Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

The large cerebral arteries of several species such as dog, pig, cow, and monkey have been shown to receive dense cholinergic innervation (Lee, 1994). At one time, it was assumed that ACh acted as the transmitter for cerebral neurogenic vasodilation. This assumption was questioned when the vasodilation in isolated cat cerebral arterial rings induced by electrical stimulation of the perivascular nerves was not blocked by atropine (Lee et al., 1975; Lee, 1980). It is now clear that nitric oxide (NO), which is released from cholinergic-nitric oxidergic nerves (Kimura et al., 1997), is the major transmitter mediating cerebral neurogenic vasodilation (Toda and Okamura, 1990; Lee and Sarwinski, 1991; Gonzalez et al., 1992). ACh released from these nerves has been suggested to act as a presynaptic modulator (Lee, 1986; Ayajiki et al., 1993) or transmitter (Liu and Lee, 1999) in inhibiting NO release. Our previous studies using in vitro tissue bath techniques demonstrated that ACh inhibition of NO-mediated neurogenic vasodilation in porcine cerebral arteries was mediated by activation of presynaptic muscarinic M₂ receptors located on perivascular nitrergic nerves, possibly resulting in suppression of voltage-dependent Ca²⁺ channels on those nerves (Liu and Lee, 1999). Because the Ca²⁺ influx from the voltage-dependent Ca²⁺ channels is crucial in activating the Ca²⁺-dependent NO synthase (NOS) in neurons (Boeckxstaens et al., 1993), suppression of these channels leads to a decreased NO formation and release from these nerves. Direct evidence for inhibition of voltage-dependent Ca²⁺ channel by activation of muscarinic receptors in the cerebral perivascular nerves, however, has not been presented.

Although inhibition of voltage-dependent Ca²⁺ channels by muscarinic agonists has been reported in several types of neurons in the central and peripheral nervous systems (Mochida and Kobayashi, 1986; Wanke et al., 1987; Beech et al., 1992; Bernheim et al., 1992; Allen and Brown, 1993; Cuevas and Adams, 1997; Jeong and Wurster, 1997), direct demonstration of a muscarinic effect on Ca²⁺ channels in the perivascular nerves is technically formidable because of difficulty in using patch-clamp techniques on these nerves. Because the major portion of cerebral perivascular nitrergic nerves originates in the sphenopalatine ganglion (SPG) in many species including cat, pig, and rat (Hara et al., 1989; Suzuki et al., 1990; Nozaki et al., 1993; Suzuki and Hardebo, 1993; Minami et al., 1994; Kadota et al., 1996; Kimura et al., 1997; Yu et al., 1997), the present study was designed to determine the role of muscarinic receptors in inhibiting voltage-dependent Ca²⁺ channels in the cultured adult rat SPG by using whole-cell patch-clamp recording. We already have shown that cultured adult rat SPG neurons (both soma and dendrites) are similar to cerebral perivascular nerves of the pig in their immunohistochemical, pharmacological, and electrophysiological characteristics (Liu et al., 2000). Specifically, both soma and dendrites of the cultured rat SPG contain N-type Ca²⁺ channels, which are the dominant voltage-dependent Ca²⁺ channels in regulating Ca²⁺ influx during membrane depolarization (Liu et al., 2000).

For comparison, the in vitro tissue bath technique was used also to continue examination of presynaptic muscarinic receptors in modulating NO-mediated neurogenic vasodilation in porcine cerebral arterial rings (Liu and Lee, 1999).

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Isolation and Culture of Rat SPG Neurons. These were carried out according to methods in our previous report (Liu et al., 2000). Sprague-Dawley rats (3 to 16 weeks old) were anesthetized with sodium pentobarbital (50 mg/kg, i.p.). The sphenopalatine ganglia of both sides were dissected (Spencer et al., 1990; Liu et al., 2000) and were placed in cold Hibernate A (Invitrogen, Carlsbad, CA) solution. Ganglia were cut into smaller pieces; transferred to Mg²⁺- and Ca²⁺-free Hanks' balanced salt solution (Invitrogen) containing papain (2 U/ml; Sigma-Aldrich, St. Louis, MO), collagenase D (0.6 mg/ml; Roche Applied Science, Indianapolis, IN), and Dispase (2.4 mg/ml, Invitrogen); and then incubated for 40 min at 37°C. Cells were released by gentle trituration at the end of the incubation. The cell suspension was centrifuged at 300g for 5 min. The pellet was gently resuspended in neurobasal culture medium (Invitrogen), containing B27 (1:50 dilution, Invitrogen), 0.5 mM L-glutamine, 25 µM L-glutamate and 50 ng/ml nerve growth factor (Alomome Laboratories Ltd, Jerusalem, Israel) (Brewer et al., 1993; Liu et al., 2000). All medium and Hanks' balanced salt solution contained 100 U/ml penicillin and 100 U/ml streptomycin (Invitrogen). The cell suspension was plated into a 24-well culture plate with a poly-D-lysine coated (50 µg/ml, Sigma-Aldrich) glass coverslip (12-mm diameter; Carolina Biological Supply Co., Burlington, NC) in each well and incubated with air containing 5% CO₂ at 37°C. The growth medium was changed every 6 days.

