VPAC receptor modulation of neuroexcitability in intracardiac neurons: dependence on intracellular calcium mobilization and synergistic enhancement by PAC1 receptor activation.

Pituitary adenylate cyclase-activating polypeptide (PACAP) and vasoactive intestinal polypeptide (VIP) have been found within mammalian intracardiac ganglia, but the cellular effects of these neuropeptides remain poorly understood. Fluorometric calcium imaging and whole cell patch clamp recordings were used to examine the effects of PACAP and VIP on [Ca2+]i and neuroexcitability, respectively, in intracardiac neurons of neonatal rats. PACAP and VIP evoked rapid increases in [Ca2+]i that exhibited both transient and sustained components. Pharmacological experiments using PAC1 and VPAC receptor-selective antagonists demonstrated that the elevations in [Ca2+]i result from the activation of VPAC receptors. The transient increases in [Ca2+]i were shown to be the product of Ca2+ mobilization from caffeine/ryanodine-sensitive intracellular stores and were not due to inositol 1,4,5-trisphosphate-mediated calcium release. In contrast, the sustained [Ca2+]i elevations were dependent on extracellular Ca2+ and were blocked by the transient receptor channel antagonist, 2-aminoethoxydiphenyl borate, which suggests that they are due to Ca2+ entry via store-operated channels. In addition to elevating [Ca2+]i, both PACAP and VIP depolarized intracardiac neurons, and PACAP was further shown to augment action potential firing in these cells. Depolarization of intracardiac neurons by the neuropeptides was dependent on activation of VPAC receptors and the concomitant increases in [Ca2+]i. Although activation of PAC1 receptors alone had no direct effects on neuroexcitability, PAC1 receptor stimulation potentiated the VPAC receptor-induced depolarizations. Furthermore, enhanced action potential firing was only observed upon concurrent stimulation of PAC1 and VPAC receptors, which indicates that these receptors act synergistically to enhance neuroexcitability in intracardiac neurons.

ropeptides that belong to the glucagon/secretin/growth hormone-releasing factor family of peptides (1)(2)(3)(4). These neuropeptides have pronounced effects on the central nervous system as well as on neurons and effector targets of the autonomic nervous system. For example, in the central nervous system, PACAP is involved in hippocampal synaptic plasticity and associative learning in mice (5), whereas VIP has been shown to regulate intrathalamic rhythm activity and may modulate information transfer through the thalamocortical circuit (6). In the peripheral nervous system, PACAP and VIP appear to play a major role in autonomic regulation of the cardiovascular system. VIP has been shown to cause positive chronotropic and inotropic effects (7)(8)(9). In contrast, PACAP has been shown to cause a transient increase in sinus rate and atrial contractility that is followed by negative chronotropic and inotropic responses in isolated canine atria (10). Also, both PACAP and VIP are potent dilators of coronary arteries (11,12). Whereas these neuropeptides can directly influence cardiac tissues, the effects of PACAP and VIP on the heart are in part due to neuroregulatory effects on parasympathetic intracardiac ganglia (9,10,13,14).
Mammalian intracardiac ganglia play a major role in neuronal control of the heart and are believed to be capable of independently monitoring and influencing cardiac function. A variety of neurotransmitters and neuromodulators, including PACAP and VIP, appear to be involved in the relaying of information within intracardiac ganglia and onto effector targets such as cardiac valves, coronary arteries, and cardiomyocytes. PACAP and VIP have been detected by immunocytochemistry in nerve fibers in the heart, coronary vasculature, and cardiac parasympathetic ganglia (15)(16)(17)(18)(19)(20). Although extrinsic fibers may be one source of these neuropeptides in the heart (21), PACAP and VIP are also found in cell bodies and nerve fibers of intrinsic cardiac neurons (15)(16)(17)(18)(19)(20). Despite the fact that data clearly demonstrate that these endogenous neuropeptides are potent regulators of the cardiovascular system, the cellular mechanisms mediating their effects remain poorly understood. There are also conflicting reports in the literature concerning the mechanisms by which PACAP and VIP regulate the function of intrinsic cardiac neurons. For example, it has been suggested that PACAP and VIP modulate neurotransmission in rat cardiac ganglia by potentiating nicotinic acetylcholine receptors (22,23), whereas studies on guinea pig intracardiac neurons indicate that PACAP, but not VIP, regulates the excitability of intrinsic cardiac neurons, possibly by influencing the K ϩ conductance, I A (19).
The functional consequences of the PAC 1 /VPAC receptor heterogeneity, which is also observed in other neuron types, such as mammalian cortical neurons, remain to be fully elucidated. The present study examined the ability of PACAP and VIP to regulate the electrical properties of intrinsic cardiac neurons from neonatal rats and to modulate [Ca 2ϩ ] i in these cells. Both PACAP and VIP depolarize intracardiac neurons and increase [Ca 2ϩ ] i by evoking Ca 2ϩ release from ryanodine-sensitive intracellular stores and promoting Ca 2ϩ entry through the plasma membrane. PACAP, but not VIP, also increased the number of action potentials fired by neurons in response to depolarizing current pulses. The changes in neuroexcitability evoked by PACAP and VIP were blocked by inhibiting the [Ca 2ϩ ] i elevations produced by the neuropeptides.

