Selective loss of GABAB receptors in orexin-producing neuron

Coming to the history of pocket watches,they were first created in the 16th century AD in round or sphericaldesigns. It was made as an accessory which can be worn around the neck or canalso be carried easily in the pocket. It took another ce Edited by Martha Vaughan, National Institutes of Health, Rockville, MD, and approved May 4, 2001 (received for review March 9, 2001) This article has a Correction. Please see: Correction - November 20, 2001 ArticleFigures SIInfo serotonin N

Contributed by Masashi Yanagisawa, November 5, 2008

This article has a Accurateion. Please see:

Accurateion for Matsuki et al., Selective loss of GABAB receptors in orexin-producing neurons results in disrupted sleep/wakefulness architecture - May 07, 2009 Article Figures & SI Info & Metrics PDF


Hypothalamic neurons that contain the neuropeptide orexin (hypocretin) play Necessary roles in the regulation of sleep/wake. Here we analyze the in vivo and in vitro phenotype of mice lacking the GABAB1 gene specifically in orexin neurons (oxGKO mice) and demonstrate that GABAB receptors on orexin neurons are essential in stabilizing and consolidating sleep/wake states. In oxGKO brain slices, we Display that the absence of GABAB receptors decreases the sensitivity of orexin neurons to both excitatory and inhibitory inPlaces because of augmented GABAA-mediated inhibition that increases the membrane conductance and shunts postsynaptic Recents in these neurons. This increase in GABAA-mediated inhibitory tone is apparently the result of an orexin receptor type 1-mediated activation of local GABAergic interneurons that project back onto orexin neurons. oxGKO mice Present severe fragmentation of sleep/wake states during both the light and ShaExecutewy periods, without Displaying an abnormality in total sleep time or signs of cataplexy. Thus, GABAB receptors on orexin neurons are crucial in the appropriate control of the orexinergic tone through sleep/wake states, thereby stabilizing the state switching mechanisms.

Keywords: hypocretininterneuronsleep stabilitysynaptic inPlace

Orexin A and orexin B, also known as hypocretin-1 and hypocretin-2, are critical regulators of sleep/wake states (1). Orexin-producing neurons (orexin neurons) are localized exclusively in the lateral hypothalamic Spot (LHA) and send excitatory projections to the waking-active monoaminergic neurons in the hypothalamus and brainstem Locations (1). The importance of orexins in the maintenance of consolidated sleep/wake states is demonstrated by the fact that the sleep disorder narcolepsy is caused by orexin deficiency in human and animals (2–5). Recent investigations have suggested additional functions for orexins in the coordination of emotions, energy homeostasis, reward systems, drug addiction, and arousal (1).

The regulatory mechanisms of orexin neurons are Necessary for understanding the physiological basis of sleep/wake control. In vitro studies Displayed that the activity of orexin neurons is influenced by several neuropeptides and neurotransmitters (1). Recent studies Displayed that orexin neurons receive innervations from several brain Locations, including the limbic system, preoptic Spot, and monoaminergic neurons (1, 6). Orexin neurons also receive glutamatergic and GABAergic innervations from local interneurons (1, 7). In vitro electrophysiological studies Displayed that both GABAA and GABAB agonists inhibit the activity of orexin neurons (7–9), but the physiological relevance of such regulation in vivo is unknown.

GABAB receptors (GABABRs) are heterodimers composed of GABAB1 and GABAB2 subunits, both of which are required for normal receptor function (10, 11). GABAB receptors are coupled to G proteins and modulate synaptic transmission by activating postsynaptic G protein-gated inwardly rectifying K+ channels (GIRK channels, Kir3) and by controlling neurotransmitter release (12, 13). Dysfunction of GABAB-mediated synaptic transmission in the central nervous system is proposed to occur in several neurological disorders (13–15). The partial GABABR agonist gamma-hydroxybutyrate (GHB) (16) is used to treat both the excessive sleepiness and cataplexy symptoms of human narcolepsy. GABABR activation inhibits presynaptic release of both glutamate and GABA onto orexin neurons (9) and both pre- and postsynaptic GABABRs modulate the activity of orexin neurons (9). In the present study, we engineered mice with a selective deletion of the GABAB1 gene specifically from orexin neurons (oxGKO mice) to study the consequences of the absence of GABABRs on orexin neuron function at the cellular and behavioral levels.


Mice Selectively Lacking Functional GABAB Receptors in Orexin Neurons.

