hypocretins on pedunculopontine tegmental neurons in rats: An in vitro study

peptides 30 (2009) 191–209

available at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/peptides

Electrophysiological effects of orexins/hypocretins on pedunculopontine tegmental neurons in rats: An in vitro study Juhyon Kim a, Kazuki Nakajima a, Yutaka Oomura b, Matthew J. Wayner c, Kazuo Sasaki a,* a

Division of Bio-Information Engineering, Faculty of Engineering, University of Toyama, 3190 Gofuku, Toyama 930-8555, Japan Department of Integrative Physiology, Graduate School of Medical Science, Kyushu University, Fukuoka 812-0054, Japan c Department of Biology, The University of Texas at San Antonio, San Antonio, TX 78249-0662, USA b

article info

abstract

Article history:

Orexin-A (ORX-A) and orexin-B (ORX-B) play critical roles in the regulation of sleep–

Received 13 June 2008

wakefulness and feeding. ORX neurons project to the pedunculopontine tegmental nucleus

Received in revised form

(PPT), which regulates waking and rapid eye movement (REM) sleep. Thus, we examined

25 September 2008

electrophysiological effects of ORXs on rat PPT neurons with a soma size of more than

Accepted 25 September 2008

30 mm. Whole cell patch clamp recording in vitro revealed that ORX-A and ORX-B depolar-

Published on line 14 October 2008

ized PPT neurons dose-dependently in normal and/or tetrodotoxin containing artificial cerebrospinal fluids (ACSFs), and the EC50 values for ORX-A and ORX-B were 66 nM and

Keywords:

536 nM, respectively. SB-334867, a selective inhibitor for ORX 1 (OX1) receptors, significantly

Orexin/hypocretin

suppressed the ORX-A-induced depolarization. The ORX-A-induced depolarization was

Pedunculopontine tegmental

reduced in high K+ ACSF with extracellular K+ concentration of 13.25 mM or N-methyl-D-

nucleus

glucamine (NMDG+)-containing ACSF in which NaCl was replaced with NMDG-Cl, and

Ascending arousal system

abolished in high K+-NMDG+ ACSF or in a combination of NMDG+ ACSF and recordings

Cholinergic neurons

with Cs+-containing pipettes. An inhibitor of Na+/Ca2+ exchanger and chelating intracellular

Sleep/wakefulness

Ca2+ had no effect on the depolarization. Most of PPT neurons studied were characterized by

REM sleep

an A-current or both A-current and a low threshold Ca2+ spike, and predominantly choli-

SB-334867

nergic. These results suggest that ORXs directly depolarize PPT neurons via OX1 receptors and via a dual ionic mechanism including a decrease of K+ conductances and an increase of non-selective cationic conductances, and support the notion that ORX neurons affect the activity of PPT neurons directly and/or indirectly to control sleep–wakefulness, especially REM sleep. # 2008 Elsevier Inc. All rights reserved.

1.

Introduction

Novel neuropeptides, orexin-A (ORX-A) and orexin-B (ORX-B), also called hypocretin-1 and hypocretin-2, respectively, were newly identified in the perifornical region of the lateral hypothalamic area (LHA) [33,37]. The ORX binds to orexin 1 (OX1) and orexin 2 (OX2) receptors which belong to the G

protein-coupled receptor superfamily. OX1 receptors have a higher affinity for ORX-A than for ORX-B, whereas OX2 receptors have an equal affinity for both ORX-A and ORX-B. The ORX was initially recognized as a regulator of feeding behavior because of the exclusive expression in the LHA known as the feeding center and the increase of food intake induced by intracerebroventricular administration of ORXs

* Corresponding author. Tel.: +81 76 445 6719; fax: +81 76 445 6723. E-mail address: [email protected] (K. Sasaki). 0196-9781/$ – see front matter # 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.peptides.2008.09.023

192

peptides 30 (2009) 191–209

[37]. However, subsequent studies demonstrated that ORXs plays a crucial role in the regulation of sleep and wakefulness. For example, Lin et al. [27] found that canine narcolepsy, that is characterized by daytime sleepiness, cataplexy and the striking transition from wakefulness into rapid eye movement (REM) sleep, is caused by mutation of the OX2 receptor gene. It is also reported that ORX knockout mice exhibit a phenotype strikingly similar to human narcolepsy patients as well as canarc-1 mutant dogs [5]. These results suggest that the intact ORX signaling system is required to maintain a proper wakefulness of animals including humans for survival [34], and deficiencies of ORX signaling are associated with the narcolepsy and cataplexy. Sleep and wakefulness is regulated by dynamic interactions between multiple neurochemically distinct systems in the brain. Wakefulness is promoted by an ascending arousal system consisting of cholinergic and monoaminergic neurons in the forebrain, midbrain and brainstem, whereas the initiation and maintenance of REM sleep and non-REM (NREM) sleep are controlled by groups of neurons in the brainstem and preoptic area of the hypothalamus [16,31,39]. Cholinergic neurons in the pedunculopontine tegmental nucleus (PPT) and laterodorsal tegmental nucleus (LDT) of the brainstem are involved in not only the maintenance of wakefulness but also the generation of REM sleep [6,25,29,35,44,45]. Indeed, electrical stimulation of the PPT caused activation of the cortical electroencephalogram (EEG), i.e., EEG arousal [41], whereas extensive loss of cholinergic mesopontine neurons by kainic acid was associated with loss of REM sleep and absent or incomplete neck muscle atonia [20,49]. In addition, many cholinergic neurons in the PPT and LDT fire rapidly during wakefulness and REM sleep, whereas they are inactive during NREM sleep [6,8,36,40]. A distinct subpopulation of PPT and LDT cholinergic neurons also specifically increases their electrical activity just prior to and during REM sleep [6,8,40]. Extracellular recording studies demonstrate that cholinergic neurons in the LDT and basal forebrain (BF) that constitute the ascending arousal system are excited by ORXs [4,7,42]. Whole cell patch clamp recording studies indicate that the excitatory effect of ORXs in LDT cholinergic neurons is both presynaptic and postsynaptic, and mediated postsynaptically by an activation of noisy cation channels [4,24]. A recent in vitro study also shows that septohippocampal cholinergic neurons which contribute to hippocampal arousal are depolarized postsynaptically through a dual ionic mechanism including an inhibition of K+ channels and an activation of Na+/Ca2+ exchanger (NCX) [51]. In contrast to these cholinergic neurons, however, the effects of ORXs on PPT cholinergic neurons have not been described. Therefore, the purpose of the present study was to examine the electrophysiological effects of ORXs on putative PPT cholinergic neurons with a soma size of more than 30 mm in rat brain slices, and to determine the ionic mechanism involved. PPT neurons that were responsive and non-responsive to ORXs were also characterized by the electrophysiological membrane properties, and some of them were labeled with biocytin and then stained with nicotinamide adenine dinucleotide phosphatediaphorase (NADPH-d), a reliable marker for mesopontine cholinergic neurons [26,47,48].

2.

Materials and methods

2.1.

Animals

Male Wistar rats 1–2 weeks of age were used (Sankyo Lab., Shizuoka, Japan). They were housed with their mothers in a light-controlled room (light on: 06:00–18:00) at a temperature of 23  2 8C for several days before the experiments. The animals and experimental procedures used were approved by the Institutional Animal Care and Use Committee of the University of Toyama.

2.2.

Slice preparations

Rats were decapitated after sevoflurane anesthesia and the brain was rapidly removed from the skull. The brain was then submerged in ice cold, oxygenated (95% O2–5% CO2) normal artificial cerebrospinal fluid (ACSF), composition in mM: NaCl 126, KCl 3, CaCl2 2.4, MgSO4 1.3, KH2PO4 1.25, NaHCO3 26 and glucose 10 with a pH of 7.4. Coronal brainstem slices 300 mm thick were cut by a microslicer (ZERO 1, Dosaka EM, Kyoto, Japan). The PPT was identified on the basis of its anatomical location adjacent to the superior cerebellar peduncle. Two or three slices including the PPT were selected from each animal, and cut with a scalpel along the midline so that two PPT slices, each including the left and right side PPT, were obtained from one coronal slice. The slices containing the PPT were then preincubated in a chamber with oxygenated normal ACSF for about 1 h at room temperature.

2.3.

Whole cell patch clamp recording

After preincubation, slices were transferred into a whole cell patch clamp recording chamber fixed to the stage of an upright microscope (BX-50WI, Olympus, Tokyo, Japan). The recording chamber was perfused with oxygenated normal ACSF at 1 ml/ min and at 34 8C. PPT neurons were visualized on a television screen through an infrared charge coupled device (CCD) camera (C2741-79, Hamamatsu Photonics, Hamamatsu, Japan) and a real-time digital video microscopy processor (XL-20, Olympus, Tokyo, Japan). Electrodes were filled with a standard internal pipette solution containing in mM: K-gluconate 120, KCl 20, HEPES 10, MgCl2 2.0, CaCl2 0.5, EGTA 1.0, Na-ATP 4.6, and Na-GTP 0.4, pH adjusted to 7.3 with KOH, and electrode resistances were 5–9 MV. Neurons were recorded in ‘‘I-clamp normal’’ current clamp or voltage clamp modes using a patch clamp amplifier (Axopatch 200B, Axon Instruments, Union City, CA, USA). An Ag/AgCl reference electrode was placed near the intermediate position between the inlet and outlet of the chamber. Membrane potentials and currents recorded via the electrodes were fed into the amplifier. Series resistance compensation and capacitive compensation were performed as much as possible by the amplifier. The output of the amplifier was digitized using an A/D converter board (Digidata 1200, Axon Instruments, Union City, USA) with a sampling rate of 10 kHz, and recorded on a hard disk by data acquisition and analysis software (pCLAMP 8, Axon Instruments, Union City, USA). Membrane potentials were low-pass filtered at 5 kHz. Furthermore, membrane potentials less than 60 mV were set to about 60 mV by current injections. Whole cell liquid

peptides 30 (2009) 191–209

junction potentials were calculated to be about 13 mV for our internal solutions, and membrane potentials were not corrected.

