[1] 20w 5-HT 2 RECEPTOR SUBTYPES MEDIATE DIFFERENT LONG-TERM CHANGES IN GABAergic ACTIVITY TO PARASYMPATHETIC CARDIAC VAGAL NEURONS IN THE NUCLEUS AMBIGUUS

[1] 237w Serotonin (5-HT), and in particular 5-HT 2 receptors, play an important role in cardiorespiratory function within the brainstem. In addition, abnormalities in the 5-HT system have been implicated in many cardiorespiratory disorders, including sudden infant death syndrome. However, little is known about the mechanisms of action of 5-HT 2 receptors in altering the activity of parasympathetic cardiac neurons in the brainstem. In this study we examined the effects of activation of different subtypes of 5-HT 2 receptors on spontaneous and respiratory-evoked GABAergic neurotransmission to cardioinhibitory vagal neurons within the nucleus ambiguus as well as rhythmic fictive inspiratoryrelated activity in rats. A single application of ␣-Me-5-hydroxytryptamine maleate (␣-Me-5-HT), a 5-HT 2 receptor agonist, did not significantly alter the frequency of spontaneous or respiratory-evoked GABAergic inhibitory postsynaptic currents (IPSCs) in cardiac vagal neurons. However, repetitive successive applications of ␣-Me-5-HT elicited a long-lasting (>1 h) decrease in the frequency of spontaneous as well as inspiratory-related GABAergic IPSCs to cardiac vagal neurons. This study demonstrates multiple, but not single applications of the 5-HT 2 receptor agonist ␣-Me-5-HT caused a long-lasting inhibition of both spontaneous and fictive inspiratory-related GABAergic neurotransmission to CVNs, which can be prevented by the 5-HT 2B receptor antagonist SB204741, but persisted with the 5-HT 2A/2C receptor antagonist ketanserin. The 5-HT 2 receptor agonist ␣-Me-5-HT also reversibly and transiently excited central fictive inspiratory activity, which was abolished by ketanserin, but was unaffected by the 5-HT 2B receptor antagonist SB204741.

[1] 77w A single application of the 5-HT 2 receptor agonist ␣-Me-5-HT (1 M) elicited a significant and transient increase in the frequency of fictive inspiratory bursts from 2.2Ϯ0.3 to 4.0Ϯ0.6 bursts/min (PϽ0.01, nϭ12, Fig. 1, left, A and B). Multiple applications of ␣-Me-5-HT (1 M) also significantly increased fictive inspiratory burst frequency from 2.0Ϯ0.2 to 3.5Ϯ0.6 bursts/min (PϽ0.05, nϭ12, Fig. 1, right, A and B). The fictive inspiratory burst frequency remained significantly elevated for 6 min (Fig. 1B).

[2] 104w Neither single nor multiple applications of ␣-Me-5-HT (1 M) caused a significant change in the fictive inspiratory burst duration (single application, 2.2Ϯ0.2 s versus 2.2Ϯ0.1 s; multiple applications, 2.4Ϯ0.2 s versus 2.1Ϯ0.1 s, nϭ12). The absence of a change in burst duration was maintained over a range of ␣-Me-5-HT concentrations because fictive inspiratory burst duration was not significantly changed by either a single application of ␣-Me-5-HT (108.9Ϯ7.8, 99.5Ϯ3.8 and 94.5Ϯ9.5% at concentrations of 0.1 (nϭ10), 1.0 (nϭ12) and 5 M (nϭ10), respectively) or with multiple applications of ␣-Me-5-HT (96.3Ϯ8.6, 97.7Ϯ5.4 and 96.3Ϯ8.8% at concentrations of 0.1 (nϭ10), 1.0 (nϭ12) and 5 M (nϭ10), respectively).

