HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) All Versions of this Article: 92/1/221    most recent 00076.2004v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (13) Citing Articles via Google Scholar Google Scholar Articles by Kohlmeier, K. A. Articles by Leonard, C. S. Search for Related Content PubMed PubMed Citation Articles by Kohlmeier, K. A. Articles by Leonard, C. S.

Hypocretin/Orexin Peptide Signaling in the Ascending Arousal System: Elevation of Intracellular Calcium in the Mouse Dorsal Raphe and Laterodorsal Tegmentum

¹Department of Physiology, New York Medical College, Valhalla, New York 10595; and ²Division of Molecular Neurobiology, The Institute of Medical Science, University of Tokyo, Tokyo 108-8639, Japan

Submitted 23 January 2004; accepted in final form 27 February 2004

	ABSTRACT

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION GRANTS ACKNOWLEDGMENTS REFERENCES

 
Dysfunction of the hypocretin/orexin (Hcrt/Orx) peptide system is closely linked to the sleep disorder narcolepsy, suggesting that it is also central to the normal regulation of sleep and wakefulness. Indeed, Hcrt/Orx peptides produce long-lasting excitation of arousal-related neurons, including those in the laterodorsal tegmentum (LDT) and the dorsal raphe (DR), although the mechanisms underlying these actions are not understood. Since Hcrt/Orx mobilizes intracellular calcium ([Ca²⁺]_i) in cells transfected with orexin receptors and since receptor-mediated Ca²⁺ transients are ubiquitous signaling mechanisms, we investigated whether Hcrt/Orx regulates [Ca²⁺]_i in the LDT and DR. Changes in [Ca²⁺]_i were monitored by fluorescence changes of fura-2 AM loaded cells in young mouse brain slices. We found Hcrt/Orx (Orexin-A, 30-1,000 nM) evoked long-lasting increases in [Ca²⁺]_i with differing temporal profiles ranging from spiking to smooth plateaus. A fragment of Hcrt/Orx (16-33) failed to evoke changes in [Ca²⁺]_i and changes were not blocked by TTX or ionotropic glutamate receptor antagonists, suggesting they resulted from specific activation of postsynaptic orexin receptors. Unlike orexin receptor–transfected cells, Hcrt/Orx-responses were not attenuated by depletion of Ca²⁺ stores with cyclopiazonic acid (CPA; 3-30 µM), thapsigargin (3 µM), or ryanodine (20 µM), although store-depletion by either CPA or ryanodine blocked Ca²⁺ mobilization by the metabotropic glutamate receptor agonist (±)-1-aminocyclopentane-trans-1,3-dicarboxylic acid (trans-ACPD; 30 µM). In contrast, Hcrt/Orx responses were strongly attenuated by lowering extracellular Ca²⁺ (20 µM) but were not inhibited by concentrations of KB-R7943 (10 µM) selective for blockade of sodium/calcium exchange. Nifedipine (10 µM), inhibited Hcrt/Orx responses but was more effective at abolishing spiking than plateau responses. Bay K 8644 (5-10 µM), an L-type calcium channel agonist, potentiated responses. Finally, responses were attenuated by inhibitors of protein kinase C (PKC) but not by inhibitors of adenylyl cyclase. Collectively, our findings indicate that Hcrt/Orx signaling in the reticular activating system involves elevation of [Ca²⁺]_i by a PKC-involved influx of Ca²⁺ across the plasma membrane, in part, via L-type calcium channels. Thus the physiological release of Hcrt/Orx may help regulate Ca²⁺-dependent processes such as gene expression and NO production in the LDT and DR in relation with behavioral state. Accordingly, the loss of Hcrt/Orx signaling in narcolepsy would be expected to disrupt calcium-dependent processes in these and other target structures.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION GRANTS ACKNOWLEDGMENTS REFERENCES

 
The recently discovered hypothalamic peptides known both as the hypocretins and orexins (Hcrt/Orx) (de Lecea et al. 1998; Kilduff and Peyron 2000; Peyron et al. 1998; Sakurai et al. 1998) were originally shown to stimulate feeding (Sakurai et al. 1998). A remarkable collection of evidence has now linked impaired signaling by this peptide system to narcolepsy (Chemelli et al. 1999; Lin et al. 1999; Nishino et al. 2000; Peyron et al. 2000; Thannickal et al. 2000), a disabling sleep disorder in which features of REM sleep intrude into wakefulness. Those findings and others suggest that the Hcrt/Orx system regulates feeding and autonomic function and is crucial for the normal regulation of wakefulness and sleep (for review, see Kilduff and Peyron 2000; Taheri et al. 2002). Since expression of REM and wakefulness is controlled in part by the reticular activating system, which includes neurons of the laterodorsal tegmentum (LDT) and dorsal raphe (DR), it is significant that these structures receive Hcrt/Orx input (Peyron et al. 1998) and express Hcrt/Orx receptors (Marcus et al. 2001). Moreover, brain slice recordings indicate that Hcrt/Orx evokes prolonged excitation of neurons in these (Brown et al. 2001; Burlet et al. 2002) and other arousal related structures such as locus ceruleus (LC) (Horvath et al. 1999; Ivanov and Aston-Jones 2000; van den Pol et al. 2002), tuberomammilary nucleus (TM) (Eriksson et al. 2001), basal forebrain (BF) (Eggermann et al. 2001), and nonspecific thalamic nuclei (Bayer et al. 2002).

Hcrt/Orx might also regulate [Ca²⁺]_i in these structures since Hcrt/Orx elevates [Ca²⁺]_i in cells transfected with orexin receptors (Lund et al. 2000; Sakurai et al. 1998; Smart et al. 1999), cultured hypothalamic and spinal cord neurons (van den Pol 1999; van den Pol et al. 1998), and dissociated ventral tegmental neurons (Uramura et al. 2001). Indeed, we have recently shown that Hcrt/Orx induces rises in [Ca²⁺]_i in TM neurons (Willie et al. 2003). This effect may be an important general action of Hcrt/Orx since elevation of free [Ca²⁺]_i regulates key neuronal functions including gene expression, neurotransmitter release, plasticity, neuronal excitability, and enzyme activity (Berridge 1998). In LDT neurons, regulation of [Ca²⁺]_i appears of particular importance since the spike-evoked Ca²⁺ influx is regulated by neurotransmitters (Kohlmeier and Leonard 2002; Leonard et al. 2000), and these neurons express high levels of the Ca²⁺-dependent enzyme NO synthase (Vincent and Kimura 1992). In fact, NO is produced locally by these neurons in a Ca²⁺-influx dependent manner (Leonard et al. 2001) and at their axonal terminals in thalamus (Williams et al. 1997) where production varies with the sleep-wakefulness state. Because of the important role Ca²⁺ may play in nuclei involved in arousal systems and the demonstrated role Hcrt/Orx plays in arousal related dysfunctions, we have examined whether Hcrt/Orx stimulates changes in [Ca²⁺]_i in neurons within the LDT and DR.