2.4.

Histochemistry

Each slice with a biocytin-filled cell was fixed in 4% paraformaldehyde in 0.1 M phosphate-buffered saline (PBS), pH 7.4, for 2 h, and immersed in a solution containing 15% sucrose in 0.1 M PBS for overnight at 4 8C. The sections were rinsed three times for 5 min in 0.1 M PBS containing 0.3% Triton X-100 (PBST, pH 7.4). According to biocytin histochemistry described previously [14], the sections were incubated with avidin-conjugated Texas Red (25 mg/ml) in 0.3% PBST for 3 h at 27 8C, and again rinsed three times for 5 min in 0.3% PBST. Cholinergic neurons were then stained using NADPH-d, a reliable marker of mesopontine cholinergic neurons [26,47,48]. The sections were incubated in a solution containing b-NADPH (1 mg/ml) and nitroblue tetrazolium (0.1 mg/ml) in 0.3% PBST for 1 h at 37 8C. After the sections were rinsed in 0.3% PBST, they were mounted on glass slides, dehydrated and coverslipped. Finally, Texas Red-positive and NADPH-dpositive neurons were identified under a microscope equipped with both epifluorescence and bright-field optics (BX-51, Olympus, Tokyo, Japan).

2.5.

Reagents

ORX-A and ORX-B were purchased from Peptide Institute (Osaka, Japan). SB-334867, an antagonist for OX1 receptors, was a gift from GlaxoSmithKline. Tetrodotoxin (TTX) was purchased from Seikagaku Corporation (Tokyo, Japan), and biocytin, b-NADPH, 1,2-bis(o-aminophenoxy)ethaneN,N,N0 ,N0 -tetraacetic acid (BAPTA), a Ca2+ specific chelator, and dichlorobenzamil (DCB), an inhibitor for a forward- and reverse-mode NCX, were purchased from Sigma (Tokyo, Japan). N-Methyl-D-glucamine (NMDG, Wako Pure Chemical Industries, Tokyo, Japan) was also used.

2.6.

193

Statistics

All data were expressed as means  S.E.M.s. For multiple comparisons, one-way or two-way analysis of variance (ANOVA) followed by a post hoc Fisher’s protected least significant difference (PLSD) test was used. For comparison of two means, Student’s t-test or Wilcoxon signed-rank test were used. P < 0.05 was taken as the level of statistical significance.

3.

Results

To examine the effect of ORXs on PPT neurons, whole cell patch clamp recordings were made in acute slice preparations of rats. In total, 236 PPT neurons with a soma size in the long axis of more than 30 mm were recorded. This criterion was used because an average diameter in the long axis of PPT cholinergic neurons is 20–30 mm in species such as rat, cat, monkey, as well as human [35]. The mean resting potential and mean input resistance measured with patch pipettes containing normal internal solution in normal and/or 0.5 mM TTX-containing ACSFs were 46.5  0.4 mV (n = 186) and 104.4  2.7 MV (n = 193), respectively.

3.1.

ORX-A and ORX-B depolarize PPT neurons

Of 236 PPT neurons, application of ORX-A and/or ORX-B at concentrations equal to or more than 300 nM depolarized 194 neurons (82.2%) in normal and/or TTX-containing ACSFs, whereas the remaining 42 neurons (17.8%) did not respond to these agents. Example records in which ORX-A at 300 nM was applied to a PPT neuron in normal and TTX-containing ACSFs were shown in Fig. 1. Application of ORX-A depolarized the membrane potential of the PPT neuron in normal ACSF, and the depolarization was accompanied by the firing of action potentials (Fig. 1A). The depolarization occurred immediately after the application of ORX-A, and returned to the resting level within about 10 min after the removal of ORX-A. The

Fig. 1 – Excitatory effect of ORX-A on PPT neurons. (A) Depolarization and action potentials induced by application of ORX-A at 300 nM in normal ACSF. (B) Depolarization induced by subsequent application of ORX-A at 300 nM to the same neuron in TTX-containing ACSF (dotted line). Note that action potentials disappeared: (") resting membrane potential of S60 mV. Solid bars, ORX-A at 300 nM.

194

peptides 30 (2009) 191–209

Fig. 2 – Dose-dependent effects of ORX-A and/or ORX-B on depolarization of membrane potentials in modified and SB334867-containing ACSFs. (A and B) Dose-dependent depolarizations in two different neurons induced by applications of ORX-A (A) and ORX-B (B) at 10 nM (left), 100 nM (middle) and 1 mM (right) in modified ACSF. (C and D) Dose-dependent depolarizations in two different neurons induced by applications of ORX-A at 10 nM (left), 100 nM (middle) and 1 mM (right) in 100 nM (C) and 1 mM (D) SB-334867-containing ACSF. (E) Dose–response curves for ORX-A and/or ORX-B in modified and SB-334867-containing ACSFs. The dose–response curve for ORX-A (&, n = 3 for 1 nM, n = 4 for 10 nM, 100 nM and 1 mM, and n = 3 for 5 mM) in modified ACSF is shifted to the left of that for ORX-B (!, n = 4 for 1 nM, 10 nM, 100 nM and 1 mM, and n = 3 for 5 mM and 10 mM). The dose-response curves for ORX-A in 100 nM (*, n = 3 for 1 nM, n = 6 for 10 nM and 100 nM, n = 3 for 1 mM and 5 mM) and 1 mM (~, n = 3 for 10 nM, n = 4 for 100 nM, 1 mM and 5 mM) SB-334867-containing ACSFs were significantly suppressed compared with that in the absence of SB-334867 (&). (F and G) Applications of ORX-A and ORX-B at 300 nM to two different neurons in modified ACSF. The first application was ORX-A (F) or ORX-B (G). (H) Mean amplitudes of the depolarization for ORX-A (open column) and for ORX-B (solid column) (n = 10): (") resting membrane potential of S60 mV. Solid bars, ORX-A or ORX-B at dose indicated. Dotted lines, SB-334867 at dose indicated. *P < 0.05. **P < 0.01. ***P < 0.001.

peptides 30 (2009) 191–209

ORX-A-induced depolarization persisted in 0.5 mM TTX-containing ACSF, although the firing of action potentials did not occur (Fig. 1B). Similar results were obtained in another 4 neurons except that in 3 out of 4 neurons the depolarization was not accompanied by action potentials even in normal ACSF. The mean latency, duration and amplitude of the depolarization to 300 nM ORX-A in normal ACSF were 0.7  0.1 min, 12.7  3.1 min and 5.4  1.2 mV (n = 5), respectively. Corresponding values in TTX-containing ACSF were 0.6  0.2 min, 9.0  0.8 min and 6.3  1.8 mV, respectively. There were no statistically significant differences in these three parameters between normal and TTX-containing ACSFs, suggesting that the depolarization is mediated by a direct postsynaptic action of ORX-A to receptors located on the recorded neurons. In subsequent experiments, TTX-containing ACSF was used to ensure the postsynaptic nature of ORXinduced responses, and hereafter 0.5 mM TTX-containing ACSF will be called modified ACSF. In 132 neurons that were challenged by 300 nM ORX-A in normal or modified ACSFs, the mean latency, duration and amplitude of the depolarization were 1.0  0.1 min, 9.0  0.3 min and 6.4  0.3 mV, respectively. A dose-dependency in the ORX-A- and ORX-B-induced depolarization was determined. Sample records in the dosedependent increase of depolarization induced by ORX-A at 10 nM, 100 nM and 1 mM are shown in the left, middle and right panels of Fig. 2A, respectively. Corresponding records for ORXB are also shown in the left, middle and right panels of Fig. 2B, respectively. In these records, the amplitude of the depolarization induced by ORX-A or ORX-B was concentration-

195

dependent. The dose-response curves for ORX-A (&) or ORX-B (!) at 1 nM, 10 nM, 100 nM, 1 mM, 5 mM and/or 10 mM were shown in Fig. 2E. Both ORX-A and ORX-B caused similar maximal increases in depolarization, but the dose-response curve for ORX-A is shifted to the left of that for ORX-B with the respective EC50 values for these ORXs being 66 nM and 536 nM, respectively. To confirm whether or not the depolarization for ORX-A is greater than that for ORX-B at a concentration between the two EC50 values even on the same neuron, ORX-A and ORX-B at 300 nM were applied to 10 PPT neurons. As shown in example records of left and right panels of Fig. 2F and G, the depolarization for ORX-A was always larger than that for ORX-B independent of the order of application of ORX-A and ORX-B. The difference between the mean amplitudes of the depolarization for ORX-A and ORX-B was statistically significant, P < 0.001 (n = 10 in each group, Fig. 2H). The relation of dose–response curves between ORX-A and ORX-B shown in Fig. 2E suggests a possibility that the ORX receptors expressed in PPT neurons are OX1 receptors, because the affinity of ORX-A to OX1 receptors is 10 times greater than that for ORX-B. Thus, the depolarizing effect of ORX-A was further examined in the presence of SB-334867, a selective antagonist for OX1 receptors. Sample records of the depolarization induced by ORX-A at 10 nM, 100 nM and 1 mM in 100 nM and 1 mM SB-334867-containing ACSFs are shown in left, middle and right panels of Fig. 2C and D, respectively. The amplitudes of the ORX-A-induced depolarization in the presence of SB-334867 were considerably smaller than those in the absence of SB-334867, and the reduction of the depolarization in the presence of SB-334867 was concentration

Fig. 3 – Effects of SB-334867 on ORX-A-induced depolarization of PPT neurons. (A) Depolarizations induced by the first (left record) and second (right record) applications of ORX-A at 300 nM in modified ACSF. (B) Depolarization induced by the first application of ORX-A at 300 nM in modified ACSF (left record), and that induced by the second application of ORX-A at the same dose in 1 mM SB-334867-containing ACSF (right record). (C) Mean amplitudes of the depolarization to the first application of ORX-A in modified ACSF (open column, n = 8), to the second application of ORX-A in modified ACSF (solid column, n = 4) and to the second application of ORX-A in 1 mM SB-334867-containing ACSF (dotted column, n = 4): (") resting membrane potential of S60 mV. Solid bars, ORX-A. Dotted line, SB-334867-containing ACSF. **P < 0.01. ***P < 0.001.