Immunocytochemistry. SPG neurons cultured for 1 to 3 weeks were fixed in 4% paraformaldehyde for 20 to 60 min at room temperature or overnight at 4°C. After rinsing three times with PBS (pH 7.4), cells were permeabilized and nonspecific sites were blocked with 5% normal goat serum in 0.2% Triton X-100 PBS for 1 h at room temperature. After rinsing once, the cells were incubated simultaneously with anti-M₂ muscarinic receptor antibody and anti-_1b-subunit of voltage-gated Ca²⁺ channel (N-type Ca²⁺ channel) antibody at room temperature overnight. The anti-M₂-receptor antibody was a monoclonal antibody raised from rat (Chemicon; 1:100 dilution). The anti-_1b-subunit of voltage-gated Ca²⁺ channel antibody was a polyclonal antibody raised from rabbit (1: 200 dilution; Alomone Laboratories). The antibodies were diluted in 0.05% Triton X-100/PBS/1.5% NGS. After incubation with the primary antibodies, the cells were rinsed with PBS three times before incubating with secondary antibodies. The secondary antibodies were fluorescein-conjugated anti-rabbit IgG (1:40 dilution; Vector Laboratories, Burlingame, CA) or rhodamine-conjugated anti-rat IgG (1:40 dilution; Jackson Immunoresearch Laboratories, Inc., West Grove, PA). After 1-h incubation with second antibodies at room temperature, cells were rinsed with PBS (pH 8.2) three times and mounted in Vectashield mounting medium (Vector Laboratories). The stained cells were observed and photographed first under a fluorescence microscope (BX50 microscope; Olympus, Tokyo, Japan) fitted with a fluorescein filter. Without changing field, cells were photographed for the second time using a rhodamine filter. Negative controls were obtained following the same incubation procedure with neutralized serum by corresponding antigen (Liu et al., 2000).

Electrophysiology. Electrophysiological study using patch-clamp recording was carried out as described previously (Liu et al., 2000). In brief, a glass coverslip containing cultured neurons was transferred from the growth medium to a 35-mm plastic Petri dish containing the extracellular recording solution (described below; Liu et al., 2000) on a phase-contract microscope (Olympus IMT2). The calcium currents were recorded in the whole-cell configuration of the patch-clamp technique (Hamill et al., 1981; Liu et al., 2000) at room temperature. Patch electrodes were pulled from 1.5-mm outer diameter with 1.0-mm inner diameter capillary glass (World Precision Instruments PG52151-4), and the tips were then fire-polished. After filling with intracellular solution (described below; Liu et al., 2000), electrode impedance in the extracellular recording solution was 3 to 4 megohms. This component of the series resistance was fully compensated using the bridge balance control of the Axoclamp 2B (Axon Instruments) used for recording. The electrode tip potential was also subtracted while in bridge mode. Tight seals of at least 2 gigaohms, but usually 5 to 10 gigaohms were obtained by light suction. Electrode series resistance increased to 6 to 8 megohms after entry into whole-cell mode but was not further compensated. Voltage protocol generation and data acquisition were performed using pClamp software (Axon Instruments, Foster City, CA) and a digital data acquisition system (Digidata 1200). To evoke calcium currents, test pulses to 0 mV for 100 ms were applied every 15 s from a 70 mV holding potential in the absence or presence of drugs. The current traces were low-pass filtered at 3 kHz and digitized at 20 kHz. The resistive and capacitative components of the leak current were subtracted using a P/4 procedure (Armstrong and Bezanilla, 1974; Evans et al., 1998). Data were stored in a computer hard drive for later analysis. The concentration that produced 50% inhibition of maximal relaxation or peak Ca²⁺ currents (EC₅₀) were calculated for each preparation. From these values, the geometric means for EC₅₀ with 95% confidence intervals were calculated (Fleming et al., 1972).

