SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. References
  7. Appendix

Recent population studies have identified important interrelationships between sleep duration and body weight regulation. The hypothalamic hypocretin/orexin neuropeptide system is able to influence each of these. Disruption of the hypocretin system, such as occurs in narcolepsy, leads to a disruption of sleep and is often associated with increased body mass index. We examined the potential interrelationship between the hypocretin system, metabolism and sleep by measuring locomotion, feeding, drinking, body temperature, sleep/wake and energy metabolism in a mouse model of narcolepsy (ataxin-ablation of hypocretin-expressing neurons). We found that locomotion, feeding, drinking and energy expenditure were significantly reduced in the narcoleptic mice. These mice also exhibited severe sleep/wake fragmentation. Upon awakening, transgenic and control mice displayed a similar rate of increase in locomotion and food/water intake with time. A lack of long wake episodes partially or entirely explains observed differences in overall locomotion, feeding and drinking in these transgenic mice. Like other parameters, energy expenditure also rose and fell depending on the sleep/wake status. Unlike other parameters, however, energy expenditure in control mice increased upon awakening at a greater rate than in the narcoleptic mice. We conclude that the profound sleep/wake fragmentation is a leading cause of the reduced locomotion, feeding, drinking and energy expenditure in the narcoleptic mice under unperturbed conditions. We also identify an intrinsic role of the hypocretin system in energy expenditure that may not be dependent on sleep/wake regulation, locomotion, or food intake. This investigation illustrates the need for coordinated study of multiple phenotypes in mouse models with altered sleep/wake patterns.

The sleep disorder narcolepsy is characterized by an inability to consolidate sleep or wake into single, long episodes, and the occurrence of ‘break-through’ rapid eye movement (REM) sleep events into wakefulness. Following on animal model discoveries (Lin et al. 1999; Chemelli et al. 1999), it has been recently demonstrated that human narcolepsy is caused by a loss of hypocretin/orexin neurons in the brain (Peyron et al. 2000; Thannickal et al. 2000). There are two hypocretin neuropeptides (hypocretin-1 and -2) that are synthesized by neurons located in the lateral hypothalamus (de Lecea et al. 1998; Sakurai et al. 1998). These neurons send neuronal projections to both cortical and subcortical neural structures and act on specific receptors (hypocretin receptor 1 and 2) that display regional variation (Sakurai et al. 1998; Marcus et al. 2001). While the loss of hypocretin neuron function leads to narcolepsy, there has been considerable work done in an attempt to determine the normal role of hypocretin in brain physiology (Beuckmann & Yanagisawa, 2002; Taheri et al. 2002; Sakurai, 2006).

Several lines of evidence indicate that hypocretin neurons are important in energy metabolism regulation. Loss of hypocretin in humans (i.e. narcolepsy) is associated with an increase in body mass index relative to that of the general population (Schuld et al. 2000; Dahmen et al. 2001; Nishino et al. 2001; Arnulf et al. 2006). This change in body mass index may be due to a fundamental change in energetics as some studies have shown reduced food intake in human narcolepsy (Lammers et al. 1996). Likewise, mouse models of narcolepsy show increased body weight and decreased food intake compared to wild-type littermates (Hara et al. 2001; Willie et al. 2001; Hara et al. 2005; Fujiki et al. 2006).

Anatomical, biochemical and electrophysiological studies have provided possible neural mechanisms for these observations. Hypocretin neurons are excitatory in areas that are responsible for feeding, drinking and thermoregulation (e.g. paraventricular, ventromedial, dorsomedial and arcuate nuclei of the hypothalamus, nucleus of the solitary tract, and glucose-sensitive neurons in the lateral hypothalamus) (Smart et al. 2002; Jobst et al. 2004). Hypocretin neurons are responsive to hormonal indicators of metabolic state such as orexigenic ghrelin and anorexigenic leptin (Olszewski et al. 2003; Yamanaka et al. 2003), and can be activated by fasting or insulin-induced hypoglycaemia (Moriguchi et al. 1999; Cai et al. 2001; Diano et al. 2003). Glucose directly inhibits hypocretin neurons through activation of tandem-pore potassium channels (Burdakov et al. 2006). These data indicate that the hypocretin system may act as a sensor of the nutritional status of the body and can respond with behavioural and physiological solutions (e.g. increase wakefulness to increase likelihood of finding food) to maintain energy homeostasis. This general concept is also supported by experiments showing increased alertness with food deprivation in wild-type but not in hypocretin neuron-ablated animals (Yamanaka et al. 2003). The role of hypocretin in mediating food anticipatory activity under restricted feeding, including as an efferent signal for locomotor and wakefulness activation by the food-entrainable circadian oscillators, is also increasingly recognized (Mieda et al. 2004; Akiyama et al. 2004).