Our data show that Hcrt/Orx elevates [Ca²⁺]_i in neurons of the LDT and DR and that these increases result from Ca²⁺ influx, in part, via L-type calcium channels rather than from store-release as suggested by studies with transfected cells and involve PKC rather than PKA-related mechanisms. Accordingly, these Ca²⁺ signals may function to couple the physiological release of Hcrt/Orx to vital regulatory processes within these key nuclei controlling behavioral state.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION GRANTS ACKNOWLEDGMENTS REFERENCES

 
Brain slices

Frontal brain slices were prepared from normal 6- to 17-day-old C57/Bl6 mice (Charles River Laboratories) as previously described (Burlet et al. 2002). All procedures complied with National Institutes of Health and institutional guidelines for ethical use of animals and were approved by NYMC. Briefly, animals were deeply anesthetized by inhalation of isofluorane and decapitated into ice-cold artificial cerebral spinal fluid (ACSF). The brain was rapidly removed, blocked, and sectioned at 250 µm with a vibratome (VT1000S, Leica). Slices containing the DR and LDT were incubated in ACSF at 36°C for 15 min and were stored in oxygenated ACSF at room temperature following this incubation. For recordings, slices were submerged in a chamber (volume: 1.4 ml) perfused at 1-1.7 ml/min with warmed ACSF (24-28°C), which was set on a fixed stage of an Olympus upright BX50WI microscope. This slow perfusion system had a larger volume and slower flow-rate than one previously used (Burlet et al. 2002). Hence, the time course of Hcrt/Orx actions measured here are slower and not precisely comparable with that previously measured.

Solutions

The standard ACSF solution for recording was gassed with 95% O₂-5% CO₂ and contained (in mM) 121 NaCl, 5 KCl, 1.2 Na₂PO₄, 2.7 CaCl₂, 1.2 MgSO₄, 26 NaHCO₃, 20 dextrose, and 4.2 lactic acid. In some experiments, extracellular Ca²⁺ concentration ([Ca²⁺]_o) was reduced to 20 µM (calculated with Patcher's Power Tools for Igor Pro) by adding 2.7 mM EGTA to the ACSF and elevating MgSO₄ to 3.9 mM. Tissue was incubated in low [Ca²⁺]_o solution for 20 min before test drug application.

Drugs

All drug-containing solutions were freshly prepared for the experiments. Hcrt/Orx (Orexin-A, Sigma or American Peptides) was diluted to 100 µM in distilled water and frozen. On the day of the experiment, minutes before the peptide was to be applied, the aliquot was unfrozen and diluted in ACSF to a final concentration of 300 nM, unless noted otherwise. Preliminary experiments indicated that following a 40-min washout of Hcrt/Orx, Ca²⁺ responses were reproducible with a second application of peptide. Therefore 40 min was the recovery time utilized for the present experiments. An amide fragment of Hcrt/Orx (Orexin-A, peptide sequence 16-33; Phoenix Pharmaceuticals) was dissolved in normal ACSF and superfused at 300 nM or 1 µM as a control for nonspecific peptide actions. TTX (Alomone Labs) was dissolved in ACSF to a final concentration of 500 nM and applied to the slice for 10 min prior to the peptide to ensure compete block of voltage-dependent sodium channels. Cyclopiazonic acid (CPA, Calbiochem) aliquots (30 mM) were prepared in dimethyl sulfoxide (DMSO) and applied at a final concentration of 3-30 µM in ACSF and 500 nM TTX. Tissue was incubated in CPA/TTX for 15-30 min prior to test drug application to ensure block of the smooth endoplasmic reticulum calcium ATPase (SERCA) pump, and washout of this reversible inhibitor was 45 min. Thapsigargin (Calbiochem) was diluted in DMSO at a stock concentration of 3 mM and delivered in ACSF containing 500 nM TTX at a final concentration of 3 µM. Slices were incubated in this irreversible blocker of the SERCA pump for 30 min prior to application of Hert/Orx. Aliquots of ryanodine (Sigma) were prepared in DMSO at a stock concentration of 20 mM and applied in ACSF at a final concentration of 20 µM. KB-R7943 mesylate (Tocris), an inhibitor of the sodium/calcium exchanger, was dissolved in water, and aliquots were added to ACSF at a final concentration of either 80 or 10 µM. Nifedipine (Sigma) and (±)-Bay K 8644 (Calbiochem) were prepared fresh on the day of the experiment in DMSO and light protected until delivered to the slice at a final concentration of 10 and 5-10 µM, respectively. Bisindolylmaleimide I (Calbiochem) was used at a final concentration of 1 µM. Aliquots of 10 mM phorbal 12, 13-dibutryate (Sigma) dissolved in DMSO were added to ACSF and delivered at a final concentration of 10 µM. 2'5'Dideoxyadenosine (DDA; Calbiochem) was initially dissolved in DMSO and applied at a final concentration of 50 µM in ACSF for 30 min before Orexin. (±)-1-Aminocyclopentane-trans-1,3-dicarboxylic acid (t-ACPD, Tocris) was delivered in the ACSF at a final concentration of 30 µM for a total application time of 2 min.

Ca²⁺ imaging

DR and LDT cells were loaded with the Ca²⁺ indicator, fura-2 by incubating slices in ACSF containing 15 µM fura-2 AM (Molecular Probes) prepared from a 3.3 mM stock of fura-2 AM in DMSO. Slices were incubated for 30 min at 36°C in a small volume equilibrated with carbogen (5% CO₂-95% O₂). Slices were transferred to the recording chamber and rinsed for 30 min to ensure fura-2 AM de-esterification and temperature equilibration. Localization of the DR and LDT was determined with a 4x objective by brightfield illumination. Individual cells were then imaged using video-enhanced DIC optics using a 40x water immersion lens (Olympus; NA 0.8). In the majority of experiments, Ca²⁺ transients were monitored by measuring the emission at 515 nm resulting from excitation of fura-2 with 380 nm (F₃₈₀) using the 71000 Chroma fura 2 filter set and a shuttered 75-W Xenon light source (Osram, Berlin-München, Germany). While Hcrt/Orx-induced changes recovered, and in many cases, showed fast rise and decay times, to verify that changes in F₃₈₀ reflected peptide-induced activity and not changes in dye concentration, or optical path length, we utilized ratiometric measures (F₃₄₀/F₃₈₀) by manually altering excitation between 340 and 380 nm. These measures corroborated the F₃₈₀ data. Optical recordings were made using a frame-transfer cooled CCD camera system (EEV 57 chip, Micromax System, Roper Scientific). Initially a full frame image (512 x 512 pixels) of the entire field that encompassed many filled cells was acquired and compared with full frame images taken under DIC-bright field illumination to identify putative neurons. Following this identification, another image was taken but binned on chip at 4 x 4 pixels. Regions of interests (ROIs) were selected to encompass identified cells and analysis of fluorescence was conducted. Images had exposures of 600 ms and were collected every 1-4 s. The cooled CCD camera was controlled by, andthe data were collected with custom software written by, one of the authors (T. Inoue) and run on a Mac OS computer.