In Vitro Tissue Bath Study on Neurogenic Vasodilation in Porcine Basilar Arterial Rings. Fresh heads of adult pigs of either sex were collected from a local slaughterhouse. The entire brain was removed and placed in a Krebs-bicarbonate solution equilibrated with 95% O₂-5% CO₂ at room temperature. Basilar arteries were dissected and cleaned of surrounding tissue under a dissecting microscope. The ring segment (4 mm long), with endothelium mechanically denuded (Lee, 1982), was cannulated with a stainless steel rod (30-gauge hemispherical section) and a short piece of platinum wire and mounted in a tissue bath containing 6 ml of Krebs-bicarbonate solution equilibrated with 95% O₂-5% CO₂ at 37°C. Tissues were equilibrated in the Krebs-bicarbonate solution for 30 min and then were mechanically stretched to a resting tension of 0.75g for another 30 min. U-46619 (0.1-1 µM) was then applied to induce an active muscle tone of about 0.75g. Tissues were electrically and transmurally stimulated with a pair of platinum electrodes through which 100 biphasic square-wave pulses of 0.6 ms in duration and 200 mA in intensity were applied at various frequencies (Lee, 1982). The neurogenic origin of this TNS-induced response was verified by its complete blockade by 1 µM TTX. Papaverine (300 µM) was applied to each tissue at the end of the experiment to induce maximal relaxation. The magnitude of a vasodilator response was expressed as a percentage of the maximal response induced by papaverine (Lee, 1982).

Solutions, Drugs Used, and Statistical Methods. The Krebs-bicarbonate solution for tissue bath contained 122.0 mM NaCl, 5.2 mM KCl, 1.33 mM CaCl₂, 1.2 mM MgSO₄, 25.0 mM NaCO₃, 0.03 mM disodium EDTA, 0.01 mM L-ascorbic acid, and 11.0 mM glucose (pH 7.4). Extracellular solution for Ca²⁺ current recording contained 130 mM NaCl, 3 mM KCl, 1 mM MgCl₂, 5.0 mM barium chloride (or 2.5 mM CaCl₂), 25 mM dextrose, and 10 mM MOPS (pH 7.25). Intracellular solution for Ca²⁺ current recording contained 140 mM cesium methane sulfonate, 1 mM MgCl₂, 5 mM EGTA, 10 mM HEPES, 2 mM MgATP, and Na₃GTP (pH 7.25).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Effect of ACh on Ca²⁺ Channel Currents in Cultured Rat SPG Neurons. ACh (0.1-1 µM) significantly inhibited the peak Ca²⁺ currents in most cultured SPG neurons (34 of 37 tested neurons). A 100-ms voltage step to 0 mV from holding potential 70 mV evoked a slowly inactivating inward Ca²⁺ current, which was largely inhibited in the presence of 1 µM ACh in the bath solution (Fig. 1A). The inhibition was observed in the tail current as well (Fig. 1A). The current-voltage (I-V) curve from 60 to +60 mV showed that ACh-induced inhibition occurred at all voltages (Fig. 1B). In a current versus time plot, which reflects the real time course of the recording, the inhibition of the Ca²⁺ currents by ACh was fast, concentration dependent, and atropine reversible (Fig. 1C). The concentration-response curve shows that the maximal inhibition is 63.1 ± 5.6% with EC₅₀ values of 2.8 (1.1-7.2) µM (Fig. 1D).

View larger version (23K): [in this window] [in a new window]   Fig. 1.   Effect of ACh on Ca2+ channel current in cultured SPG neurons of the rat. A, Ca2+ channel current evoked in response to a depolarizing step from holding potential of 70 to 0 mV in the absence (control) and presence of 1 µM ACh. B, whole-cell I-V relationship of peak Ca2+ channel current in the absence () and presence () of 1 µM ACh. C, peak Ca2+ channel current amplitude () plotted as a function of time when 0.1 and 1 µM ACh and 3 µM atropine were successively applied to the bath as indicated. The peak Ca2+ channel currents were evoked every 15 s by a depolarizing step to 0 mV from holding potential 70 mV. D, concentration-response curve of Ca2+ channel current inhibition by ACh. Data represent means ± S.D. and n indicates number of tested cells.