196

peptides 30 (2009) 191–209

dependent. The dose-response curves for ORX-A at 1 nM (except in the case of 1 mM SB-334867), 10 nM, 100 nM, 1 mM and 5 mM in the presence of 100 nM SB-334867 (*) and 1 mM SB-334867 (~) are also displayed in Fig. 2E. When the statistical significance on the dose-response curves for ORX-A was tested by the two-way ANOVA with repeated measure in the range from 10 nM to 5 mM, there was a significant difference in the absence and presence of SB-334867 (F2,36 = 25.7, P < 0.001), and the post hoc Fisher’s PLSD test further revealed that the mean amplitudes of the depolarization at 100 nM, 1 mM and 5 mM in the presence of 100 nM and 1 mM SB-334867 were significantly smaller than corresponding values in the absence of SB334867 (P < 0.001 at 100 nM, P < 0.01 at 1 mM and P < 0.05 at 5 mM for 100 nM SB-334867; P < 0.001 at 100 nM, P < 0.001 at 1 mM and P < 0.01 at 5 mM for 1 mM SB-334867). To further confirm the decrease of ORX-A-induced depolarization in the presence of SB-334867, the depolarization to ORX-A at 300 nM in the same PPT neurons was compared in the absence and presence of SB-334867. ORX-A was first applied to 8 neurons in modified ACSF. Each of two examples of the depolarization induced by ORX-A is shown in the left

record of Fig. 3A and B. In 4 out of 8 neurons, the second application of ORX-A in modified ACSF produced a depolarization similar to that in the first application of ORX-A (right record of Fig. 3A). In the remaining 4 neurons, however, the second application of ORX-A in 1 mM SB-334867-containing ACSF produced a considerably smaller depolarization than that in the first application of ORX-A (right record of Fig. 3B). The mean amplitude of the depolarization to the first application of ORX-A in modified ACSF was 8.1  1.0 mV (n = 8), and those to the second application in modified and SB334867-containing ACSFs were 8.0  1.2 mV (n = 4) and 1.6  0.6 mV (n = 4), respectively (Fig. 3C). Although there was no significant difference between the mean amplitudes of the depolarization to the first and second applications of ORX-A in modified ACSF, the mean amplitude of the depolarization to the second application of ORX-A in SB334867-containing ACSF was significantly smaller than those to the first and second applications of ORX-A in modified ACSF (P < 0.001 for the first application and P < 0.01 for the second application), indicating again that SB-334867 effectively decreases the excitatory effect of ORX-A.

Fig. 4 – Increase of membrane resistance in ORX-A-induced depolarization of PPT neurons. (A) Depolarization induced by application of 500 nM ORX-A (solid bar). Downward deflections indicate a decrease in electrotonic potential amplitude due to constant-current hyperpolarizing pulses (40 pA, 500 ms, 30 s interval) passed through the patch pipette to measure membrane resistance. When membrane potential during the depolarization is manually returned to the resting level, an increase in the membrane resistance is seen. (B) Membrane potential changes induced by hyperpolarizing currents at ‘‘a’’ and ‘‘b’’ indicated by downward arrows in A were enlarged, and then rearranged to compare the membrane resistance before (‘‘a’’) and during (‘‘b’’) the effect of ORX-A. Membrane resistance during the depolarization is obviously lager than that before the depolarization. Hyperpolarizing current is shown in lower panel. (C) Percent increase in mean membrane resistance during the depolarization (solid column) compared with that before the depolarization (open column) which was considered to be 100% (n = 6): depo., depolarization. *P < 0.05.

peptides 30 (2009) 191–209

3.2. Increase of membrane resistance and involvement of K+ conductances in ORX-A-induced depolarization of PPT neurons In 6 neurons, a constant hyperpolarizing current (40 pA, 500 ms) was injected through a patch pipette to measure membrane resistance before and during a depolarization induced by application of ORX-A at 500 nM. As shown in Fig. 4A and B, when membrane resistance at ‘‘a’’ before the depolarization is considered to be 100% (control), membrane resistance at ‘‘b’’ during the depolarization induced by ORX-A increased by 22%. The mean increase of membrane resistance during the depolarization obtained from 6 neurons was 18.8  2.8% (Fig. 4C), and the increase was statistically significant compared to that in control before the depolarization, P < 0.05 (Wilcoxon signed-rank test). These results suggest that the increase in membrane resistance during the ORX-A-induced depolarization might be attributable to a closure of K+ channels, that is, a decrease of K+ conductances. To test this possibility, extracellular K+ concentration ([K+]o) was increased in 6 neurons whose membrane potentials were maintained at about 60 mV. As shown in the upper record of Fig. 5A, application of ORX-A at 300 nM depolarized a PPT neuron in modified ACSF with 4.25 mM of [K+]o, estimated equilibrium potential for K+, 92 mV (control). When [K+]o was increased to 13.25 mM, estimated equilibrium potential for K+, 62 mV, the depolarization to the second application of ORX-A in the same neuron considerably decreased, but never disappeared (middle record). Following a recovery period in which [K+]o was returned to 4.25 mM, the third application of ORX-A again produced a depolarization with an amplitude similar to that in control (lower record). The recovery period was 20–30 min in the present and other subsequent experi-

197

ments. The mean amplitudes of the depolarization in control, high K+ ACSF and recovery were 8.9  1.3 mV, 3.4  0.8 mV and 7.1  1.3 mV (n = 6), respectively (Fig. 5B). The mean depolarization obtained in high K+ ACSF was significantly lower than those obtained in control (P < 0.01) and recovery (P < 0.05). These results suggest a possibility that the depolarization of PPT neurons induced by ORX-A is partly due to a decrease of K+ conductances, but that other ionic mechanism(s) might still be involved in the depolarization.

3.3. Effects of substitution of NMDG-Cl for NaCl in ORXA-induced depolarization of PPT neurons Previous studies have shown that ORXs-induced excitatory responses in neurons of some brain regions are mediated via dual ionic mechanisms including K+ conductances. In brainstem neurons of the locus coeruleus (LC) [32], dorsal motor nucleus of the vagus (DMNV) [18] and nucleus of the solitary tract (NST) [53], for example, ORXs modulates non-selective cationic conductances (NSCC) as well as K+ conductances, and thereby produce changes in excitability of these neurons. Another study further demonstrates that septohippocampal cholinergic neurons are excited by ORXs via an activation of NCX and a reduction of K+ conductances [51]. It is well known that both NSCC and NCX are dependent on extracellular Na+. Thus, to test whether or not NSCC and/or NCX are involved in the ORX-A-induced depolarization of PPT neurons, we used NMDG+ ACSF (composition in mM: NMDG-Cl 126, KCl 3, CaCl2 2.4, MgSO4 1.3, KH2PO4 1.25, NaHCO3 26 and glucose 10 with a pH of 7.4) in which NaCl in modified ACSF was replaced with NMDG-Cl. ORX-A at 300 nM was applied 3 times to the same PPT neuron (Fig. 6A). The first application of ORX-A produced a representative depolarization in modified ACSF (control,

Fig. 5 – Involvement of K+ conductances in ORX-A-induced depolarization of PPT neurons. (A) ORX-A at 300 nM was applied three times to the same PPT neuron in modified and high K+ ACSFs. Upper record: depolarization induced by the first application of ORX-A in modified ACSF (control, [K+]o = 4.25 mM). Middle record: depolarization induced by the second application of ORX-A in high K+ ACSF ([K+]o = 13.25 mM). Lower record: depolarization induced by the third application of ORX-A in modified ACSF (recovery, [K+]o = 4.25 mM). (B) Mean amplitudes of the depolarization induced by ORX-A in control (open column), high K+ ACSF (solid column) and recovery (dotted column) (n = 6): (") resting membrane potential of S60 mV. Solid bars, ORX-A. Dotted line, high K+ ACSF. *P < 0.05. **P < 0.01.

198

peptides 30 (2009) 191–209

upper record). In NMDG+ ACSF, however, the depolarization induced by the second application of ORX-A was largely attenuated as compared to control (middle record). The third application of ORX-A following a recovery period in which NMDG+ ACSF was retuned to modified ACSF again induced a depolarization similar to that in control (lower record). The average amplitudes of the depolarization induced by ORX-A in control, NMDG+ ACSF and recovery were 7.9  1.0 mV, 3.3  0.7 mV and 6.3  1.1 mV (n = 5), respectively, and the mean amplitude of the ORX-A-induced depolarization during the substitution of NMDG-Cl for NaCl was significantly lower than those in control (P < 0.01) and recovery (P < 0.05) (Fig. 6B), indicating that the ORX-A-induced depolarization is extracellular Na+-dependent, and indeed NSCC and/or NCX are involved in the mechanism of the ORX-induced depolarization.

