representative recordings of I_Ca²⁺ showing a reduction by nifedipine and a further inhibition by addition of UK 14,304. B, Effect of UK14,304 (1 μM) to I_Ca²⁺ in the presence of nifedipine (1 μM). Currents recorded in the presence of nifedipine and UK 14,304 were normalized to the control current in the same neuron. The inhibitory effects of UK 14,304 were reversible on washing with nifedipine-containing Krebs’ solution (n = 4). I_Ca²⁺ was activated by depolarizing the membrane potential from −70 mV to −10 mV.

ATX (0.1 μM), reduced I_Ca²⁺ by 20 ± 4% from −0.9 ± 0.1 to −0.7 ± 0.06 nA (P < 0.01, n = 5; Fig. 3). In the presence of ATX, UK 14,304 (1 μM) reversibly inhibited I_Ca²⁺ by 50 ± 3% to 0.4 ± 0.05 nA (P < 0.001, n = 5; Fig. 3A,B).

Figure 3.

Inhibitory effect of UK 14,304 on I_Ca²⁺ in the presence of ω-agatoxin IVA (ATX), a P/Q-type channel toxin. A, UK 14,304 inhibited I_Ca²⁺ in the presence ATX. I_Ca²⁺ was activated by depolarizing the membrane potential from −70 mV to −10 mV. B, UK14,304 inhibits I_Ca²⁺ in the presence of ATX. Currents recorded in the presence of ATX and UK 14,304 were normalized to the control current in the same neuron. ATX alone reduced I_Ca²⁺ significantly (P < 0.05). In the presence of ATX, UK 14,304 (1 μM) further inhibited I_Ca²⁺ (P < 0.05). The inhibitory effect of UK 14,304 was reversible by washing with ATX-containing Krebs’ solution (n = 4).

CTX (0.1 μM), reduced I_Ca²⁺ by ∼35%. In the presence of CTX, UK 14.304 (1 μM) reversibly reduced I_Ca²⁺ by 22 ± 3 % from −0.96 ± 0.1 to −0.7 ± 0.1 nA (P < 0.05, n = 8; Fig. 4A,B).

Figure 4.

Inhibitory effect of UK 14,304 on I_Ca²⁺ in the presence of an N-type Ca²⁺ channel toxin. A, UK 14,304 inhibited I_Ca²⁺ in the presence of CTX. I_Ca²⁺ was activated by depolarizing the membrane potential from −70 mV to −10 mV. B, Inhibitory effect of UK 14,304 on I_Ca²⁺ in the presence of CTX. CTX inhibited I_Ca²⁺ (P < 0.05) and UK 14.304 (1 μM) produced a further reduction in current amplitude (P < 0.05, n = 8). Currents recorded in the presence of CTX with 14,304 were normalized to the current recorded in the presence of CTX in the same neuron.

NiCl₂ (50 μM), inhibited I_Ca²⁺ by 47 ± 3% from −1.6 ± 0.2 to −0.9 ± 0.1 nA (P < 0.05, n = 6; Fig. 5A,B). In the presence of NiCl₂ (50 μM), UK 14,304 (1 μM) did not further reduce I_Ca²⁺ (P > 0.05, n = 6; Fig. 5B). The effect of UK 14,304 on I_Ca²⁺ in the presence of the specific R-type Ca²⁺ channel toxin, SNX-482 (0.1 μM), was also tested. In these experiments the mean control I_Ca²⁺ was −1.1 ± 0.2 nA (n = 5) and SNX-482 reduced I_Ca²⁺ by 44 ± 4% to −0.6 ± 0.1 nA (P < 0.05, n = 5; Fig. 5C,D). Subsequent application of UK 14,304 did not further reduce I_Ca²⁺ (P > 0.05, n = 5; Fig. 5C,D). I_Ca²⁺ recovered after washing with drug-free Krebs’ solution (Fig. 5D).

Figure 5.

UK 14,304 does not inhibit I_Ca²⁺ in the presence of R-type Ca²⁺ channel blockers. A and C, Original recordings of I_Ca²⁺ show that UK 14,304 did not alter I_Ca²⁺ in the presence NiCl₂ (A) or SNX-482 (C). B and D, Lack of effect of UK14,304 on I_Ca²⁺ in the presence of R-type Ca²⁺ channel toxins. I_Ca²⁺ was inhibited by NiCl₂ (B, P < 0.05, n = 6) and SNX-482 (D, P < 0.05, n = 5). In the presence of NiCl₂ or SNX-482 (D, n = 5), UK 14,304 did not further reduce I_Ca²⁺ (P > 0.05). In both cases, washing with blocker free Krebs’ solution restored I_Ca²⁺ (n = 3 for NiCl₂ in B, n = 4 for SNX 482 in D). I_Ca²⁺ in the presence of blockers was normalized to the current amplitude recorded under control conditions.