[3] 81w Application of the 5-HT 2B receptor antagonist SB204741 (10 M) by itself did not significantly change the frequency of fictive inspiratory bursts (2.7Ϯ0.4 bursts/min versus 2.5Ϯ0.2 bursts/min; nϭ11). SB204741(10 M) also did not alter the ␣-Me-5-HT (1 M) elicited increase in the fictive inspiratory burst frequency, because fictive inspiratory burst frequency increased from 2.5Ϯ0.2 to 4.5Ϯ0.7 bursts/ min in response to a single application (PϽ0.05, nϭ11) and from 2.5Ϯ0.2 to 5.1Ϯ0.7 bursts/min in response to multiple applications (PϽ0.05, nϭ11; Fig. 1C).

[4] 83w In contrast, although application of the 5-HT 2A/2C receptor antagonist ketanserin (1 M) did not by itself significantly change the basal frequency of fictive inspiratory bursts (2.7Ϯ0.3 bursts/min versus 2.4Ϯ0.3 bursts/min; nϭ10), ketanserin (1 M) abolished the facilitation of fictive inspiratory activity evoked by ␣-Me-5-HT. In the presence of ketanserin (1 M) applications of ␣-Me-5-HT (1 M) failed to significantly change the fictive inspiratory burst frequency (2.4Ϯ0.3 versus 2.6Ϯ0.2 bursts/min (single application; nϭ10), and 2.4Ϯ0.3 versus 1.9Ϯ0.3 bursts/min (multiple applications; nϭ10; Fig. 1D)).

[5] 182w Consistent with previous reports (Neff et al., 2004;Huang et al., 2006) the frequency of GABAergic IPSCs increased during fictive inspiratory bursts (Fig. 2A, 4A, and 5A). A single application of ␣-Me-5-HT (1 M) did not produce a significant change in the frequency of either spontaneous or fictive inspiratory-related GABAergic IPSCs (6.4Ϯ1.1 Hz versus 8.0Ϯ1.0 Hz and 9.9Ϯ1.7 Hz versus 8.8Ϯ1.2 Hz, respectively; nϭ10 CVNs; Fig. 2A and B). In contrast, multiple applications of ␣-Me-5-HT (1 M) caused a significant and long-lasting inhibition in the frequency of both spontaneous and fictive inspiratory-related GABAergic IPSCs. The frequency of spontaneous GABAergic IPSCs decreased from 6.4Ϯ0.8 Hz (control) to 3.8Ϯ0.9 Hz (1-2 min after applications, PϽ0.01, nϭ10) and remained significantly decreased for at least 60 min after the applications of ␣-Me-5-HT (3.0Ϯ0.6 Hz; PϽ0.001, nϭ10; Fig. 2A and B). Similar to spontaneous GABAergic neurotransmission, the frequency of fictive inspiratory-related GABAergic IPSCs decreased from 9.1Ϯ1.4 Hz (control) to 3.9Ϯ0.7 Hz (1-2min after applications; PϽ0.001, nϭ10) and remained significantly decreased for at least 60 min after the applications of ␣-Me-5-HT (4.3Ϯ1.2 Hz; PϽ0.001, nϭ10; Fig. 2A and B).

[6] 81w The lack of responses of GABAergic frequency to a single ␣-Me-5-HT application were consistent over a range of different concentrations of ␣-Me-5-HT including 0.1, 1.0 and 5 M (Fig. 3). Neither spontaneous GABAergic frequency (112Ϯ20, 125Ϯ16 and 94Ϯ24% at concentrations of 0.1 (nϭ6), 1 (nϭ10) and 5 M (nϭ6), respectively) or fictive inspiratory-related GABAergic frequency (95Ϯ13, 89Ϯ12 and 85Ϯ25% at concentrations of 0.1 (nϭ6), 1 (nϭ10) and 5 M (nϭ6), respectively) were changed by a single application of ␣-Me-5-HT (Fig. 3).