Relation of [Ca²⁺]_i levels to changes in fluorescence

Changes in fluorescence (F) were measured as dF/F, where F is the fluorescence at rest within a ROI following subtraction of background fluorescence that was defined as fluorescence measured from a region of the slice without labeled cells. dF/F is the change in fluorescence following subtraction of F measured before stimulation. dF/F was corrected for bleaching that occurred during the run. Using fura-2, rises in Ca²⁺ are reflected as a decrease in F₃₈₀ and an increase in F₃₄₀. dF/F ratios in all figures are presented such that upgoing traces represent rises in [Ca²⁺]_i. Ratio measurements were not converted to absolute values of [Ca²⁺]_i due to several well known uncertainties (as discussed in Connor and Cormier 2000). Differences between means were determined by utilization of a paired Student's t-test or a one-way ANOVA. Nonparametric comparisons were conducted using ² analysis.

Whole cell patch-clamp recordings

To verify that dF/F signals arose from neurons, we established whole cell patch recordings from fura-2–loaded cells that had first been identified as responding to Hcrt/Orx. Patch pipettes were fabricated from borosilicate glass (A-M Systems, 8250) and were filled with a solution containing (in mM) 144 K-gluconate, 0.2 EGTA, 3 MgCl₂, 10 HEPES, 0.3 NaGTP, and 4 Na₂ATP. Gigaseals were obtained under visual control using an Axoclamp 2A (Axon Instruments) operated in continuous voltage-clamp mode to monitor seal resistance (3-KHz output filter). After establishing whole cell recordings, the amplifier was switched to current-clamp mode (10-KHz output filter), and constant-current pulses were delivered to determine if the cells fired action potentials and were hence neurons. Pipettes also contained 50 µM of Alexa-594 (Molecular Probes) to visualize the recorded cell by using a Chroma 41004 Texas Red filter set.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION GRANTS ACKNOWLEDGMENTS REFERENCES

 
Hcrt/Orx elevated [Ca²⁺]_i in the DR and LDT

Changes in cellular fluorescence were monitored from regions determined to be in DR and LDT by initial visual inspection of the slices at low power (4x objective). Fura-2 loading of these slices resulted in a discrete pattern of cellular fluorescence readily observable using a 40x objective (Fig. 1, A and B). Comparing DIC images with fluorescence images verified that the fluorescence arose from cells that appeared to be neurons based on their size and processes (13-18 µm; Fig. 1, A and B). Application of Hcrt/Orx (Orexin-A, 300 nM) induced consistent, long-lasting but reversible changes in dF/F in many of these cells, suggesting that Hcrt/Orx evokes somatic elevations of [Ca²⁺]_i (Fig. 1, A and B). To control for possible alternative explanations of these fluorescence changes, we monitored fura-2 fluorescence ratiometrically (F₃₄₀/F₃₈₀) in a subset of cells before, during, and after this effect. Hcrt/Orx also reversibly increased the F₃₄₀/F₃₈₀ ratios, strongly indicating that Hcrt/Orx stimulates rises in [Ca²⁺]_i in these cells (Fig. 1C).

View larger version (80K): [in this window] [in a new window]   FIG. 1. Hypocretin/orexin (Hcrt/Orx) induced [Ca2+]i rises in both the dorsal raphe (DR) and the lateral dorsal tegmental (LDT) nucleus. A and B: left: full frame image of fura-2–loaded cells (F380) within the DR and LDT from which fluorescent changes were monitored. White boxes indicate the regions shown expanded to right. F380 of fura-2–loaded cells are shown in the middle panels, and the same cells are shown under bright field DIC conditions in right panels. Regions of interest (ROI) from which fluorescence signals were averaged are indicated by squares over the somata of these cells. Hcrt/Orx-induced [Ca2+]i rises are reflected by changes in dF/F as indicated in the right panels. C: histogram of ratiometric measures (F340/F380) showing the percent response following Hcrt/Orx application normalized to the baseline level of [Ca2+]i. Data corroborate the single wavelength measures and indicate that Hcrt/Orx elevates [Ca2+]i. Scale bars are 20 µm on full frame images and 10 µm on expanded images.

The transients induced by Hcrt/Orx varied from cell to cell but could be broadly categorized into three main profiles: 1) "Plateau" responders had relatively slow rise times (10-90% rise time, 90.0 ± 11.3 s, 2 DR slices, n = 5, 3 LDT slices, n = 6) that rose to a peak and returned to baseline over the course of several minutes; 2) "Spiker" responders had shorter transients with more rapid onsets (10-90% rise time, 11.2 ± 3.5 s, 2 DR slices n = 4, 2 LDT slices, n = 4), and shorter durations with faster recovery times (seconds) that exhibited a burst or oscillation behavior arising from a more stable baseline; and 3) "Plateau/Spiker" responders displayed a combination of the others in that briefer transients arose from a slower rising baseline. Figure 2, A and B, shows the heterogeneity of these Hcrt/Orx responses observed within single image fields in the DR and LDT. Response latency from the onset of Hcrt/Orx exposure was also variable (typical range: 2.7-5 min) but did not appear to be correlated with response profile. Different response types were often found to co-exist within the same field along with cells that failed to respond. Although we did not explicitly examine this issue due to differences in focal planes of each cell in relation to the chosen best imaged focal plane, our impression was that spiker responses appeared more prevalent among smaller cells. The peak amplitudes of transients exhibited a wide range; however, the average change in dF/F measured from DR cells displaying plateau responses was 13.5 ± 1.1% (32 DR slices, n = 199) and from the LDT was 11.6 ± 1.0% (24 LDT slices, n = 155; P > 0.05). We also found that the effects of Hcrt/Orx were repeatable after allowing a recovery time of 40-60 min between applications (P > 0.05, 2 DR slices, n = 4, 3 LDT slices, n = 6). In some cases, imaging from proximal dendrites was also possible, and there, Hcrt/Orx induced changes in dF/F as well with similar kinetics to those seen in the soma (Fig. 2C).