EXPERIMENTAL PROCEDURES
Membrane currents and [Ca 2ϩ ] i were investigated in isolated intracardiac ganglion neurons of neonatal rats (4 -7 days old). The preparation and culture of isolated intrinsic cardiac neurons have been previously described (38). Dissociated neurons were plated onto poly-L-lysine-coated glass coverslips and incubated at 37°C for 36 -72 h under a 95% air, 5% CO 2 environment prior to the experiments.
Microfluorometric Measurements-The calcium-sensitive dye fura-2/AM was used for measuring intracellular free calcium concentrations in intracardiac neurons, as previously described (39). Cells plated on coverslips were incubated for 1 h at room temperature in physiological saline solution (PSS) consisting of 140 mM NaCl, 3 mM KCl, 2.5 mM CaCl 2 , 1.2 mM MgCl 2 , 7.7 mM glucose, and 10 mM HEPES (pH to 7.2 with NaOH), which also contained 1 M fura-2/AM and 0.1% Me 2 SO. The coverslips were then washed in PSS (fura-2/AM-free) prior to the experiments being carried out. All solutions were applied via a rapid application system identical to that previously described (40).
A DG-4 high speed wavelength switcher (Sutter Instruments Co., Novato, CA) was used to apply alternating excitation with 340-and 380-nm UV light. Fluorescent emission at 510 nm was captured using a Sensicam digital CCD camera (Cooke Corp., Auburn Hills, MI) and recorded with Slidebook Version 3.0 software (Intelligent Imaging Innovations, Denver, CO) on a Pentium IV computer. Changes in [Ca 2ϩ ] i were calculated using the Slidebook 3 software (Intelligent Imaging Innovations, Denver, CO) from the intensity of the emitted fluorescence following excitation with 340-and 380-nm light, respectively, using the equation, where R represents the fluorescence intensity ratio (F 340 /F 380 ) as determined during experiments, Q is the ratio of F min to F max at 380 nm, and K d is the Ca 2ϩ dissociation constant for fura-2. Calibration of the system was performed using a fura-2 calcium imaging calibration kit (Molecular Probes, Inc., Eugene, OR) and values were determined to be as follows: F min /F max ϭ 23.04; R min ϭ 0.31; R max ϭ 8.87. Electrophysiology-Coverslips containing the dissociated neurons were transferred to a 0.5-ml recording chamber mounted on a phasecontrast microscope (400ϫ). The extracellular recording solution was identical to the Ca 2ϩ imaging PSS. The Ca 2ϩ -free PSS contained no CaCl 2 and included 1 mM EGTA. All drugs were bath-applied in PSS. In some electrophysiological and Ca 2ϩ imaging experiments, the order of drug administration was reversed, whereby PACAP was given in the presence of other drugs (or in Ca 2ϩ -free PSS) prior to being given alone. This alteration in the protocol was done as a control to ensure that any change in response was not due to rundown. Currents were amplified and filtered (5 kHz) using an Axoclamp-200B amplifier, digitized with a 1322A DigiData digitizer (20 kHz), and collected on a Pentium IV computer using the Clampex 8 program (Axon Instruments, Inc., Foster City, CA). Data analysis was conducted using the pClamp 8 program, Clampfit.
The whole cell perforated patch variation of the patch clamp recording technique was used as previously described (41,42). This configuration preserves intracellular integrity, preventing the loss of cytoplasmic components and subsequent alteration of functional responses of these neurons (22,43). The pipette solution contained 75 mM K 2 SO 4 , 55 mM KCl, 5 mM MgSO 4 , 10 mM HEPES, 198 g/ml amphotericin B, and 0.4% Me 2 SO.
Membrane potential responses to hyperpolarizing and depolarizing current pulses (Ϫ100 to ϩ200 pA) were determined in the absence and presence of VIP/PACAP receptor ligands. The mean resting membrane potential, latency for action potential firing, action potential duration, amplitude of afterhyperpolarization, and the number of action potentials elicited by depolarizing current pulses were determined in the absence and presence of PAC 1 and VPAC receptor ligands. Membrane input resistance was also determined as described previously (42).

PACAP and VIP Directly Increase Free [Ca 2ϩ
] i -The most abundant PAC 1 receptor isoform expressed in neonatal rat intrinsic cardiac neurons, PAC 1 -HOP, has been shown to elicit elevations in intracellular calcium in cortical neurons (44). However, the relationship between PACAP and intracellular calcium has not been investigated in the autonomic neurons. The effect of PACAP on Ca 2ϩ handling in intracardiac neurons was studied using fura-2 Ca 2ϩ -imaging techniques. Fig. 1A shows a representative trace of change in [Ca 2ϩ ] i (⌬[Ca 2ϩ ] i ) as a function of time recorded from a neuron before, during and following a 2-min application of 100 nM PACAP-27. Intracellular Ca 2ϩ concentrations increased by Ͼ400 nM when PACAP-27 was applied and returned to near control levels upon washout of drug. In similar experiments, PACAP-27 significantly increased intracellular calcium by ϳ400%, from a baseline of 65.3 Ϯ 5.5 nM to a peak value of 335.1 Ϯ 65.4 nM (n ϭ 23) (Fig.  1B). In all cells responding to PACAP, the elevations in [Ca 2ϩ ] i were reversed when PACAP was removed from the bath, and subsequent applications of PACAP to the same cell produced [Ca 2ϩ ] i increases of similar magnitude.