[7] 137w The inhibition of spontaneous GABAergic IPSCs 59 -60 min after multiple applications of ␣-Me-5-HT was similar with ␣-Me-5-HT applied at a range of different concentrations including 0.1, 1.0 and 5 M (Fig. 3). The inhibition of spontaneous GABAergic IPSCs after multiple applications of ␣-Me-5-HT were 63Ϯ12, 47Ϯ10 and 34Ϯ9% at concentrations of 0.1 (nϭ6), 1 (nϭ10) and 5 M (nϭ6) (Fig. 3). However, the inhibition of inspiratory-related GABAergic IPSCs after multiple applications of ␣-Me-5-HT did show some dose dependence because the most profound decrease in the GABAergic IPSCs frequency was produced by applications of ␣-Me-5-HT at a concentration of 5 M (39%Ϯ14%, nϭ6), which was significantly greater (PϽ0.05) than the inhibition elicited by a concentration of 0.1 M (82Ϯ12%, nϭ6) but not significantly different from the responses at a concentration at 1 M (48Ϯ13%, nϭ10; (Fig. 3).

[8] 141w Application of the 5-HT 2A/2C receptor antagonist ketanserin, over a range of concentrations including 1, 10, and 100 M, did not significantly change the basal frequency of spontaneous GABAergic IPSCs (control, 8Ϯ1 Hz, 1 M, 8Ϯ1 Hz, nϭ7; control, 5Ϯ1 Hz, 10 M, 6Ϯ1 Hz, nϭ7; control, 4Ϯ1 Hz, 100 M, 4Ϯ1 Hz, nϭ6). Similarly, ketanserin, over a range of concentrations including 1, 10 and 100 M, did not significantly alter the inhibitory responses to multiple applications of ␣-Me-5-HT (1 M) because the spontaneous GABAergic IPSCs decreased upon multiple exposures of ␣-Me-5-HT with ketanserin concentrations of 1, 10 and 100 M from 8Ϯ1 to 5Ϯ1 Hz (PϽ0.05, nϭ7), from 6Ϯ1 to 2Ϯ1 Hz (PϽ0.01, nϭ7) and from 4Ϯ1 to 0.6Ϯ0.2 Hz (PϽ0.01, nϭ6), respectively (only the responses in the presence of 1 M ketanserin are shown in Fig. 4A and C).

[9] 69w Ketanserin did, however, prevent the ␣-Me-5-HT-mediated inhibition of inspiratory-evoked GABAergic activity upon multiple applications of ␣-Me-5-HT. The long-lasting inhibition of inspiratory-related GABAergic activity that occurs with multiple applications of ␣-Me-5-HT exposure was prevented by 1 M ketanserin because the frequency of inspiratory-related GABAergic IPSCs remained statistically unchanged from control levels after ␣-Me-5-HT application in the presence of ketanserin (1 M, 12Ϯ2 vs 8Ϯ2 Hz, nϭ7; Fig. 4A and C).

[10] 73w Application of the 5-HT2 B receptor antagonist SB204741 did not evoke any significant change in the frequency of spontaneous or inspiratory-related GABAergic IPSCs at SB204741 concentrations of both 1 and 10 M (from 4Ϯ1 to 4Ϯ1 (nϭ7), and 9Ϯ2 to 8Ϯ2 (nϭ7), respectively). Similarly, inspiratory-related GABAergic IPSC responses to a single application of ␣-Me-5-HT were not altered by SB204741 at a concentration of 10 M (Fig. 5B; from 11Ϯ2 to 11Ϯ2 Hz, nϭ7).

[11] 124w However, the inhibition of spontaneous GABAergic IPSCs with multiple applications of ␣-Me-5-HT was completely prevented by SB204741 at a concentration of 10 M (Fig. 5A and C), but SB204741 at a concentration of 1 M did not alter the inhibition of spontaneous GABAergic IPSCs by multiple applications of ␣-Me-5-HT. Spontaneous GABAergic IPSC frequency decreased from 4Ϯ1 to 0.8Ϯ0.2 (PϽ0.0001, nϭ7) in the presence of 1 M SB204741 after multiple applications of ␣-Me-5-HT, but in the presence of 10 M SB204741 the inhibition of GABAergic IPSCS by multiple applications of ␣-Me-5-HT was prevented (control: 8Ϯ2; SB204741: 8Ϯ2 Hz, nϭ7). Similarly, the ␣-Me-5-HT mediated inhibition of inspiratory related GABAergic IPSCs was prevented by 10 M SB204741 (control: 11Ϯ2, SB204741 9Ϯ2 Hz, nϭ7; (Fig. 5A and C).