View larger version (60K): [in this window] [in a new window]   FIG. 2. Hcrt/Orx-induced [Ca2+]i rises exhibited multiple temporal profiles in both the DR and LDT. F380 images of cells within the (A) DR and (B) LDT from which changes in fluorescence were monitored. Numbers by fluorescence images correspond to each numbered dF/F trace on right, which reflects changes in fluorescence within the ROIs. Responses to Hcrt/Orx could be broadly categorized into "Plateau," "Spiker," or Plateau/Spiker" as indicated. C: dF/F traces showing that spiking responses also occurred in dendrites and that the transients recorded at dendrites exhibited similar kinetics to those recorded at the soma. Lack of response in ROI 4 that is adjacent to the soma and dendrite indicates the similarity of the dendritic signals is not due to scatter from the soma. D: histogram comparing response types between the DR and LDT reveal that the likelihood of Hcrt/Orx eliciting a particular response type was similar in each nucleus. Scale bar = 10 µm and applies to each image group.

Hcrt/Orx actions were dose-dependent and specific

Previous experiments suggested that maximum Hcrt/Orx-induced excitation of LDT neurons occurred at 300 nM (Burlet et al. 2002). We therefore examined concentrations of Hcrt/Orx ranging from 3-1,000 nM. We found that 30 nM produced just barely detectable fluorescence changes (0.1 ± 0.1%, 2 raphe slices, n = 4, 1 LDT slice, n = 3), while 1 µM produced the largest responses (Fig. 3). We also found that response profile was stable across a wide range of Hcrt/Orx concentrations. For example, we never found a case in which a plateau response turned into a transient response at higher doses over the range of 100 nM–1 µM (2 DR slices, n = 5, 3 LDT slices, n = 5). These data suggest that the different response types were not simply due to ligand concentration differences between cells.

View larger version (16K): [in this window] [in a new window]   FIG. 3. Hcrt/Orx-induced responses were dose-dependent and were not elicited by an Hcrt/Orx amide fragment. A and B: responses of 2 different cells to 4 concentrations of Hcrt/Orx-A shows that the response category appeared independent of the concentration of Hcrt/Orx. For example, cells that exhibited a plateau response at 100 nM also exhibited plateau responses at 300 nM and 1 µM (A). (C) Histogram detailing the response of DR and LDT cells to 4 concentrations of Hcrt/Orx. Responses are normalized to the response elicited at 1 µM. (D1) An amide fragment of Hcrt/Orx (Hcrt/Orx (16-33) did not elicit changes in dF/F in DR or LDT cells in which subsequent applications of Hcrt/Orx did elicit changes (D2).

Hcrt/Orx-induced rises in Ca²⁺ persisted in TTX and ionotropic glutamate receptor antagonists

Since Hcrt/Orx application evokes prolonged repetitive firing in LDT (Burlet et al. 2002) and DR neurons (Brown et al. 2001), we tested whether the changes in dF/F produced by Hcrt/Orx were sensitive to action potential blockade by TTX (0.5 µM). In the first group of cells studied (9 DR slices n = 33, 5 LDT slices n = 26), we found that all three response profiles were encountered in the presence of TTX, suggesting that none of these profiles required sodium-dependent action potentials. Moreover, the likelihood of encountering each type of response was not different in TTX.

In a separate population of cells, we also compared the effect of Hcrt/Orx applied after action potentials were blocked by TTX and again following 1 h of wash out of TTX, which was sufficient for recovery of action potentials as determined in separate experiments. Using plateau responses which were most easily quantifiable, we found that the response increased from 10.3 ± 0.4 to 13.5 ± 0.3% following washout of TTX (P < 0.05, 3 DR slices, n = 15, 3 LDT slices, n = 20; Fig. 4). TTX also had minor effects on spiking and plateau/spiking responses indicating that action potential-evoked Ca²⁺ influx contributed little to the elevation of [Ca²⁺]_i we measured following Hcrt/Orx application.

View larger version (36K): [in this window] [in a new window]   FIG. 4. All response categories were still evident following the blockade of voltage-gated sodium channels with TTX. Hcrt/Orx-evoked changes in dF/F were similar during spike blockade with TTX (top) and after washout of TTX. However, note that while 4 of the 5 cells that responded to Hcrt/Orx in control artificial cerebrospinal fluid (ACSF) also responded in TTX, the response was reduced in amplitude and spikes were reduced in amplitude and frequency during the blockade.

Hcrt/Orx-induced rises in [Ca²⁺]_i were resistant to depletion of intracellular Ca²⁺ stores

To determine if these Hcrt/Orx-induced rises in [Ca²⁺]_i involved the release of Ca²⁺ from intracellular stores, as observed in heterologous expression systems (Sakurai et al. 1998; Smart et al. 1999), we depleted intracellular Ca²⁺ stores with the SERCA pump inhibitors, thapsagargin (3 µM), or the reversible cyclopiazonic acid (CPA; 3-30 µM). Continuous application of thapsigargin or CPA in the presence of TTX produced a transient rise in dF/F, suggesting depletion of Ca²⁺ from SERCA pump–dependent intracellular stores was successful (latency: 85 ± 4.2 s). Nevertheless, all three response profiles observed prior to depletion were observed in the presence of CPA or thapsigargin. Indeed, the average amplitude of the plateau response (12.3 ± 2.3%) was not different from that in control conditions (12.5 ± 2.6%; 4 DR slices, n = 12, 4 LDT slices, n = 13; P > 0.05; Fig. 5, A, C, and D).

View larger version (26K): [in this window] [in a new window]   FIG. 5. Depletion of intracellular Ca2+stores with CPA or thapsigargin did not attenuate Hcrt/Orx-induced rises in [Ca2+]i. A: dF/F values for 5 cells from the same field are displayed following Hcrt/Orx application, prior to (left) and following CPA application (right). Rises in dF/F were seen following CPA application (middle). B: CPA effectively depleted intracellular Ca2+ stores since t-ACPD-mediated mobilization of Ca2+ from these stores was abolished by CPA. C: baseline spiking was also present in CPA-store depleted conditions. D: responses to Hcrt/Orx were not attenuated following depletion of intracellular stores of Ca2+ by thapsigargin. Thus smooth endoplasmic reticulum calcium ATPase (SERCA) pump–mediated intracellular Ca2+ stores are not involved in Hcrt/Orx induced [Ca2+]i rises. TTX was present throughout these experiments.