3.4. Involvement of NSCC, but not NCX, in ORX-Ainduced depolarization of PPT neurons To assess the involvement of NCX, DCB, a specific inhibitor for a forward- and reverse-mode NCX [11,23], was used. Application of ORX-A at 300 nM produced a depolarization in modified ACSF as shown in the upper record of Fig. 7A (control). Further application of ORX-A at 300 nM to the same neuron in 60 mM DCB-containing ACSF again produced a depolarization similar to that in the first application of ORX-A (lower record of Fig. 7A). The mean amplitudes of the depolarization induced by ORX-A in control and 30–60 mM DCB-containing ACSF were 9.5  1.5 mV and 9.1  1.2 mV (n = 4), respectively, and there was no significant difference between them, P > 0.05 (Fig. 7B). To further test the involvement of NCX, BAPTA, a high-affinity

Ca2+ chelator, was used because chelating intracellular Ca2+ prevents activation of NCX without affecting NSCC [10,28]. BAPTA at 10 mM was added to the standard internal pipette solution and loaded to neurons via the pipette. The effect of BAPTA on chelating intracellular Ca2+ is confirmed by a change in firing mode of breakaway Ca2+ spikes [38]. The left, middle and right records of Fig. 7C were obtained immediately, 1 min and 10 min after establishing the whole cell configuration, respectively. Breakaway Ca2+ spike (~) was induced once by a voltage step from a holding potential of 60 mV to 0 mV (not shown), but then ceased because of Ca2+-dependent inactivation of Ca2+ channels (left record). The number of breakaway Ca2+ spikes increased with time after breakthrough (middle record), and consequently breakaway Ca2+ spikes occurred repetitively during the voltage step because of an abolishment of the Ca2+-dependent inactivation of Ca2+ channels and an attenuation of Ca2+-dependent K+ currents induced by chelating of intracellular Ca2+ (right record). The depolarization to ORX-A at 300 nM in the same BAPTA-loaded PPT neurons was compared in modified and NMDG+ ACSFs. ORX-A was first applied to 9 BAPTA-loaded neurons in modified ACSF. Each of two examples of the depolarization induced by ORX-A is shown in the left upper and lower records of Fig. 7D. In 5 out of 9 neurons, the second application of ORX-A in modified ACSF produced a depolarization similar to that in the first application of ORX-A (right upper record of Fig. 7D). In the remaining 4 neurons, however, the second application of ORXA in NMDG+ ACSF produced a considerably smaller depolarization than that in the first application of ORX-A (right lower record of Fig. 7D). The mean amplitude of the depolarization to the first application of ORX-A in modified ACSF was 7.5  1.1 mV (n = 9), and those to the second application in

Fig. 6 – Effects of replacement of NaCl in modified ACSF with NMDG-Cl on ORX-A-induced depolarization of PPT neurons. (A) ORX-A at 300 nM was applied three times to the same PPT neuron in modified or NMDG+ ACSFs. Upper record: depolarization induced by the first application of ORX-A in modified ACSF (control). Middle record: depolarization induced by the second application of ORX-A in NMDG+ ACSF. Lower record: depolarization induced by the third application of ORX-A in modified ACSF (recovery). (B) Mean amplitudes of the depolarization induced by ORX-A in control (open column), NMDG+ ACSF (solid column) and recovery (dotted column) (n = 5): (") membrane potential of S60 mV. Solid bars, ORX-A at 300 nM. Dotted line, NMDG+ ACSF. *P < 0.05. **P < 0.01.

peptides 30 (2009) 191–209

199

Fig. 7 – Involvement of NSCC, but not NCX, in ORX-A-induced depolarization of PPT neurons. (A) ORX-A at 300 nM was applied twice to the same PPT neuron in modified and 30–60 mM DCB-containing ACSFs. Upper record: depolarization induced by the first application of ORX-A in modified ACSF (control). Lower record: depolarization induced by the second application of ORX-A in 60 mM DCB-containing ACSF. (B) Mean amplitudes of the depolarization induced by ORX-A in control (open column) and DCB-containing ACSF (solid column) (n = 4). (C) Effects of chelating intracellular Ca2+ by BAPTA on breakaway Ca2+ spikes. BAPTA was contained in the internal pipette solution, and loaded to PPT neurons via the pipette. Breakaway Ca2+ spikes (~) were induced by a voltage step from a holding potential of S60 mV to 0 mV. Left, middle and right records were obtained in the same neuron 0 min, 1 min and 10 min after establishing a whole cell configuration, respectively. (D) ORX-A at 300 nM was applied twice to the same BAPTA-loaded neuron. Upper records: depolarizations induced by the first (left record) and second (right record) applications of ORX-A in modified ACSF. Lower records: depolarization induced by the first application of ORX-A in modified ACSF (left record), and that induced by the second application of ORX-A in NMDG+ ACSF (right record). (E) Mean amplitudes of the depolarization to the first application of ORXA in modified ACSF (open column, n = 9), to the second application of ORX-A in modified ACSF (solid column, n = 5) and to the second application of ORX-A in NMDG+ ACSF (dotted column, n = 4): (") resting membrane potential of S60 mV. Solid bars, ORX-A. Dotted lines in A and D, 60 mM DCB-containing and NMDG+ ACSFs, respectively. *P < 0.05. **P < 0.01.

200

peptides 30 (2009) 191–209

modified and NMDG+ ACSFs were 7.0  0.8 mV (n = 5) and 2.3  0.7 mV (n = 4), respectively (Fig. 7E). Although there was no significant difference between the mean amplitudes of the depolarization to the first and second applications of ORX-A in modified ACSF, the mean amplitude of the depolarization to the second application of ORX-A in NMDG+ ACSF was significantly smaller than those to the first and second applications of ORX-A in modified ACSF (P < 0.01 for the first application and P < 0.05 for the second application). These results suggest that the ORX-induced depolarization in PPT neurons is partly mediated by NSCC, rather than NCX. Liu et al. [28] demonstrated that serotonin neurons in the dorsal raphe nucleus (DR) are directly activated by ORXs via NSCC, and the non-selective cationic current is enhanced in Ca2+-free/high-Mg2+ ACSF than in modified ACSF. To confirm whether or not NSCC in PPT neurons has such a property, we tested the effect of Ca2+-free/high-Mg2+ ACSF on the ORX-A-induced depolarization. The depolarization elicited by application of ORX-A at 300 nM in modified ACSF (control, upper record of Fig. 8A) was enhanced by the subsequent application of ORX-A at the same dose in Ca2+-free/high-Mg2+ ACSF (middle record of Fig. 8A). When Ca2+-free/high-Mg2+ ACSF was changed to modified ACSF, the amplitude of the depolarization induced by the third application of ORX-A was recovered to a level similar to that in the control (lower record of Fig. 8A). The mean magnitudes of the depolarization in control, Ca2+-free/ high-Mg2+ ACSF and recovery were 6.2  1.0 mV, 11.5  1.6 mV and 5.1  1.1 mV (n = 5), respectively, and the depolarization in Ca2+-free/high-Mg2+ ACSF was significantly larger than those in control (P < 0.05) and recovery (P < 0.01), suggesting that NSCC in PPT neurons has a similar

property with that seen in serotonin neurons in the DR, and that extracellular calcium ions do not contribute to a significant part of the ORX-A-induced depolarization.

3.5. ORX-A-induced depolarization depends upon a dual ionic mechanism including K+ conductances and NSCC The results described above show that the ORX-A-induced depolarization is mediated by a decrease of K+ conductances and an increase of NSCC. To test this possibility, application of ORX-A at 300 nM was repeated three times to the same PPT neuron in modified or high K+-NMDG+ ACSFs (Fig. 9A). In high K+-NMDG+ ACSF, [K+]o was increased to 13.25 mM from 4.25 mM and NaCl in modified ACSF was replaced with NMDG-Cl. The upper record in Fig. 9A shows a representative depolarization to the first application of ORX-A in modified ACSF (control), and the depolarization to the second application of ORX-A was eliminated in high K+-NMDG+ ACSF (middle record). When high K+-NMDG+ ACSF was returned to modified ACSF, the third application of ORX-A produced a depolarization similar to that in the first application of ORX-A (lower record). The means of the depolarization in control, high K+NMDG+ ACSF and when recovered were 9.9  1.0 mV, 0.3  0.3 mV and 7.3  1.1 mV (n = 5), respectively, and the decrease in depolarization in high K+-NMDG+ ACSF was statistically significant compared to those in control (P < 0.001) and modified ACSF during recovery (P < 0.001) (Fig. 9B). In the following experiments, K-gluconate and KCl in the standard internal pipette solution were replaced with Csgluconate and CsCl, respectively, and Cs+ was loaded to neurons via the pipette to block K+ channels [28]. The effect of Cs+ in blocking K+ channels was confirmed by the fact that breakaway Ca2+ spikes (~ in Fig. 9C) induced by a voltage step

Fig. 8 – Effects of Ca2+ removal from extracellular solution on ORX-A-induced depolarization of PPT neurons. (A) ORX-A at 300 nM was applied three times to the same PPT neuron in modified and Ca2+-free/high Mg2+ ACSFs. Upper record: depolarization induced by the first application of ORX-A in modified ACSF (control). Middle record: depolarization induced by the second application of ORX-A in Ca2+-free/Mg2+ ACSF. Lower record: depolarization induced by the third application of ORX-A in modified ACSF (recovery). (B) Mean amplitudes of the depolarization induced by ORX-A in control (open column), Ca2+-free/Mg2+ ACSF (solid column) and recovery (dotted column) (n = 5): (") membrane potential of S60 mV. Solid bars, ORX-A. Dotted line, Ca2+-free/high Mg2+ ACSF. *P < 0.05. **P < 0.01.