Whereas increased emphasis has been on the upstream effects of metabolic indicators on the hypocretin system itself, the potential role of hypocretin in regulating metabolism has been less understood. Central administration of hypocretins increases food and water intake and enhances oxygen consumption and body temperature (Sakurai et al. 1998; Lubkin & Stricker-Krongrad, 1998; Kunii et al. 1999; Edwards et al. 1999; Sweet et al. 1999; Haynes et al. 1999; Wang et al. 2001; Monda et al. 2001; Asakawa et al. 2002; Kiwaki et al. 2004). However, due to complex behavioural association of these effectors with wakefulness, motivation, and locomotion, it is difficult to disentangle the specific role of hypocretin in energy metabolism. In this study, we took a comprehensive approach to investigate the relationship between the hypocretin system and energy metabolism by measuring a number of physiological and behavioural parameters simultaneously, using a mouse model of narcolepsy under a homogeneous genetic background.

Methods

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. References
  7. Appendix

Animals

Orexin-ataxin-3 transgenic mice expressing a truncated Machado–Joseph disease gene product (ataxin-3) with an expanded polyglutamine stretch specifically in hypocretin neurons (Hara et al. 2001) were backcrossed with C57Bl/6. The expression of ataxin-3 in hypocretin-expressing neurons causes cell death, with over 99% of hypocretin neurons lost by 12 weeks postnatal age (Hara et al. 2001). Adult male littermates with (TG) or without (WT) the transgene and matched for age and body weight (Table 1) were selected from a congenic line (generations 9–10) for all experiments. Mice were housed in a temperature (21–23°C) and humidity (40–60%) controlled environment with a 12 h–12 h light–dark schedule. Food (Prolab RMH 3000, PMI Nutrition International, St Louis, MO, USA) and water were available ad libitum. At the end of the experiment, animals were killed using carbon dioxide with cervical dislocation under CO2 narcosis, if necessary. The entire study was approved and conducted in accordance with the guidelines of Stanford's Administrative Panel for Laboratory Animal Care.

Table 1.  Animal age, weight and 24 h measurement of locomotor activity, feeding, drinking, and core body temperature in wild-type (WT) and transgenic (TG) mice
 WTTG
  1. Data are expressed as means ±s.d. (n) for age and weight, and means ±s.e.m. (n) for other measurements. Locomotor activity (Group 1, measured using infrared monitors), feeding and drinking (Group 1, 2, and 3), and core body temperature (Group 2, 3, and 4) were collected and averaged over 3 days. ‘Day’ represents the 12 h light-on period, and ‘night’ the 12 h light-off period. Due to device errors, drinking and feeding data of two animals from each genotype were excluded. *P < 0.05, unpaired two-tailed t test, compared with wild-type.

Age (d)203 ± 14 (23) 202 ± 13 (23)
Body weight (g)34 ± 3 (23) 34 ± 4 (23)
Locomotor activity (count h−1)
 24 h1455 ± 88 (8)   1132 ± 55 (8)* 
 Day1091 ± 120 (8)  803 ± 57 (8)*
 Night1819 ± 86 (8)   1461 ± 103 (8)*
Food intake (g)
 24 h4.9 ± 0.1 (14)  4.4 ± 0.1 (14)*
 Day1.56 ± 0.07 (14) 1.32 ± 0.09 (14)
 Night3.33 ± 0.07 (14)  3.12 ± 0.07 (14)*
Water intake (ml)
 24 h5.6 ± 0.3 (14)  4.8 ± 0.2 (14)*
 Day1.8 ± 0.2 (14) 1.4 ± 0.2 (14)
 Night3.8 ± 0.1 (14)  3.4 ± 0.1 (14)*
Core body temperature (°C)
 24 h36.58 ± 0.06 (12) 36.53 ± 0.07 (12)
 Day36.03 ± 0.07 (12) 35.84 ± 0.08 (12)
 Night37.12 ± 0.08 (12) 37.21 ± 0.08 (12)

Surgery

Under 3–5% gaseous isoflurane anaesthesia, a telemetry transmitter (ETA-F20, 3.9 g weight, Data Science International, St Paul, MN, USA) capable of acquiring and sending electroencephalograph (EEG), temperature, and movement data was implanted intra-abdominally. The two EEG electrodes were tunnelled subcutaneously and screwed into the skull at the following coordinates: anterior/posterior from bregma (AP) 1.5 mm, lateral (ML) 1.5 mm and AP −3.5 mm, ML 3 mm. Depth of anaesthetization was monitored through withdrawal reflex, respiratory pattern, and mucous membrane colour. An analgesic (5 mg kg−1 carprofen) was given subcutaneously immediately and one day after the surgery, and antibiotics (5 mg kg−1 day−1 enrofloxacin) was given subcutaneously for 5 days after surgery. Surgical wounds, animal behaviours and body weight were monitored to ensure sufficient analgesics and antibiotics were supplied as needed. Mice were allowed to fully recover for a minimum of 2 weeks before the experiments.