While the data indicate that Hcrt/Orx-induced Ca²⁺ transients in the CNS are SERCA pump independent, it has been reported in other cell types that calcium-induced calcium-release (CICR) channels (ryanodine receptors) can trigger Ca²⁺ release from stores that are resistant to depletion by thapsigargin or CPA (Golovina and Blaustein 1997; Murphy and Miller 1989; Thayer et al. 1988) suggesting that in some cell types, SERCA pump and ryanodine receptor-mediated stores are separate (Murphy and Miller 1989; Thayer et al. 1988). We therefore examined the possibility that Hcrt/Orx-induced Ca²⁺ rises were influenced by ryanodine receptors (calcium-induced calcium release channels) located on thapsigargin/CPA-insensitive stores. Responses to Hcrt/Orx were examined following preincubation of the slices in 20 µM ryanodine that blocks the ryanodine receptor in the open configuration (Rousseau et al. 1987). Application of ryanodine alone did not cause detectable changes in fluorescence, as has been previously reported (McPherson et al. 1991). In the presence of ryanodine, each response type previously observed following Hcrt/Orx application were also observed, and there was no significant difference in the amplitude of the plateau responses observed in control (18.2 ± 4.5%) and ryanodine conditions (23.1 ± 7.6%; P > 0.05; 2 DR slices, n = 10, 2 LDT slices, n = 14; Fig. 6 A), indicating that CICR is not necessary for Hcrt/Orx-induced Ca²⁺ transients.

View larger version (29K): [in this window] [in a new window]   FIG. 6. Blockade of the ryanodine receptor did not inhibit Hcrt/Orx responses. A: control responses (left) were not attenuated after application of ryanodine (right). B: in contrast, control responses to t-ACPD (left) were nearly abolished (right) by application of ryanodine. Thus Hcrt/Orx-induced rises in [Ca2+]i were not dependent on mobilization of Ca2+ from these intracellular stores. TTX was present throughout these experiments.

Hcrt/Orx-induced rises in [Ca²⁺]_i were nearly abolished by lowering extracellular [Ca²⁺]_o

To determine if Hcrt/Orx induced rises required external Ca²⁺, we compared responses in control and low extracellular Ca²⁺ ([Ca²⁺]_o) conditions. At a calculated [Ca²⁺]_o of 20 µM, we found that all three profiles of Hcrt/Orx-induced responses were significantly attenuated from those elicited in the same DR and LDT cells under control conditions (Fig. 7, A and B). The average plateau response was reduced by 83.8 ± 0.7% from that seen in control conditions (P < 0.05, 9 DR slices, n = 17, 5 LDT slices, n = 18; Fig. 7C). Moreover, spiking responses observed in control conditions were abolished in low [Ca²⁺]_o conditions in the same cells. Thus lowering the [Ca²⁺]_o strongly attenuated the response which recovered to near 100% following reintroduction of [Ca²⁺]_o (P > 0.05, 1 DR slice, n = 2, 1 LDT slice, n = 1; Fig. 7A).

View larger version (31K): [in this window] [in a new window]   FIG. 7. Hcrt/Orx-induced rises in [Ca2+]i were dependent on influx of Ca2+ from the extracellular space. A: control responses in the DR (left) were significantly attenuated by low Ca2+ ACSF (middle). Recovery of responses occurred following return to normal Ca2+ ACSF (right). B: Hcrt/Orx-induced responses in LDT cells were also attenuated in low Ca2+ ACSF. Data from 2 different slices show that low Ca2+ ACSF attenuated responses for all categories of responders C: summary histogram showing the percent change in fluorescence in control solution, low Ca2+ solution, and following washout of low Ca2+ solution normalized to responses obtained in control conditions. Data indicate that Hcrt/Orx-induced [Ca2+]i rises depend on extracellular Ca2+ influx.

A possible explanation for the dependence on [Ca²⁺]_o is that the Hcrt/Orx receptor(s) couple to the sodium/calcium exchanger (NCE). Indeed, it has recently been reported that a drug that interferes with the NCE attenuates Hcrt/Orx-induced inward currents (Burdakov et al. 2003; Eriksson et al. 2001; Wu et al. 2002). While forward operation of the pump that would generate an inward current and extrude Ca²⁺ from the cell could not be responsible for Hcrt/Orx-induced responses observed here, it is possible that reverse operation of the pump could contribute to the observed responses. Therefore we examined the effect of the NCE inhibitor, KB-R7943, at two concentrations on responses to Hcrt/Orx. We found that at 10 µM, a concentration reported to be at the high end of specificity of inhibition of the pump (Iwamoto et al. 1996), all three response types to Hcrt/Orx were similar to those seen in control conditions (plateau response in 10 µM KB-R7943, 98.9 ± 1.0% of control, P > 0.05, 1 DR slice, n = 2; 2 LDT slices, n = 4, Fig. 8). At 80 µM, responses to Hcrt/Orx were significantly attenuated (plateau response, 49.8 ± 0.5% of control, P < 0.05, 4 DR slices, n = 4, 3 LDT slices, n = 4); however, at this concentration the drug is probably not acting specifically (Iwamoto et al. 1996). Hence, these data suggest that the Hcrt/Orx-induced calcium influx in DR and LDT cells does not depend on operation of the NCE.

View larger version (18K): [in this window] [in a new window]   FIG. 8. Selective inhibition of the Na+/Ca2+ exchanger (NCE) did not attenuate Hcrt/Orx-induced rises in [Ca2+]i. A1: Hcrt/Orx responses in control conditions and with 2 concentrations of KB-R7943, an inhibitor of NCE. At 10 µM, a concentration reported to specifically inhibit the NCE, Hcrt/Orx responses, including "spiker" responses (A2), were not significantly different from control. At the higher concentration, which has been reported to have nonspecific effects, Hcrt/Orx responses were partially inhibited. B: histogram summarizes the effect of these two KB-R7943 concentrations on Hcrt/Orx-induced changes in [Ca2+]i that are normalized to control responses.

Antagonists of L-type calcium channels have been reported to reduce the Hcrt/Orx evoked increase in [Ca²⁺]_i in ventral tegmental neurons (Uramura et al. 2001). Since L-type channels are present on LDT neurons (Kohlmeier and Leonard 2002) and DR neurons (Penington et al. 1991), we examined the effects of nifedipine, an L-channel inhibitor on responses induced by Hcrt/Orx in DR and LDT neurons. Ten micromolar nifedipine attenuated Hcrt/Orx-induced plateau responses to 44.5 ± 5.6% of control (P < 0.05; 8 DR slices, n = 16; 6 LDT slices, n = 12; Fig. 9) and abolished (n = 13/22) or attenuated (n = 5/22) the majority of spiker responses (Fig. 9A). Since nifedipine was more effective at abolishing baseline spiking than plateau responses (P < 0.05; ² test), it appears L-type calcium channels play a greater role in generating Hcrt/Orx-induced spiking than plateau responses. Nevertheless, nifedipine-insensitive spontaneous spiking was also observed in some cells, suggesting that nifedipine-sensitive channels are not necessary for spiking behavior in some cells.