peptides 30 (2009) 191–209

201

Fig. 9 – Disappearance of ORX-A-induced depolarization by preventing a dual ionic mechanism including K+ conductances and NSCC. (A) ORX-A at 300 nM was applied three times to the same PPT neuron in modified and high K+-NMDG+ ACSFs. In high K+-NMDG+ ACSF, [K+]o was increased to 13.25 mM from 4.25 mM and NaCl was replaced with NMDG-Cl. Upper record: depolarization induced by the first application of ORX-A in modified ACSF (control). Middle record: disappearance of ORX-Ainduced depolarization in high K+-NMDG+ ACSF. Lower record: depolarization induced by the third application of ORX-A in modified ACSF (recovery). (B) Mean amplitudes of the depolarization in control (open column), high K+-NMDG+ ACSF (solid column) and recovery (dotted column) (n = 5). (C) Broadening of breakaway Ca2+ spikes induced by K+ channel blocking due to replacement of K+ with Cs+ in the internal pipette solution. Breakaway Ca2+ spikes (~) were induced by a voltage step from a holding potential of S60 mV to 0 mV in modified ACSF. Left, middle and right records were obtained in the same PPT neuron 0 min, 1 min and 10 min after establishing a whole cell configuration, respectively. It should be noted that the duration of breakaway Ca2+ spikes is broadened with time after breakthrough. (D) Internal pipette solution contains Cs+ instead of K+. ORX-A at 300 nM was applied three times to the same PPT neuron in modified and NMDG+ ACSFs. Upper record: depolarization induced by the first application of ORX-A in modified ACSF (control). Middle record: disappearance of ORX-A-induced depolarization in NMDG+ ACSF. Lower record: depolarization induced by the third application of ORX-A in modified ACSF (recovery). (E) Mean amplitudes of the depolarization induced by ORX-A in control (open column), NMDG+ ACSF (solid column) and recovery (dotted column) (n = 5): (") membrane potential of S60 mV. Solid bars, ORX-A. Dotted lines in A and D, high K+-NMDG+ and NMDG+ ACSFs, respectively. *P < 0.05. **P < 0.01. ***P < 0.001.

202

peptides 30 (2009) 191–209

from a holding potential of 60 mV to 0 mV show marked broadening with time after establishing the whole cell configuration as illustrated by the left (0 min after breakthrough), middle (1 min after breakthrough) and right (10 min after breakthrough) records of Fig. 9C. Again, application of ORX-A at 300 nM was repeated three times to the same PPT neuron in modified or NMDG+ ACSFs (Fig. 9D). The depolarization to the first application of ORX-A was recorded in modified ACSF (control, upper record). When modified ACSF was changed to NMDG+ ACSF, the depolarization to the second application of ORX-A did not occur (middle record). The depolarization induced by the third application of ORX-A in a recovery period with modified ACSF was similar to that in control (lower record). The mean magnitudes of the depolarization in control, NMDG+ ACSF and recovery were 6.1  1.4 mV, 0.1  0.5 mV and 4.1  1.2 mV (n = 6), respectively, and the decrease of the depolarization in NMDG+ ACSF was statistically significant compared to those in control (P < 0.01) and recovery (P < 0.05) (Fig. 9E). These results indicate that indeed the ORX-induced depolarization is mediated by a dual ionic mechanism; that is, the reduction of K+ conductances and the increase of NSCC.

3.6. Current–voltage relationship in ORX-A-induced depolarization and reversal potential To further confirm the dual ionic mechanism underlying the ORX-A-induced depolarization, current (I)–voltage (V) curves before and during the ORX-A-induced depolarization were recorded using patch pipettes filled with the standard internal solution and NMDG+ ACSF or using patch pipettes containing Cs+ and modified ACSF. In this series of experiments, PPT neurons were voltage clamped to a holding potential of 60 mV. Whole cell current traces in response to a series of voltage step commands (250 ms, 100 mV to 40 mV in 10 mV increments) before and during the ORX-A effect were recorded in a PPT neuron in NMDG+ ACSF (left and right panels in Fig. 10A, respectively). The steady-state current was calculated from each current trace as a mean for 50 ms just before the termination of voltage step command, and plotted against the corresponding command voltage by Clampfit software (Axon Instruments, Union City, USA) to obtain steady-state I–V curves before (*) and during (!) the ORX-A effect (Fig. 10B). The I–V relationship of ORX-A-induced net currents (5) obtained by subtracting these two I–V curves in the potential range from 100 mV to 40 mV was approximately linear over this range of voltages, and reversed the polarity at 96.5 mV ( , reversal potential), close to the equilibrium potential of K+. The mean reversal potential obtained from 5 neurons was 86.3  4.5 mV. Whole cell current traces before and during the application of ORX-A were further recorded in a PPT neuron using the patch pipette containing Cs+ in modified ACSF (left and right panels in Fig. 10C, respectively). In these experiments, a pre-pulse of +70 mV was added for 50 ms before a series of voltage step commands (250 ms, 90 mV to 10 mV in 10 mV increments) because of the suppression of Ca2+ spikes generated during the depolarizing voltage command pulses (Fig. 10C). The steady-state current in each current trace was determined as described above. To

compensate an influence of Ca2+-tail currents induced by pre-pulses, the steady-state current obtained at the command voltage of 60 mV (holding potential) was subtracted from each steady-state current in left and right records of Fig. 10C, and then the steady-state I–V curves before (*) and during (!) the ORX-A effect were plotted (Fig. 10D). The I–V relationship of ORX-A-induced net currents (5) obtained by subtracting these two I–V curves between 90 mV to 10 mV was almost linear over this range of voltages, and reversed the polarity at 37.5 mV ( , reversal potential), close to the equilibrium potential of non-selective cationic channel. The mean reversal potential obtained from 5 neurons was 36.0  4.3 mV. Recently, it has been reported that ORX-A enhances voltage-dependent somatic Ca2+ transients by augmenting the contribution from L-type Ca2+ channels in DR and LDT neurons in mice [24]. If this is the case in PPT neurons in rats, it might be possible that reversal potentials obtained above are disturbed by an increase of Ca2+ current and an enhancement of Ca2+ influx followed by an activation of Ca2+-dependent currents such as Ca2+-dependent K+ current. Thus, reversal potentials were again determined in experiments as shown in Fig. 10 except that Ca2+-free ACSF was used to limit current from voltage-gated Ca2+ channels. Again, whole cell current traces in response to a series of voltage step commands (250 ms, 100 mV to 40 mV in 10 mV increments) before and during the ORX-A effect were recorded in a PPT neuron in Ca2+-free/high-Mg2+-NMDG+ ACSF (left and right panels in Fig. 11A, respectively). The I–V relationship of ORX-A-induced net currents (5) obtained by subtracting two I–V curves before (*) and during (!) the ORX-A effect between 100 mV and 40 mV was approximately linear over this range of voltages, and reversed the polarity at 97.5 mV ( , reversal potential) (Fig. 11B). The mean reversal potential obtained from 5 neurons was 84.4  5.1 mV, and not significantly different from that in NMDG+ ACSF. Whole cell current traces before and during the effect of ORX-A were further recorded in a PPT neuron using Cs+-containing electrodes in Ca2+-free/high-Mg2+ ACSF (left and right panels in Fig. 11C, respectively). For comparison with the reversal potential obtained in modified ACSF in the same condition, the pre-pulse of +70 mV (duration: 50 ms) was also added before a series of voltage step commands (250 ms, 90 mV to 10 mV in 10 mV increments) (Fig. 11C). The steady-state I–V curves before (*) and during (!) the ORX-A effect were constructed described above (Fig. 11D). The I–V relationship of ORX-A-induced net currents (5) obtained by subtracting these two I–V curves between 90 mV and 10 mV was almost linear over this range of voltages, and reversed the polarity at 19.8 mV ( , reversal potential). The mean reversal potential in 5 neurons was 21.0  3.8 mV. In other experiments, whole cell current traces in response to a series of voltage steps (250 ms, 60 mV to 0 mV in 10 mV increments) without pre-pulses were recorded in 5 PPT neurons using Cs+-containing pipettes in Ca2+-free/high-Mg2+ ACSF (data not shown), and the mean reversal potential obtained was 21.9  1.8 mV (n = 5). There was no significant difference between the mean reversal potentials obtained with and without pre-pulses in Ca2+-free/high-Mg2+ ACSF. However,

peptides 30 (2009) 191–209

203

Fig. 10 – I–V relationships in ORX-A-induced depolarization and reversal potentials. PPT neurons were voltage-clamped with a holding potential of S60 mV. (A) Whole cell current traces recorded in response to voltage steps (250 ms, from S100 mV to S40 mV) in 10 mV increments before (control, left panel) and during (right panel) the ORX-A effect in NMDG+ ACSF. (B) Steady-state I–V curves before (*) and during (!) the ORX-A effect and I–V relationship of ORX-A-induced net currents (5) obtained from current traces in A. ( ) reversal potential of S96.5 mV. (C) K+ in the internal pipette solution was replaced with Cs+ to block K+ channels. Whole cell currents were recorded in response to voltage steps (250 ms, from S90 mV to S10 mV) in 10 mV increments before (control, left panel) and during (right panel) the ORX-A effect in modified ACSF. A pre-pulse of +70 mV was applied for 50 ms just before each voltage step to suppress Ca2+ spikes or oscillation. (D) Steady-state I–V curves before (*) and during (!) the ORX-A effect and I–V relationship of ORX-A-induced net currents (5) obtained from current traces in C. ( ) reversal potential of S37.5 mV. ( ) and dotted lines in A and C, holding current for the holding potential of S60 mV (S4.4 pA for A and S91.4 pA for C).

these two mean reversal potentials obtained in Ca2+-free/ high-Mg2+ ACSF were significantly greater than that obtained in modified ACSF (P < 0.01 for mean reversal potential obtained with pre-pulses in Ca2+-free/high-Mg2+ ACSF; P < 0.05 for mean reversal potential obtained without pre-pulses in Ca2+-free/high-Mg2+ ACSF). These results indicate that the current limitation from voltage-dependent Ca2+ channels does not affect the reversal potential for K+ channels, whereas it shifts significantly the reversal potential for non-selective cationic channels to a positive potential direction. It appears that the positive shift of the reversal potential for non-selective cationic channels seen here is due to an enhancement of non-selective cationic inward currents by Ca2+ removal, but not due to an abolishment of the increase of Ca2+ currents and the enhancement of Ca2+ influx (see Section 4). Thus, present results confirm the involvement of the dual ionic mechanism described above in the ORX-A-induced depolarization.