In a series of experiments, the PAC 1 receptor-selective agonist, maxadilan, and VPAC receptor-selective agonist, VIP, were used to determine the PACAP/VIP receptor subtype(s) mediating the PACAP-induced elevations in [Ca 2ϩ ] i. Fig. 1A shows representative traces of change in [Ca 2ϩ ] i recorded from two neurons as a function of time before, during and following a 2-min application of maxadilan and VIP, respectively. Maxadilan, at a concentration (10 nM) specific for PAC 1 Fig. 2D and reveals that application of either L-8-K (n ϭ 9) or YF-GRF (n ϭ 5) significantly attenuated PACAPevoked elevations in [Ca 2ϩ ] i in isolated intracardiac neurons, whereas M65 (n ϭ 5) had no effects on the neuropeptideinduced changes in [Ca 2ϩ ] i . These data further support the conclusion that the PACAP and VIP evoked increases in [Ca 2ϩ ] i are mediated by VPAC receptors and not PAC 1 receptors in these cells.
PACAP and VIP are known to modulate nicotinic acetylcholine receptors (nAChRs) in intrinsic cardiac neurons (23). This raises the possibility that the observed PACAP-and VIP-induced elevations in [Ca 2ϩ ] i may be due to regulation of presynaptic or postsynaptic nAChRs by the neuropeptides. Thus, the Na ϩ channel blocker, TTX (400 nM), and ganglionic nAChR blocker, mecamylamine (25 M), were used to block nicotinic neurotransmission pre-and postsynaptically, respectively, to verify that the effects of exogenously applied PACAP and VIP on PACAP and VIP Mobilize Ca 2ϩ from Intracellular Stores and Evoke Ca 2ϩ Entry through the Plasma Membrane-PAC 1 and VPAC receptors have been linked to both Ca 2ϩ entry through the plasma membrane via the activation of L-type calcium channels (36,46) and Ca 2ϩ release from intracellular stores (47). Experiments were conducted to determine the source of Ca 2ϩ responsible for the VPAC-mediated increase in [Ca 2ϩ ] i . Application of PACAP-27 produced an increase in [Ca 2ϩ ] i that exhibited both a transient and a sustained component (Fig. 3A). However, when PACAP was applied in Ca 2ϩ -free PSS, the initial transient increase in [Ca 2ϩ ] i was observed, but the sustained component was abolished (Fig. 3A) . Similar results were observed when VIP was used as the agonist (data not shown). Fig. 3B shows a bar graph of mean peak and sustained [Ca 2ϩ ] i recorded under normal and Ca 2ϩ -free conditions from 12 isolated intracardiac neurons. Although the magnitude of the transient peak [Ca 2ϩ ] i remained unchanged (normal PSS: 179.2 Ϯ 17.6 nM; Ca 2ϩ -free PSS: 173.4 Ϯ 17.8 nM), the sustained component significantly decreased from 103.6 Ϯ 11.5 to 45.7 Ϯ 6.7 nM following the removal of extracellular Ca 2ϩ . This latter [Ca 2ϩ ] i value was equivalent to control baseline [Ca 2ϩ ] i . Thus, the sustained component is dependent on extracellular Ca 2ϩ and is probably due to Ca 2ϩ flux through the plasmalemma.
The presence of a PACAP-induced transient elevation in [Ca 2ϩ ] i in the absence of extracellular calcium suggested that activation of VPAC receptors mobilized Ca 2ϩ from intracellular stores in these neurons. In order to test this hypothesis, we used the endoplasmic reticulum Ca 2ϩ -ATPase inhibitor, thapsigargin, which specifically blocks Ca 2ϩ reuptake, leading to depletion of the endoplasmic reticulum Ca 2ϩ stores (48). Prior to the experiments, dissociated intracardiac neurons were incubated for 1 h in 20 M thapsigargin (in PSS, 37°C). Experiments were then conducted to determine whether PACAP increased [Ca 2ϩ ] i under these conditions. Untreated neurons collected from either the same neonatal rat or animals from the same litter were used as positive controls. Fig. 3C shows traces . In contrast, thapsigargin preincubation had no effect on the peak [Ca 2ϩ ] i elevations produced by rapid focal application of 500 M ACh (Fig. 3D). Thapsigargin also eliminated the rise in [Ca 2ϩ ] i elicited by bath application of 5 mM caffeine (Fig. 3D), which has been shown to evoke release of Ca 2ϩ from intracellular stores in these cells (49). Therefore, depleting intracellular Ca 2ϩ stores via thapsigargin pretreatment abolished both the transient and sustained PACAP-induced changes in [Ca 2ϩ ] i .