Locomotion, feeding, drinking, and metabolic measurements

Each mouse was introduced to an 11 × 31 × 12 cm3 Plexiglas recording chamber for indirect calorimetry measurement of oxygen consumption and carbon dioxide production (Oxymax, Columbus Instruments, Columbus, OH, USA). Room air was supplied to the chamber at a flow rate of 0.57 l min−1, and was referenced every hour during measurement. Each chamber was also equipped with real-time monitoring of powdered food intake, cyclic measurements of water intake through volumetric detectors and of locomotor activity (LMA) through infrared monitors. This system allows simultaneous measurement of 16 animals at a time with an indirect calorimetry measurement scheduled to cycle through the chambers once every hour, measuring one chamber at a time. Mice spent at least 48 h adapting to the environment, and the data collected after this adaptation period were used in the analyses presented herein. Body weight before and after the experiments was measured to ensure animal health.

Locomotion, feeding, drinking and metabolic measurements linked with sleep and core body temperature recordings

Mice were monitored in the Oxymax recording chambers. A receiver (RPC-1, Data Science International) was mounted beneath each of eight chambers for telemetry measurement of EEG, LMA, and core body temperature (Tb). EEG was sampled at 250 Hz, and the other parameters were sampled at 50 Hz. The infrared activity monitors were disabled as they increased noise in the telemetry data. This setup allows simultaneous measurement of eight animals at a time, with real time recording of EEG, LMA, Tb and feeding, and cyclic drinking and calorimetry measurements with a time resolution of 0.5 h. For instantaneously linked measurements, mice were examined individually and data were collected at 10 s intervals for each parameter, except for 3 min interruptions in collection of calorimetry data due to hourly reference air measurement. These interruptions resulted in a negligible loss of metabolic data and we equivalently reduced the number of sleep/wake bouts used in our analyses. Because a 10 s sampling of the chamber air represents only about 4% of the housing volume, energy expenditure (EE) measured at a specific time point is influenced by the levels of prior period (i.e. equilibration). Thus, the changes of actual EE of an individual are under-represented by the changes of measured EE. Our device was, however, sensitive enough to detect differences in EE during different sleep states and across genotypes.

Sleep stage analyses

We used ETA-F20 transmitters that provided EEG and signal strength signals. Signal strength was based on the relative distance and orientation between the transmitter and the receiver, and our system displayed signal strength continuously (sampling rate of 50 Hz) as a waveform on the y-axis. By analysing data recorded from a EET-F20 telemetry transmitter, which provided EEG, signal strength and electromyography (EMG) signals, we found that the change in signal strength provided a function almost identical to that provided by EMG in sleep scoring (data not shown). For example, continuous large fluctuations occur during active wake, while fluctuations were subtle or absent during quiet wake and absent during nonrapid eye movement (NREM) sleep. The use of ETA-F20 in sleep analysis has also been validated independently (Tang & Sanford, 2002). We visually scored sleep stages based on EEG (filtered with low-band pass at 25–30 Hz) and signal strength waveforms in 10 s epochs using SleepSign (Kissei America, Irvine, CA, USA). Wake was defined as mixed frequency (predominantly > 4 Hz), low amplitude EEG with (mostly during active wake) or without (mostly during quiet wake) changes in signal strength. During active wake, rhythmic alpha waves (8–9 Hz) may have also appeared. During quiet wake, slowing of EEG activity and theta trains (4–5 Hz) generally occurred. NREM sleep was defined as high amplitude, low frequency (0.25–4 Hz) EEG with an absence of change in signal strength. EEG during REM sleep was similar to active wake, but rhythmic alpha waves (8–9 Hz) often predominated, with relatively low amplitude EEG and no changes in signal strength. Of note, we did not differentiate ‘cataplexy’ from ‘sleep onset REM sleep from wake’ (direct REM transition from wake, or DREM) in narcoleptic mice, and elected to use the more conservative term, DREM. With the aid of LMA counts, we separated wakefulness into two states: active wake and quiet wake; active wake was defined as a wake epoch with greater than 1 count of LMA, while quiet wake was defined as a wake epoch with 0 or 1 count of LMA. A ‘bout’ (or ‘episode’) of a given stage was defined as at least three consecutive 10 s epochs of the respective stage that was not interrupted by at least 30 s (three consecutive 10 s epochs) of another stage.