View larger version (18K): [in this window] [in a new window]   Fig. 2.   Effect of subtype-selective muscarinic receptor antagonists on ACh-induced inhibition of Ca2+ channel current in cultured SPG neurons. With 1 µM ACh in the bath, superimposed Ca2+ channel current elicited by depolarizing steps from 70 to 0 mV in the absence and presence of selective muscarinic receptor antagonists: 0.1 to 10 µM pirenzepine (A), 0.1 to 1 µM methoctramine (B), 0.1 to 10 µM 4-DAMP (C), and 0.1 to 10 µM tropicamide (D). Only methoctramine reversed ACh-induced inhibition of Ca2+ channel current.

Effects of Cholinergic Agents on NO-Mediated Neurogenic Vasodilation in Porcine Basilar Arteries. Porcine basilar arterial rings denuded of endothelial cells in the presence of active muscle tone induced by U-46619 exclusively relaxed upon TNS at 2 Hz (Fig. 3A). The TNS-elicited vasodilation was significantly enhanced by 0.1 µM atropine. The enhanced vasodilation was completely inhibited by 30 µM L-NNA and reversed by L-arginine. The LNNA- and TTX-sensitive TNS-elicited neurogenic vasodilation, however, was inhibited by exogenous 1 µM ACh and 1 µM ABET, a muscarinic M₂-receptor agonist (Fig. 3B). These results were reproducible in the basilar arteries from 20 pigs and were consistent with results reported previously (Liu and Lee, 1999).

View larger version (17K): [in this window] [in a new window]   Fig. 3.   Effects of muscarinic agonists and antagonist on NO-mediated, TNS-elicited neurogenic vasodilation in porcine basilar arteries. A, a representative tracing of relaxation in a porcine basilar artery without endothelial cells elicited by TNS at 2 Hz. The active muscle tone in the arterial ring was induced by 0.3 µM U-46619. The TNS-elicited relaxation was enhanced by atropine and abolished by L-NNA. The inhibition was reversed by L-arginine (L-Arg). The relaxation was abolished by TTX. B, a representative tracing showing that vasodilation induced by TNS at 2 Hz was decreased in the presence of 1 µM ACh or 1 µM ABET.

Comparable Effects of Preferential Muscarinic M₁ and M₂ Agonists on Ca²⁺ Channel Currents in Rat SPG Neurons and Porcine Cerebral Neurogenic Vasodilation. In porcine basilar arteries, ABET significantly inhibited TNS (2 Hz)-elicited neurogenic vasodilation, with EC₅₀ values of 0.5 (0.1-2.5) µM and maximal inhibition of 52.2 ± 11.0% (n = 6; Fig. 4A). On the other hand, McN-A-343, a preferential muscarinic M₁-receptor agonist, inhibited TNS (2 Hz)-elicited neurogenic vasodilation with lower efficacy (maximal inhibition was 36.0 ± 12.5%; n = 6) and significantly higher EC₅₀ values (>10 µM) (Fig. 4A).

View larger version (18K): [in this window] [in a new window]   Fig. 4.   Effects of preferential muscarinic M1- and M2-receptor agonists on cerebral neurogenic vasodilation (A) and Ca2+ channel currents in the SPG neurons (B). A, effect of different concentrations of McN-A-343 and ABET on TNS (2 Hz)-elicited neurogenic vasodilation in porcine basilar arteries without endothelial cells. ABET is more potent than McN-A-343 in blocking relaxation induced by TNS. B, effects of different concentrations of ABET, McN-A-343, and oxotremorine (Oxo-M) in inhibiting Ca2+ channel current in the cultured SPG. The Ca2+ channel current was the peak current elicited by a depolarization step from 70 to 0 mV. All data represent means ± S.D.M. n represents number of cells varying from six to eight in each point.

Colocalization of Muscarinic M₂ Receptors and N-Type Ca²⁺ Channels in the Rat SPG Neurons. Results from double-labeling immunocytochemical study indicated that the cultured rat SPG neurons (1-3 weeks) were immunoreactive for both N-type Ca²⁺ channels and muscarinic M₂ receptor (Fig. 5, A and B). The N-type Ca²⁺ channel immunoreactivity was labeled with fluorescein isothiocyanate (A) and muscarinic M₂-receptor immunoreactivity with rhodamine (B). Both N-type Ca²⁺ channel- and muscarinic M₂-receptor immunoreactive fibers including soma and dendrites were completely coincident. In negative control, neurons were not immunoreactive to the antigen preabsorbed N-type Ca²⁺ channel antibodies or muscarinic M₂-receptor antibodies (data not shown).