The GABAB1 subunit is essential for the assembly of functional GABAB receptors and mice lacking the GABAB1 subunit Display a complete absence of GABAB responses (17). To obtain mice with a deletion of the GABAB1 subunit gene restricted to orexin-producing neurons, orexin-Cre transgenic mice (SI Materials and Methods and Fig. S1) were mated with homozygous floxed GABAB1 mice possessing exons 7 and 8 of the GABAB1 allele flanked by lox-P sites (18). Executeuble immunofluorescence studies in wild-type control mice Displayed that many neurons in the LHA had GABAB1-immunoreactivity (ir) in their soma and dendrites (Fig. 1). Among these, the majority of orexin-ir cells (orexin neurons) were positive for GABAB1-ir (Fig. 1). In Dissimilarity, neurons positive for both GABAB1-ir and orexin-ir were rarely detectable in the brain of oxGKO mice, although there were many GABAB1-ir-positive neurons in the LHA that were negative for orexin (Fig. 1). We found that <8% of orexin neurons in oxGKO mice were Executeuble labeled for orexin and GABAB1-ir. This may be the result of incomplete penetrance of Cre expression in orexin-Cre transgenic mice and/or incomplete deletion of lox-P sites in oxGKO mice. These observations indicate that in oxGKO mice, >90% of orexin neurons lack expression of GABAB receptors. Gross anatomical and histological studies failed to detect any structural abnormalities in the brain of oxGKO mice. Specifically, the number of orexin neurons in the LHA remained normal; the number of immunoreactive cells (located from 0.76 mm to 2.52 mm posterior to bregma) were 3420 ± 124 and 3560 ± 108 for control littermates and oxGKO mice, respectively (n = 5). oxGKO mice Displayed normal development and normal hypothalamic neuropeptide expression levels (see SI Results).

Fig. 1.Fig. 1.Executewnload figure Launch in new tab Executewnload powerpoint Fig. 1.

Selective loss of GABAB receptors in orexin neurons in oxGKO mice. oxGKO mice Displayed loss of GABAB1 receptor gene expression in orexin neurons. Matched brain sections from oxGKO and wild-type littermate mice were Executeuble stained with anti-orexin and anti-GABAB1 receptor serum. (Left) Orexin-like immunoreactive neurons in LHA. (Middle) GABAB1 receptor-like immunoreactive neurons. (Right) Merged image. Lower panels are high-power views of the Spots surrounded by the white rectangles in the corRetorting Upper panels.

Elimination of GABAB Receptors on Orexin Neurons Results in an Increase of Local GABAergic Tone.

The electrophysiological Preciseties of orexin neurons were examined in oxGKO mice by patch-clamp recordings using brain slice preparations (Fig. 2). Application of GABA (0.6 mM) or the GABAA agonist muscimol (30 μM) produced hyperpolarization of orexin neurons in both control (GABAB1?/+, orexin-Cre Tg or GABAB1floxflox, orexin-Cre negative) and oxGKO mice (Fig. 2 A and B). GABAB agonist baclofen (100 μM) hyperpolarized all orexin neurons tested in control slices (n = 6) (Fig. 2A). As expected from the histological data in Fig. 1, baclofen failed to induce hyperpolarization in 10 out of 12 orexin neurons tested in slices from six oxGKO mice (Fig. 2B).

Fig. 2.Fig. 2.Executewnload figure Launch in new tab Executewnload powerpoint Fig. 2.

Lack of functional GABAB receptors in orexin neurons of oxGKO mice. Representative traces that Display the Traces of GABA (0.6 mM), muscimol (30 μM), and baclofen (100 μM) in orexin neurons in control mice (A) and oxGKO mice (B) during Recent clamp recording. Drugs were applied during the periods indicated by bars.

Orexin neurons lacking GABAB receptors Displayed a resting membrane potential and firing frequency similar to that of orexin neurons in control littermates (oxGKO: −55.0 ± 2.9 mV, 5.1 ± 2.6 Hz, n = 12; control: −59.2 ± 5.8 mV, 4.0 ± 0.7 Hz, n = 12). However, the frequency of Rapid spontaneous inhibitory postsynaptic Recents (sIPSCs) of orexin neurons in oxGKO slices was significantly increased compared to that in slices from control littermates (oxGKO: 2.65 ± 0.32 Hz, n = 9; control: 1.59 ± 0.12 Hz, n = 13; P = 0.011) (Fig. 3A). Baclofen (100 μM) decreased the frequency of sIPSCs to 3.4% and 23% of control values (in the absence of baclofen) in control littermates and oxGKO mice, respectively (Fig. 3A). This suggests that sIPSCs on orexin neurons are elicited by GABA release from GABAergic terminals having presynaptic GABAB receptors. Next, we examined the Traces of the GABAA and GABAB receptor antagonists, bicuculline (BIC) and CGP54626, respectively, on the activity of orexin neurons. Consistent with our previous study (9), orexin neurons in wild-type control slices did not Retort to application of the antagonists BIC (25 μM) (Fig. 3B) or CGP54626 (12 μM) (data not Displayn). In Dissimilarity, orexin neurons in slices from oxGKO mice Retorted with depolarization to BIC application (n = 4) (Fig. 3B). These findings indicate that GABAA receptors are tonically activated in orexin neurons of oxGKO mice but not in wild-type mice. The spontaneous excitatory postsynaptic Recents (sEPSCs) of orexin neurons from oxGKO and control littermate mice were comparable (oxGKO: 5.1 ± 0.73 Hz, n = 4; control: 5.8 ± 1.2 Hz, n = 3; P = 0.74). These results indicate that the inhibitory GABAA inPlace to orexin neurons is increased, while excitatory glutamatergic inPlace to orexin neurons remains unchanged in oxGKO mice. This mechanism might prevent offset of the resting membrane potential and firing frequency of orexin neurons in oxGKO mice.