Data analyses

Feeding, drinking and EE data were all adjusted for a body weight of 30 g. EE was calculated from measured oxygen consumption (inline image) and carbon dioxide production (inline image): EE = 3.815 ×inline image+ 1.232 ×inline image. Respiratory quotient (RQ) was defined as: RQ =inline image/inline image. Through an analysis of changes in EE associated with large, discrete bouts of locomotor activity, we determined that the system had a 90 s delay between behavioural changes and recorded changes in metabolism. This delay is negligible with hourly sampling, but significant with high time resolution (10 s) measurement. Thus, EE and RQ data were shifted backward by 90 s to temporally match the LMA, feeding, drinking, and Tb measurements. EE changes at sleep stage transitions were measured during those periods characterized by at least 60 s of a single bout before and after the transitions, and a z-score transformation was used to normalize the values in each measured period. Data were analysed using Microsoft Excel 2000 and GraphPad Prism 3.0, and the statistical comparisons between the genotypes were analysed by two-way analysis of variance (ANOVA), analysis of covariance (ANCOVA), Student's t test, Mann–Whitney test, and Kolmogorov–Smirnov test, as appropriate. Data are expressed as means ±s.e.m. except as noted. Statistical significance was set at P < 0.05.

Experimental protocols

Experiment 1 Data were collected hourly for each parameter from four different groups of mice: Group 1 (8 WT and 8 TG) were examined for locomotor activity, feeding and drinking. Group 2 and 3 (each consisted of 4 WT and 4 TG) were examined for feeding, drinking and core body temperature. Group 4 (4 WT and 4 TG) were examined for core body temperature. The results are shown in Table 1 and Figs 1 and 3. Data were collected from the 60th to the 132nd hour after animals entered the chambers (i.e. from ZT0 on the 4th day to ZT0 on the 7th day). Each parameter was binned at 2 h intervals and 24 h average over the 3 days was calculated for each animal. Finally, group means were calculated with statistical analyses.

image

Figure 1. Diurnal fluctuation of locomotor activity (LMA, A), core body temperature (Tb, B), and wakefulness (C) in wild-type (WT) and transgenic (TG) mice Data shown are from different groups of mice. LMA was measured using infrared monitors. Two-way ANOVA indicates a significant difference in LMA between the WT and TG mice (n= 8 in each genotype), and no difference in mean Tb (n= 12 in each genotype) or wake time (n= 5 WT, 6 TG). The P-values from two-way ANOVA are presented above the curves. The dashed lines with open circles represent group averages of the WT mice and the continuous lines with filled circles represent group averages of the TG mice. Bars underneath indicate light and dark phases. *Significant difference (P < 0.05, unpaired, two-tailed t test) between the WT and TG mice at a specific time point. Data of 24 h are double plotted side by side to better illustrate the diurnal rhythms.

Download figure to PowerPoint

image

Figure 3. Diurnal fluctuation of energy expenditure (EE, A) and respiratory quotient (RQ, B) in wild-type (WT) and transgenic (TG) mice Two-way ANOVA indicates a significant difference in EE between the WT and TG mice (n= 8 in each genotype), and no difference in RQ (n= 8 in each genotype). The P-values from two-way ANOVA are presented above the curves. The dashed lines with open circles represent group averages of the WT and the continuous lines with filled circles represent group averages of the TG. Bars underneath indicate light and dark phases. *Significant difference (P < 0.05, unpaired, two-tailed t test) between the WT and TG mice at a specific time point. Data are double plotted.

Download figure to PowerPoint

Experiment 2 Five WT and six TG mice were measured individually with data collected every 10 s for each parameter. The results are shown in Tables 2 and 3 and Figs 2, 4, 5 and 6. Data were collected from the 60th to the 84th hour after animals entered the chambers (i.e. from ZT0 on the 4th day to ZT0 on the 5th day).

Table 2.  Sleep parameters in wild-type (WT) and transgenic (TG) mice
 WT (n= 5)TG (n= 6)
  1. REM in transgenic mice includes REM with transitions from both NREM and wake. *P < 0.05, unpaired two-tailed t test, compared with wild-type.

