Selective enhancement of excitatory synaptic activity in the rat nucleus tractus solitarius by hypocretin 2

PII: S 0 3 0 6 - 4 5 2 2 ( 0 2 ) 0 0 4 8 8 - 8

Neuroscience Vol. 115, No. 3, pp. 707^714, 2002 A 2002 IBRO. Published by Elsevier Science Ltd All rights reserved. Printed in Great Britain 0306-4522 / 02 $22.00+0.00

www.neuroscience-ibro.com

SELECTIVE ENHANCEMENT OF EXCITATORY SYNAPTIC ACTIVITY IN THE RAT NUCLEUS TRACTUS SOLITARIUS BY HYPOCRETIN 2 B. N. SMITH,a
S. F. DAVIS,a A. N. VAN DEN POLb and W. XUa a

Department of Cell and Molecular Biology, Tulane University, 6400 Freret Street, New Orleans, LA 70118, USA b

Department of Neurosurgery, Yale University School of Medicine, New Haven, CT 06520, USA

Abstract5Hypocretin 2 (orexin B) is a hypothalamic neuropeptide thought to be involved in regulating energy homeostasis, autonomic function, arousal, and sensory processing. Neural circuits in the caudal nucleus tractus solitarius (NTS) integrate viscerosensory inputs, and are therefore implicated in aspects of all these functions. We tested the hypothesis that hypocretin 2 modulates fast synaptic activity in caudal NTS areas that are generally associated with visceral sensation from cardiorespiratory and gastrointestinal systems. Hypocretin 2-immunoreactive ¢bers were observed throughout the caudal NTS. In whole-cell recordings from neurons in acute slices, hypocretin 2 depolarized 48% and hyperpolarized 10% of caudal NTS neurons, e¡ects that were not observed when Csþ was used as the primary cation carrier. Hypocretin 2 also increased the amplitude of tractus solitarius-evoked excitatory postsynaptic currents (EPSCs) in 36% of neurons and signi¢cantly enhanced the frequency of spontaneous EPSCs in most (59%) neurons. Spontaneous inhibitory postsynaptic currents (IPSCs) were relatively una¡ected by the peptide. The increase in EPSC frequency persisted in the presence of tetrodotoxin, suggesting a role for the peptide in regulating glutamate release in the NTS by acting at presynaptic terminals. These data suggest that hypocretin 2 modulates excitatory, but not inhibitory, synapses in caudal NTS neurons, including viscerosensory inputs. The selective nature of the e¡ect supports the hypothesis that hypocretin 2 plays a role in modulating autonomic sensory signaling in the NTS. A 2002 IBRO. Published by Elsevier Science Ltd. All rights reserved. Key words: glutamate, neuropeptide, orexin, patch-clamp, vagus, viscerosensory.

lators of arousal. This diverse set of functions has led to the proposal that the hypocretins may coordinate autonomic tone with arousal (Mieda and Yanagisawa, 2002; Sutcli¡e and De Lecea, 2002). Although cellular e¡ects of hcrt2 have been reported in hypothalamic regions associated with autonomic and endocrine function (van den Pol et al., 1998; Shirasaka et al., 2001; Samson et al., 2002), few studies have examined the peptide’s actions on neurons in brainstem autonomic areas (Hwang et al., 2001). Principal among these brainstem autonomic centers, the nucleus tractus solitarius (NTS) is the site of ¢rst central synaptic contact for ¢bers of the vagus nerve mediating sensory information from the thoracic and abdominal viscera (Kalia and Sulivan, 1982; Spyer et al., 1984; Shapiro and Miselis, 1985). Viscerosensory signals are integrated in the NTS, with local circuits and inputs from other brain regions contributing to signal processing. Although direct vagal and solitary complex involvement in hypocretinmediated autonomic regulation has been proposed (Takahashi et al., 1999; Krowicki et al., 2002), these results are controversial (Dube et al., 1999). At the cellular level, hcrt2 tends to enhance fast excitatory and inhibitory (i.e. glutamatergic and GABAergic) neurotransmission, and depolarizes neurons in several CNS regions (van den Pol et al., 1998; Horvath et al., 1999; Eggermann et al., 2001; Hwang et al., 2001; Shirasaka et al., 2001). In two recent studies, hcrt2 has been demon-

The hypocretins (orexins) are neuropeptides produced in and near the lateral hypothalamus with a widespread distribution in the CNS (De Lecea et al., 1998; Peyron et al., 1998; Date et al., 1999; Nambu et al., 1999). Anatomical and behavioral evidence suggests that the peptides enhance feeding behavior when infused into the ventricles (Sakurai et al., 1998), and also regulate cardiovascular and sympathetic function (Shirasaka et al., 1999; Chen et al., 2000). A role for hypocretin 2 (hcrt2) in modulating somatic sensory input in the dorsal horn has been proposed (van den Pol, 1999; Grudt et al., 2002). De¢ciencies in hypocretin peptides or their receptors result in disruption of normal sleep patterns (Chemelli et al., 1999; Lin et al., 1999), supporting the hypothesis that the hypocretins are also important regu-