[1] 83w There are two major findings from this study: (1) multiple, but not single applications of the 5-HT 2 receptor agonist ␣-Me-5-HT caused a long-lasting inhibition of both spontaneous and fictive inspiratory-related GABAergic neurotransmission to CVNs within the NA. The 5-HT 2 -mediated inhibitory effect was completely prevented by the 5-HT 2B receptor antagonist SB204741, whereas the 5HT 2 -mediated inhibition of spontaneous GABAergic neurotransmission persisted with the 5-HT 2A/2C receptor antagonist ketanserin, although ketanserin abolished the fictive inspiratory-related GABAergic inhibition evoked by ␣-Me-5-HT.

[2] 47w (2) Both single and multiple applications of the 5-HT 2 receptor agonist ␣-Me-5-HT reversibly and transiently excited central fictive inspiratory activity. The excitatory effect on fictive inspiratory activity was abolished by ketanserin, a 5-HT 2A/2C receptor antagonist, but was unaffected by the 5-HT 2B receptor antagonist SB204741.

[3] 517w The results in this study demonstrate that multiple, but not single, applications of the 5-HT 2 receptor agonist ␣-Me-5-HT inhibit both spontaneous and fictive inspiratoryrelated GABAergic neurotransmission to CVNs within the NA, and this inhibition persisted for at least 1 h. The mechanisms of this long-lasting inhibition are unknown. 5-HT 2 receptors have a role in synaptic plasticity (Fuller et al., 2001a;Chen and Bazan, 2003;Shay et al., 2005), and repetitive 5-HT 2 receptor activation is hypothesized to initiate intracellular signaling events leading to the rapid synthesis of proteins, particularly brain-derived neurotrophic factor (BDNF), necessary to maintain long-term plasticity (Mitchell et al., 2001;Feldman et al., 2003). It is possible BDNF contributes to the long-lasting inhibition of GABAergic neurotransmission observed in this study because BDNF mRNA is expressed in neurons located within the NA Fig. 3. There was no significant difference in the frequency of either spontaneous or inspiratory-related GABAergic IPSCs during a single ␣-Me-5-HT application at concentrations of 0.1 (nϭ6), 1 (nϭ10), or 5 M (nϭ6). Similarly, the decrease of the frequency of spontaneous GABAergic IPSCs 60 min after multiple applications of ␣-Me-5-HT was not concentration dependent (nϭ6, 10 and 6 CVNs at concentrations of 0.1, 1 and 5 M, respectively). However, the inhibition of frequency of fictive inspiratory-related GABAergic IPSCs 60 min after multiple applications of ␣-Me-5-HT at a concentration of 5 M (nϭ6) was significantly lower (PϽ0.05) than the responses at a concentration of 0.1 M (nϭ6) but not significantly different from the responses at a concentration of 1 M (nϭ10). * PϽ0.05. (Zaidi et al., 2005) and BDNF suppressed GABAergic synaptic neurotransmission (Henneberger et al., 2002) in other neurons. Because ␣-Me-5-HT does not discriminate between subtypes of 5-HT 2 receptors (Baxter et al., 1995), we used two 5-HT 2 receptor antagonists to isolate the subtypes of the 5-HT 2 receptors involved: ketanserin, which is a specific 5-HT 2A/2C antagonist, and SB204741, a selective 5-HT 2B antagonist. The inhibitory effect of ␣-Me-5-HT was blocked by 5-HT 2B receptor antagonist SB204741, suggesting a critical role of 5-HT 2B receptors in 5-HT 2 -mediated inhibition of GABAergic neurotransmission, while the 5-HT 2A/2C receptor antagonist ketanserin blocked only the fictive inspiratory-related GABAergic inhibition evoked by ␣-Me-5-HT but ketanserin did not abolish the 5HT 2 -mediated inhibition of spontaneous GABAergic neurotransmission. Interestingly, in the presence of ketanserin the inhibitory effect of ␣-ME-5-HT on the spontaneous IPSCs to CVNs did not persist for 60 min, suggesting that 5-HT 2A and 5-HT 2C receptors likely contribute to the long-lasting, but not initial, 5-HT 2 -mediated inhibition of spontaneous GABAergic neurotransmission to CVNs. This work indicates 5-HT 2A/2C receptors are likely involved in selectively activating the GABAergic neurotransmission to cardiac vagal neurons that occurs during inspiration, and blocking these 5-HT 2A/2C receptors would reduce respiratory sinus arrhythmia and heart rate variability. The site of action with bath-applied 5-HT 2 agonist and antagonistsin this study is unknown. Because 5-HT 2B receptors modulated GABAergic neurotransmission to CVNs but did not influence central fictive inspiratory burst frequency, it is likely that there are different sites of 5-HT actions for CVNs and respiratory neurons, including hypoglossal motorneurons.