3.7. Electrophysiological classification, soma size and ORXs-induced response of PPT neurons Based on electrophysiological membrane properties of PPT neurons reported in the previous studies [21,26,43], PPT neurons were classified into 4 groups. The first type (Type 1) was characterized by a low-threshold Ca2+ spike (LTS), the second type (Type 2) by an A-current, the third type (Type 3) by the presence of both LTS and A-current, and the fourth type (Type 4) by having neither LTS nor A-current. In the present study, the electrophysiological membrane properties of PPT neurons were examined by current and voltage clamp modes. The left panels in Fig. 12A–C show membrane voltage changes in current clamp mode in response to a series of hyperpolarizing current step pulses (500 ms, 120–0 pA in 30 pA increments for A and B, 240–0 pA in 60 pA increments for C), whereas the right panels in Fig. 12A–C show current changes in voltage clamp mode in response to a series of hyperpolar-

204

peptides 30 (2009) 191–209

Fig. 11 – I–V relationships in ORX-A-induced depolarization and reversal potentials in Ca2+-free ACSF. PPT neurons were voltage-clamped with a holding potential of S60 mV. (A) Whole cell current traces recorded in response to voltage steps (250 ms, from S100 mV to S40 mV) in 10 mV increments before (control, left panel) and during (right panel) the ORX-A effect in Ca2+-free/high Mg2+-NMDG+ ACSF. (B) Steady-state I–V curves before (*) and during (!) the ORX-A effect and I–V relationship of ORX-A-induced net currents (5) obtained from current traces in A: ( ) reversal potential of S97.5 mV. (C) K+ in the internal pipette solution was replaced with Cs+ to block K+ channels. Whole cell currents were recorded in response to voltage steps (250 ms, from S90 mV to S10 mV) in 10 mV increments before (control, left panel) and during (right panel) the ORX-A effect in Ca2+-free/high Mg2+ ACSF. To compare with the reversal potential in modified ACSF in the same condition, the pre-pulse of +70 mV was applied for 50 ms just before each voltage step. (D) Steady-state I–V curves before (*) and during (!) the ORX-A effect and I–V relationship of ORX-A-induced net currents (5) obtained from current traces in C: ( ) reversal potential of S19.8 mV. and dotted lines in A and C, holding current for the holding potential of S60 mV (+33.5 pA for A and S164.2 pA for C).

izing and depolarizing voltage step pulses (250 ms, 100 mV to 0 mV in 10 mV increments). The holding potential for voltage clamp was 80 mV. In 199 out of 236 PPT neurons, 63 (31.7%), 135 (67.8%) and 1 (0.5%) were classified into Type 2, Type 3 and Type 4 groups, respectively (Table 1). None of the neurons was classified into Type 1. Fig. 12A–C shows sample records for Type 2, Type 3 and Type 4, respectively. In Fig. 12A, LTS was not seen in the left record, but A-current (solid arrow) was

induced as seen in the right record. In Fig. 12B, both LTS (open arrow in the left record) and A-current (solid arrow in the right record) were observed, but neither LTS nor A-current was seen in Fig. 12C. The mean soma sizes in the long axis of Type 2 and Type 3 neurons were 40.3  1.0 mm (n = 63) and 41.2  0.7 mm (n = 135), respectively, and that of the one Type 4 neuron was 43.7 mm. Of 63 Type 2 and 135 Type 3 neurons, 46 (46/ 63 = 73.0%) and 112 (112/135 = 83.0%) were depolarized by

Table 1 – Summary of ORX effects on each type of PPT neurons.

Depolarization No response Total

Type 1

Type 2

Type 3

Type 4

0 (0%) 0 (0%) 0 (0%)

46 (23.1%) 17 (8.5%) 63 (31.7%)

12 (56.3%) 23 (11.6%) 135 (67.8%)

1 (0.5%) 0 (0%) 1 (0.5%)

Total 159 (79.9%) 40 (20.1%) 199 (100%)

205

peptides 30 (2009) 191–209

Fig. 12 – Electrophysiological identification of PPT neurons by membrane properties. (A–C) Type 2, Type 3 and Type 4 neurons, respectively. Left records in A–C: responses of each Type neuron to injection of hyperpolarizing current pulses (500 ms, 120–0 pA in 30 pA increments for A and B, 240–0 pA in 60 pA increments for C) in current clamp mode. Open arrow in B indicates LTS. Right records in A–C: responses of each Type neuron to voltage step pulses (250 ms, S100 mV to 0 mV in 10 mV increments) in voltage clamp mode with a holding potential of S80 mV. Solid arrows in A and B indicate A-current: (") membrane potential of S60 mV.

ORX-A and/or ORX-B, respectively. The remainder in Type 2 and Type 3 neurons did not respond to these agents. The one neuron classified into Type 4 was depolarized by ORX-A. Of 37 neurons that were not classified into 4 groups, 35 were depolarized by ORX-A or ORX-B, and the remaining 2 were not.

3.8. Cholinergic nature and ORX-A-induced response of Type 2 and Type 3 PPT neurons Of 199 PPT neurons described above, 26 including 9 (34.6%) Type 2 and 17 (65.4%) Type 3 neurons were labeled by biocytin

and stained histochemically by NADPH-d to determine the cholinergic nature of PPT neurons. Of 9 Type 2 neurons, 6 (66.7%) were cholinergic and 3 (33.3%) non-cholinergic, and ORX-A-induced depolarization was recorded in 4 (66.7%) of 6 cholinergic neurons and 2 (66.7%) of 3 non-cholinergic neurons (Table 2). Of 17 Type 3 neurons, 15 (88.2%) were cholinergic and 2 (11.8%) non-cholinergic, and the depolarization to ORX-A was induced in 14 (93.3%) of 15 cholinergic neurons and 1 (50.0%) of 2 non-cholinergic neurons (Table 2). Fig. 13A and B shows a PPT neuron that is Texas red-positive and NADPH-d-positive, respectively. This PPT neuron was

Table 2 – Summary of cholinergic nature and ORX effects in Type 2 and Type 3 PPT neurons. Type 2

Depolarization No response Total

Type 3

Total

Cholinergic

Non-cholinergic

Cholinergic

Non-cholinergic

4 (15.4%) 2 (7.7%) 6 (23.1%)

2 (7.7%) 1 (3.8%) 3 (11.5%)

14 (53.8%) 1 (3.8%) 15 (57.7%)

1 (3.8%) 1 (3.8%) 2 (7.7%)

21 (80.8%) 5 (19.2%) 26 (100%)

206

peptides 30 (2009) 191–209

Fig. 13 – Histochemical identification of cholinergic PPT neurons and ORX-A-induced depolarization. (A) A biocytin-filled neuron. (B) Neuron in A was NADPH-d-positive. Scale bar in A is also applicable in B. (C) Schematic representation of the PPT. *Location of the PPT neuron shown in A and B. (D) Depolarization induced by ORX-A at 300 nM (solid bar) in the neuron shown in A and B: (") membrane potential of S60 mV.

located at the place shown by asterisk ( ) in Fig. 13C, and was depolarized by ORX-A at 300 nM (Fig. 13D). In addition, the neuron was classified as Type 3 (data not shown).

4.

Discussion

Several lines of evidence have shown that ORXs acts in the brain to modulate sleep–wakefulness as well as feeding [34,37]. Indeed, electrophysiological studies have demonstrated that ORXs excites cholinergic neurons in the BF [7] and LDT [4,24,42], histaminergic neurons in the tuberomammillary nucleus [1,9], noradrenergic neurons in the LC [13,15,19,46] and serotonergic neurons in the DR [2,3,28] that constitute an ascending arousal system. Our present whole cell recordings in current-clamp mode also demonstrate clearly that ORXs rapidly and reversibly depolarizes the membrane potential of the majority of PPT neurons including cholinergic and non-cholinergic neurons, and in some neurons the depolarization was accompanied by the firing

of action potentials. The facts that the depolarization occurs in the presence of TTX or in Ca2+-free/high-Mg2+ ACSF suggest that the effect of ORXs is due to a direct postsynaptic action on PPT neurons. Thus, present results provide electrophysiological evidence for the previous notion that hypothalamic ORX neurons affect the activity of PPT neurons directly and/or indirectly to control sleep–wakefulness, especially REM sleep [34]. The ORXs-induced depolarization is concentration-dependent, and the dose-response curve for ORX-A is shifted to the left of that for ORX-B with the respective EC50 values being 66 nM and 536 nM. In agreement with this, the amplitude of the depolarization for ORX-A was significantly greater than that for ORX-B when both ORX-A and ORX-B are applied to the same PPT neurons in an equimolar concentration between the two EC50 values. As Sakurai et al. [37] reported, using a radioligand binding assay in transfected cell lines expressing the human ORX receptors, OX1 receptors have 10 times the affinity for ORX-A than for ORX-B, whereas OX2 receptors have an equal affinity for ORX-A and ORX-B. Thus, the present