PACAP Induces Mobilization of [Ca 2ϩ ] i from Caffeine-and Ryanodine-sensitive Stores-Previous studies have demonstrated that rat intrinsic cardiac neurons have two distinct endoplasmic reticulum Ca 2ϩ stores, one that is sensitive to inositol 1,4,5-trisphosphate (IP 3 ) and a second that is sensitive to ryanodine and caffeine (39,49). The PACAP/VIP receptors have been shown to mobilize Ca 2ϩ from both IP 3 -and ryanodine/caffeine-sensitive stores in other cell types (44,46). To determine whether PACAP evokes release of Ca 2ϩ from ryan-odine/caffeine-sensitive stores we used two approaches: 1) depleting the ryanodine/caffeine-sensitive stores of Ca 2ϩ by bathapplying 5 mM caffeine and 2) blocking the release of Ca 2ϩ from these stores by bath-applying 10 M ryanodine.  (Fig. 4B). Similarly, ryanodine depressed the peak elevation in [Ca 2ϩ ] i from 162.2 Ϯ 27.0 to 32.7 Ϯ 10.9 nM (Fig. 4D). The initial application of caffeine produced an increase in [Ca 2ϩ ] i (236.8 Ϯ 34.1 nM) similar to that elicited by PACAP but, unlike PACAP, was often associated with Ca 2ϩ oscillations (see Fig. 4A). Ryanodine, however, did not promote calcium release in any of the cells tested, which is consistent with the drug acting as an inhibitor of the ryanodine receptor at this concentration (50).
Results from our experiments with ryanodine, however, cannot rule out the possibility that neuropeptide-stimulated IP 3 production evokes small, local changes in [Ca 2ϩ ] i that initiate the release of Ca 2ϩ from the caffeine/ryanodine stores. The IP 3 receptor antagonist 2-APB (51) was used to determine whether IP 3 -mediated Ca 2ϩ -induced Ca 2ϩ release is responsible for the transient elevation in [Ca 2ϩ ] i observed following VPAC recep-  Fig. 5C), which is primarily due to release of Ca 2ϩ from IP 3 -sensitive stores in these cells (49). Fig. 5D shows a bar graph of the mean peak increase in [Ca 2ϩ ] i evoked by PACAP, caffeine, and muscarine in the absence and presence of 2-APB. Whereas 2-APB significantly depressed the peak increases in [Ca 2ϩ ] i evoked by muscarine from 107 Ϯ 25 to 31 Ϯ 8 nM, it had no effect on those elicited by PACAP or caffeine.
Whereas the transient component of the PACAP-induced elevation in [Ca 2ϩ ] i was unaffected by 2-APB, the sustained component was depressed in the presence of the drug (Fig. 5A). In six identical experiments, the PACAP-induced sustained increase in [Ca 2ϩ ] i decreased from 66 Ϯ 7 to 30 Ϯ 6 nM (Fig.  5E). Similarly, the sustained elevations in [Ca 2ϩ ] i observed following application of caffeine or muscarine were also attenuated by 2-ABP (Fig. 5E). At the concentration used here (50 M), 2-APB also directly blocks several subtypes of transient receptor potential ion channels (52). Therefore, these data suggest that capacitive Ca 2ϩ entry through transient receptor potential channels contributes to the sustained elevation in [Ca 2ϩ ] i evoked by VPAC receptor activation.

PACAP-and VIP-induced Excitability in Rat Intracardiac
Neurons-Earlier studies in intracardiac neurons from adult guinea pigs suggest that PACAP, but not VIP, modulates neuroexcitability in these cells (19). However, no such changes have been reported in neonatal rat intracardiac neurons (22,23). Data suggest that intracardiac neurons of these two species express different PACAP and VIP receptor subtypes (19,37), and thus the neuropeptides may have distinct cellular effects. For example, in contrast to observations reported here, application of PACAP has not been shown to affect [Ca 2ϩ ] i in guinea pig intracardiac neurons. Thus, it seemed prudent to further investigate the effect of PACAP and VIP on the electrical properties of rat intrinsic cardiac neurons. Fig. 6A shows a family of representative voltage responses elicited by 500-ms, 150-pA depolarizing current pulses from an isolated intracardiac neuron held under current clamp mode using the whole cell perforated patch method. Membrane responses were recorded in the absence (Control) and presence of bath applied PACAP-27 (100 nM), and following washout of the neuropeptide (Wash). Application of PACAP depolarized the neuron and increased the number of action potentials elicited by the current pulse. The PACAP evoked membrane depolarization was observed in a majority of the cells tested (54 of 73 cells) (Fig.   FIG. 3. The transient, but not  In most neurons tested, the depolarizing current pulses evoked short, adapting trains of action potentials (46 of 73 neurons), whereas 15 neurons fired single action potentials and 12 neurons exhibited tonic firing when challenged with a similar current pulses. In a population of neurons (42 of 73), the number of action potentials elicited was increased by 115%, from 3.2 Ϯ 0.3 to 6.9 Ϯ 0.5, by 100 nM PACAP-27 (Fig. 6C). Following a 15-min washout of the neuropeptide, the number of action potentials induced by the current pulses returned to near control levels (3.9 Ϯ 0.5). The increase in action potentials observed in the presence of PACAP was statistically significant when compared with either control or washout. The population of neurons exhibiting augmented action potential firing included phasic, adapting, and tonic neurons. Thus, the effect of PACAP was not associated with a cell type with distinct active membrane properties. Examining the action potential characteristics of neurons that showed increased firing revealed that PACAP significantly decreased the action potential latency period from 6.0 Ϯ 0.7 to 4.7 Ϯ 0.5 ms (n ϭ 7; p Ͻ 0.01) and increased the afterhyperpolarization from Ϫ13.3 Ϯ 6.1 to Ϫ21.4 Ϯ 6 mV (n ϭ 7; p Ͻ 0.05). However, other parameters including action potential duration and overshoot were not altered (data not shown). It is important to note that the membrane depolarization and increase in action potential firing were not correlated, and only 30 of 73 cells showed both a depolarization and increased action potential firing. This observation suggests that two distinct mechanisms are mediating these membrane responses.