[4] 168w In addition to 5-HT 2 modulation of cardiac vagal neurons, multiple lines of evidence suggest excitatory 5-HT 2 receptors are also important in respiratory function in the brainstem. In decerebrate adult cats (Rose et al., 1995), anesthetized adult rats (Fenik and Veasey, 2003), and decerebrate newborn rats ((Khater-Boidin et al., 1999;Glerant et al., 2005) 5-HT induced an increase in hypoglossal activity due to activation of 5-HT 2 receptors. The results of this study are in agreement with most other studies because a single application of the 5-HT 2 receptor agonist ␣-Me-5-HT produced a significant increase in fictive inspiratory burst frequency, similar to that observed in a majority of other studies (Schwarzacher et al., 2002). The excitatory effect on fictive inspiratory activity was prevented by ketanserin, but not altered by SB204741. The site(s) of action of ␣-Me-5-HT is not fully known. ␣-Me-5-HT may act directly on hypoglossal motorneurons (Berger et al., 1992); in addition, ␣-Me-5-HT may also act on respiratory neurons within the pre-Botzinger complex (Schwarzacher et al., 2002).

[5] 247w In addition to the rapid responses, 5-HT 2 receptors may be involved in long-term changes of synaptic and central respiratory activity (Fuller et al., 2001a;Chen and Bazan, 2003;Shay et al., 2005). A prolonged 5-HT-dependent augmentation of respiratory motor output is often referred to as LTF (Mitchell and Johnson, 2003). LTF can be elicited by episodic hypoxia in anesthetized and con-scious animals (Mitchell and Johnson, 2003) and episodic activation of 5-HT receptors (Johnson et al., 2001;Bocchiaro and Feldman, 2004). In this study multiple applications of the 5-HT 2 receptor agonist ␣-Me-5-HT (1 M) produced a significant increase in hypoglossal nerve burst frequency that only persisted for 6 min. These results are different from the results of Bocchiaro and Feldman (Bocchiaro and Feldman, 2004), who, using medullary slice preparation from neonatal rats, demonstrated that multiple applications of ␣-Me-5-HT induced persistent (Ͼ1 h) increases in hypoglossal nerve activity. However, this LTF may be dependent on the strain of rats used because there appear to be substantial differences in hypoglossal LTF between substrains of Sprague-Dawley rats (Fuller et al., 2001b;Feldman et al., 2003). Indeed, hypoglossal LTF was present in Charles River Laboratories/Sasco but not Harlan rats (Fuller et al., 2001b). In this study we used Sprague-Dawley rats from a Hilltop supplier. Because Fuller et al. (2001b) did not mention in their study which substrain of Sprague-Dawley rats was used, differences between strains of animals may be responsible for the differences in the duration of LTF evoked by the 5-HT 2 agonist.