*Corresponding author. Tel. : +1-504-862-3150; fax: +1-504-8656785. E-mail address: [email protected] (B. N. Smith). Abbreviations : ACSF, arti¢cial cerebrospinal £uid; CNQX, 6-cyano-7-nitroquinoxaline-2,3-dione; EGTA, ethyleneglycolbis-(L-aminoethylether)-N,N,NP,NP-tetraacetic acid ; EPSC, excitatory postsynaptic current ; hcrt2, hypocretin 2; HEPES, N-2hydroxyethylpiperazine-N-2-ethanesulfonic acid; IPSC, inhibitory postsynaptic current ; mEPSC, miniature EPSC; NTS, nucleus tractus solitarius ; PBS, phosphate-bu¡ered saline; PSC, postsynaptic current ; sEPSC, spontaneous EPSC; sIPSC, spontaneous IPSC; TS, tractus solitarius; TTX, tetrodotoxin. 707

NSC 5854 1-11-02

708

B. N. Smith et al.

strated to have relatively selective e¡ects on synaptic transmission to inhibitory interneurons of the spinal cord (Grudt et al., 2002) or neurons of the tegmentum (Burlet et al., 2002), e¡ects which are likely related to the depolarizing action of hcrt2 in nearby neurons. Regardless of reported autonomic and sensory functions for the peptide and the generally excitatory actions of hcrt2 in speci¢c brain regions associated with autonomic and sensory functions, its e¡ects on synaptic transmission in the caudal NTS, a principal center for autonomic viscerosensory processing, have not been de¢ned. We tested the hypothesis that hcrt2 modulates synaptic transmission in the caudal NTS. Our results indicate that hcrt2 signi¢cantly in£uences neurons in the NTS, depolarizing neuronal membranes and increasing excitatory, but not inhibitory synaptic input to most neurons. The selective nature of these e¡ects on pre- and postsynaptic elements in the nucleus is consistent with the hypothesis that hcrt2 plays a speci¢c role in regulating autonomic function, acting to modulate the initial processing of viscerosensory information.

EXPERIMENTAL PROCEDURES

Animals and slice preparation Adult (4^6 weeks) male Sprague^Dawley rats (Harlan, Indianapolis, IN, USA) were housed in a vivarium (12 h light/dark cycle) under the care of a veterinary sta¡. All procedures were approved by the Tulane University Animal Care and Use Committee. Rats were anesthetized by sodium pentobarbital injection (100 mg/kg, i.p.) and killed by decapitation while anesthetized. Brains were rapidly removed and immersed in ice-cold (0^4‡C), oxygenated (95% O2 /5% CO2 ) arti¢cial cerebrospinal £uid (ACSF) containing (in mM): 124 NaCl, 3 KCl, 26 NaHCO3 , 1.4 NaH2 PO4 , 11 glucose, 1.3^2 CaCl2 , and 1.3 MgCl2 , pH = 7.3^7.4, with an osmolality of 290^315 mOsm/kg. Transverse brainstem slices (300^400 Wm) containing the caudal NTS were made using a vibrating microtome (Vibratome Series 1000, Technical Products Intl, St. Louis, MO, USA). The slices were then transferred to a storage chamber, where they were allowed to equilibrate for at least an hour prior to recording. In most cases, a single brain slice was then transferred to a submersion style recording chamber on a ¢xed stage mounted under an upright microscope (Olympus, Model BX50WI; Melville, NY, USA) and continuously perfused with warmed (32^ 35‡C) and oxygenated ACSF. In some cases, slices were transferred to an interface-type recording chamber, where they were perfused with ACSF. The ACSF used for recordings was identical to that used in the dissection. Whole-cell recording After an equilibration period of 1^2 h, whole-cell recordings were obtained in the NTS using patch pipettes with open resistance of 3^5 M6, as described previously (Smith et al., 1998). Recording pipettes ¢lled with (in mM): 130 K-gluconate (or Csgluconate), 1 NaCl, 5 EGTA, 10 HEPES, 1 MgCl2 , 1 CaCl2 , 3 KOH, 2^4 Mg-ATP, and 0.2% biocytin (all from Sigma); pH = 7.2, adjusted with 5 M KOH or CsOH. In most cases, neurons were targeted for recording under a 40U water immersion objective (NA = 0.8) using infrared di¡erential interference contrast (IR-DIC) optics (Olympus) and a Spot RT Slider CCD camera (Diagnostic Instruments, Sterling Heights, MI, USA). Electrophysiological signals were recorded using an Axopatch 200B or Axopatch 1D ampli¢er (Axon Instruments, Union City, CA, USA), low-pass ¢ltered at 5 kHz and digitized at 88 kHz (Neuro-corder, Cygnus Technology Inc., Delaware Water