View larger version (20K): [in this window] [in a new window]   FIG. 9. L-type calcium channels contribute to Hcrt/Orx-induced responses in the DR and LDT. A: nifedipine (10 µM; in ACSF with TTX) attenuated or abolished both plateau and spiking responses evoked by Hcrt/Orx in the DR (A1) and LDT (A2). Spiking responses were more consistently abolished than plateau response (see RESULTS). B: Bay-K 8644, an L-type Ca2+ channel agonist, augmented Hcrt/Orx-induced Ca2+ rises in the DR (B1) and LDT (B2). In some cases, Bay-K 8644 alone induced spiking that could obscure Hcrt/Orx-induced responses. C: Hcrt/Orx-evoked calcium transients do not result from depolarization alone. Strong depolarization by AMPA (1 µm) induced Ca2+ transients (C1). Following blockade of action potentials with TTX, the AMPA-mediated depolarization failed to elicit Ca2+ transients (C2). Nevertheless, subsequent application of Hcrt/Orx in TTX induced both plateau and spiking Ca2+ transients (C3). Collectively, data suggest that mechanisms in addition to depolarization are required to generate Hcrt/Orx-evoked calcium transients.

To investigate the possibility that Hcrt/Orx-mediated depolarization alone could account for the Ca²⁺ transients, we monitored the response of DR and LDT neurons to bath application of 1 µM AMPA, which produces greater inward somatic current in LDT neurons than does Hcrt/Orx (compare Burlet et al. 2002; Sanchez and Leonard 1996) and evokes repetitive firing in quiescent DR and LDT neurons (data not shown). AMPA produced an increase in somatic [Ca²⁺]_i that was completely abolished by TTX, indicating that the Ca²⁺ signal resulted from a spike-evoked Ca²⁺ influx rather than the AMPA-mediated depolarization (n = 8). In contrast, in the presence of TTX, Hcrt/Orx evoked both plateau and spiker responses in these same cells (Fig. 9C₃). These data suggest that depolarization alone is not the mechanism by which Hcrt/Orx induces somatic Ca²⁺ transients.

Hcrt/Orx-induced rises in [Ca²⁺]_i depend in part on PKC

Hcrt/Orx receptors have been shown to stimulate PLC (Lund et al. 2000). Because the [Ca²⁺]_i rises in the LDT and DR did not appear to be dependent on IP₃-mediated release from intracellular stores, we examined the DAG branch of the PLC pathway using two different PKC inhibitors: bisindolylmaleimide I and chelerythrine chloride. We found that these inhibitors significantly attenuated all types of Hcrt/Orx-evoked Ca²⁺ rises (range of reduction: 100-11%; mean reduction: 54.8 ± 0.9%, P < 0.05, 9 DR slices, n = 20, 7 LDT slices, n = 17; Fig. 10). Since bisindolylmaleimide I also has been reported to have weak inhibitory effects on PKA (Toullec et al. 1991), we examined the effect of adenylyl cyclase inhibition by 2'5'dideoxyadenosine (DDA) as a control. In eight cells in which Hcrt/Orx was found to induce rises in [Ca²⁺]_i, responses to Hcrt/Orx were not significantly attenuated in the presence of DDA (plateau response control amplitude, 13.9 ± 4.4 vs. 13.5 ± 4.4% in DDA, 2 DR slices, n = 3, 2 LDT slices, n = 5, P > 0.05, Fig. 10). Cells exhibiting the spiker response also appeared unaffected. Supporting a role of PKC mechanisms in Hcrt/Orx-induced rises in [Ca²⁺]_i, we found that application of a phorbal ester induced patterns of Ca²⁺ rises similar to those elicited by Hcrt/Orx (n = 23; Fig. 10D).

View larger version (21K): [in this window] [in a new window]   FIG. 10. Inhibition of protein kinase C (PKC) attenuated Hcrt/Orx-induced rises in [Ca2+]i. A1 and A2: responses to Hcrt/Orx under control conditions (left) were significantly attenuated in the presence of the PKC inhibitor bisindolylmaleimide I (1 µM). B: similarly, Hcrt/Orx responses were attenuated in the presence of the structurally dissimilar PKC inhibitor, chelerythrine chloride (10 µM). Because bisindolylmaleimide I has been reported to also have actions on protein kinase A, responses to Hcrt/Orx were also examined in the presence of an adenylyl cyclase inhibitor, DDA. Hcrt/Orx-induced Ca2+ transients were not attenuated in the presence of 50 µM DDA (C). D: a phorbal ester that results in production of PKC-induced patterns of rises in [Ca2+]i similar, albeit larger in amplitude, to those elicited by Hcrt/Orx. E: histogram comparing percentage of reduction of changes in dF/F from control in the presence of nifidipine, bisindolylmaleimide I, and a combination of the two. Data indicate a PKC-regulated mechanism is involved in Hcrt/Orx responses and activation of L-type calcium channels probably involves this 2nd messenger.

Hcrt/Orx elevated [Ca²⁺]_i in neurons rather than glia

To verify that Hcrt/Orx-induced changes in fluorescence arose from neurons rather than glia, we patch-clamped 14 fura-2AM loaded cells that exhibited Hcrt/Orx induced transients (4 DR slices, n = 5; 7 LDT slices, n = 9) and 2 that did not respond. Since fura-2 fluorescence rapidly washed out of recorded cells after establishing the whole cell configuration, we included the dye Alexa-594 in the pipettes so that recorded cells could be visualized while recording. In all cases, depolarizing current steps (100- to 500-ms duration) evoked one or more overshooting action potentials (peak > +10 mV), indicating the imaged cells were neurons (Fig. 11).