[6] 233w The long-lasting withdrawal of GABAergic neurotransmission to CVNs elicited by repetitive 5-HT 2 receptor activation provides a potential neurochemical mechanism that may be important in sudden infant death syndrome (SIDS). Episodic hypoxia, often resulting from an unstable breathing pattern during sleep, would likely result in repetitive activation of raphe 5-HT neurons (Bodineau and Larnicol, 2001;Feldman et al., 2003), which could lead to repetitive 5-HT 2 receptor activation and persistent inhibition of spontaneous and fictive inspiratory-related GABAergic neurotransmission to parasympathetic CVNs evoking a bradycardia. Progressive bradycardia and respiratory failure are considered the most likely causes of death in SIDS (Meny et al., 1994;Thach, 2005). These responses may be exaggerated in SIDS victims because these infants have a higher number of 5-HT neurons (Paterson et al., 2006). The cerebrospinal fluids of infant victims of sudden death show a very significant increase in the metabolites of 5-HT (Caroff et al., 1992), and evidence from infants that succumb to SIDS suggests that the mechanisms of death involve a progressive bradycardia, even in the presence of continued breathing movements (Meny et al., 1994). In addition, a recent report demonstrates that gasping activity is dependent on 5-HT 2A receptors (Tryba et al., 2006). It is possible that 5-HT 2 receptors are responsible for the bradycardia that occurs with hypoxia and that an exaggeration of these 5-HT 2 receptor-mediated responses increases the excitation of CVNs and the risk of SIDS.

[1] 395w To identify cardiac vagal neurons in vitro, a two-stage procedure was utilized. In an initial surgery, Sprague-Dawley rats (postnatal days 2-6; Hilltop, Scottdale, PA, USA) were anesthetized with hypothermia and received a right thoracotomy. The heart was exposed, and 0.05 ml of rhodamine (Molecular Probes, Eugene, OR, USA) was injected into the pericardial sac to retrogradely label CVNs. The labeling of CVNs using these procedures was previously described (Bouairi et al., 2006). Specificity of the cardiac vagal labeling was confirmed by the absence of any labeled neurons in the brainstem when rhodamine is injected either outside the pericardial sac or within the pericardial sac if the cardiac branch of the vagus nerve is sectioned (nϭ4). Recent work demostrated this method does not label neurons in the compacta formation but identifies cardiac vagal neurons localized in the external formation of the NA (Bouairi et al., 2006). In other control experiments (nϭ10), i.v. injection of up to 10 mg of rhodamine failed to label any neurons in the medulla except for rare labeling of neurons in the area postrema, an area with a deficient bloodbrain barrier. On the day of experiment (2-4 days later), the animals were anesthetized with halothane and killed by rapid cervical dislocation. The brain was submerged in cold (4 °C) buffer composed of 140 mM NaCl, 5 mM KCl, 2 mM CaCl 2 , 5 mM glucose, and 10 mM Hepes and continually gassed with 100% O 2 . Under a dissection microscope, the cerebellum was removed, and the hindbrain was isolated. A single slice of the medulla (800-m thickness) that included CVNs, the rostral hypoglossal nucleus and rootlets and the pre-Botzinger complex was obtained and submerged in a recording chamber, which allowed perfusion (5-10 ml/min) of artificial cerebrospinal fluid at room temperature containing 125 mM NaCl, 3 mM KCl, 2 mM CaCl 2 , 26 mM NaHCO 3 , 5 mM glucose, and 5 mM Hepes equilibrated with carbogen (95% O 2 and 5% CO 2 , pH 7.4). All animal procedures were performed in compliance with the institutional guidelines at George Washington University and are in accordance with the recommendations of the Panel on Euthanasia of the American Veterinary Medical Association and the National Institutes of Health publication Guide for the Care and Use of Laboratory Animals. All efforts were made to minimize the number of animals used and their suffering.

[2] 56w The thick medullary slice preparation generates rhythmic fictive inspiratory-related motor discharge in hypoglossal cranial nerves. Spontaneous fictive inspiratory-related activity was recorded by monitoring motorneuron population activity from hypoglossal nerve rootlets using a suction electrode. Hypoglossal rootlet activity was amplified 50,000 times, filtered (10 -300 HZ bandpass; CWE, Ardmore, PA, USA) and electronically integrated (ϭ50 ms; CWE).