Gap, PA, USA) for storage on videotape and for on-line capture (Digidata 1320A, Axon Instruments). Data were analyzed o¡-line on a computer using the pCLAMP program suite (Axon Instruments) or Mini-analysis (Synaptosoft, Decatur, GA, USA). Hcrt2 (or orexin B; Sigma, St. Louis, MO, USA) was bath applied for 2^4 min at a concentration of 0.1^3 WM. Seal resistances were typically 2^4 G6 and series resistance was typically 6 20 M6, uncompensated. Series resistance, measured from brief voltage steps (5 mV, 5 ms) applied through the recording pipette, was monitored at several times during the recording prior to, during, and after drug application. Recordings in which a s 10% change in series resistance was measured during drug application were excluded from the analysis. In some cases tetrodotoxin (TTX; 2 WM) was used to block action potential-dependent synaptic activity. Bicuculline methiodide (10^30 WM) and 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX; 10 WM) were used in some recordings to block GABAA receptor-mediated inhibitory postsynaptic currents (IPSCs) and AMPA/kainate receptor-mediated excitatory postsynaptic currents (EPSCs), respectively (all chemicals from Sigma). Changes in whole-cell slope conductance were assessed using a ramp stimulus protocol which ¢rst stepped to 3100 mV for 500 ms and then linearly depolarized the membrane to +20 mV over the course of 16 s. Responses to ¢ve such voltage ramps were averaged before, during, and after drug application. Peptide-induced changes in resting membrane potential or holding current were determined by comparing the membrane potential at the x-axis crossing point and/or by periodically monitoring the voltage at which no current was measured (i.e. removing voltage-clamp control of the neuron by switching to I = 0) during the recording. Electrical stimulation (300 Ws, 0.1 Hz) of the tractus solitarius (TS) was performed using a pair of te£on-coated platinum^ iridium wires (75 Wm diameter, V100 Wm tip separation). The criteria used for detecting and measuring evoked and spontaneous postsynaptic currents (PSCs) have been described (Smith et al., 1998). When e¡ects were measured in both EPSCs and IPSCs in the same neuron, the membrane potential was clamped at 370 to 380 mV during EPSC measurements and at 330 to 0 mV during IPSC measurements. At least 2 min (typically 100^ 250 events) of continuous spontaneous PSC activity was measured for each condition. E¡ects of hypocretin on spontaneous PSC frequency and amplitude were analyzed within a recording using the Kolmogorov^Smirnov test; e¡ects on evoked EPSC amplitudes were analyzed using a paired two-tailed Student’s t-test (signi¢cance at P 6 0.05). Numbers are reported as the mean R standard error of the mean. Hcrt2 immunohistochemistry Hypocretin-containing elements in the brainstem were identi¢ed using a polyclonal antibody as previously described (van den Pol et al., 1998). The antiserum was raised against the active peptide, hcrt2, and appeared speci¢c in adsorption tests in the presence of antigen (van den Pol et al., 1998; van den Pol, 1999). Similar staining patterns were found with antisera against preprohypocretin. In brief, animals were perfused transcardially with 4% paraformaldehyde in 0.15 M NaPO4 bu¡er (pH = 7.3), post-¢xed 4^8 h at 4‡C, embedded in 30% sucrose in 0.01 M phosphate-bu¡ered saline (PBS; pH = 7.4) until they had equilibrated, and sectioned at nominally 30 Wm on the freezing stage of a sliding microtome. After several rinses in PBS, the antibody was applied at a concentration of 1:5000 in PBS containing 0.1% Triton X-100; overnight at 4‡C. The sections were subsequently reacted with a biotin-conjugated secondary antibody (IgG; Vector Laboratories, Burlingame, CA, USA; 1:200; 4^8 h, 20‡C). Hcrt2-immunoreactive ¢bers were identi¢ed using the ABC method (ABC Elite; 1:100, 2 h to overnight; Vector Laboratories). After rinsing again in PBS (three times 5 min), the labeled neurons were visualized with diaminobenzidine (DAB) at a concentration of 0.06% with 0.003% H2 O2 in 0.01 M PBS. The slices were then mounted on slides (Superfrost/ Plus, Fisher Scienti¢c), air dried overnight, then dehydrated in alcohol and covered in permount. Images of individual sections were obtained using a Spot RT CCD camera (Diagnostic Instru-

NSC 5854 1-11-02

Hypocretin modulation of NTS neurons

709

Fig. 1. Example of the distribution of hcrt2-immunoreactive ¢bers in the NTS. Three levels of the brainstem are shown from caudal (A) to rostral (E) at low power (A, C, and E). For each level, the boxed area is shown at higher power to the right (B, D, and F). Scale bar = 100 Wm for low power and 25 Wm for higher power images. AP, area postrema ; cc, central canal; ts, tractus solitarius; XII, hypoglossal nucleus.

ments) mounted on a Leica DMLB microscope. Except for matching brightness of individual panels for aesthetics post hoc using Photoshop v5.5 (Adobe), images shown are unaltered.

RESULTS

Hcrt2-immunoreactive ¢bers in the caudal NTS Concentrations of hcrt2-immunoreactive ¢bers were observed throughout the rostro-caudal extent of the caudal NTS, especially in the medial and dorsomedial divisions of the nucleus (Fig. 1). Relative to nearby areas such as area postrema, nucleus gracilis, and the hypoglossal nucleus, ¢bers in these regions of the NTS appeared to be dense. No neuronal or glial somata were labeled by the antibody in the NTS. Staining was restricted to ¢bers and terminals. Immunoreactive axons had both small and large boutons, suggestive of presynaptic endings where hypocretin might be released. Electrophysiological recordings were made from neurons in the medial and dorsomedial divisions of the NTS in slices at the same rostro-caudal levels of the nucleus shown in Fig. 1 (i.e. corresponding to rostro-caudal NTS levels within 600 Wm of the area postrema). The regions from which recordings were made were highly innervated by hcrt2-immunoreactive axons.