View larger version (119K): [in this window] [in a new window]   FIG. 11. Hcrt/Orx-evoked calcium transients arose from neurons in the DR and LDT. Fura-2–loaded cells (F380) within the DR (A1) and LDT (B1) from which fluorescence changes were monitored (scale in top right A1 = 20 µm and applies to all panels). Overlaid traces show changes in dF/F induced by Hcrt/Orx in the indicated cells (scale bars = 10% dF/F and 2 min). DIC images of the corresponding cells are shown in A2 and B2 (arrows). Whole cell recordings were obtained from these cells that are shown during the recordings after filling with Alexa-594 (560 ± 27 nm; A3 and B3) from the patch pipette (right). Overlaid traces show action potentials elicited by depolarizing current steps, showing that recorded cells were neurons (scale bars = 10 mV and 250 ms).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION GRANTS ACKNOWLEDGMENTS REFERENCES

Ca²⁺ Influx versus release from intracellular stores

Orexin receptors (Ox1R and Ox2R) were originally discovered through screening an orphan G protein–coupled receptor library for ligand-evoked cytoplasmic Ca²⁺ transients (Sakurai et al. 1998). Results from subsequent studies with CHO cells transfected with Ox1Rs implicated activation of phospholipase C (PLC) and the IP₃-mediated release of Ca²⁺ from SERCA pump–dependent intracellular stores that activated a Ca²⁺ influx pathway that was inhibited following thapsigargin-mediated depletion of internal stores, suggesting a role of store-operated channels (Lund et al. 2000; Smart et al. 1999). Subsequently, this influx pathway was pharmacologically distinguished from store-operated channels in CHO cells (Kukkonen and Akerman 2001). In contrast, we found that the Ca²⁺ transients evoked by Hcrt/Orx in the LDT and DR were blocked by lowering [Ca²⁺]_o, but were insensitive to depletion of intracellular stores, even at Hcrt/Orx concentrations that produce robust store-dependent Ca²⁺ transients in Ox1R-transfected CHO cells. The apparent store independence of these transients did not result from a lack of CPA/thapsigargin-sensitive stores since Ca²⁺ mobilization by t-ACPD was blocked by CPA. Thus activation of native OxRs in LDT and DR appears to engage a Ca²⁺ influx pathway without the store-dependent release mechanism observed in transfected cells. This agrees with previous observations in cultured hypothalamic neurons where orexin-B evokes Ca²⁺ transients that are thapsigargin-insensitive but dependent on [Ca²⁺]_o (van den Pol 1999). It is also consistent with observations from acutely isolated ventral tegmentum neurons (Uramura et al. 2001) where low concentrations of orexin-A (10 nM) evoked Ca²⁺ transients that depended on extracellular Ca²⁺, although in that study, the sensitivity to store depletion was not tested.

Second messengers and effectors

Our results and other published evidence indicate that the Ca²⁺ influx pathway(s) activated by Hcrt/Orx is partially mediated by activation of PKC. Evidence from CHO-Ox1R cells indicate that inhibition of PLC abolishes Hcrt/Orx-evoked Ca²⁺ transients (Lund et al. 2000) but that inhibition of IP₃ receptors only inhibited responses that lead to Ca²⁺ release from intracellular stores (Kukkonen and Akerman 2001). Consistent with this, experiments using cultured hypothalamic neurons found that Hcrt/Orx-evoked Ca²⁺ transients were strongly attenuated by compounds inhibiting PKC but not by those inhibiting PKA (van den Pol et al. 1998) and PKC involvement was also found in Hcrt/Orx-induced transients in isolated VTA neurons (Uramura et al. 2001). Our results using PKC and PKA inhibitors in LDT and DR were consistent with these studies. Additionally, we found that a phorbal ester induced patterns of rises in [Ca²⁺]_i that were similar to those elicited by Hcrt/Orx. Thus there is an emerging consensus that a Ca²⁺ influx pathway activated by native OxRs involves PKC.

Several effectors might contribute to the observed Ca²⁺ influx. The sodium/calcium exchanger (NCE) appears to mediate the depolarization produced by Hcrt/Orx in septal and arcuate nucleus neurons (Burdakov et al. 2003; Wu et al. 2002) and in the tuberomammillary nucleus (Eriksson et al. 2001) where Hcrt/Orx also elevates [Ca²⁺]_i (Willie et al. 2003). Since reverse operation of the NCE, which elevates [Ca²⁺]_i in cardiac muscle (Bridge et al. 1990; Wier 1990) and is found in neurons (Li et al. 1994), might account for the Ca²⁺ influx, we examined whether the NCE inhibitor KB-R7943 attenuated Hcrt/Orx-evoked Ca²⁺ transients. At a concentration of 10 µM, which inhibits >80% of the reverse mode operation of the NCE (Iwamoto et al. 1996), the Hcrt/Orx-evoked Ca²⁺ transients were not attenuated, although at 80 µM, these transients were reduced by about 50%. However, since KB-R7943 at 30 µM also inhibits dihydropyridine-sensitive Ca²⁺ channels and voltage-dependent sodium channels (Iwamoto et al. 1996), this effect of high concentrations of KB-R7943 is unlikely to be specific. Hence, the NCE does not appear to mediate the Ca²⁺ influx evoked by Hcrt/Orx in DR and LDT neurons.

It is more likely that the Ca²⁺ influx is mediated by the opening of ion channels. Accordingly, we observed that Hcrt/Orx-induced Ca²⁺ transients were significantly attenuated by an L-type calcium channel antagonist and enhanced in the presence of an L-type channel agonist. These findings are consistent with results from dissociated VTA neurons (Uramura et al. 2001) and cultured somatotrophs where Hcrt/Orx augments an L-type calcium current (Xu et al. 2002). Nevertheless, blockade of L-type calcium channels did not abolish all Hcrt/Orx responses. While this may have resulted from incomplete inhibition of L-type calcium channels, this seems unlikely since nifedipine abolished the Ca²⁺ transients in many cells, especially those with spiking responses. Hence, it is likely that additional calcium permeable channels are involved in mediating the Hcrt/Orx evoked Ca²⁺ influx in many DR and LDT cells.

Although we did not measure whole cell currents in this study, we previously found that Hcrt/Orx evokes an inward current with an increase in membrane conductance in LDT neurons (Burlet et al. 2002), and it is known to depolarize DR neurons (Brown et al. 2001). Since both these actions appear to involve a nonselective cation current (Brown et al. 2002; Liu et al. 2002; Tyler et al. 2001), it is possible that this current is partly carried by Ca²⁺ ions and may contribute to the nifedipine-resistant responses. Indeed, such a nonspecific cation current was suggested to account for the orexin-evoked Ca²⁺ influx in CHO cells (Lund et al. 2000). In contrast, even though large Ca²⁺ transients were evoked in cultured hypothalamic neurons by Hcrt/Orx, no whole cell currents were observed when these cells were recorded under whole cell conditions (van den Pol 1999). A possible explanation of this discrepancy is that recordings were made using a high chloride pipette solution in the van den Pol study that we found to inhibit the Hcrt/Orx-evoked current (Burlet et al. 2002), probably by uncoupling G protein–coupled receptors (Lenz et al. 1997). Thus accumulating evidence suggests one or more Ca²⁺-permeable ion channels are involved. Future experiments using simultaneous ion imaging and current recording will be necessary to determine the relative roles played by each possible effector.