Time (% total)
 24 h
  Wake53 ± 3 53 ± 2  
  NREM43 ± 2 41 ± 2  
  REM3.6 ± 0.65.6 ± 0.6 
 Day
  Wake39 ± 3 42 ± 2  
  NREM55 ± 2 53 ± 2  
  REM6 ± 15.3 ± 0.9 
 Night
  Wake67 ± 3 64 ± 2  
  NREM31 ± 2 30 ± 2  
  REM1.6 ± 0.5 6.0 ± 0.3*
Episode count
 24 h
  Wake98 ± 4 165 ± 7*  
  NREM115 ± 6  180 ± 11* 
  REM36 ± 5 50 ± 7  
 Day
  Wake55 ± 5 80 ± 4* 
  NREM69 ± 5 94 ± 5* 
  REM28 ± 5 26 ± 5  
 Night
  Wake44 ± 2 85 ± 6* 
  NREM47 ± 2 85 ± 8* 
  REM8 ± 224 ± 3* 
Episode duration (m)
 24 h
  Wake7.5 ± 0.74.5 ± 0.3*
  NREM5.7 ± 0.23.6 ± 0.2*
  REM1.47 ± 0.041.63 ± 0.09 
 Day
  Wake4.8 ± 0.73.5 ± 0.2 
  NREM6.1 ± 0.24.4 ± 0.3*
  REM1.50 ± 0.031.5 ± 0.1 
 Night
  Wake10.9 ± 0.6 5.5 ± 0.7*
  NREM5.1 ± 0.42.7 ± 0.1*
  REM1.4 ± 0.11.75 ± 0.07*
Table 3.  Detailed analyses of rapid eye movement (REM) sleep in wild-type (WT) and transgenic (TG) mice
 WT (n= 5)TG (n= 6)
  1. *P < 0.05, unpaired two-tailed t test, compared with wild-type

  2. P < 0.05, two-tailed Mann–Whitney test, compared with wild-type. ‡P < 0.05, paired two-tailed t test, compared with NREM–REM transition during the same period.

NREM–REM
 24 h
  REM episode36 ± 5 44 ± 7 
  REM Duration (m)1.47 ± 0.041.58 ± 0.09
  Prior NREM Duration (m)7.7 ± 0.5 4.1 ± 0.3*
 Day
  REM episode28 ± 5 26 ± 5 
  REM Duration (m)1.51 ± 0.031.51 ± 0.09
  Prior NREM Duration (m)7.5 ± 0.3 4.5 ± 0.5*
 Night
  REM episode8 ± 218 ± 3*
  REM Duration (m)1.4 ± 0.11.65 ± 0.09
  Prior NREM Duration (m)8.3 ± 1.0 3.3 ± 0.2*
Wake–REM (DREM)
 24 h
  REM episode0 6 ± 2†
  REM Duration (m) 2.02 ± 0.08‡
  Prior Wake Duration (m)12 ± 4 
 Day
  REM episode00.3 ± 0.3
  REM Duration (m)2.4 ± 0.7
  Prior Wake Duration (m)8 ± 4
 Night
  REM episode0 6 ± 2†
  REM Duration (m) 2.01 ± 0.07‡
  Prior Wake Duration (m)12 ± 4 
image

Figure 2. Histograms of wake (A) and NREM sleep (B) duration in wild-type (WT) and transgenic (TG) mice For equal bin histograms (left), the wake data are binned in 4 min intervals, and the NREM sleep data are binned in 1 min intervals. The dashed lines with open circles represent group averages of the WT mice and the continuous lines with filled circles represent group averages of the TG mice. The distributions of wake and NREM sleep durations were significantly different between the two genotypes (P < 0.001, Kolmogorov–Smirnov test). The bar graphs (right) show the differences between the WT and TG mice in three arbitrary bins of state duration. The open bars represent the WT mice and the filled bars represent the TG mice. *Significant difference (P < 0.05, unpaired, two-tailed t test) between the WT and TG mice.

Download figure to PowerPoint

image

Figure 4. Energy expenditure during different states and transitions in wild-type (WT) and transgenic (TG) mice A, hypnograms coupled with energy expenditure (EE) and core body temperature (Tb) measurements. One representative animal from each genotype is shown. EE and Tb are expressed in four colours with each colour representing the values at each of the four states: active wake (AW), quiet wake (QW), NREM and REM. B, average EE during AW, QW, NREM and REM. Note the differences among the states are underrepresented due to frequent state transition and measurement carry-over. Two-way ANOVA showed significant difference in overall EE between the WT and TG mice (P < 0.05, n= 5 for WT, n= 6 for TG). The open bars represent the WT mice and the filled bars represent the TG mice. Data are expressed as mean ±s.e.m.*Significant difference between AW or QW and NREM or REM in the WT mice; #significant difference between AW and NREM in the TG mice (P < 0.05, unpaired, two-tailed t test). C, z-score transform of EE at state transitions. The dashed lines with open circles represent the WT mice and the continuous lines with filled circles represent the TG mice. Data are selected to avoid using those from transient periods (e.g. < 60 s of REM). Numbers in parentheses represent the numbers of data points used in analyses: (WT, TG). Data are expressed as means ±s.d.

Download figure to PowerPoint

image

Figure 5. Progression of locomotor activity (LMA, A), feeding (B), and drinking (C) upon awakening in wild-type (WT) and transgenic (TG) mice LMA was measured using telemetry. Two-way ANOVA shows no significant difference in LMA (P= 0.32), feeding (P= 0.96) or drinking (P= 0.97) between the WT and TG mice (n= 5 for WT, n= 6 for TG). The open bars represent the WT mice and the filled bars represent the TG mice. *Significant difference (P < 0.05, unpaired, two-tailed t test) between the WT and TG mice at a specific period.