Postsynaptic responses in NTS neurons The e¡ects of hcrt2 on membrane potential were examined in whole-cell patch-clamp recordings from 37 caudal NTS neurons. Hcrt2 depolarized 10 cells (5 R 2 mV), hyperpolarized two cells (4, 5 mV), and had no e¡ect on membrane potential in nine of 21 neurons recorded using Kþ in the pipette. Steady-state current evoked by a depolarizing ramp revealed a small and variable reduction in whole-cell conductance underlying the depolarization in ¢ve of these neurons (Fig. 2). The reversal potential for the hcrt2 e¡ect in these cells was also inconsistent, ranging from 380 to 345 mV. When Csþ was used as the primary cation carrier, a depolarization (4 mV) was observed in only one of 16 neurons (Fig. 2), with the remaining 15 neurons being una¡ected (0.1 R 0.07 mV; P s 0.05). These data suggested that hcrt2 depolarized some NTS neurons by a Csþ -sensitive mechanism, perhaps involving reduction of a Kþ current. Blockade of EPSCs and IPSCs by amino acid receptor antagonists Synaptic input to most NTS neurons was recorded at holding potentials which allowed separation of outward and inward currents, based both on direct measurements and previously published results in these neurons (e.g.

NSC 5854 1-11-02

710

B. N. Smith et al.

increased by 17% in another cell, with amplitude in the remaining 15 neurons deviating by 6 5% (mean = 1 R 0.25%; P s 0.05). Hcrt2 (1 WM) reversibly increased the amplitude of EPSCs evoked by electrical stimulation of the TS in four of 11 neurons (Fig. 4), having no signi¢cant e¡ect on TS-evoked EPSCs in the remaining cells. In those four neurons, mean amplitude of evoked EPSCs was increased by 39 R 7% (P 6 0.05). E¡ects of hcrt2 on evoked and spontaneous EPSCs were observed in both Kþ and Csþ -loaded neurons, suggesting that the e¡ects were not simply due to a change in resistance in the postsynaptic membrane of the recorded cell, which could conceivably in£uence the detection limit for the smallest sEPSCs. Rather, they appeared to be due to activity of the peptide either on other neurons in the slice (i.e. that may have been depolarized by hcrt2) or on terminals of a¡erent neurons, including terminals of visceral a¡erent ¢bers. Fig. 2. E¡ect of hcrt2 on ramp conductance and membrane potential. (A) Linear range of the current response of an NTS neuron to a slow (16 s) voltage ramp from 3100 mV to +20 mV. Hcrt2 applied to the neuron resulted in a decrease in the whole-cell slope conductance and a small depolarization (black trace). The reversal for the e¡ect in this neuron was near 360 mV (double arrowheads). Averages of ¢ve consecutive ramps are shown for each condition ; arrow indicates the control (gray) trace. The e¡ect was reversible, but the trace is omitted from the ¢gure for clarity. (B) Average hcrt2-induced change in membrane potential (measured at nominally 0 pA holding current) for NTS neurons recorded with Kþ (n = 21) or Csþ (n = 16) as the primary cation carrier in the pipette. The number of neurons that were depolarized by s 2 mV under each condition is indicated next to each bar.

Smith et al., 1998). Fast inward postsynaptic currents were observed in all NTS neurons. The amplitude of these currents was generally larger at more negative potentials and they were blocked by 10 WM CNQX (n = 5; Fig. 3). They were therefore considered to be glutamatergic EPSCs. At membrane potentials positive to about 350 mV, fast outward postsynaptic currents were observed in all neurons. These currents were well resolved at holding potentials positive to 330 mV, especially when Csþ was used as the primary cation carrier in the recording pipette to reduce membrane noise (Smith et al., 1998). These synaptic currents were blocked by 30 WM bicuculline methiodide (n = 5; Fig. 3) and were therefore considered to be GABAA receptor-mediated IPSCs. E¡ects of hcrt2 on synaptic currents The e¡ects of hcrt2 on synaptic input were examined to determine if overall synaptic regulation was altered in the caudal NTS. The frequency of spontaneous EPSCs (sEPSCs) was reversibly increased by hcrt2 in 10 of 17 neurons (Fig. 4). In those 10 neurons, 1 WM hcrt2 increased the frequency of sEPSCs by 22^206%, with mean frequency being increased by 92 R 13% (P 6 0.05). No e¡ect on sEPSC frequency was observed in the remaining seven neurons (30.9 R 0.1%; P s 0.05). The amplitude of sEPSCs was not signi¢cantly altered by hcrt2, being decreased by 18% in one neuron and