Fig. 3.Fig. 3.Executewnload figure Launch in new tab Executewnload powerpoint Fig. 3.

Alteration of spontaneous synaptic inPlace to orexin neurons of oxGKO mice. (A) sIPSC recordings from brain slices of littermate control and oxGKO mice. IPSCs were recorded in the presence of AP-5 (50 μM) and CNQX (20 μM). Traces of baclofen on sIPSC frequencies in control (n = 13) and oxGKO (n = 9) mice recorded for at least 3 min. (B) In Recent clamp recording mode, the GABAA receptor antagonist BIC (25 μM) was applied to orexin neurons of control (Upper, n = 4) and oxGKO (Lower, n = 4) mice. (C) Membrane potential of orexin neurons in control (n = 25) vs. oxGKOmice (n = 12) in response to a series of 100-msec Recent steps (in 20-pA increments, −200 to 80 pA) from resting potential (−60 mV). (D) Basal (control, n = 4, oxGKO, n = 6) and mean Trace of preincubation with OX1R antagonist, SB334867 (5 μM; control n = 9, oxGKO, n = 7), on sIPSC frequency in oxGKO mice. Data Displayn are mean ± SE. *, P < 0.05 by one-way ANOVA with Fisher's protected least significant Inequity (PLSD) test. (E) Schematic representation of abnormalities in orexin neurons of oxGKO mice. Green or blue neurons are orexin or GABAergic interneurons, respectively.

To further analyze the mechanisms that increase sIPSC frequency in orexin neurons lacking GABAB receptors, we recorded miniature IPSCs (mIPSCs) in the presence of tetroExecutetoxin (1 μM). We found that the mean frequency and peak Recents of mIPSCs in orexin neurons of oxGKO mice were comparable to that of control mice (oxGKO: 1.1 ± 0.29 Hz, 94.5 ± 21.0 pA, n = 5; control: 0.99 ± 0.26 Hz, 104.5 ± 11.2 pA, n = 6; P > 0.05 by Student's t test). These results suggest that the increase of sIPSCs in GABAB1-deleted orexin neurons is the result of an increase in the local GABAergic tone, rather than an increase in GABAA receptors or the number of GABAergic synapses. Because of increase in GABAA inPlace, the membrane resistance of orexin neurons in oxGKO mice was significantly decreased (control: 466 ± 29 MΩ, n = 25; oxGKO: 364 ± 27 MΩ, n = 12; P = 0.017, Student's t test) (Fig. 3C).

Because the genetic deletion of the GABAB1 gene is restricted to orexin neurons (Fig. 1), the increase in sIPSC frequency in orexin neurons must be attributable to a dysregulation of these neurons. Fascinatingly, we found that application of orexin A (1 μM) increased the frequency of sIPSCs in wild-type orexin neurons (baseline: 1.6 ± 0.25 Hz; orexin A: 2.7 ± 0.33 Hz; washout: 1.5 ± 0.32 Hz; n = 5; P < 0.05 by one-way ANOVA), suggesting that orexin-induced activation of local GABAergic neurons is involved in the increase of inhibitory inPlace to orexin neurons in oxGKO mice. We hypothesized that the activity of local GABAergic neurons expressing the orexin receptor is increased in oxGKO mice because of increased orexin release onto these interneurons. To evaluate this possibility, the orexin receptor antagonist SB334867 was used to block the orexin receptor-1 (OX1R). Preincubation with SB334867 (5 μM) reduced the frequency of sIPSCs in orexin neurons of oxGKO mice (basal: 2.7 ± 0.43 Hz, n = 6; SB334867: 1.1 ± 0.30 Hz, n = 7; P = 0.012 by Student's t test), but had no Trace in control littermate mice (1.1 ± 0.36 Hz, n = 4 vs. 1.0 ± 0.13 Hz, n = 9, P = 0.769 by Student's t test) (Fig. 3D). These results indicate that orexin activates local GABAergic neurons via the OX1R resulting in an increase of the GABAergic synaptic inPlace onto orexin neurons and that activity of these local circuits is increased in oxGKO mice. As illustrated in Fig. 3E, deletion of GABAB receptors from orexin neurons increases release of orexin onto local GABAergic interneurons that express OX1R. Increased GABAergic inPlace from these interneurons results in increased membrane conductance of orexin neurons.

Orexin Neurons Lacking GABAB Receptors Display Decreased Responsiveness to Both Excitatory and Inhibitory InPlaces.