The effects of PACAP were in part mimicked by the related neuropeptide, VIP (Fig. 7). Bath application of VIP depolarized a population of neurons (8 of 13) from Ϫ47.3 Ϯ 1.7 to Ϫ44.5 Ϯ 1.5 mV (Fig. 7B). This depolarization was of lesser magnitude than the PACAP-induced depolarization but was statistically significant and reversed upon washout of the peptide (Ϫ45.9 Ϯ 1.6). VIP affected a population of neurons similar to PACAP (ϳ60%). However, unlike PACAP, VIP had no effects on action potential firing in these cells (Fig. 7C). The number of action potentials fired in response to 150-pA depolarizing current pulses before, during, and after washout of 100 nM VIP were 2.8 Ϯ 0.5, 3.2 Ϯ 0.6, and 3.4 Ϯ 0.6, respectively.
Membrane input resistance was investigated to determine whether the neuropeptide-induced changes in excitability were associated with opening and/or closing of ion channel(s). Current pulses (Ϫ100 pA) were applied to elicit hyperpolarizations before, during, and after PACAP or VIP (100 nM) application. The amplitude of the sustained hyperpolarization was used to determine input resistances. Neither PACAP nor VIP significantly altered the input resistance. This observation suggests that the neuropeptides are probably exerting their effects by simultaneously opening and closing multiple membrane chan- nels, resulting in no net change in input resistance. Likewise, there was no correlation between changes in input resistance and PACAP-induced modulation of neuroexcitability in guinea pig intrinsic cardiac neurons (19).  (Fig. 8B). Fig. 8C shows a bar graph of the number of action potentials elicited by identical depolarizing current pulses in cells exposed to PACAP prior to or following incubation in [N-Ac-Tyr 1 ,D-Phe 2 ]GRF-(1-29). Whereas PACAP increased firing from 2.7 Ϯ 0.3 to 6.3 Ϯ 1.5 action potentials under control conditions, the neuropeptide had no effect on action potential firing following application of [N-Ac-Tyr 1 , D-Phe 2 ]GRF- . Although VIP in part mimicked the depolarizing effects of PACAP, the magnitude of the VIP-induced depolarization was only ϳ60% of that elicited by PACAP. Also, VIP had no effects on action potential firing in these cells. This observation suggests that PAC 1 receptors may also contribute to the PACAPinduced changes in the electrical properties of intracardiac neurons. To examine this hypothesis, the PAC 1 -selective antagonist, M65, was used in a series of experiments to determine whether inhibition of PAC 1 receptors alters the PACAP-evoked increase in excitability. Co-application of M65 (10 nM) depressed both the PACAP-induced depolarization and increase in action potential firing in neurons shown to respond to 100 nM PACAP-27 (Table I). Surprisingly, bath application of maxadilan (10 -100 nM) failed to produce depolarizations in these neurons (Table I). Taken together, these data suggest that stimulation of VPAC receptors increases neuroexcitability in intrinsic cardiac neurons and that this increase in neuroexcitability is enhanced by concurrent activation of PAC 1 receptors. association of increases in the amplitude of the [Ca 2ϩ ] i -dependent action potential afterhyperpolarization with augmented action potential firing in the cells. To determine the role of PACAP-induced Ca 2ϩ mobilization in intracardiac neuroexcitability, whole cell perforated patch experiments were conducted in current clamp mode under Ca 2ϩ -free extracellular conditions. In rat intracardiac neurons, removal of extracellular Ca 2ϩ over a short period of time (minutes) not only prevents Ca 2ϩ conductance through the plasma membrane but also results in depletion of Ca 2ϩ from internal stores. 2 Thus, PACAPinduced [Ca 2ϩ ] i mobilization is blocked by washing the intracardiac neurons with Ca 2ϩ -free PSS. Fig. 9A shows representative action potentials evoked from a neuron by depolarizing current pulses (300 ms, 150 pA) in the absence and presence of PACAP-27 (100 nM) under control conditions (2.5 mM Ca 2ϩ ) and in Ca 2ϩ -free conditions (0 Ca 2ϩ , 1 mM EGTA). In the presence of external Ca 2ϩ , PACAP depolarized the neuron from Ϫ51.5 to Ϫ47.8 mV and increased the number of action potentials fired in response to the current pulse. However, PACAP failed to produce either of these changes when extracellular Ca 2ϩ was removed. In similar experiments, Ca 2ϩ -free conditions inhibited the ability of PACAP to alter the resting membrane potential (Fig. 9B) and the action potential firing properties (Fig. 9C) of neurons shown to respond to the neuropeptide under control conditions. In normal extracellular Ca 2ϩ (2.5 mM), PACAP-27 (100 nM) depolarized the cells from Ϫ51.1 Ϯ 1.6 to Ϫ46.8 Ϯ 1.9 mV, whereas in Ca 2ϩ -free PSS the resting membrane potential was Ϫ50.9 Ϯ 1.0 and Ϫ51.9 Ϯ 1.8 mV in the absence and presence of PACAP, respectively (n ϭ 6). Similarly, in five neurons that exhibited increased action potential firing when PACAP was bath-applied in PSS containing normal Ca 2ϩ (2.5 mM), removal of extracellular Ca 2ϩ abolished the neuropeptide-mediated effects (Fig. 9C). Thus, PACAPevoked changes in excitability are dependent on the neuropeptide-elicited elevation in [Ca 2ϩ ] i .