E¡ects of hcrt2 on miniature synaptic currents To determine if the increase in EPSC frequency elicited by hcrt2 was due to increased action potential ¢ring in local a¡erent neurons, the peptide was applied to ¢ve cells in the presence of TTX, two of which previously responded to hcrt2 in normal ACSF with an apparent increase in sEPSC frequency. The frequency of action potential-independent miniature EPSCs (mEPSCs) was enhanced by hcrt2 in TTX (2 WM) in ¢ve of ¢ve neurons (Fig. 5). Application of 1 WM hcrt2 increased mEPSC frequency by 16^49%, with mean frequency being increased by 26 R 6% (P 6 0.05). Amplitude of mEPSCs was unchanged by hcrt2 (35 R 6%; P s 0.05). The increased frequency in the absence of a change in amplitude further suggested that hcrt2 acted at receptors located on terminals of glutamatergic neurons. E¡ects of hcrt2 on spontaneous IPSCs (sIPSCs) were examined in eight NTS neurons, including four cells in which sEPSC frequency was increased by the peptide. Overall, neither sIPSC frequency (13 R 4.6%; P s 0.05) nor amplitude (0 R 0.05%; P s 0.05) was signi¢cantly altered by the peptide, even during the same application in which enhanced sEPSC activity was observed. In two cases, sIPSC frequency was mildly increased by hcrt2 (33 and 15%). In these neurons, TTX prevented the increase in mIPSC frequency during a second application, even when mEPSC frequency was enhanced in the same hcrt2 application to the same neuron (Fig. 5). These data suggested that hcrt2 had little direct e¡ect on inhibitory synaptic input to NTS neurons, and supported the hypothesis that the peptide selectively enhances excitatory synaptic input to a subset of caudal NTS neurons by acting, at least in part, at receptors located on presynaptic terminals.

DISCUSSION

Numerous anatomical, behavioral, and physiological studies have suggested that the hypocretins can modify autonomic function, sensory integration, arousal, and

NSC 5854 1-11-02

Hypocretin modulation of NTS neurons

711

Fig. 3. E¡ects of amino acid receptor antagonists on sIPSCs and sEPSCs in NTS neurons. (A) A segment of sIPSCs (a1) recorded at 310 mV and sEPSCs (a2) recorded at 370 mV. Inset: Arrow points to boxed area in a2 (500 ms) shown temporally expanded. (B) Recordings at the same holding potentials as in (A) in the presence of the glutamate AMPA/kainate receptor antagonist, CNQX (10 WM). In the top traces (b1), sIPSCs remain evident at a holding potential of 310 mV, whereas sEPSCs are absent when the cell was voltage-clamped at 370 mV (b2). Inset : Arrow points to boxed area in b2 (500 ms) shown temporally expanded. No sEPSCs were observed in the presence of CNQX. All traces in (A) and (B) are from the same neuron. (C) A segment of spontaneous synaptic activity recorded at a holding potential of 325 mV. Both sEPSCs and sIPSCs were evident at this holding potential. Inset : Boxed area in (C) (500 ms) shown temporally expanded. (D) The sIPSCs were blocked by bicuculline (30 WM), whereas the sEPSCs remained evident. Inset: Boxed area in (D) (500 ms) shown temporally expanded. Traces in (C) and (D) are from the same neuron. Continuous traces are shown for each data set; Csþ was used as the primary intracellular cation carrier in both recordings.

energy homeostasis (Mieda and Yanagisawa, 2002; Sutcli¡e and De Lecea, 2002). The present results indicate that hcrt2 ¢bers richly innervate the caudal NTS and that the peptide has signi¢cant e¡ects on excitatory, but not inhibitory, fast synaptic activity in NTS neurons. Together with the small depolarization of some NTS neurons, the selective nature of the e¡ects of hcrt2 suggest an important role for the peptide in regulating integration of visceral sensory signals in the caudal NTS, a brainstem site that is critically involved in processing autonomic information. Hcrt2-immunoreactive ¢bers in the caudal NTS Behavioral and physiological studies have implicated the hypocretins in the regulation of feeding behaviors and cardiorespiratory function, acting at the level of neurons in the caudal brainstem (Takahashi et al., 1999; Chen et al., 2000; Krowicki et al., 2002). The positions of neurons recorded in this study corresponded with the locations of hcrt2 ¢bers and of gastric- and cardiovascular-related visceral a¡erent ¢bers (Shapiro and Miselis, 1985; Spyer et al., 1984), further supporting the hypothesis that the hypocretins are positioned to in£uence autonomic behaviors. Previous studies indicated that hcrt

¢bers were present in the vagal complex (Peyron et al., 1998; Date et al., 1999; Nambu et al., 1999), but did not emphasize the distribution of hcrt2 ¢bers within the vagal complex or the rostro-caudal distribution of the ¢bers in the caudal NTS. The concentration of hcrt2 ¢bers we observed in medial and dorsomedial NTS regions is consistent with the hypothesized involvement of the peptide in viscerosensory processing. Hcrt2-induced membrane depolarization Hcrt2 caused a small depolarization in a subset of NTS neurons. A postsynaptic decrease in resting Kþ current has been observed in other neurons in response to hypocretin (Eggermann et al., 2001; Hwang et al., 2001; Shirasaka et al., 2001). Since the depolarization in NTS neurons was sensitive to Csþ , a decreased Kþ current may have contributed to the depolarization observed in some NTS neurons and could therefore have complicated the analysis of EPSC amplitude. However, the reversal potential for the hcrt2 e¡ect was positive to the expected reversal for a purely Kþ -mediated e¡ect and was not consistent from neuron to neuron. Modulation of a mixed cation current (Hwang et al., 2001) or electrogenic pump (Eriksson et al., 2001) have