Diverse temporal profiles of Hcrt/Orx-evoked Ca²⁺ transients

We also found that Hcrt/Orx evoked Ca²⁺ transients with diverse temporal profiles in the LDT and DR. While a few previous studies of mammalian neurons have reported Ca²⁺ transients evoked by Hcrt/Orx (Uramura et al. 2001; van den Pol 1999; van den Pol et al. 1998), none have reported such diverse patterns. Such diverse profiles are common in nonneuronal cells and can arise from multiple mechanisms (Berridge and Dupont 1994). While spiking and oscillating Ca²⁺ transients are often associated with the release of Ca²⁺ from intracellular stores (Berridge and Dupont 1994), such behavior can also be produced by Ca²⁺ influx through membrane channels (Byron and Taylor 1993), as appears to be the case for all the responses we observed. While it is possible that a thapsigargin/CPA and ryanodine-insensitive store (e.g., NAADP-sensitive store) may have contributed (Genazzani et al. 1997; Lee 1997), such a store has not been observed in neurons and is unlikely to account for the immediate attenuation of Hcrt/Orx responses produced by lowering extracellular Ca²⁺. Furthermore, all response types were sensitive to nifedipine, which reduced or abolished the spiking responses more effectively than the plateau responses. This indicates a role of L-type calcium channels in each response pattern and an even greater role in spiking. Interestingly, these responses did not appear to result simply from membrane depolarization. We found that concentrations of AMPA expected to produce a larger somatic depolarization than Hcrt/Orx did not produce any somatic Ca²⁺ transients in the absence of TTX-sensitive action potentials although Hcrt/Orx did evoke somatic Ca²⁺ transients in the same cells in the presence of TTX. Since intracellular calcium diffusion is highly restricted (Connor and Nikolakopoulou 1982), these transients most likely arose from calcium permeable channels, including L-type calcium channels located at the soma. Since the AMPA-mediated depolarization failed to activate these channels, additional modulation by Hcrt/Orx receptor activation may be necessary. Our finding that PKC inhibition was not additive with L-channel inhibition is consistent with the possibility that Hcrt/Orx receptor activation of PKC promotes L-channel activity in LDT and DR neurons. Ca²⁺ entry through L-type calcium channels may be of considerable functional significance since it can preferentially couple to immediate early gene transcription (Deisseroth et al. 2003). Moreover, the particular temporal dynamics of the Ca²⁺ signals appears to be important in determining which transcription pathways become activated (Dolmetsch et al. 1997). Thus one intriguing possibility is Hcrt/Orx can engage different transcription pathways within subpopulations of LDT and DR neurons depending on the temporal profile of the evoked Ca²⁺ response.

Neuronal heterogeneity of the LDT and DR

Both the LDT and DR contain neurons expressing different transmitter phenotypes. Evidence from whole cell recordings in the LDT (Burlet et al. 2002) and DR (Liu et al. 2002) indicate that both the principal neurons (cholinergic in LDT and serotonergic in DR) and the other neurons, presumably including GABAergic neurons, are excited by Hcrt/Orx. A different situation has been observed in some other structures. For example, in the basal forebrain, Hcrt/Orx excites cholinergic neurons but not GABAergic neurons (Eggermann et al. 2001) and in the arcuate nucleus, Hcrt/Orx excites a population of electrophysiologically distinct GABA neurons (Burdakov et al. 2003). While we were unable to identify the histochemical nature of the neurons in our study, we expect that both principal and other neurons were imaged in the LDT and DR. Given previous electrophysiological findings, it is therefore likely that all types of neurons in these structures exhibit Ca²⁺ transients elicited by Hcrt/Orx. Thus Hcrt/Orx may provide similar biophysical signals to both the principal and GABAergic cells of these nuclei that also appear to be involved in behavioral state control (Maloney et al. 1999). Nevertheless, future simultaneous recording and imaging studies will be necessary to explicitly examine this point. Such studies should also prove useful in determining if Ca²⁺ signals with particular temporal profiles are associated with specific transmitter phenotypes.

Metabotropic glutamate receptors in the LDT

Another new observation was that the activation of metabotropic glutamate receptors with t-ACPD evokes rises in [Ca²⁺]_i via store-mediated release in LDT neurons. These data indicate for the first time that metabotropic glutamate receptors (mGluRs) are present in LDT and that they are capable of mobilizing Ca²⁺ from intracellular stores—presumably via production of IP₃ (Sugiyama et al. 1987). LDT neurons receive fast glutamatergic synaptic input (Sanchez and Leonard 1996), which is important for generating REM sleep and wakefulness (Datta et al. 2001). Moreover, glutamatergic input can both be inhibited (Arrigoni et al. 2001) and enhanced (Burlet et al. 2002), suggesting that regulation of excitatory synaptic pathways are important in the state-dependent control of these and other neurons (Peever et al. 2003). The fact that Hcrt/Orx enhances synaptic release of glutamate also suggests that Hcrt/Orx might regulate [Ca²⁺]_i both by enhancing synaptic activation of mGluRs and by directly elevating [Ca²⁺]_i by receptor stimulated influx. Such a dual action could enhance the temporal concordance of these two Ca²⁺ signals. This may be of considerable functional interest since the coincidence of Ca²⁺ transients arising from mGluR activation and either action potentials (Nakamura et al. 1999) or NMDA receptors (Nakamura et al. 2002) can result in very large propagating Ca²⁺ waves. Depending on when Hcrt/Orx release occurs in these nuclei (Estabrooke et al. 2001; Kiyashchenko et al. 2002; Torterolo et al. 2001), temporal concordance of these calcium signals might occur at different times since presumed cholinergic LDT and serotonergic DR neurons have different firing patterns across the sleep-wake cycle. During waking, firing in both groups is high, while during REM sleep, firing in a subpopulation of presumed cholinergic neurons is high but the serotonergic neurons are silent (Saper et al. 2001). Future studies will be necessary to examine both the interactions and biological consequences of such calcium signals in these neurons which appear so critical for sleep and wakefulness.

	GRANTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION GRANTS ACKNOWLEDGMENTS REFERENCES

 
This study was supported by National Institutes of Health Grants NS-10877 to K. A. Kohlmeier and NS-27881 and HL-64150 to C. S. Leonard.

	ACKNOWLEDGMENTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION GRANTS ACKNOWLEDGMENTS REFERENCES

 
We thank Drs. C. Tyler, N. Ross, and W. Ross for many useful discussions and comments on a previous version of this manuscript.

	FOOTNOTES

 
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

	REFERENCES

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION GRANTS ACKNOWLEDGMENTS REFERENCES

 
Arrigoni E, Rainnie DG, McCarley RW, and Greene RW. Adenosine-mediated presynaptic modulation of glutamatergic transmission in the laterodorsal tegmentum. J Neurosci 21: 1076–1085, 2001.[Abstract/Free Full Text]

Bayer L, Eggermann E, Saint-Mleux B, Machard D, Jo