View larger version (10K): [in this window] [in a new window]   Fig. 5.   Colocalization of M2 receptors and N-type Ca2+ channels in a cultured SPG neuron of the rat. A, a cultured SPG neuron (14th day) was immunoreactive to fluorescein isothiocyanate-conjugated N-type Ca2+ channel antibodies. The cell was photographed using the fluorescein isothiocyanate filter (BA515; Olympus). B, the cell in A was also immunoreactive to rhodamine-conjugated M2-receptor antibodies. The cell was photographed using the rhodamine filter (BA590; Olympus). The bar represents 25 µm.

Effect of CTX on Muscarinic Inhibition of Ca²⁺ Channel Currents in the Rat SPG Neurons. Previous pharmacology studies demonstrated that TNS-elicited neurogenic vasodilation in porcine basilar arteries was significantly inhibited by 0.1 µM -conotoxin, a selective N-type Ca²⁺ channel blocker. In the presence of the inhibition produced by CTX, the residual vasodilation was not further inhibited by ABET or enhanced by atropine (Liu and Lee, 1999). Consistent with this report, 1 µM -conotoxin, but not 0.1 mM nifedipine (a selective L-type Ca²⁺ channel blocker), largely inhibited Ca²⁺ currents in SPG neurons. In the presence of 1 µM -conotoxin, the residual Ca²⁺ current was not affected by 10 µM ACh or 10 µM ABET (Fig. 6, A and B, n = 4).

View larger version (18K): [in this window] [in a new window]   Fig. 6.   Effect of -conotoxin on muscarinic receptor-mediated inhibition of Ca2+ channel current in cultured SPG neurons. A, peak Ca2+ channel current amplitude () plotted as a function of time when 10 µM ABET and 10 µM ACh were applied in the presence of 1 µM -conotoxin in the bath. The peak Ca2+ channel currents were evoked every 15 s by a depolarizing step to 0 mV from holding potential 70 mV. B, a summary of peak amplitude of Ca2+ channel currents inhibited by 1 µM CTX. The difference between CTX and control group is significant (*, p < 0.01). There is no significant difference among CTX, CTX + ACh, and CTX + ABET groups (p > 0.05). Data represent means ± S.D. n indicates number of cells examined.

Effect of 8-Bromo-cAMP on Muscarinic Inhibition of Ca²⁺ Channel Currents in the Rat SPG Neurons. Because muscarinic M₂-receptor-mediated effects often are coupled to a decrease in intracellular cAMP, 8-bromo-cAMP, a membrane permeable cAMP analog, was used to examine the involvement of cAMP in the muscarinic inhibition of Ca²⁺ channels. In cultured rat SPG neurons, 1 to 10 mM 8-bromo-cAMP did not affect the peak Ca²⁺ currents (n = 3, data not shown), nor did it affect ACh-induced inhibition of Ca²⁺ currents, which, however, were reversed by atropine (n = 6, Fig. 7).

View larger version (16K): [in this window] [in a new window]   Fig. 7.   Effect of 8-bromo-cAMP on muscarinic receptor-mediated inhibition of Ca2+ channel current in cultured SPG neurons. Peak Ca2+ channel current amplitude () plotted as a function of time when 1 µM ACh, 1 to 10 mM 8-bromo-cAMP, and 10 µM atropine were successively applied to the bath as indicated. The peak Ca2+ channel currents were evoked every 15 s by a depolarizing step to 0 mV from holding potential 70 mV.

Effect of Pertussis Toxin (PTX) on Muscarinic Inhibition of Ca²⁺ Channel Currents in the Rat SPG Neurons. Because muscarinic M₂ receptors are coupled to G_i/o protein, PTX, which uncouples G_i/o protein from receptors by ADP ribosylation (West et al., 1985), was used to examine the involvement of G proteins in inhibition of Ca²⁺ currents in the SPG neurons. PTX (1 µg/ml) was added to the cell culture dish 24 to 36 h before patch-clamp recording. High concentration (10 µM) of ACh was used to produce the maximal inhibition of Ca²⁺ currents. In control SPG neurons (n = 8), the maximal inhibition produced by ACh was 64.5 ± 5.8%. In PTX-pretreated SPG neurons (n = 9), the maximal inhibition however was 19.4 ± 16.4% (n = 9). The difference of the maximal inhibition of Ca²⁺ currents between the PTX-treated and untreated neuron was significant (p < 0.01; Fig. 8).