Download figure to PowerPoint

image

Figure 6. Progression of core body temperature (Tb, A) and energy expenditure (EE, B) in wake and NREM sleep in wild-type (WT) and transgenic (TG) mice Two-way ANOVA indicates no significant difference in Tb during wake (P= 0.78) and NREM sleep (P= 0.10) between the WT and TG mice (n= 5 for WT, n= 6 for TG). The EE values during both wake and NREM sleep were significantly lower in the TG than those in the WT mice. The rate change during wake was significantly lower in the TG mice (P < 0.05, ANCOVA for the first 3 min and beyond), and was similar during NREM sleep in both genotypes (P > 0.11, ANCOVA for data within the first 8 min). The open bars represent the WT mice and the filled bars represent the TG mice. *Significant difference (P < 0.05, unpaired, two-tailed t test) between the WT and TG mice at a specific period.

Download figure to PowerPoint

Results

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. References
  7. Appendix

Locomotor activity, feeding, drinking, core body temperature and sleep/wake over 24 h

Male wild-type (WT) and hemizygotic transgenic (TG) mice 6–7 months old, matched for age and body weight (Table 1), were introduced to the recording chambers at Zeitgeber time (ZT) 10 (i.e. 2 h before light-off). Upon entering the chambers, all mice displayed a brief period of exploratory behaviour. During the dark period (ZT 12–24), however, the WT mice showed significantly higher LMA compared to home cage baselines, while the TG mice had a less prominent enhancement of activity. The hyperactivity in the WT mice lasted two nights before returning to a stable level (see Supplemental material, Fig. S1). Due to such a large difference between the WT and TG mice in response to a novel environment, we did not use data collected from the first two days.

LMA, feeding, drinking, and Tb (Experiment 1)

Data were collected from the 60th to the 132nd hour after animals entered the chambers (i.e. from ZT0 on the 4th day to ZT0 on the 7th day). Each parameter was binned at 2 h intervals and the 24 h average over the 3 days was calculated for each animal. Finally, group means were calculated with statistical analyses.

Both WT and TG mice displayed robust diurnal rhythms in LMA, food intake and water intake (P < 0.0001, ANOVA), with these measurements being higher during the night than during the day (Table 1, Fig. 1A). The WT mice were significantly more active, and consumed more food and water than the TG mice during both day and night, though the intake measurements were not statistically significant during the day (Table 1).

Tb displayed similar diurnal rhythms in both genotypes (Table 1, Fig. 1B). There was no significant difference in 24 h mean between the two genotypes (P= 0.26, ANOVA). Data binned at 2 h showed that TG mice had significantly higher Tb values between ZT14 and ZT18, and lower values between ZT22 and ZT4 (P < 0.05, unpaired two-tailed t tests). This difference is likely to be due to larger sleep-state dependent fluctuations in Tb in the WT (see below and discussion).

Sleep analysis (Experiment 2)

Data were collected from the 60th to the 84th hour after animals entered the chambers (i.e. from ZT0 on the 4th day to ZT0 on the 5th day).

Sleep analyses of the two genotypes revealed no significant differences in total wake or NREM sleep amounts or diurnal rhythmicity (Table 2, Fig. 1C, NREM sleep diurnal pattern not shown). There were, however, significantly more episodes of both wake and NREM sleep in the TG mice, and these episodes were significantly shorter than those in the WT mice. Examination of episode duration (Fig. 2) reveals different distributions of wake and NREM sleep durations in the two genotypes. Kolmogorov–Smirnov tests show that the maximum difference between the cumulative distributions is 0.14%ile for wake (P < 0.001), and 0.29%ile for NREM (P < 0.001). The TG mice had significantly more episodes of wake and NREM sleep with short durations (< 9 min wake, < 6 min NREM), but fewer episodes with longer durations (> 27 min wake, > 8 min NREM) than the WT mice (P < 0.05, unpaired two-tailed t tests).

There was more REM sleep during the night in the TG than in the WT mice (Table 2), which is likely to be due to the presence of direct transitions from wake to REM sleep (DREM) in the TG mice; we observed no such direct transitions in the WT mice (Table 3). The TG mice showed shorter latency in transitioning from NREM to REM and had more REM episodes during the night than the WT mice. DREM in the TG mice occurred mostly during the night and had longer durations than normal REM.

Spectral analysis revealed higher EEG power densities at the theta and alpha band (4–12 Hz) during DREM in the TG mice than those during normal REM in both groups (P < 0.05, ANOVA with post hoc t test). Compared to the WT mice, there was also a trend of lower delta (0.5–4 Hz) power during wake and NREM sleep in the TG mice, although these differences did not reach statistical significance (see Supplemental material, Fig. S2). These findings are consistent with the hypothesis that DREM, at least in part, represents a cataplexy-like state in the TG mice.