Effects of Ca 2ϩ -free Extracellular Conditions on PACAP-induced
In order to confirm that in the absence of extracellular Ca 2ϩ intracardiac neurons were capable of firing multiple action potentials, 1 mM Ba 2ϩ was bath-applied in Ca 2ϩ -free PSS. Barium (1 mM) has previously been shown to block membrane K ϩ channels and to increase neuroexcitability in intracardiac neurons (42). Under these calcium-free conditions, barium significantly depolarized (p Ͻ 0.01) the intracardiac neurons from Ϫ51.1 Ϯ 1.9 to Ϫ43.8 Ϯ 1.5 mV and increased action potential firing (p Ͻ 0.01) from a mean of 1.8 Ϯ 0.3 to 5.8 Ϯ 0.5 action potentials (n ϭ 5). These effects were reversible upon washout of Ba 2ϩ . Thus, removal of extracellular Ca 2ϩ does not prevent intracardiac neurons from exhibiting increased neuroexcitability.
Effects of Ryanodine on PACAP-induced Intracardiac Neuroexcitability-Electrophysiological experiments conducted under Ca 2ϩ -free conditions suggested that PACAP-induced changes in excitability were at least in part due to the effect of the neuropeptide on [Ca 2ϩ ] i . Further experiments were conducted to determine if PACAP-induced mobilization of Ca 2ϩ from ryanodine-sensitive intracellular stores contributed to the increased neuroexcitability observed in the presence of the neuropeptide. Fig. 10A shows representative action potentials evoked in response to a depolarizing current pulse (300 ms, 150 pA) during bath application of 100 nM PACAP-27 before and after pretreatment with ryanodine (10 M). Whereas PACAP increased action potential firing in this cell under control conditions, preincubation in ryanodine eliminated the PACAPinduced changes in firing characteristics. Moreover, preincubation in ryanodine also prevented PACAP from depolarizing the 2 W. I. DeHaven and J. Cuevas, unpublished observation. cell, since PACAP depolarized the cell by 3.7 mV under control conditions but failed to alter the resting membrane potential in the presence of ryanodine. In similar experiments, preincubation in ryanodine blocked PACAP-induced depolarizations in cells that were shown to depolarize in the presence of the neuropeptide (Fig. 10B). Fig. 10C shows a bar graph of the number of action potentials elicited by identical depolarizing current pulses in cells exposed to PACAP prior to and following incubation in ryanodine. Although PACAP increased firing from 1.7 Ϯ 0.3 to 5.3 Ϯ 0.9 action potentials under control conditions, PACAP had no effects on action potential firing following application of ryanodine. This observation further establishes elevations in [Ca 2ϩ ] i as a obligatory component in the PACAP/VIP enhancement of excitability in these cells. DISCUSSION The major finding reported here is that activation of VPAC receptors increases neuroexcitability in intrinsic cardiac neurons and that concurrent activation of VPAC and PAC 1 receptors results in a synergistic augmentation of neuroexcitability. Furthermore, the PACAP-and VIP-mediated increase in excitability is dependent on VPAC receptor-induced Ca 2ϩ release from caffeine-and ryanodine-sensitive intracellular stores.