[3] 136w Individual CVNs in the NA were identified by the presence of the fluorescent tracer using a Zeiss Axioskop upright microscope (Carl Zeiss Inc., Thornwood, NY, USA) using a 40ϫ water immersion objective. These identified CVNs were then imaged with differential interference contrast optics, infrared illumination, and infrared-sensitive video detection cameras to gain better spatial resolution. Patch pipettes (2.5-3.5 M⍀) containing 150 mM KCl, 4 mM MgCl 2 , 2 mM EGTA, 2 mM Na-ATP, and 10 mM Hepes (pH 7.4) were guided to the surface of individual CVNs. This pipette solution resulted in inward Cl Ϫ currents upon activation GABA receptors (calculated reversal potential of Cl Ϫ ϭϩ4 mV) at a holding potential of Ϫ80 mV. Voltage clamp whole-cell recordings were made with an Axopatch 200B and pClamp 8 software (Axon Instruments, Union City, CA, USA).

[4] 184w GABAergic neurotransmission was isolated by continuous focal application of strychnine (1 M), D-2-amino-5-phosphonovalerate (50 M), and 6-cyano-7-nitroquinoxaline-2,3-dione (50 M) to block glycine, N-methyl-D-asparate (NMDA) and non-NMDA receptors, respectively. Continual focal drug applications were performed using a pneumatic picopump pressure system (WPI, Sarasota, FL, USA). Drugs were continuously and focally released throughout the experiments using a picrospritzer and pressure ejected from a patch pipette positioned within 30 m of the patched CVN. The maximum range of drug application was determined previously to be 100 -120 m downstream from the drug pipette and was considerably less behind the drug pipette (Wang et al., 2002). The following drugs were applied by inclusion in the perfusate: the 5-HT 2 receptor agonist ␣-methyl-5-hydroxytryptamine maleate (␣-Me-5-HT, 1 M); and two 5-HT 2 receptor antagonists with differential affinity for different 5-HT 2 receptors: 5-HT 2B receptor antagonist SB204741 (10 M) and 5-HT 2A/2C receptor antagonist ketanserin (1 M). The pK values of the drugs are, respectively, on 5-HT 2A , 5-HT 2B and5-HT 2C receptors: ␣-Me-5-HT: 7.4, 8.8, 6.2; SB204741: Ͻ5.3, 7.8, Ͻ6; ketanserin 8.9, 5.4, 7.0 (Baxter et al., 1995).

[5] 123w Control rhythmic fictive inspiratory-related activity and GABAergic IPSCs were recorded simultaneously for 4 min. Slices were initially exposed to a single 4-min application of ␣-Me-5-HT and then after a 30-min washout of the agonist were re-exposed to multiple applications of ␣-Me-5-HT (three applications of 4-min duration every 9 min). In experiments including antagonists, the antagonist was applied for 4 min followed by the same protocol for agonist application. After the end of each experiment the GABAergic inhibitory postsynaptic currents (IPSCs) were abolished by gabazine (25 M) focal application to the patched CVN. Only one experiment was conducted per preparation. This protocol was designed to mimic previous studies in the literature that examined respiratory responses to intermittent hypoxia and elicit LTF of respiratory activity.

[6] 220w Synaptic events were detected using MiniAnalysis (version 5.6.12; Synaptosoft, Decatur, GA, USA). The frequency of IPSCs that occurred in CVNs was grouped in 1-s bins and cross-correlated with onset of fictive inspiratory-related hypoglossal activity. The 1-s period before the fictive inspiratory burst onset was considered spontaneous activity, while the 1-s period immediately after the fictive inspiratory burst onset was analyzed as the inspiratory period. Burst frequency and duration were measured using pClamp 8 software (Molecular Devices, Sunnyvale, CA, USA). Data were analyzed during the last 2 min of the control period, during last 2 min of single drug applications, during 2 min immediately after the end of multiple drug applications and every 2 min during the washout condition after single and multiple drug applications, as appropriate. The results from studying ␣-Me-5-HT concentration-dependent responses are presented as the mean percentage of controlϮSEM and were statistically compared with analysis of variance (ANOVA) and Kruskal-Wallis nonparametric test followed by Dunn's posttest. Otherwise, results were not presented as a percentage of control but were presented as meansϮSEM and were statistically compared using ANOVA with repeated measures and Tukey's post test to examine the responses throughout the time course of the experiments and paired Student's t test when comparing the results from control to antagonist applications. Significant difference for all data were set at PϽ0.05.