Metabolic measurement over 24 h (Experiment 1)

Hourly measurement of indirect calorimetry revealed that the oxygen consumption (inline image), carbon dioxide production (inline image), and total energy expenditure (EE, Fig. 3A) exhibited a significant diurnal rhythm in both genotypes. After lights-on, all the measurements reduced rapidly within 2 h. The levels remained low with small fluctuations for 8 h, and then rapidly increased and reached their highest levels after lights-off. During the first 2 h of darkness, values returned to levels that remained relatively low with small fluctuations until the last two hours when a second peak was reached. The diurnal patterns of inline image, inline image, and EE in each genotype followed closely with one another, and were similar to the pattern of LMA (Fig. 1A). Figure 3A shows the EE data in 2 h bins. TG mice had consistently reduced EE across 24 h, as compared to the WT mice, with the difference being larger during the day. The 24 h EE (expressed as the area under curve) in the WT mice was 11 ± 4% more than that in the TG mice (P < 0.05, unpaired two-tailed t test, n= 8 in each genotype).

The respiratory quotient (RQ), an indicator of the type of substrate being utilized, exhibited a significant diurnal rhythm in both genotypes (Fig. 3B). The RQ was lower during the day, indicating a higher fat utilization, when animals are mostly resting and not feeding. There was no statistically significant difference in RQ between the WT and TG mice.

Simultaneous measurement of all parameters with high time resolution over 24 h (Experiment 2)

Simultaneous examination of metabolism, locomotion and sleep reveals that EE is closely correlated to sleep staging in both genotypes. Visual inspection of the aligned time series (see examples in Fig. 4A) reveal that entering into the wake-state caused EE to rise and entry into NREM sleep caused EE to fall. LMA during wakefulness caused the largest overall increase in EE and was associated with the highest peaks in EE. Entering REM sleep from NREM sleep was associated with an increase in EE. Figure 4B summarizes the mean EE during different states. Data were averaged from all epochs for each state over 24 h. The EE during either active or quiet wake was significantly higher than that during NREM or REM sleep. While the overall EE was lower in the TG mice, there was no difference in the pattern of EE changed with sleep and wake between the two genotypes.

In order to compare the changes in EE in response to state transitions between the WT and TG mice, we selected transition points where two sleep stages on either side lasted at least 60 s and normalized the EE values in each transition. Figure 4C depicts the z-score transforms of EE at various transitions. The patterns of EE change at state transitions between the WT and TG mice were similar, regardless of the lower overall levels of EE in the TG mice. Thus, both genotypes show similar patterns of energetic changes in response to shifts in sleep state.

Although the diurnal patterns of EE changes in the WT and TG mice were of similar shape, there was a difference in the instantaneous change in EE in response to LMA and state transition. EE in the TG mice fluctuated with smaller amplitudes than in the WT, as reflected in their average changes from peak to trough (CPT). CPT is defined as the difference between two closest data points at which EE values change directions (from increasing to decreasing, or vice versa). CPT in the TG was significantly lower than that in the WT mice (0.070 ± 0.003 kcal h−1versus 0.083 ± 0.002 kcal h−1, P < 0.05, unpaired two-tailed t test, n= 8 in each genotype, measured over 24 h), and analyses of the CPT distribution showed that there were more small fluctuations (CPT < 0.07 kcal h−1) and fewer large fluctuations (CPT > 0.13 kcal h−1) in the TG than in the WT mice (P < 0.05, unpaired two-tailed t test). (See Supplemental material, Fig. S3A)

Tb change in response to locomotor activity and state transition showed a similar pattern to EE; however, these changes were not as instantaneous, and brief state transitions often caused no change in temperature (see examples in Fig. 4A). This is probably due to temperature buffering by the body mass of the animals. The temperature fluctuation in the TG mice also had significantly lower amplitudes than that in the WT mice, with the average CPT (definition similar to that of EE) of 0.25 ± 0.01°C in the TG and 0.33 ± 0.01°C in the WT mice (P < 0.05, unpaired two-tailed t test, n= 12 in each genotype, measured over 24 h). The CPT distribution showed that there were more small fluctuations (CPT < 0.05°C) and fewer large fluctuations (CPT > 0.09°C) in the TG than in the WT mice (P < 0.05, unpaired two-tailed t test). (See Supplemental material, Fig. S3B).