NSC 5854 1-11-02

712

B. N. Smith et al.

Fig. 4. E¡ect of hcrt2 on excitatory input. (A) Hcrt2 increased sEPSC frequency. (A1) Control ACSF. (A2) Hcrt2 (1 WM). (A3) Wash to control ACSF (20 min). Vm = 378 mV. (B) Cumulative frequency plot indicates hcrt2 increased sEPSC frequency (P 6 0.0001). (C) Hcrt2 (1 WM) increased the amplitude of the EPSC evoked by electrical stimulation of the TS. Evoked IPSCs were not observed in this neuron. Arrows point to the peaks of averaged EPSCs (n = 7). Labels for overlapped traces are c, control solution ; hcrt2, hypocretin ; w, wash (15 min). Pipettes contained Csþ in both recordings.

also been suggested as mechanisms underlying the postsynaptic depolarization caused by hypocretin receptor activation. These mechanisms may also be involved in the hcrt2-mediated depolarization of a subset of NTS neurons in the present study. In light of our data suggesting that the depolarization was Csþ -sensitive, changes in other cation currents or pumps would presumably occur in conjunction with a change in Kþ conductance. The phenotype and connections of the subset of NTS neurons that were depolarized by hcrt2 are not known. However, the relatively small number of responding neurons suggests that hcrt2 may a¡ect distinct components of viscerosensory processing in the NTS rather than acting as a non-speci¢c ‘excitatory’ molecule. E¡ects of hcrt2 on synaptic currents The most consistent e¡ect of hcrt2 in the NTS was the enhancement of EPSC frequency. These data support the hypothesis that the peptide modi¢es local and/or visceral a¡erent excitatory inputs to most, but not all, NTS neurons. Glutamate is the predominant neurotransmitter released by both visceral a¡erents and local excitatory neurons in the NTS (e.g. Andresen and Yang, 1990; Smith et al., 1998), and this is supported by results of amino acid receptor antagonists in these studies. The e¡ect of hcrt2 on evoked EPSC amplitude and sEPSC frequency may have been due to enhanced glutamate release from terminals or an increase in e¡ect of postsynaptic glutamate receptors. Activation of hypocretin receptors that are closely coupled to postsynaptic glutamate receptors that are activated by vagal input could enhance the amplitude of glutamatergic EPSCs, regard-

less of changes in whole-cell conductance. This could be an important confound if the receptors are located on dendrites, which may be electrotonically distant from the presumably somatic recording site. However, the lack of consistent e¡ect on sEPSC amplitude argues against the hypothesis that hcrt2 acts speci¢cally at receptors coupled to a postsynaptic glutamate receptor. The increase in sEPSC frequency also appeared to be independent of the decrease in whole-cell conductance seen in some neurons, since EPSCs were enhanced in neurons that were not depolarized by hcrt2 and/or were loaded with Csþ , which putatively blocked Kþ currents throughout the neuron and e¡ectively isolated e¡ects of hcrt2 to those occurring on a¡erent neurons and terminals. Further, at least a portion of the e¡ect on EPSC frequency was maintained in the absence of action potential-dependent synaptic release in the presence of TTX, suggesting activity at hypocretin receptors on glutamatergic axon terminals. Since NTS neurons are a heterogeneous set of cells that may di¡erentially express receptors on dendrites, somata, or axon terminals, it remains possible that a portion of the enhancement of vagal and other glutamatergic inputs are speci¢cally modulated at the site of the synapse. Regardless of the depolarizing e¡ects on some NTS neurons, the present data indicate that hcrt2 can act presynaptically to enhance glutamate release at some synapses in the NTS, perhaps including glutamate released from terminals of primary visceral a¡erent processes. Enhancement of sEPSC frequency was observed in the absence of an e¡ect on sIPSCs, a signi¢cant di¡erence from most previous reports of hcrt activity in other brain regions. The receptors bound by hypocretin are thought to increase Ca2þ in£ux and enhance GABAergic and

NSC 5854 1-11-02

Hypocretin modulation of NTS neurons

713

Fig. 5. E¡ect of hcrt2 on miniature IPSCs and EPSCs in an NTS neuron. (A) In the presence of TTX (2 WM) action potential-independent miniature IPSCs (mIPSCs) were observed at a holding potential of 310 mV. (B) Application of hcrt2 had little e¡ect on mIPSC frequency. (C) Washout of hypocretin to TTX-containing ACSF. (D) In the same neuron (Vm = 375 mV), mEPSCs were observed in the presence of TTX. (E) Hcrt2 increased the frequency of mEPSCs. Same hcrt2 application as in (B). (F) Partial reversal of the e¡ect after V10 min wash to TTX-containing ACSF. (G) Cumulative frequency plot indicates hcrt2 increased mEPSC frequency (P 6 0.0001). Pipette contained Csþ .