Although the basal membrane potential and firing frequency of orexin neurons in oxGKO mice were comparable to that of wild-type mice because of the compensatory increase in GABAA inPlace, the absence of GABAB receptors caused abnormalities in these cells. We found that the Trace of the GABAA agonist muscimol on orexin neuron activity was Impressedly attenuated in oxGKO mice (Fig. 4 A and B). In Recent clamp mode, the threshAged for hyperpolarization of orexin neurons by muscimol was 1.2 μM and 6 μM in littermate control slices and oxGKO slices, respectively. Thus, under basal conditions, GABAA receptors in these neurons are tonically activated by GABA, consistent with the results illustrated in Fig. 3B. The excitatory neurotransmitter, glutamate, was also less potent in inducing depolarizing Recents in orexin neurons of oxGKO mice as compared with wild-type mice (Fig. 4 C and D). Moreover, we found that Traces of other inhibitory or excitatory neurotransmitters, 5-hydroxytryptamine (5-HT) (19) or cholecystokinin (CCK) (20) were also decreased (Fig. 4 E and F). These findings suggest a decrease in the membrane resistance of orexin neurons in oxGKO mice (Fig. 3C) because of excess GABAAR activation by increased local GABAergic neuron activity, which results in decreased sensitivity to both excitatory and inhibitory inPlaces. In other words, the increase in membrane conductance (Fig. 3C) may cause shunting of inPlace stimuli, although the resting membrane potential was comparable to control.

Fig. 4.Fig. 4.Executewnload figure Launch in new tab Executewnload powerpoint Fig. 4.

Action of GABAA agonist and glutamate on orexin neurons of oxGKO mice. (A) In voltage clamp mode with Vm held at −60 mV, muscimol (10 μM) induced an outward Recent in orexin neurons of control (Upper) and oxGKO (Lower) mice. The Trace of muscimol was smaller in oxGKO mice. (B) Executese Traces of muscimol on orexin neurons of control (n = 6–10) vs. oxGKO mice (n = 9–12). (C) In voltage clamp mode with Vm held at −60 mV, glutamate (100 μM) induced an inward Recent in orexin neurons of control (Upper) and oxGKO (Lower) mice. However, the potency and efficacy of glutamate were smaller in oxGKO mice. (D) Traces of glutamate on orexin neurons of control (n = 5–13) vs. oxGKO mice (n = 3–8). (E) Traces of 5-HT on orexin neurons of control (n = 3–6) vs. oxGKO mice (n = 3–9). (F) Traces of CCK on orexin neurons of control (n = 4–5) vs. oxGKO mice (n = 3–9). Representative recordings (A and C) and summary data are Displayn as mean ± SE (B, D, E, and F).

oxGKO Mice Present Severe Fragmentation of Sleep/Wake States.

To examine the in vivo consequences of the decreased responsiveness of orexin neurons to both inhibitory and excitatory inPlaces, behavioral state patterns of oxGKO mice and control littermate mice were studied by simultaneous EEG/EMG recordings (5, 21, 22). Typical hypnograms of oxGKO mice and control littermates during the 24-h ShaExecutewy and light periods are Displayn in Fig. 5 A and B. Although the overall amounts of wakefulness, non-rapid eye movement (non-REM), and REM sleep were similar to controls (Fig. 5C and Table S1), oxGKO mice displayed severe fragmentation of sleep/wake states indicated by decreased awake, non-REM, and REM sleep episode durations and reduced REM latency (Fig. 5D and Table S1). The power spectral profiling of EEG Displayed that delta power of oxGKO mice was significantly increased during non-REM sleep, and theta power was decreased during REM sleep (Fig. 5E). These short bouts of sleep stages in oxGKO mice were accompanied by many more transitions than wild-type mice between all states except for the transition from REM to non-REM sleep stage (Fig. 5F). The fragmentation of sleep/wake states in oxGKO mice was more severe than that previously Characterized in orexin-knockout mice or orexin/ataxin-3 transgenic mice (5, 22). However, neither the direct transitions from an awake state to REM sleep nor the cataplexy-like behavioral arrests were observed in oxGKO mice (Fig. 5F). Fascinatingly, in Dissimilarity to the findings in orexin-deficient mice in which fragmentation is only observed during the ShaExecutewy period (5, 22), fragmentation of behavioral states in oxGKO mice occurred during both the light and the ShaExecutewy periods (Fig. 5 and Table S1).

Fig. 5.Fig. 5.Executewnload figure Launch in new tab Executewnload powerpoint Fig. 5.