View larger version (26K): [in this window] [in a new window]   Fig. 8.   Summary of the maximal inhibition of Ca2+ channel current in the SPG neurons in the absence (control) and presence of PTX-pretreatment. PTX (1 µg/ml) was added to cultured SPG neurons 24 to 36 h before recording. Ca2+ channel currents were elicited by depolarizing steps from 70 mV to 0 mV. Maximal inhibition of peak Ca2+ channel currents was produced by 10 µM ACh. Data represent means ± S.D. Number of tested cells in each group is indicated in parentheses. *, p < 0.01 indicates significant difference from control.

Effect of GTPS on Muscarinic Inhibition of Ca²⁺ Channel Currents in the Rat SPG Neurons. To further investigate the involvement of G protein in M-receptor-mediated inhibition of Ca²⁺ currents in the SPG neurons, GTPS was used to replace GTP in the intracellular recording solution in some studies. GTPS, a nonhydrolyzable GTP analog, binds tightly to the -subunit of G proteins and confers a receptor-independent, irreversible G protein action. When GTPS was present in the intracellular solution, the Ca²⁺ currents in all recorded SPG neurons were rapidly depressed without adding muscarinic agonist. After 10 min of dialysis, the residual current was 28.3 ± 5.5% of the original current, and no further inhibition induced by ACh was observed in the majority of the neurons examined (seven of nine) (Fig. 9, A and B). The inhibition of Ca²⁺ currents produced by 10 µM ACh in two neurons was small and was not reversed by atropine (data not shown).

View larger version (21K): [in this window] [in a new window]   Fig. 9.   Effects of GTPS on muscarinic receptor-mediated inhibition of Ca2+ channel current in cultured SPG neurons. Peak Ca2+ channel current amplitude () plotted as a function of time when ACh was applied to the bath after 10 min of internal dialysis of GTPS (A) as indicated. GTPS (0.5 mM) was used instead of GTP in the intracellular solution. The peak Ca2+ channel currents were evoked every 30 s by a depolarizing step to 0 mV from holding potential 70 mV. ACh was added when no more significant decrease of current was seen (usually around 10 min after whole-cell formation). B, summary of the inhibition of peak Ca2+ channel currents by 10 µM ACh in the absence and presence of GTPS in the intracellular solution. The currents were recorded 10 min after GTPS dialysis. Data represent means ± S.D. n indicates number of tested cells.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Atropine has been shown to increase NO-mediated cerebral neurogenic vasodilation in several species including the pig (Liu and Lee, 1999), cow (Ayajiki et al., 1993), and monkey (Toda et al., 1997). This effect of atropine has been suggested to be caused by blockade of presynaptic muscarinic receptors, thereby blocking the inhibitory effect of endogenously released ACh on NO release (Ayajiki et al., 1993; Liu and Lee, 1999) from perivascular cholinergic-nitrergic nerves (Kimura et al., 1997; Yu et al., 1998). Our results found in porcine cerebral arteries further suggested that the inhibitory effect of ACh was mediated by presynaptic M₂ receptors, which were coupled to suppression of N-type Ca²⁺ channels (Liu and Lee, 1999). In the present study, we further demonstrated that ACh and ABET (a muscarinic M₂-receptor agonist) inhibited NO-mediated neurogenic vasodilation in porcine cerebral arterial rings and provided new evidence that these two muscarinic agonists inhibited N-type Ca²⁺ currents in the rat SPG neurons, which are partial sources of cerebrovascular nitrergic nerves. These similar results found in the pig and rat support our hypothesis that ACh inhibits neurogenic vasodilation by suppressing Ca²⁺ influx through N-type Ca²⁺ channels on nitrergic nerves, thereby causing decreased formation and release of NO and diminished vasodilation (Fig 10).