glutamatergic synaptic transmission in some hypothalamic and brainstem neurons (van den Pol et al., 1998). In the two cases where IPSC frequency was slightly increased by hcrt2 in the present analysis, the e¡ect was prevented in further applications by blocking action potentials with TTX, suggesting that the enhancement of IPSCs in those cells arose from increased activity (either depolarization or synaptically induced) in an intact local GABAergic neuron. The e¡ect on IPSCs is congruent with recent reports indicating selective enhancement of EPSCs in tegmental neurons (Burlet et al., 2002) or of IPSCs in the dorsal horn of the spinal cord (Grudt et al., 2002), both of which were predominantly due to postsynaptic depolarization of local neurons. Unlike those studies, the present results indicate that a hcrt2 enhances EPSCs in NTS by increasing glutamate (but not GABA) release, at least in part by acting at receptors on axon terminals. Regardless of the receptor location(s), the relative speci¢city for modulation of excitatory input suggests that the peptide acts in a functionally relevant manner in the NTS to increase responsiveness to selected inputs. Hcrt2 function in the NTS Consistent with the hypothesis that hcrt2 modi¢es sensory input (van den Pol, 1999; Grudt et al., 2002), the

present data suggest that the peptide plays a role in modulating viscerosensory information. This may include, for example, blood pressure information arising from cardiac baroreceptors or satiety signals arising from the stomach. Understanding how hcrt2 regulates synaptic circuitry in the NTS will be important for interpreting results from behavioral studies relating to the reported e¡ects of the hypocretins on gastrointestinal or cardiorespiratory regulation, as well as other aspects of viscerosensory integration. The hypocretins have primarily excitatory e¡ects in several hypothalamic and brainstem regions (van den Pol et al., 1998; Horvath et al., 1999; Eriksson et al., 2001; Hwang et al., 2001). The present ¢ndings indicating a hcrt2-mediated enhancement of excitatory neurotransmission in the NTS are consistent with the hypothesis that the hypocretin system participates in integrating neural systems involved in energy homeostasis, autonomic function, arousal, and sensory processing.

Acknowledgements+This research was supported by funds from the National Science Foundation (B.N.S. and A.N.v.d.P.), the American Heart Association (B.N.S.), and the Louisiana Board of Regents (B.N.S.). We thank D. Liu for technical assistance and to Dr. F.E. Dudek for his contribution to preliminary aspects of the study.