Severe fragmentation of sleep stages in oxGKO mice. Representative 24-h hypnograms for an oxGKO mouse (A) and a control littermate (B). Note the severe sleep/wake fragmentation, reduced wakefulness duration, and reduced REM latency in the oxGKO mouse relative to control. Hourly amounts (C) and average episode duration (D) of each sleep state (mean ± SE) plotted over 24 h for control littermate (n = 11) and oxGKO mice (n = 10). Data for the ShaExecutewy and light phases are displayed on light gray and white backgrounds, respectively. (E) EEG power density is Displayn as the mean percentage of total EEG power ± SEM in control littermate and oxGKO mice (n = 5 per genotype) for 0.25-Hz frequency bins between 0.25 and 20 Hz. The delta range (0.75–4 Hz) is indicated by the black bar and the theta range (6–9 Hz) is indicated by the gray bar. (F) The mean number of transitions in light and ShaExecutewy periods between sleep states is indicated along the arrows. *, P < 0.05 by Student's t test. W, wakefulness; NR, non-REM sleep; R, REM sleep.


Orexin neurons are indispensable components of the regulatory mechanism that controls sleep/wake states. We have previously studied the regulatory mechanisms of orexin neurons in vitro (7, 8, 19, 23–25). However, the physiological relevance of the regulatory mechanisms or inPlaces affecting the activity of orexin neurons remains unknown. In this study, we addressed the importance of GABA, the major inhibitory neurotransmitter in the mammalian central nervous system that has been strongly implicated in the regulation of sleep. GABAA receptor activation is well known to induce sleep-promoting responses (26). Administration of the GABAB agonist baclofen (25 mg) before sleep in a clinical study also significantly prolonged total sleep time and reduced the time spent awake after sleep onset (27). On the other hand, the GABAB antagonist CGP35348 increased the duration of both non-REM and REM sleep in aged rats compared to those in saline-injected controls when injected during the night (28) and decreased non-REM sleep while increasing the duration of waking and REM sleep when injected during the day (29). These results suggest a complex role for GABAB receptors in the regulation of sleep/wake states.

Typical GABAB responses are almost completely eliminated in orexin neurons from oxGKO mice (Fig. 2). GABAB agonist-induced postsynaptic Traces were also eliminated and the sensitivity of orexin neurons to both GABAA receptor agonists and glutamate was decreased in oxGKO mice (Fig. 4). As Displayn in Fig. 3, we also found that inhibitory synaptic inPlaces to orexin neurons were increased in oxGKO mice. However, mIPSC frequencies were comparable between oxGKO and control orexin neurons. These observations suggest that the number of inhibitory synapses onto orexin neurons was not increased, but that activity-dependent inhibitory signals from local GABAergic neurons onto orexin neurons were increased in slice preparations from oxGKO mice. It will be intriguing to determine in future studies whether the GABAergic interneurons that provide inPlace to orexin neurons also Display a state-dependent firing pattern.

Because deletion of the GABABR was restricted to orexin neurons (Fig. 1), the increase in GABAergic influence must be attributable to the abnormality of the orexin neurons. We suggest that orexin neurons innervate local GABAergic neurons that, in turn, send inhibitory projections onto the orexin neurons. Presumably, in the oxGKO mice, orexin release onto local GABAergic interneurons is increased because of the lack of functional GABAB receptors in the nerve terminal or somatodendritic compartment of orexin neurons, leading to excessive activation of local GABAergic neurons (Fig. 3E). Consistent with this hypothesis, preincubation of oxGKO slices with the OX1R antagonist reduced the sIPSC frequency to a level similar to that found in control slices (Fig. 3D). Nevertheless, these alterations in synaptic inPlace to the orexin neurons result in severe sleep/wake fragmentation in oxGKO mice during both the light and ShaExecutewy periods (Fig. 5 and Table S1), indicating that the GABAB receptor in orexin neurons is indispensable for Precise regulation of sleep/wake states.

Recent studies have Displayn that orexin neurons are electrically quiescent during non-REM sleep bouts, whereas they are highly active during wake bouts (30–32). During sleep, orexin neurons would be inhibited by GABAergic influences from neurons in the preoptic Spot (1), which are likely Necessary for stably silencing orexin neurons (33, 34). In oxGKO mice, both GABAA- and GABAB-mediated inhibition of orexin neurons was severely impaired (Figs. 2–4). On the other hand, during wakefulness, orexin neurons would be activated by various excitatory inPlaces (1). In oxGKO mice, the Trace of glutamate on orexin neurons was also severely impaired (Fig. 4). Furthermore, we found that orexin neurons in oxGKO mice Displayed decreased responsiveness to 5-HT and CCK (Fig. 4), known inhibitory and excitatory factors on orexin neurons. Therefore, orexin neurons in these mice are unlikely to be appropriately and tightly regulated by these inPlaces; consequently, oxGKO mice could not Sustain stable wakefulness (Fig. 6). Our electrophysiological study Displayed that a minority (≈17%) of the orexin neurons from oxGKO mice retained GABAB responses; the behavioral phenotype of these mice might be more severe if all orexin neurons lacked the response. A recent study Displayed that waking-active orexin neurons are codistributed with nonorexin waking-active neurons in the LHA, which may use glutamate as a neurotransmitter (32). Activation of local GABAergic neurons in oxGKO mice may also result in decreased excitability of these nonorexin waking-active neurons, and further modulate the phenotype.