Inhibitory effect of UK 14,304 on I_Ca²⁺ is blocked by pertussis toxin

After pertussis toxin treatment, activation of α₂-adrenoceptor by UK 14,304 (1 μM) failed to alter I_Ca²⁺ significantly (P > 0.05, n = 5, Fig. 6A,B).

Figure 6.

PTX blocks the inhibitory effect of UK 14,304 on I_Ca²⁺. A, The inhibitory effect of UK 14,304 (1 μM) on I_Ca²⁺ was blocked by PTX. I_Ca²⁺ activated by depolarizing the membrane potential to − 10 mV from a holding potential of −70 mV. B, PTX pretreatment blocks UK 14,304-induced inhibition of I_Ca²⁺. I_Ca²⁺ was plotted as the mean current normalized to current amplitude before UK 14,304 application. Data are mean ± s.e.m. (n = 5).

UK 14,304 inhibits R-type Ca²⁺ channels in the soma and varicosities

Δ[Ca²⁺]_i in the soma and varicosities was evoked by depolarizing the membrane potential with KCl (60 mM). Varicosites were identified visually as periodic swellings in nerve fibers (Fig. 7A). Many of these varicosities were in close proximity to individual soma suggesting that these might be the sites of synaptic contact (Fig. 7B). We verified that Δ[Ca²⁺]_I in varicosities did not require axonal conduction by testing the effects of the sodium channel blocker, tetrodotoxin (TTX, 0.3 μM) on Δ[Ca²⁺]_i. These data show that TTX reduced the calcium signal by about 20% indicating that sodium channel dependent action potentials contributed to the calcium signal. However, the Δ[Ca²⁺]_I is due predominately to action potential independent activation of voltage-gated Ca²⁺ channels (Fig. 7C,D).

Figure 7.

Δ[Ca²⁺]_i caused be elevated KCl (60 mM) in cell soma and in varicose nerve fibers. A-C, Images of Fluo-4 fluorescence in two neurons and in a varicose nerve fiber. A shows baseline fluorescence, B shows Δ[Ca²⁺]_i in the presence of KCl (60 mM) and C shows the Δ[Ca²⁺]_i response in the presence of tetrodotoxin (0.3 μM, TTX). Arrows in “B” show the position of varicosities. D, Mean data from 3 experiments similar to that shown in A-C. Data were obtained from 6 neurons and 14 varicosities. TTX significantly reduced Δ[Ca²⁺]_i but did not block this response (P < 0.05, Student's t-test for paired data). These data indicate axonal conduction contributed to the Δ[Ca²⁺]_i response axonal conduction was not an absolute requirement. The Δ[Ca²⁺]_i response was due largely to depolarization-induced activation of voltage-gated Ca²⁺ channels in the cell soma and in varicosities.

In the next set of experiments, we verified the stability of the KCl-induced Δ[Ca²⁺]_i over the time course of our studies. It was found that Δ[Ca²⁺]_i did not decline significantly in either nerve cell bodies or in varicosities during 3 successive KCl applications over a 20 minute period, the time course of subsequent experiments (Fig. 8A,B).

Figure 8.

Time control study for stability of Δ[Ca²⁺]_i during 3 successive KCl (60 mM) applications. Data are Δ[Ca²⁺]_i caused by KCl depolarization at 10 minute intervals. Data were normalized to the first Δ[Ca²⁺]_i response and are mean + s.e.m. Δ[Ca²⁺]_i did not decline significantly over the time course of this protocol (P > 0.05).

Δ[Ca²⁺]_i was measured in the soma and in varicosities of individual myenteric neurons (Fig. 9A,B). While drug-induced effects on I_Ca²⁺ can be measured directly in the soma it is more difficult to measure drug-induced changes in I_Ca²⁺ varicosities. We used Δ[Ca²⁺]_i as an indirect measure of I_Ca²⁺ in varicosities. UK 14,304 (1 μM) inhibited Δ[Ca²⁺]_i in the cell soma by 48 ± 8% (from 56 ± 11 to 30 ± 7 AFU; P < 0.05, n = 6; Fig. 9B). UK 14,304 (1 μM) also inhibited Δ[Ca²⁺]_i in varicosities by 44 ± 5% (from 10 ± 2 to 6 ± 2 AFU; P < 0.05, n = 7; Fig. 9B). The Δ[Ca²⁺]_i evoked by KCl (60 mM) was abolished by CdCl₂ (100 μM) in both the cell soma and in varicosities (Fig. 9A,B).