peptides 30 (2009) 191–209

results suggest that the receptors expressed on the PPT neurons are presumably OX1. The fact showing that the depolarization induced by ORX-A is significantly decreased in the presence of SB-334867, a selective OX1 receptor antagonist, further supports the predominant expression of OX1 receptors on PPT neurons. Anatomical studies also demonstrate that PPT neurons express predominantly OX1 receptors although mRNAs for both OX1 and OX2 receptors are detectable in the PPT [12,30]. Interestingly, recent studies suggest that inhibition of wake/NREM sleep transition depends upon OX2 receptor activation, while normal regulation of REM sleep relies not only upon OX2 receptor activation, but also upon other signaling pathways mediated by OX1 receptors [34,50]. In the present study, application of ORX-A to PPT neurons in high K+-NMDG+ ACSF did not produce a depolarization, and in NMDG+ ACSF the ORX-A-induced depolarization of PPT neurons was also not observed by using patch pipettes filled with an internal pipette solution containing Cs+. Since DCB, an inhibitor of the NCX, or chelating intracellular Ca2+ by BAPTA does not affect the ORX-A-induced depolarization, these results suggest that the ORX-A-induced depolarization in PPT neurons is mediated by a dual ionic mechanism, that is, an inhibition of K+ channels and an activation of the nonselective cationic channels. Other confirming evidence in support of the dual ionic mechanism for the ORX-A-induced depolarization is derived from the data on reversal potentials in the I–V relationships of ORX-A-induced net currents. In NMDG+ ACSF, the reversal potential of the depolarization was about 86 mV, and close to the equilibrium potential of K+, whereas with the use of the Cs+-containing patch electrodes the reversal potential of the depolarization in modified ACSF was about 36 mV, and close to a reversal potential expected for non-selective cationic channels [22,28,52]. Recently, it has been reported that ORX-A depolarizes LDT and DR neurons in mice by activating non-selective cationic channels, and enhances somatic Ca2+ transients by augmenting the contribution from voltage-gated L-type Ca2+ channels [24]. If the latter is the case in PPT neurons, it might be possible that reversal potentials described above are disturbed by an increase of Ca2+ current and by an enhancement of Ca2+ influx. Thus, reversal potentials were again tested in Ca2+-free situations to limit current from voltage-dependent Ca2+ channels. The reversal potential of the depolarization in Ca2+-free/high-Mg2+-NMDG+ ACSF was about 84 mV, and not significantly different from that in NMDG+ ACSF. The reversal potential of the depolarization obtained using Cs+-containing patch pipettes in Ca2+-free/high-Mg2+ ACSF was about 21 mV regardless of whether pre-pulses were used, and shifted significantly to more positive potential direction than that in modified ACSF (about 36 mV). Liu et al. [28] have shown that in DR serotonin neurons the ORXs-induced inward current with a holding potential of 65 mV in Ca2+-free/high-Mg2+ ACSF is approximately twice compared to that in normal ACSF, and suggested that Ca2+ removal enhances nonselective cationic current, possibly by removal of the Ca2+ inhibition of the channel. In the present study, we also demonstrated that the ORX-A-induced depolarization in Ca2+free/high-Mg2+ ACSF is almost two times greater than that in modified ACSF. As seen in Fig. 10D and Fig. 11D, furthermore, the I–V relationship of ORX-A-induced net currents in Ca2+-

207

free/high-Mg2+ ACSF shifted to more negative current direction than that in modified ACSF, indicating an increase of nonselective cationic inward current. The mean non-selective cationic inward current at 60 mV in Ca2+-free/high-Mg2+ ACSF was 66.1  25.3 pA (n = 5), and almost twice compared to that in modified ACSF (33.3  11.8 pA, n = 5). Thus, it seems likely that the increase of non-selective cationic current in Ca2+-free/high-Mg2+ ACSF shifts the I–V relationship of ORX-Ainduced net currents to a negative current direction and the reversal potential for non-selective cationic current to a positive potential direction. Kohlmeier et al. [24] demonstrated that an enhancement of Ca2+ influx during the ORX-A effect is produced approximately 2 s after the onset of voltage clamp step from 60 mV to 30 mV (see Fig. 5A of [24]). In our recordings of whole cell current traces, the durations of the voltage step pulse and the pre-pulse plus voltage step pulse were 250 ms and 300 ms, respectively. Thus, it is unlikely that an enhancement of Ca2+ influx and an increase of Ca2+ current which carries Ca2+ occur during these very short periods. Although it is unclear whether ORXs augment the Ca2+ current and Ca2+ entry by enhancing the contribution of voltage-gated Ca2+ channels in PPT neurons of rats, it is hard to conceive that reversal potentials measured in the present study are disturbed by such an augmentation of Ca2+ current and Ca2+ entry, if any. Thus, present results on the I–V relationships of the ORX-A-induced depolarization confirm that the ORX-Ainduced depolarization in PPT neurons depends upon both the reduction of K+ conductances and the activation of NSCC. It has been reported that in neurons of the brainstem nuclei such as the LC, DMNV and NTS the ORXs-induced depolarization is produced by the same dual ionic mechanism [18,32,53]. Hwang et al. [18] have pointed out a possibility that the degree of contribution from two ionic mechanisms on the ORXinduced depolarization may vary in different DMNV neurons. In the present study, the reductions of the amplitude in the ORX-A-induced depolarization in high K+ ACSF and in NMDG+ ACSF ranged from 36.3% to 85.8% (n = 6) and from 48.1% to 72.4% (n = 5), respectively, when the amplitude of the ORX-Ainduced depolarization in modified ACSF was considered to be 100%. The mean reduction was 60.2  9.3% (n = 6) in high K+ ACSF (Fig. 5B) and 58.9  4.5% (n = 5) in NMDG+ ACSF (Fig. 6B), and there was no significant difference between them. These results indicate that the degree of contribution from two ionic mechanisms may vary in different PPT neurons, but on average they are approximately equal. According to previous studies, PPT neurons can be classified into four types: Type 1 neurons are characterized by a LTS, Type 2 neurons by a transient outward current (Acurrent), Type 3 neurons by the A-current plus LTS, and Type 4 neurons by the lack of both A-current and LTS [21,26]. Of the 199 out of 236 PPT neurons studied, 63 (31.7%), 135 (67.8%), and 1 (0.5%) were classified into Type 2, Type 3 and Type 4, respectively. Type1 neuron was not observed. In a previous study, the percentages of Type 1, Type 2 and Type 3, and Type 4 neurons in the PPT were 36.1%, 33.3%, and 30.6%, respectively, and these values were different from those obtained in the present study [21]. The difference might be due to the fact that PPT neurons with a soma size of more than 30 mm are recorded in the present study. Indeed, it has been reported that the somatic area of the Type 2 and Type 3 neurons is almost

208

peptides 30 (2009) 191–209

always more than twice as large as that of Type 1 and Type 4 neurons [21]. Another anatomical study further demonstrated that the somatic area of PPT cholinergic neurons in rats was significantly greater than that of non-cholinergic neurons [17]. In agreement with these findings, a combined rhodamine labeling and NADPH-d histochemistry [26] or a combined biocytin labeling and choline acetyltransferase immunohistochemistry [21,43] have shown that 50–60% of Type 2 and Type 3 neurons were identified as cholinergic, whereas Type 1 and Type 4 neurons were non-cholinergic. The present results combining biocytin labeling and NADPHd histochemistry also indicated that 66.7% of Type 2 neurons and 88.2% of Type 3 neurons are cholinergic. The remainder of the Type 2 and Type 3 neurons was not cholinergic. Application of ORX-A on these cholinergic and non-cholinergic neurons further revealed that 85.7% of cholinergic neurons and 60.0% of non-cholinergic neurons are depolarized by ORX-A, indicating that ORX excites both cholinergic and non-cholinergic neurons. In conclusion, PPT neurons were directly depolarized by ORXs via presumably OX1 receptors. Ionic mechanisms involved in the depolarization include a decrease of K+ conductances and an increase of NSCC. Most of PPT neurons recorded were characterized by A-current or both A-current and LTS, and predominantly cholinergic when determined by a combined biocytin labeling and NADPH-d histochemistry. These results suggest that PPT neurons might be involved in cellular mechanisms through which ORXs participate in the regulation of sleep–wakefulness, especially REM sleep.

[7]

[8]

[9]

[10]

[11]

[12]

[13]

[14]

[15]

[16]

Acknowledgements [17]

This work was supported in part by a Grant-in-Aid for Scientific Research (No. 20590227 to K.S.) from the Japan Society for the Promotion of Science. One of the authors (K.S.) offers special thanks to Mr. Chikamitsu Nakayama for his encouragement throughout this work.

[18]

[19]

references [20] [1] Bayer L, Eggermann E, Serafin M, Saint-Mileux B, Machad D, Jones BE, et al. Orexins (hypocretins) directly excite tuberomammillary neurons. Eur J Neurosci 2001;14: 1571–5. [2] Brown RE, Sergeeva O, Eriksson KS, Haas HL. Orexin A excites serotonergic neurons in the dorsal raphe nucleus of the rat. Neuropharmacology 2001;40:457–9. [3] Brown RE, Sergeeva OA, Eriksson KS, Haas HL. Convergent excitation of dorsal raphe serotonin neurons by multiple arousal systems (orexin/hypocretin, histamine and noradrenaline). J Neurosci 2002;22:8850–9. [4] Burlet S, Tyler CJ, Leonard CS. Direct and indirect excitation of laterodorsal tegmental neurons by hypocretin/orexin peptides: implications for wakefulness and narcolepsy. J Neurosci 2002;22:2862–72. [5] Chemelli RM, Willie JT, Sinton CM, Elmquist JK, Scammell T, Lee C, et al. Narcolepsy in orexin knockout mice: molecular genetics of sleep regulation. Cell 1999;98:437–51. [6] Datta S, Siwek DF. Single cell activity patterns of pedunculopontine tegmentum neurons across the

[21]

[22]

[23] [24]