Our studies show that bath application of VIP or PACAP evokes an elevation in intracellular free Ca 2ϩ that is completely blocked by the VPAC receptor antagonists L-8-K and [N-Ac-Tyr 1 ,D-Phe 2 ]GRF- . Furthermore, the effects of PACAP and VIP were not mimicked by the PAC 1 -selective agonist maxadilan. Taken together, these data suggest that the effects of PACAP and VIP on [Ca 2ϩ ] i are mediated by activation of VPAC receptors in these cells. Previous studies have shown that rat intracardiac neurons express three PAC 1 receptor isoforms (PAC 1 -short, PAC 1 -HOP1, and PAC 1 -HOP2) as well as VPAC 1 and VPAC 2 receptors (37). The predominant PAC 1 receptor, PAC 1 -HOP, has been shown to evoke release of calcium from intracellular stores via both cAMP and IP 3 pathways in various cell types including neuroepithelial cells, chromaffin cells, astrocytes, and cortical neurons (44,53,54). However, our data indicate that stimulation of PAC 1 has no direct effect on [Ca 2ϩ ] i in neonatal rat intracardiac neurons. Earlier studies using single-cell reverse transcription-PCR detected VPAC 1 receptor transcripts in Ͻ20% of neonatal rat intracardiac neurons, whereas VPAC 2 receptor transcripts were expressed in Ͼ90% of the cells (37). In light of the fact that PACAP increased [Ca 2ϩ ] i in over Ͼ95% of the cells tested, VPAC 2 receptors are likely to be mediating these increases.
To date, there have been no reports of VPAC receptors evoking elevations of [Ca 2ϩ ] i in neurons, and there is limited information concerning VPAC receptor-mediated regulation of [Ca 2ϩ ] i in other cell types. Stimulation of VPAC receptors with low concentrations of VIP (1 nM) has been shown to evoke calcium oscillations in pancreatic acinar cells (55). In contrast to our observations, however, higher concentrations of VIP (100 nM) failed to elicit elevations in [Ca 2ϩ ] i and had an inhibitory effect on ACh-evoked [Ca 2ϩ ] i responses in acinar cells (55). VPAC-mediated increases in [Ca 2ϩ ] i have also been reported in the lymphoblastic MOLT4 and SUPT1 cells (56,57). Exogenously expressed VPAC 1 and VPAC 2 receptors have also been shown to stimulate [Ca 2ϩ ] i in COS7 cells and Chinese hamster ovary cells (47,58).
The sient and a sustained component. The transient component was observed in the absence of extracellular Ca 2ϩ but was blocked by ryanodine (10 M) or depletion of the Ca 2ϩ stores by caffeine or thapsigargin. These results suggest that the transient component is due to VPAC mobilization of intracellular Ca 2ϩ and, more specifically, due to Ca 2ϩ release from ryanodine-sensitive Ca 2ϩ stores. In pancreatic and submandibular gland cells, VPAC receptor activation promotes Ca 2ϩ oscillations via a phospholipase C/IP 3 cascade (59). Thus, in those secretory cells, VIP is probably increasing [Ca 2ϩ ] i by evoking Ca 2ϩ release from IP 3 -sensitive Ca 2ϩ stores. Although rat intracardiac neurons also contain IP 3 -sensitive stores (49), our results with ryanodine and 2-APB suggest that a phospholipase C-IP 3 -IP 3 -receptor pathway is not involved in the VPAC receptor-induced increases in [Ca 2ϩ ] i seen in these cells.
Unlike the transient component, the VPAC-mediated sustained elevations in [Ca 2ϩ ] i were abolished by removal of extracellular Ca 2ϩ , suggesting that they are due to calcium entry through the plasma membrane. In various cell types, including ␣T3-1 gonadotrophs (60) and HIT-T15 insulinoma cells (61)   voltage-activated calcium channels in intracardiac neurons (22). Furthermore, VPAC-mediated modulation of voltagegated Ca 2ϩ channels appears to involve a cAMP/protein kinase A-dependent mechanism (60, 61), but in intrinsic cardiac neurons the activator of adenylate cyclase, forskolin (100 M), failed to mimic the effects of PACAP and VIP on [Ca 2ϩ ] i (data not shown). Moreover, PACAP and VIP depolarize intracardiac neurons to a membrane potential of approximately Ϫ45 mV, which is negative to the threshold for activation of voltagegated Ca 2ϩ channels in these cells (42). Thus, the influx of Ca 2ϩ observed following VPAC receptor stimulation is unlikely to be mediated by Ca 2ϩ entry through such Ca 2ϩ channels. Given that the sustained component is only observed following Ca 2ϩ release from the ryanodine-sensitive stores and that it may also be produced by depletion of these stores by caffeine, it is probably mediated by store-operated channels. This conclusion is supported by our observation that the sustained elevation in [Ca 2ϩ ] i produced by PACAP is blocked by the transient receptor potential channel antagonist 2-APB. Depletion of caffeine/ ryanodine-sensitive Ca 2ϩ stores has previously been shown to evoke capacitive Ca 2ϩ entry in both excitable and nonexcitable cells (62,63).