Figure 9.

Inhibitory effect of UK 14,304 on Δ[Ca²⁺]_i. A, Recordings Δ[Ca²⁺]_i before and during application of KCl (60 mM) in the absence and presence of UK 14,304 (1 μM) and CdCl₂ (100 μM). UK 14,304 reduced Δ[Ca²⁺]_i in both soma (upper traces) and a varicosity (lower traces). Δ[Ca²⁺]_i evoked by KCl depolarization was blocked CdCl₂. B, Pooled data from experiments similar to that shown in “A”. Control measurements made in individual soma or varicosities were set to a value of “1” and data obtained in the presence of UK 14,304 or CdCl₂ were expressed as a fraction of that control value. Data are mean ± s.e.m. *indicates significantly different from Control (P <0.05). #indicates significantly different from measurements made in the presence of UK 14,304 (P <0.05).

NiCl₂ (50 μM) inhibited soma Δ[Ca²⁺]_i evoked by KCl depolarization by 55 ± 5% from 45 ± 6 to 20 ± AFU (P < 0.05, n = 4; Fig. 10A,B). In the presence of NiCl₂ (50 μM), UK 14,304 (1 μM) did not alter Δ[Ca²⁺]_i (P > 0.05, n = 4; Fig. 10A,B). In 4 of 5 varicosities, NiCl₂ (50 μM) inhibited Δ[Ca²⁺]_i by 47 ± 7% from 5.2 ± 0.8 AFU to 2.6 ± 0.3 AFU (P < 0.05, n = 4; Fig. 10A,B). In the presence of NiCl₂ (50 μM) further application of UK 14,304 (1 μM) also did not alter Δ[Ca²⁺]_i (P > 0.05, n = 4; Fig. 10A,B). Similarly, SNX- 482 (0.1 μM) inhibited Δ[Ca²⁺]_i in the cell soma and in varicosities and UK 14,304 failed to inhibit Δ[Ca²⁺]_i in the presence of the R-type Ca²⁺ channel blocker (Fig. 11).

Figure 10.

UK 14,304 does not inhibit on Δ[Ca²⁺]_i in the presence of NiCl₂ in soma and varicosities. A, Original recordings of Δ[Ca²⁺]_i evoked by KCl (60 mM). NiCl₂ (50 μM) inhibited Δ[Ca²⁺]_i in the soma (upper tracings) and varicosities (lower traces). Subsequent application of UK 14,304 (1 μM) did not alter Δ[Ca²⁺]_i evoked by KCl depolarization. B, Pooled from experiments illustrated in “A”. NiCl₂ inhibited Δ[Ca²⁺]_i evoked by KCl in the soma and in varicosities while subsequent addition of UK 14,304 did not further reduce Δ[Ca²⁺]_i evoked by KCl depolarization. Control measurements made in individual soma or varicosities were set to a value of “1” and data obtained in the presence of UK 14,304 or NiCl₂ were expressed as a fraction of that control value. Data are mean ± s.e.m. *indicates significantly different from Control (P <0.05).

Figure 11.

UK 14,304 does not inhibit on Δ[Ca²⁺]_i in the presence of SNX-482 in soma and varicosities. Original recordings of Δ[Ca²⁺]_i evoked by 60 mM KCl. SNX-482 (0.1 μM) inhibited Δ[Ca²⁺]_i in a soma (upper traces) and a varicosity (lower traces). Further application of UK 14,304 (1 μM) did not alter Δ[Ca²⁺]_i evoked by KCl depolarization.