[25]

sleep–wake cycle in the freely moving rats. J Neurosci Res 2002;70:611–21. Eggermann E, Serafin M, Bayer L, Machad D, Saint-Mileux B, Jones BE, et al. Orexins/hypocretins excite basal forebrain cholinergic neurons. Neuroscience 2001;108:177–81. el Mansari M, Sakai K, Jouvet M. Unitary characteristics of presumptive cholinergic tegmental neurons during the sleep–waking cycle in freely moving cats. Exp Brain Res 1989;76:519–29. Eriksson KS, Sergeeva O, Brown RE, Haas HL. Orexin/ hypocretin excites the histaminergic neurons of the tuberomammillary nucleus. J Neurosci 2001;21:9273–9. Farkas RH, Chien PY, Nakajima S, Nakajima Y. Properties of a slow nonselective cation conductance modulated by neurotensin and other neurotransmitters in midbrain dopaminergic neurons. J Neurophysiol 1996;76:1968–81. Frelin C, Barbry P, Vigne P, Chassande O, Cragoe Jr EJ, Lazdunski M. Amiloride and its analogs as tools to inhibit Na+ transport via the Na+ channel, the Na+/H+ antiport and the Na+/Ca2+ exchanger. Biochimie 1988;70:1285–90. Greco MA, Shiromani PJ. Hypocretin receptor protein and mRNA expression in the dorsolateral pons of rats. Mol Brain Res 2001;88:176–82. Hagan JJ, Leslie RA, Patel S, Evans ML, Wattam TA, Holmes S, et al. Orexin A activates locus coeruleus cell firing and increases arousal in the rat. Proc Natl Acad Sci USA 1999;96:10911–6. Horikawa K, Armstrong WE. A versatile means of intracellular labeling: injection of biocytin and its detection with avidin conjugates. J Neurosci Methods 1988;25:1–11. Horvath TL, Peyron C, Diano S, Ivanov A, Aston-Jones G, Kilduff TS, et al. Hypocretin (orexin) activation and synaptic innervation of the locus coeruleus neurons. J Comp Neurol 1999;415:145–59. Hobson JA, Pace-Schott EF. The cognitive neuroscience of sleep: neuronal systems, consciousness and learning. Nat Rev Neurosci 2002;3:679–93. Honda T, Semba K. An ultrastructural study of cholinergic and non-cholinergic neurons in the laterodorsal and pedunculopontine tegmental nuclei in the rat. Neuroscience 1995;68:837–53. Hwang LL, Chen CT, Dun NJ. Mechanisms of orexininduced depolarizations in rat dorsal motor nucleus of vagus neurons in vitro. J Physiol (Lond) 2001;537:511–20. Ivanov A, Aston-Jones G. Hypocretin/orexin depolarizes and decreases potassium conductance in the locus coeruleus neurons. Neuroreport 2000;11:1755–8. Jones BE, Webster HH. Neurotoxic lesions of the dorsolateral pontomesencephalic tegmentum-cholinergic cell area in the cat. I. Effects upon the cholinergic innervation of the brain. Brain Res 1988;451:13–32. Kang Y, Kitai ST. Electrophysiological properties of pedunculopontine neurons and their postsynaptic responses following stimulation of substantia nigra reticulata. Brain Res 1990;535:79–95. Kawanabe Y, Hashimoto N, Masaki T, Miwa H. Ca2+ influx through nonselective cation channels plays an essential role in noradrenalin-induced arachidonic acid release in Chinese hamster ovary cells expressing a1A-, a1B-, or a1Dadrenergic receptors. J Pharmacol Exp Ther 2001;299:901–7. Kleyman TR, Cragoe Jr EJ. Amiloride and its analogs as tools in the study of ion transport. J Membr Biol 1988;105:1–21. Kohlmeier KA, Watanabe S, Tyler CJ, Burlet S, Leonard CS. Dual orexin actions on dorsal raphe and laterodorsal tegmentum neurons. J Neurophysiol 2008;100:2265–81. Koyama Y, Sakai K. Modulation of presumed cholinergic mesopontine tegmental neurons by acetylcholine and monoamines applied iontophoretically in unanesthetized cats. Neuroscience 2000;96:723–33.

peptides 30 (2009) 191–209

[26] Leonard CS, Llina´s R. Electrophysiology of mammalian pedunculopontine and laterodorsal tegmental neurons in vitro: implications for the control of REM sleep. In: Steriade M, Biesold D, editors. Brain cholinergic systems. New York: Oxford University Press; 1990. p. 205–23. [27] Lin L, Faraco J, Li R, Kadotani H, Rogers W, Lin X, et al. The sleep disorder canine narcolepsy is caused by a mutation in the hypocretin (orexin) receptor 2 gene. Cell 1999;98:365–76. [28] Liu R-J, van den Pol AN, Aghajanian GK. Hypocretins (orexins) regulate serotonin neurons in the dorsal raphe nucleus by excitatory direct and inhibitory indirect actions. J Neurosci 2002;22:9453–64. [29] Maloney KJ, Mainville L, Jones BE. Differential c-Fos expression in cholinergic, monoaminergic, and GABAergic cell groups of the pontomesencephalic tegmentum after paradoxical sleep deprivation and recovery. J Neurosci 1999;19:3057–72. [30] Marcus JN, Aschkenasi CJ, Lee CE, Chemelli RM, Saper CB, Yanagisawa M, et al. Differential expression of orexin receptors 1 and 2 in the rat brain. J Comp Neurol 2001;435:6–25. [31] McCarley RW. Neurobiology of REM and NREM sleep. Sleep Med 2007;8:302–30. [32] Murai Y, Akaike T. Orexins cause depolarization via nonselective cationic and K+ channels in isolated locus coeruleus neurons. Neurosci Res 2005;51:55–65. [33] Nambu T, Sakurai T, Mizukami K, Hosoya Y, Yanagisawa M, Goto K. Distribution of orexin neurons in the adult rat brain. Brain Res 1999;827:243–60. [34] Ohno K, Sakurai T. Orexin neuronal circuitry: role in the regulation of sleep and wakefulness. Front Neuroendocrinol 2008;29:70–87. [35] Reese NB, Garcia-Rill E, Skinner RD. The pedunculopontine nucleus—auditory input, arousal and pathophysiology. Prog Neurobiol 1995;47:105–33. [36] Sakai K, el Mansari M, Jouvet M. Inhibition by carbachol microinjections of presumptive cholinergic PGO-on neurons in freely moving cats. Brain Res 1990;527:213–23. [37] Sakurai T, Amemiya A, Ishii M, Matsuzaki I, Chemeli RM, Tanaka H, et al. Orexins and orexin receptors: a family of hypothalamic neuropeptides and G protein-coupled receptors that regulate feeding behavior. Cell 1998;92: 573–85. [38] Sanchez RM, Surkis A, Leonard CS. Voltage-clamp analysis and computer simulation of a novel cesium-resistant Acurrent in guinea pig laterodorsal tegmental neurons. J Neurophysiol 1998;79:3111–26. [39] Saper CB, Chou TC, Scammell TE. The sleep switch: hypothalamic control of sleep and wakefulness. Trends Neurosci 2001;24:726–31. [40] Steriade M, Datta S, Pare´ D, Oakson G, Curro´ Dossi RC. Neuronal activities in brain-stem cholinergic nuclei related

[41]

[42]

[43]

[44]

[45]

[46]

[47]

[48]

[49]

[50]

[51]

[52]

[53]

209

to tonic activation processes in thalamocortical systems. J Neurosci 1990;10:2541–59. Steriade M, Dossi RC, Pare´ D, Oakson G. Fast oscillations (20–40 Hz) in thalamocortical systems and their potentiation by mesopontine cholinergic nuclei in the cat. Proc Natl Acad Sci USA 1991;88:4396–400. Takahashi K, Koyama Y, Kayama Y, Yamamoto M. Effects of orexin on the laterodorsal tegmental neurones. Psychiatry Clin Neurosci 2002;56:335–6. Takakusaki K, Kitai ST. Ionic mechanisms involved in the spontaneous firing of tegmental pedunculopontine nucleus neurons of the rat. Neuroscience 1997;78:771–94. Takakusaki K, Saitoh K, Harada H, Kashiwayanagi M. Role of basal ganglia-brainstem pathways in the control of motor behaviors. Neurosci Res 2004;50:137–51. Takakusaki K, Saitoh K, Harada H, Okumura T, Sakamoto T. Evidence for a role of basal ganglia in the regulation of rapid eye movement sleep by electrical and chemical stimulation for the pedunculopontine tegmental nucleus and the substantia nigra pars reticulata in decerebrate cats. Neuroscience 2004;124:207–20. van den Pol AN, Ghosh PK, Liu RJ, Li Y, Aghajanian GK. Hypocretin (orexin) enhances neuron activity and cell synchrony in developing mouse GFP-expressing locus coeruleus. J Physiol (Lond) 2003;541:169–85. Vincent SR, Kimura H. Histochemical mapping of nitric oxide synthase in the rat brain. Neuroscience 1992;46: 755–84. Vincent SR, Satoh K, Armstrong DM, Fibiger HC. NADPHdiaphorase: a selective histochemical marker for the cholinergic neurons of the pontine reticular formation. Neurosci Lett 1983;43:31–6. Webster HH, Jones BE. Neurotoxic lesions of the dorsolateral pontomesencephalic tegmentum-cholinergic cell area in the cat. II. Effects upon sleep–waking states. Brain Res 1988;458:285–302. Willie JT, Chemelli RM, Sinton CM, Tokita S, Williams SC, Kisanuki YY, et al. Distinct narcolepsy syndromes in orexin receptor-2 and orexin null mice: molecular genetic dissection of non-REM and REM sleep regulatory processes. Neuron 2003;38:715–30. Wu M, Zaborszky L, Hajszan T, van den Pol AN, Alreja M. Hypocretin/orexin innervation and excitation of identified septohippocampal cholinergic neurons. J Neurosci 2004;24:3527–36. Yang B, Ferguson AV. Orexin-A depolarizes dissociated rat area postrema neurons through activation of a nonselective cationic conductance. J Neurosci 2002;22:6303–8. Yang B, Ferguson AV. Orexin-A depolarizes nucleus tractus solitarius neurons through effects on nonselective cationic and K+ conductances. J Neurophysiol 2003;89:2167–75.