We have now shown that, in addition to increasing [Ca 2ϩ ] i , VIP and PACAP enhance neuronal excitability in intrinsic cardiac neurons. However, the effects on excitability produced by these neuropeptides are not identical. These differences may shed light onto possible distinct roles for the two neuropeptides in the regulation of intracardiac neurons and thus of the car-diovascular system. Bath application of PACAP or VIP results in a depolarization of intrinsic cardiac neurons. Whereas PACAP has previously been shown to depolarize guinea pig intracardiac neurons, VIP had no effect on the resting membrane potential of those cells (19). The observation that VIP depolarizes intracardiac neurons and that VPAC-specific antagonists block PACAP-induced depolarizations suggests that VPAC receptors directly influence neuroexcitability in rat intracardiac neurons. VPAC receptor activation has been shown to depolarize thalamic relay and medial pontine reticular formation neurons by 2-3 mV (6, 64), which is similar to the VIP-induced depolarizations reported here (2.4 Ϯ 0.4 mV).
The observation that maxadilan, a PAC 1 receptor-specific agonist (45), does not evoke a depolarization of intracardiac neurons or increase action potential firing suggests that PAC 1 receptors do not have a direct effect on neuroexcitability in these cells. PAC 1 receptor activation, however, potentiates VPAC receptor-induced depolarizations. Evidence for this conclusion is provided by two observations reported here: 1) PACAP depolarizes neurons to a greater extent than VIP, and 2) the PAC 1 receptor antagonist, M65, depressed PACAP-induced depolarizations. Furthermore, increases in action potential firing in response to depolarizing current pulses were only observed under conditions in which both PAC 1 and VPAC receptors were stimulated (i.e. when PACAP was used as the agonist and neither PAC 1 nor VPAC receptor-specific antagonists were applied). Thus, simultaneous PAC 1 and VPAC receptor stimulation elicits a synergistic FIG. 9. PACAP-induced changes in intracardiac neuroexcitability are blocked by removal of extracellular Ca 2؉ and depletion of internal Ca 2؉ stores. A, action potentials evoked in response to a depolarizing current pulse (300 ms, 150 pA) from a single intracardiac neuron in the absence and presence of 100 nM PACAP in normal PSS (Control; 100 nM PACAP) or in Ca 2ϩ -free PSS (0 mM Ca 2ϩ , 1 mM EGTA, 0 mM Ca 2ϩ , 1 mM EGTA, PACAP), respectively. Cells were exposed to Ca 2ϩ -free PSS for ϳ10 min prior to the Ca 2ϩ -free experiments. B, scatter plot of mean RMP Ϯ S.E. recorded from six neurons under control and Ca 2ϩ -free conditions (0 Ca 2ϩ ) in the absence (Control, Wash) and presence of bath-applied PACAP-27 (100 nM; PACAP). C, bar graph of the average number of action potentials Ϯ S.E. elicited by depolarizing current pulse (300 ms, 150 pA) in intracardiac neurons studied under the same conditions as described for B. The asterisks denote significant difference from control (p Ͻ 0.01). enhancement of neuroexcitability and produces changes in the active membrane properties that are not seen with stimulation of either receptor alone.
The VPAC-mediated increases in [Ca 2ϩ ] i appear to be critical for the enhanced neuroexcitability of intracardiac neurons by PACAP. PACAP failed to depolarize intracardiac neurons and to increase action potential firing when applied in the presence of ryanodine, when the intracellular stores were depleted with caffeine, or in the absence of extracellular Ca 2ϩ . It is interesting to note that caffeine had no direct effects on neuroexcitability (Table I), and thus Ca 2ϩ release from intracellular stores is necessary but not sufficient to promote enhanced neuroexcitability. Muscarinic receptor activation has been shown to depolarize intracardiac neurons and to augment action potential firing in these cells (42). Our studies and those of others (49) have shown that muscarinic receptor activation mobilizes Ca 2ϩ from intracellular stores in intracardiac neurons. Although inhibition of the K ϩ conductance, I M , appears to be a major factor contributing to muscarinic receptor-induced neuroexcitability in these neurons (42), it would be of interest to determine whether muscarine-induced increases in [Ca 2ϩ ] i also play a role in the enhanced neuroexcitability and thus if mobilization of intracellular [Ca 2ϩ ] i is a mechanism by which various excitatory neurotransmitters regulate the electrical properties of intrinsic cardiac neurons.
In conclusion, the present study demonstrates the first evidence of VPAC receptor-mediated regulation of intracellular Ca 2ϩ in neurons and shows that by increasing [Ca 2ϩ ] i , VPAC receptors enhance neuroexcitability. Furthermore, our results demonstrate that the heterogeneity in PAC 1 and VPAC receptors observed in intracardiac neurons has significant implications for the effects of the neuropeptides on cellular function. The ability of PACAP to depolarize the neurons and increase action potential firing is dependent on PACAP stimulation of both PAC 1 and VPAC receptors. VIP, by acting exclusively on VPAC receptors, evokes a depolarization of lesser magnitude than PACAP and fails to increase action potential firing. The ability of PACAP and VIP to differentially influence the electrical activity of intracardiac neurons may also help explain why these related neuropeptides have distinct effects on the cardiovascular system. Whereas both PACAP and VIP have been shown to dilate coronary arteries and to produce positive chronotropic effects (8,10,65), PACAP, but not VIP, evokes pronounced negative chronotropic and negative inotropic effects that are mediated by activation of intrinsic cardiac neurons (10).