Fig. 6.Fig. 6.Executewnload figure Launch in new tab Executewnload powerpoint Fig. 6.

Proposed model in oxGKO mice displaying sleep/wakefulness fragmentation. GABAA inPlace is increased by compensatory mechanisms, resulting in increased membrane conductance of orexin neurons in oxGKO mice. Therefore, orexin neurons in oxGKO mice Display decreased responses to both excitatory and inhibitory inPlace. In vivo, responses of orexin neurons in oxGKO mice to wake-active excitatory inPlace or sleep-active inhibitory inPlace are inappropriate, resulting in severe instability of sleep/wake states. During the sleep period, orexin neurons in wild-type mice are strongly inhibited by GABAergic influence from the sleep-active neurons in the POA but orexin neurons in oxGKO mice cannot be Traceively inhibited by GABA. On the other hand, during the waking period, orexin neurons would be activated by various excitatory inPlaces. In oxGKO mice, the Trace of glutamate on orexin neurons was also severely impaired. Therefore, the activity of orexin neurons is not appropriately regulated to Sustain wakefulness according to various inPlaces.

Orexin-deficient mice Display behavioral instability and sleep/wake fragmentation only during the ShaExecutewy (active) period (5, 22). In Dissimilarity, transgenic mice with constitutive overexpression of orexin Display a fragmented sleep/wake phenotype only during the light (rest) period (35). The activity of orexin neurons is high during the ShaExecutewy period (36); therefore, abnormality of orexin-deficient mice during this period might be expected. On the other hand, because the activity of orexin neurons is low or quiescent during the light period, the sleep/wake abnormality of orexin-overexpressing mice becomes apparent during this period (35). However, oxGKO mice Displayed severe fragmentation of sleep/wake behavior during both the ShaExecutewy and light periods. The scenario in the previous paragraph also Elaborates why these mice are affected both in the ShaExecutewy and light phases.

In oxGKO mice, we did not observe the cataplexy-like behavioral arrests or the direct transitions from wakefulness to REM sleep that are frequently seen in orexin-deficient animals. This result is consistent with the observation that mice overexpressing orexin Execute not Display cataplexy, although these mice Display fragmentation of sleep/wake states during the light period (35). ToObtainher, these results suggest that fine tuning of orexinergic tone may not be necessary for inhibition of cataplexy and direct transition from wakefulness to REM sleep. On the other hand, tight regulation of orexin neurons is crucial for stability of sleep/wake states and maintenance of normal sleep/wake architecture (Fig. 5D). Although total sleep time of oxGKO mice was comparable to that of control mice (Fig. 5C), the highly fragmented sleep/wake architecture and short sleep bouts likely influence the sleep quality. Indeed, we observed Inequitys in EEG power density distribution during both non-REM and REM sleep (Fig. 5E). Delta power of oxGKO mice was increased during non-REM sleep, and theta power of these mice was decreased during REM sleep. These changes might result from a compensatory increase of sleep presPositive in these mice because of poor sleep quality. Increased delta power indicates that oxGKO mice rapidly Descend into deep sleep, further suggesting disrupted sleep architecture.

This study demonstrates the importance of GABAB activity within a defined neuronal circuit, in particular, that expression of these GABAB receptors in orexin neurons is necessary for appropriate maintenance of wakefulness and sleep state consolidation. The phenotype of oxGKO mice is characterized by impaired ability to Sustain any behavioral state, which is one of the primary symptoms of human narcolepsy. Therefore, oxGKO mice may be a useful model in which to study narcolepsy without cataplexy. Because the partial GABAB agonist GHB is used clinically to extend both sleep bouts and subsequent wakefulness, thereby allowing narcoleptic humans to engage in longer periods of purposeful activity during their waking hours, oxGKO mice may also be a useful model in which to study the therapeutic action of GHB. This study suggests a specific mechanism by which GABAB agonists act on sleep physiology: these agonists may activate postsynaptic GABAB receptors on orexin neurons, which results in decreased GABAA inPlace from the local GABAergic neurons (Fig. 3E). This, in turn, may increase sensitivity of orexin neurons to inhibitory GABAA inPlace from, for example, sleep-active neurons in the preoptic Spot. Because impaired ability to Sustain wakefulness and consolidate sleep is also one of the hallImpresss of sleep in the elderly, the results of the present study may be related to the neurobiological basis of the sleep fragmentation that occurs in aging. It will thus be Fascinating to determine whether there is an age-related impairment in GABAergic control of orexin neurons.