[1] 32w Heart rate is dominated by the activity of premotor cardioinhibitory vagal neurons (CVNs) located in the nucleus ambiguus. Within the nucleus ambiguus (NA) premotor neurons receive a high number of axosomatic serotonin

[2] 45w (5-HT) contacts, and the 5-HT contacts surrounding neurons in the nucleus ambiguus are among the most dense in the brainstem (Takeuchi et al., 1983). 5-HT fibers also specifically surround CVNs, which have been described as "ensheathed in 5-HT immunoreactive axonal boutons" (Izzo et al., 1993).

[3] 98w A multitude of different 5-HT receptors influence cardiorespiratory function in the brainstem. Central 5-HT 7 receptors play an important role in the reflex activation of parasympathetic outflow to the heart upon stimulating cardiopulmonary afferent fibers, arterial baroreceptors and chemoreceptor afferents (Kellett et al., 2005). Central 5-HT 1A receptors are also involved in mediating both cardiopulmonary and baroreceptor reflex-evoked vagal bradycardia, but may not be involved in chemoreceptor-elicited responses in cardiac vagal neurons (Skinner et al., 2002). While activation of 5-HT 1A receptors potentiate, 5-HT 1B/D agonists depress chemoreceptor reflex activation of parasympathetic cardiac neurons (Dando et al., 1998).

[4] 131w Microinjection or ionphoretic application of different 5-HT agonists into the NA also provided mixed responses. While low doses of the 5-HT 1A agonist 8-OH-DPAT generally inhibit CVNs, higher doses elicited an excitation of CVNs (Wang and Ramage, 2001). Other work demonstrated microinjection of the 5-HT 1A agonist 8-OH-DPAT excited CVNs to evoke a bradycardia (Chitravanshi and Calaresu, 1992). Additional 5-HT receptor agonists have yet to be tested. A limitation of these microinjection studies is that the sites of action and mechanisms responsible for these responses are unknown. The heart rate responses to microinjection of 5-HT agonists may have been evoked by stimulation of local polysynaptic pathways, interneurons, activation of presynaptic terminals that synapse upon cardiac vagal neurons, modification of postsynaptic synaptic currents or direct alterations in the membrane properties of CVNs.

[5] 85w In contrast, 5-HT receptors, and in particular 5-HT 2 receptors, are critical to respiratory function in the brainstem. 5-HT 2 receptors are essential in the responses to intermittent hypoxia and evoked long-term changes of synaptic and central respiratory activity (Fuller et al., 2001a;Chen and Bazan, 2003;Shay et al., 2005). A prolonged 5-HT-dependent augmentation of respiratory motor output, long-term facilitation (LTF), can be elicited by both episodic hypoxia (Mitchell and Johnson, 2003) and episodic activation of 5-HT 2 receptors (Johnson et al., 2001;Bocchiaro and Feldman, 2004).

[6] 116w In summary, despite the strong evidence from anatomic work that 5-HT fibers abundantly surround parasym-pathetic cardiac vagal neurons (Izzo et al., 1993) as well as the evidence that 5-HT 2 receptors play an essential role in cardiorespiratory network function, there is a lack of information on the role of acute and intermittent activation of 5-HT 2 receptors in modulating the activity of cardiac vagal neurons in the NA. In this study we tested the hypothesis that single as well as multiple applications of 5-HT 2 receptor agonists modulate the important respiratory-related and spontaneous GABAergic neurotransmission to cardiac vagal neurons and investigated the relative contribution of different 5-HT 2 receptor subtypes (5-HT 2B and 5-HT 2A/C ).