DISCUSSION

Previous work showed that, in guinea pig small intestinal myenteric neurons, total I_Ca²⁺ is composed of currents contributed by L- N- P/Q- and R-type voltage-gated calcium channels (12,13). In the present study, we found that I_Ca²⁺ was inhibited in a concentration-dependent manner by the selective α₂-adrenoceptor agonist, UK 14,304. UK 14,304-induced inhibition of I_Ca²⁺ persisted after L- N- and P/Q-type channels were blocked indicating that the α₂-adrenoceptor does not couple to inhibition of those Ca²⁺ channel subtypes. However, when the R-type Ca²⁺ channel was blocked by low concentrations of NiCl₂ or by SNX-482, UK 14,304 no longer inhibited I_Ca²⁺. Previous work showed that α2-adrenoceptors couple to inhibition of Ca²⁺ channels in guinea pig small intestinal submucosal neurons where α₂-adrenoceptors couple specifically to inhibition of N-type Ca²⁺ channels (36). However, our data demonstrate that α₂-adrenoceptors selectively couple to inhibition of R-type calcium channels in guinea pig small intestinal myenteric neurons. The differential coupling of α₂-adrenoceptors to N- and R-type Ca²⁺ channels may be related to differential expression of Ca²⁺ channel subtypes in the two enteric plexuses. Immunohistochemical studies have shown that the subunits forming N-type channels (α1B) are expressed by myenteric and submucosal neurons (36) while α_1E subunits (forming R-type Ca²⁺ channels) are expressed only in the myenteric plexus (37). Therefore, R-type Ca²⁺ channels would be available for modulation by α₂-adrenoceptors only in the myenteric plexus.

It is possible that our recording conditions favored R-type calcium channel modulation by α2 adrenoceptors as the high extracellular Ca²⁺ concentration (5 mM) could favor the function of one channel subtype over another. The single channel conductance of the calcium channel subtypes (10−20 pS) expressed by myenteric neurons are similar, so elevated Ca²⁺ does not favor one channel over another in terms of peak current. However, there are differences in Ca²⁺-dependent mechanisms controlling channel activation and inactivation (19). L-type Ca²⁺channels in muscle exhibit a rapid calcium dependent inactivation while N and P/Q type channels exhibit both Ca²⁺-dependent inhibition and excitation under different activation conditions. The protocol we used in which Ca²⁺ currents were activated by single voltage commands applied at 10 s intervals would minimize the contribution of these Ca²⁺ dependent mechanisms modulating channel activity.

R-type Ca²⁺ channels are localized to cell bodies of neurons throughout the nervous system (23,36-38) and these Ca²⁺ channels have several functions. For example, R-type Ca²⁺ channels regulate action potential firing in hippocampal CA1 pyramidal neurons (39). While R-type Ca²⁺ channels are responsible for 50% of the total voltage-gated I_Ca²⁺ in myenteric neurons (13), their function has not yet been established. R-type calcium channels in myenteric neurons require strong depolarizations and currents carried by R-channels have fast activation kinetics (3). These attributes would allow them to contribute to Ca²⁺ entry during action potentials in the somatodendritic region of myenteric neurons. Ca²⁺ entry through R-type channels may play a role in action potential firing patterns in the myenteric plexus. α₂-adrenoceptors are localized to the soma of guinea pig ileum myenteric neurons (34,40). These receptors are targets for norepinephrine released by sympathetic nerves supplying the small intestine and activation of sympathetic nerves inhibits enteric neurons (25,34,35). α₂-adrenoceptor-mediated inhibition of R-type Ca²⁺ currents could modulate action potential duration or firing rate in myenteric neurons.

α₂-Adrenoceptors couple to the inhibitory G_O/G_I G-proteins (28,35,36,41,42). In the present study, the inhibitory effect of UK 14,304 on I_Ca²⁺ was blocked by pertussis toxin, a selective G_O/G_I G-protein toxin. This result confirms previous studies done in submucosal neurons showing that α₂-adrenoceptors couple to inhibition of voltage-gated Ca²⁺ channels via pertussis toxin sensitive G-protein (36). However, we have shown for the first time that R-type Ca²⁺ channels are a specific target for α₂-adrenoceptor-mediated inhibition. The signaling cascade that links the α₂-adrenoceptor to the R-type Ca²⁺ channel inhibition in guinea-pig myenteric neurons has yet to be elucidated. However, it has been established that in submucosal neurons α₂-adrenoceptors activate a membrane-delimited, G-protein-dependent pathway where G_O interacts directly with Ca²⁺ channels to decrease their open probability (43). It is likely that R-type Ca²⁺ channels in myenteric neurons are inhibited by α₂-adrenoceptors via a similar mechanism. In addition to coupling to inhibition of Ca²⁺ channels, α₂-adrenoceptors can also couple via a G_o-dependent mechanism to activation of K⁺ channels and membrane hyperpolarization (36). α₂-adrenoceptor mediated hyperpolarizations are very prominent in submucous plexus neurons and the hyperpolarization accounts for a large part of the inhibitory effect of norepinephrine on these cells (36). However, α₂-adrenoceptor-mediated hyperpolarization is detected only occasionally in studies using intracellular electrodes to record from neurons in the acutely isolated myenteric plexus preparation (34). Therefore, the most prominent effect of α₂-adrenoceptors activation in the myenteric plexus is likely to be direct inhibition of Ca²⁺ channels via the G_o/G_i-dependent mechanism.