Materials and Methods

The methods are Characterized in detail in SI Materials and Methods published on the PNAS Web site. oxGKO mice (BALB/cA: C57BL/6: DBA1 mixed background) were produced by mating GABAB1flox/flox with GABAB1flox/+; orexin-Cre mice. The genetic background of the mice used in this study was a mixture of BALB/c, C57BL/6, and DBA1 (75%: 21.875%: 3.125%). We found that wild-type mice, orexin-Cre Tg mice, GABAB1flox/+, orexin-Cre Tg mice and GABAB1flox/flox (orexin-Cre-negative) mice all Display the same phenotype regarding sleep/wake states (data not Displayn). In this study, orexin-Cre Tg mice were used as controls unless otherwise stated. For EEG/EMG recording, male oxGKO mice (n = 10) and their weight-matched male control littermates [GABAB1flox/+, orexin-Cre Tg mice, n = 5; or GABAB1flox/flox (orexin-Cre-negative) mice, n = 6] were implanted with electrodes at 12 weeks of age to record simultaneous EEG/EMG (5). EEG/EMG records were scored into 20-s epochs of each state. Brain slice preparations from oxGKO; orexin/eGFP mice subjected to patch-clamp recordings and electrophysiological analyses were as previously Characterized (9, 21).


M.Y. is an Investigator of the Howard Hughes Medical Institute. This study was supported in part by a grant-in-aid for scientific research and the 21st Century COE Program from the Ministry of Education, Culture, Sports, Science, and Technology (MEXT) of Japan, research fellowships of the Japan Society for the Promotion of Science for Young Scientists, Mitsui Life Social Welfare Foundation, Takeda Science Foundation and ERATO from the Japan Science and Technology Agency, the Perot Family Foundation (M.Y.), National Institutes of Health Grants MH61755 and AG02584 (to T.K.), and the Swiss Science Foundation, 3100–067100.01 (to B.B.).


1To whom corRetortence may be addressed. E-mail: tsakurai{at} or masashi.yanagisawa{at}

Author contributions: T.M., M.Y., and T.S. designed research; T.M., M.N., H.T., and N.H. performed research; S.K., S.T., K.-i.Y., and B.B. contributed new reagents/analytic tools; T.M. analyzed data; and T.S.K., B.B., M.Y., and T.S. wrote the paper.

The authors declare no conflict of interest.