NSC 5854 1-11-02

714

B. N. Smith et al. REFERENCES

Andresen, M.C., Yang, M.Y., 1990. Non-NMDA receptors mediate sensory a¡erent synaptic transmission in medial nucleus tractus solitarius. Am. J. Physiol. 259, H1307^1311. Burlet, S., Tyler, C.J., Leonard, C.S., 2002. Direct and indirect excitation of laterodorsal tegmental neurons by Hypocretin/Orexin peptides : implications for wakefulness and narcolepsy. J. Neurosci. 22, 2862^2872. Chemelli, R.M., Willie, J.T., Sinton, C.M., Elmquist, J.K., Scammell, T., Lee, C., Richardson, J.A., Williams, S.C., Xiong, Y., Kisanuki, Y., Fitch, T.E., Nakazato, M., Hammer, R.E., Saper, C.B., Yanagisawa, M., 1999. Narcolepsy in orexin knockout mice: molecular genetics of sleep regulation. Cell 98, 437^451. Chen, C.T., Hwang, L.L., Chang, J.K., Dun, N.J., 2000. Pressor e¡ects of orexins injected intracisternally and to rostral ventrolateral medulla of anesthetized rats. Am. J. Physiol. 278, R692^R697. Date, Y., Ueta, Y., Yamashita, H., Yamaguchi, H., Matsukara, S., Kangawa, K., Sakurai, T., Yanagisawa, M., Nakazato, M., 1999. Orexins, orexigenic hypothalamic peptides, interact with autonomic, neuroendocrine and neuroregulatory systems. Proc. Natl. Acad. Sci. USA 96, 748^ 753. De Lecea, L., Kildu¡, T.S., Peyron, C., Gao, X.-B., Foye, P.E., Danielson, P.E., Fukuhara, C., Battenberg, E.L.F., Gautvik, V.T., Bartlett, F.S., Frnkel, W.N., van den Pol, A.N., Bloom, F.E., Gautvik, K.M., Sutcli¡e, J.G., 1998. The hypocretins: hypothalamus-speci¢c peptides with neuroexcitatory activity. Proc. Natl. Acad. Sci. USA 95, 322^327. Dube, M.G., Kalra, S.P., Kalra, P.S., 1999. Food intake elicited by central administration of orexins/hypocretins : identi¢cation of hypothalamic sites of action. Brain Res. 842, 473^477. Eggermann, E., Sera¢n, M., Bayer, L., Machard, D., Saint-Mleux, B., Jones, B.E., Muhlethaler, M., 2001. Orexins/hypocretins excite basal forebrain cholinergic neurones. Neuroscience 108, 177^181. Eriksson, K.S., Sergeeva, O., Brwon, R.E., Haas, H.L., 2001. Orexin/hypocretin excites the histaminergic neurons of the tuberomammillary nucleus. J. Neurosci. 21, 9273^9279. Grudt, T.J., van den Pol, A.N., Perl, E.R., 2002. Hypocretin-2 (orexin-B) modulation of super¢cial dorsal horn activity in rat. J. Physiol. (Lond.) 538, 517^525. Horvath, T.L., Peyron, C., Diano, S., Ivanov, A., Aston-Jones, G., Kildu¡, T.S., van den Pol, A.N., 1999. Hypocretin (orexin) activation and synaptic innervation of the locus coeruleus noradrenergic system. J. Comp. Neurol. 415, 145^159. Hwang, L.L., Chen, C.T., Dun, N.J., 2001. Mechanisms of orexin-induced depolarizations in rat dorsal motor nucleus of vagus neurones in vitro. J. Physiol. (Lond.) 537, 511^520. Kalia, M., Sullivan, J.M., 1982. Brainstem projections of sensory and motor components of the vagus nerve in the rat. J. Comp. Neurol. 211, 248^ 265. Krowicki, Z.K., Burmeister, M.A., Berthoud, H.-R., Scullion, R.T., Fuchs, K., Hornby, P.J., 2002. Orexins in rat dorsal motor nucleus of the vagus potently stimulate gastric motor function. Am. J. Physiol. Gastrointest. Liver Physiol. 283, G465^G472. Lin, L., Faraco, J., Li, R., Kadotani, H., Rogers, W., Lin, X., Qiu, X., de Jong, P.J., Nishino, S., Mignot, E., 1999. The sleep disorder canine narcolepsy is caused by a mutation in the hypocretin (orexin) receptor 2 gene. Cell 98, 365^376. Mieda, M., Yanagisawa, M., 2002. Sleep, feeding, and neuropeptides: roles of orexins and orexin receptors. Curr. Opin. Neurobiol. 12, 339^345. Nambu, T., Sakurai, T., Mizukami, K., Hosoya, Y., Yanagisawa, M., Goto, K., 1999. Distribution of orexin neurons in the adult rat brain. Brain Res. 827, 243^260. Peyron, C., Tighe, D.K., van den Pol, A.N., de Lecea, L., Heller, H.C., Sutcli¡e, J.G., Kildu¡, T.S., 1998. Neurons containing hypocretin (orexin) project to multiple neuronal systems. J. Neurosci. 18, 9996^10015. Sakurai, T., Amemiya, A., Ishii, M., Matsuzaki, I., Chemelli, R.M., Tanaka, H., Williams, S.C., Richardson, J.A. et al., 1998. Orexins and orexin receptors : A family of hypothalamic neuropeptides and G protein-coupled receptors that regulate feeding behavior. Cell 92, 573^585. Samson, W.K., Taylor, M.M., Follwell, M., Ferguson, A.V., 2002. Orexin actions in hypothalamic paraventricular nucleus: physiological consequences and cellular correlates. Regul. Pept. 104, 97^103. Shapiro, R.E., Miselis, R.R., 1985. The central organization of the vagus nerve innervating the stomach of the rat. J. Comp. Neurol. 238, 473^488. Shirasaka, T., Miyahara, S., Kunitake, T., Quing-Hua, J., Kato, K., Takahashi, M., Kannan, H., 2001. Orexin depolarizes rat hypothalamic paraventricular nucleus neurons. Am. J. Physiol. 281, R1114^R1118. Shirasaka, T., Nakazato, M., Matsukura, S., Takasaki, M., Kannan, H., 1999. Sympathetic and cardiovascular actions of orexins in conscious rats. Am. J. Physiol. 277, R1780^R1785. Smith, B.N., Dou, P., Barber, W.D., Dudek, F.E., 1998. Vagally-evoked synaptic activity in the immature rat nucleus tractus solitarius in an intact in vitro preparation. J. Physiol. (Lond.) 512, 149^162. Spyer, K.M., Donoghue, S., Felder, R.B., Jordan, D., 1984. Processing of a¡erent inputs in cardiovascular control. Clin. Exp. Hypertens. A 6, 173^184. Sutcli¡e, J., De Lecea, L., 2002. The hypocretins: Setting the arousal threshold. Nat. Rev. 3, 339^349. Takahashi, N., Okumura, T., Yamada, H., Kohgo, Y., 1999. Stimulation of gastric acid secretion by centrally administered orexin-A in conscious rats. Biochem. Biophys. Res. Commun. 254, 623^627. van den Pol, A.N., 1999. Hypothalamic hypocretin (orexin) : Robust innervation of the spinal cord. J. Neurosci. 19, 3171^3182. van den Pol, A.N., Gao, X.-B., Obrietan, K., Kildu¡, T.S., Belousov, A.B., 1998. Presynaptic and postsynaptic actions and modulation of neuroendocrine neurons by a new hypothalamic peptide, hypocretin/orexin. J. Neurosci. 18, 7962^7971. (Accepted 26 July 2002)

NSC 5854 1-11-02