We did not establish the electrophysiological classification of the neurons from which recordings were obtained because TTX and K⁺ channel blockers (TEA and Cs⁺) were present in the recording solutions throughout these studies. TTX will block synaptic transmission; therefore, we were unable to record the fast excitatory postsynaptic potentials that are a property of S neurons (3). The K⁺ channel blockers we used would block the slow action potential afterhyperpolarization that is characteristic of AH-type neurons (3). Our previous study (9) showed that R-type Ca²⁺ currents were recorded from 90% of myenteric neurons maintained in primary culture. However, in the absence of data obtained from phenotypically-identified neurons, it is not possible for us to conclude that R-type Ca²⁺ channels are expressed by both S and AH type neurons. Similarly, we showed that the inhibitory affect of α₂-adrenoceptor activation on R-type Ca²⁺ currents was observed in almost all neurons but in the absence of phenotypic identification we can not conclude that this effect occurs in both S and AH type neurons.

The data described above indicate that α₂-adrenoceptors couple selectively to inhibition of R-type Ca²⁺ channels located to nerve cell bodies. However, the principal site of action for norepinephrine in the myenteric plexus is on nerve endings where norepinephrine acts to inhibit neurotransmitter release (25,34). Detailed characterization of Ca²⁺-dependent mechanisms in myenteric neuronal varicosities is difficult because their small size makes them inaccessible to studies using micro- or patch clamp electrodes. However, imaging techniques using Ca²⁺-sensitive fluorescent probes, such as Fluo-4, and photometry provide an opportunity to study Ca²⁺ in individual varicosities. In these Ca²⁺ imaging studies, activation of voltage-gated Ca²⁺ channels was accomplished by raising extracellular KCl from 5 to 60 mM. This would change the membrane potential of neurons and varicosities to approximately −20 mV, a level similar to that used to activate Ca²⁺ channels in the patch clamp studies. KCl-induced depolarization evoked a CdCl₂-sensitive increase in Fluo-4 fluorescence in the soma and varicosities of myenteric neurons indicating that this signal was due to Ca²⁺ influx through voltage-gated calcium Ca²⁺. This provided an opportunity to study the contribution of R-type Ca²⁺channels to Ca²⁺ entry into varicosities and to study α₂-adrenoceptor modulation of these channels.

NiCl₂, at a concentration that is selective for R-type Ca²⁺ channels, and SNX-482 reduced the Ca²⁺ signal in the soma and nerve terminals by about 50%. Our whole-cell patch clamp data showed that the R-type Ca²⁺ channel carried up to 50% of the total I_Ca²⁺. Therefore, half of the KCl-evoked calcium signal measured in the cell soma and in varicosities is due to activation of R-type Ca²⁺ channels. The KCl-evoked Ca²⁺ signals in the cell soma and in varicosities were reduced by approximately 50% by UK 14,304. In addition, when the Ca²⁺ signal was reduced by NiCl₂ or SNX-482, UK 14,304 did not produce a further inhibition of this response. These data are similar to the UK 14,304-induced inhibition of I_Ca²⁺ measured in the cell soma using whole-cell recording. Based on these data we conclude that α₂-adrenoceptors expressed by myenteric neuronal varicosities couple selectively to inhibition of R-type Ca²⁺ channels.

α₂-adrenoceptors mediate presynaptic inhibition of fast and slow excitatory synaptic transmission in the ENS (34,44,45). The α₂-adrenoceptors are the target for sympathetic nervous system modulation of synaptic transmission and R-type Ca²⁺ channels may be one target for nerve-released norepinephrine. However, previous studies have shown that N, and P/Q type Ca²⁺ channels make major contributions to the calcium entry into nerve terminals required for neurotransmitter release from myenteric neurons (14-18). The contribution of R-type Ca²⁺ channels to the release of fast or slow excitatory synaptic transmitters in the myenteric plexus remains to be determined.

CONCLUSION

In guinea-pig small intestinal myenteric neurons, activation of the α₂-adrenoceptor selectively inhibits the R-type Ca²⁺ channels via a G_i/G_o-protein-linked pathway. As sympathetic nerve fibers contact the soma-dendritic region and varicosities in the myenteric plexus (46,47), α₂-adrenoceptor-mediated inhibition of R-type Ca²⁺ channels may be a mechanism by which norepinephrine can regulate myenteric neuron excitability and/or neurotransmitter release.