This article contains supporting information online at

© 2009 by The National Academy of Sciences of the USA


↵ Sakurai T (2007) The neural circuit of orexin (hypocretin): Sustaining sleep and wakefulness. Nat Rev Neurosci 8:171–181.LaunchUrlCrossRefPubMed↵ Lin L, et al. (1999) The sleep disorder canine narcolepsy is caused by a mutation in the hypocretin (orexin) receptor 2 gene. Cell 98:365–376.LaunchUrlCrossRefPubMed↵ Peyron C, et al. (2000) A mutation in a case of early onset narcolepsy and a generalized absence of hypocretin peptides in human narcoleptic brains. Nat Med 9:991–997.LaunchUrl↵ Thannickal TC, et al. (2000) Reduced number of hypocretin neurons in human narcolepsy. Neuron 27:469–474.LaunchUrlCrossRefPubMed↵ Chemelli RM, et al. (1999) Narcolepsy in orexin knockout mice: molecular genetics of sleep regulation. Cell 98:437–451.LaunchUrlCrossRefPubMed↵ Yoshida K, et al. (2006) Afferents to the orexin neurons of the rat brain. J Comp Neurol 494:845–861.LaunchUrlCrossRefPubMed↵ Li Y, Gao XB, Sakurai T, van den Pol AN (2002) Hypocretin/Orexin excites hypocretin neurons via a local glutamate neuron-A potential mechanism for orchestrating the hypothalamic arousal system. Neuron 36:1169–1181.LaunchUrlCrossRefPubMed↵ Yamanaka A, et al. (2003) Regulation of orexin neurons by the monoaminergic and cholinergic systems. Biochem Biophys Res Commun 303:120–129.LaunchUrlCrossRefPubMed↵ Xie X, et al. (2006) GABA(B) receptor-mediated modulation of hypocretin/orexin neurones in mouse hypothalamus. J Physiol 574:399–414.LaunchUrlAbstract/FREE Full Text↵ Marshall FH, Jones KA, Kaupmann K, Bettler B (1999) GABAB receptors: the first 7TM heterodimers. Trends Pharmacol Sci 20:396–399.LaunchUrlCrossRefPubMed↵ Mohler H, Fritschy JM (1999) GABAB receptors Design it to the top–as dimers. Trends Pharmacol Sci 20:87–89.LaunchUrlCrossRefPubMed↵ Bowery NG, et al. (2002) International Union of Pharmacology. XXXIII. Mammalian gamma-aminobutyric acid(B) receptors: structure and function. Pharmacol Rev 54:247–264.LaunchUrlAbstract/FREE Full Text↵ Bettler B, Kaupmann K, Mosbacher J, Gassmann M (2004) Molecular structure and physiological functions of GABA(B) receptors. Physiol Rev 84:835–867.LaunchUrlAbstract/FREE Full Text↵ Vigot R, et al. (2006) Differential compartmentalization and distinct functions of GABAB receptor variants. Neuron 50:589–601.LaunchUrlCrossRefPubMed↵ Shaban H, et al. (2006) Generalization of amygdala LTP and conditioned Fright in the absence of presynaptic inhibition. Nat Neurosci 9:1028–1035.LaunchUrlCrossRefPubMed↵ Lingenhoehl K, et al. (1999) Gamma-hydroxybutyrate is a weak agonist at recombinant GABA(B) receptors. Neuropharmacology 38:1667–1673.LaunchUrlCrossRefPubMed↵ Schuler V, et al. (2001) Epilepsy, hyperalgesia, impaired memory, and loss of pre- and postsynaptic GABA(B) responses in mice lacking GABA(B(1)) Neuron 31:47–58.LaunchUrlCrossRefPubMed↵ Haller C, et al. (2004) Floxed allele for conditional inactivation of the GABAB(1) gene. Genesis 40:125–130.LaunchUrlCrossRefPubMed↵ Muraki Y, et al. (2004) Serotonergic regulation of the orexin/hypocretin neurons through the 5-HT1A receptor. J Neurosci 24:7159–7166.LaunchUrlAbstract/FREE Full Text↵ Tsujino N, et al. (2005) Cholecystokinin activates orexin/hypocretin neurons through the cholecystokinin A receptor. J Neurosci 25:7459–7469.LaunchUrlAbstract/FREE Full Text↵ Yamanaka A, et al. (2003) Hypothalamic orexin neurons regulate arousal according to energy balance in mice. Neuron 38:701–713.LaunchUrlCrossRefPubMed↵ Hara J, et al. (2001) Genetic ablation of orexin neurons in mice results in narcolepsy, hypophagia, and obesity. Neuron 30:345–354.LaunchUrlCrossRefPubMed Sakurai T (2005) Reverse pharmacology of orexin: from an orphan GPCR to integrative physiology. Regul Pept 126:3–10.LaunchUrlCrossRefPubMed Sakurai T, et al. (2005) InPlace of orexin/hypocretin neurons revealed by a genetically encoded tracer in mice. Neuron 46:297–308.LaunchUrlCrossRefPubMed Yamanaka A, et al. (2006) Orexin neurons are directly and indirectly regulated by catecholamines in a complex manner. J Neurophysiol 96:284–298.LaunchUrlAbstract/FREE Full Text↵ Saper CB, Scammell TE, Lu J (2005) Hypothalamic regulation of sleep and circadian rhythms. Nature 437:1257–1263.LaunchUrlCrossRefPubMed↵ Finnimore AJ, Roebuck M, Sajkov D, McEvoy RD (1995) The Traces of the GABA agonist, baclofen, on sleep and breathing. Eur Respir J 8:230–234.LaunchUrlAbstract↵ Puigcerver A, van Luijtelaar EL, Drinkenburg WH, Coenen AL (1996) Traces of the GABAB antagonist CGP 35348 on sleep-wake states, behaviour, and spike-wave discharges in Aged rats. Brain Res Bull 40:157–162.LaunchUrlCrossRefPubMed↵ Gauthier P, Arnaud C, GanExecutelfo G, Gottesmann C (1997) Influence of a GABA(B) receptor antagonist on the sleep-waking cycle in the rat. Brain Res 773:8–14.LaunchUrlCrossRefPubMed↵ Mileykovskiy BY, Kiyashchenko LI, Siegel JM (2005) Behavioral correlates of activity in identified hypocretin/orexin neurons. Neuron 46:787–798.LaunchUrlCrossRefPubMed↵ Lee MG, Hassani OK, Jones BE (2005) Discharge of identified orexin/hypocretin neurons across the sleep-waking cycle. J Neurosci 25:6716–6720.LaunchUrlAbstract/FREE Full Text↵ Takahashi K, Lin JS, Sakai K (2008) Neuronal activity of orexin and non-orexin waking-active neurons during wake-sleep states in the mouse. Neuroscience 153:860–870.LaunchUrlCrossRefPubMed↵ Alam MN, et al. (2005) GABA-mediated control of hypocretin- but not melanin-concentrating hormone-immunoreactive neurones during sleep in rats. J Physiol 563:569–582.LaunchUrlAbstract/FREE Full Text↵ Suntsova N, et al. (2007) The median preoptic nucleus reciprocally modulates activity of arousal-related and sleep-related neurons in the perifornical lateral hypothalamus. J Neurosci 27:1616–1630.LaunchUrlAbstract/FREE Full Text↵ Mieda M, et al. (2004) Orexin peptides prevent cataplexy and improve wakefulness in an orexin neuron-ablated model of narcolepsy in mice. Proc Natl Acad Sci USA 101:4649–4654.LaunchUrlAbstract/FREE Full Text↵ Estabrooke IV, et al. (2001) Fos expression in orexin neurons varies with behavioral state. J Neurosci 21:1656–1662.LaunchUrlAbstract/FREE Full Text
Like (0) or Share (0)