View larger version (19K): [in this window] [in a new window]   Fig. 10.   A diagram showing a cholinergic-nitrergic nerve terminal innervating large cerebral arteries at the base of the rat and pig brain. NO, which is not stored in vesicles, is Colocalized and co-released with ACh (), which is stored in the vesicles. NO is synthesized from L-arginine (L-Arg) in the presence of nitric-oxide synthase (NOS). L-citrulline (L-Cit), the byproduct of NO synthesis, is converted to argininosuccinate (AS) in the presence of argininosuccinate synthase (ASS) and is further converted to L-Arg by argininosuccinate lyase (ASL) (Yu et al., 1997). This L-Cit-L-Arg cycle for continuing production of NO (Chen and Lee, 1995) provides direct evidence for the neuronal source of NO. The neuronal NO plays a major role in cerebral neurogenic vasodilation, which is mediated by activation of guanylate cyclase (GC) and cGMP synthesis in the smooth muscle cell (SMC). ACh released from cholinergic-nitrergic nerves acts on presynaptic muscarinic M2 receptors (M2R) resulting in Gi-protein-mediated negative coupling () to the N-type Ca2+ channels (N-Ca2+ch). This leads to suppression of Ca2+ influx through this type of Ca2+ channels and decreased NOS activity accompanied by a decrease in NO formation and release and diminished relaxation of the smooth muscle cell. The endogenous ACh is not a postsynaptic transmitter because it does not directly affect the SMC tone (Lee, 1982). +, indicates activation.

Muscarinic Inhibition of Neurogenic Vasodilation in Cerebral Arteries. Electrical stimulations of perivascular nerves exclusively cause vasodilation in the large cerebral arteries at the base of the brain of several species including the pig (Lee, 1994). The vasodilation is mediated mainly by NO (Lee, 1994; Zhang et al., 1998). In our previous and the present studies, the increase in neurogenic vasodilation by atropine was completely abolished by L-NNA, an NOS inhibitor, and fully recovered by further application of L-arginine, the precursor of NO synthesis. Because atropine did not affect the vasodilation produced by sodium nitroprusside, an NO donor (data not shown), the enhancement of neurogenic vasodilation induced by atropine was most likely attributable to its presynaptic inhibition of muscarinic receptors resulting in increased NO release. That is, atropine antagonizes the tonic inhibition by endogenous ACh of NO release (Ayajiki et al., 1993; Liu and Lee, 1999) because both ACh and NO are co-released in the perivascular cholinergic-nitrergic nerves (Kimura et al., 1997; Yu et al., 1998). Further pharmacological and immunocytochemical studies have shown that the muscarinic M₂ receptor is the specific subtype located on the nitrergic nerves mediating the inhibition of neurogenic vasodilation induced by ACh in porcine cerebral arteries (Liu and Lee, 1999). This is further supported by results of the present study.

Muscarinic M₂ Receptors Mediate Inhibition of N-Type Ca²⁺ Channels in Cultured Rat SPG Neurons. Due to difficulties of using patch clamp on the perivascular nerves, the effects of muscarinic receptor activation or inhibition on voltage-dependent Ca²⁺ channels were investigated in cultured SPG neurons. The SPG is one of the major sources of cerebral perivascular nitrergic nerves in the rat, cat, and pig (Suzuki and Hardebo, 1993; Kimura et al., 1997; Yu et al., 1998). SPG neurons were obtained from adult rats instead of pigs because of inconsistency in porcine SPG neuronal culture. Cultured rat SPG neurons have been found to contain neurotransmitters and N-type Ca²⁺ channels like those found in cerebral perivascular nitrergic nerves in whole-mount preparations from the pig (Liu et al., 2000).

Possible Mechanisms of Muscarinic Inhibition of Ca²⁺ Channel Currents in the Rat SPG Neurons. Although activation of muscarinic M₂ receptor is coupled to an inhibition of adenylate cyclase (Ui, 1984; Peralta et al., 1988), a decrease in intracellular cAMP level was not a major factor causing inhibition of cerebral neurogenic vasodilation (Liu and Lee, 1999). This was supported by the present findings that 8-bromo-cAMP did not affect calcium currents, suggesting that cAMP was not the mediator for the muscarinic inhibition of Ca²⁺ channel current in the SPG neurons. This result is consistent with reports by others that intracellular cAMP is not involved in muscarinic receptor-mediated inhibition of Ca²⁺ channels (Anwyl, 1991; Dolphin et al., 1991; Allen and Brown, 1993; Jeong and Wurster, 1997), although an unidentified second messenger has been reported to partially mediate the muscarinic inhibition of Ca²⁺ channels in the superior cervical ganglionic neurons (Bernheim et al., 1991; Hille, 1992).