Analyses of LMA, feeding, drinking, Tb and EE in an average sleep/wake bout (Experiment 2)

Since sleep/wake is more fragmented in the TG mice (greater number of short wake/NREM bouts and smaller number of long wake/NREM bouts – more transitions between sleep stages), we analysed the relationship between our measured parameters and the bout length of the stages, as well as the progression of these parameters in each stage. LMA, feeding, or drinking data during all wake episodes were aligned at the stage onset and averaged to generate a time series. Because the duration of wake ranged from 0.5 to 75 min, each data point in the time series represented the mean of a decreasing number of values. In order to reduce skewed data representation caused by rare events occurring at the end of a few long bouts, the time points were binned unequally such that there were approximately the same number of values in each bin.

Figure 5 shows that upon awakening, both genotypes displayed a gradual increase in LMA, feeding and drinking. When the animals stayed awake for long time, drinking and LMA peaked and started to decrease at 20 min after wake, while feeding peaked around 14 min after wake. The WT and TG mice showed similar progression in these parameters, although there was a significant difference in drinking shortly after waking up, with the TG mice consuming more water than the WT during the first 4 min, and the WT equalling and passing the level of the TG mice at 7.5 min. Cumulative LMA, feeding and drinking over wake time were not significantly different between the WT and TG mice (LMA P= 0.11, feeding P= 0.42, drinking P= 0.43, ANOVA for data during the first 20 min of wakefulness). Analyses separating day and night yielded similar results (data not shown), except for higher variability in feeding and drinking during the day, due to the rarity of these events. These results indicate that the levels of LMA, feeding and drinking are dependent on the wake bout length such that they increase with longer bout lengths, and that there is no difference in the rate of increase in these parameters when comparing the WT and TG mice.

Similar analyses were applied to Tb and EE during wake and NREM sleep. Figure 6A shows that there were no significant differences in the progression of temperature during wake or NREM sleep between the WT and TG mice. Two-way ANOVA for the net changes in Tb calculated by subtracting the initial value of each episode showed no significant difference in the rate of Tb rise during wake (P= 0.45) or fall during NREM sleep (P= 0.41) between the WT and TG mice (n= 5 for WT, n= 6 for TG). Figure 6B shows that the changes in EE during NREM sleep were similar in both genotypes (P= 0.75, ANOVA). There was, however, a significantly steeper increase of EE during wake in the WT than in the TG mice (P < 0.05, ANCOVA for the first 3 min and beyond). Analysis of net changes in EE showed that the significant difference between the WT and TG mice started at 2 min after waking up, and gradually increased until 20 min (P < 0.05, ANOVA with post hoc t test). Analyses separating day and night revealed a greater difference between the two genotypes during the day than that during the night. These results indicate that the changes in Tb and EE during wake and NREM sleep are dependent on the bout length, and that there is a difference between the WT and TG in energy metabolism that is not accounted for by the difference in bout duration.

The duration of REM sleep in both the TG and WT mice was too short to provide meaningful analysis of changes in body temperature and EE during this stage and these data are not presented.

Discussion

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. References
  7. Appendix

In this study, we report on the simultaneous recording of sleep, locomotion, energy metabolism, drinking and feeding in wild-type and narcoleptic mice. As in other species with longer sleep bouts (most typically humans), we found expected changes in EE that are sleep-state and locomotion dependent, validating for the first time the use of rodent models for the study of sleep-dependent changes in EE. The study also allowed us to explore the specific involvement of hypocretin in the mediation of differences previously reported in various physiological parameters when studied in isolation.

Reduced LMA, feeding and drinking in the hypocretin-deficient mice

We found a significant reduction in LMA, feeding, and drinking in mice with ablated hypocretin neurons, compared to their normal littermates (Table 1). These results are in accordance with other studies using narcoleptic animals as well as normal animals given hypocretin or hypocretin receptor antagonists (Sakurai et al. 1998; Lubkin & Stricker-Krongrad, 1998; Kunii et al. 1999; Edwards et al. 1999; Sweet et al. 1999; Haynes et al. 1999; Chemelli et al. 1999; Hara et al. 2001). We also explored the interrelationship of the sleep-wake disruption with these variables, as they are tightly linked. We found that the daily amount of sleep and wake were not different between the two genotypes. There exists, however, severe sleep fragmentation in the TG mice (Table 2, Fig. 2). These results are in agreement with data from other animal models of narcolepsy (Kaitin et al. 1986; Chemelli et al. 1999; Hara et al. 2001; Mochizuki et al. 2004; Beuckmann et al. 2004). By analysing the behavioural progression after animals awoke, we demonstrate that the changes in LMA, feeding and drinking are likely to be secondary to a change in the duration of wake episodes (Fig. 5) and are not proximally due to a lack of hypocretin. The TG mice are capable of performing all these activities to the same extent as the WT mice, if they stay awake as long as the WT mice do, and the rates of change in these variables are indistinguishable from the WT mice. It is likely that the reduced daily LMA, feeding and drinking are due to the inability of the TG animals to remain awake long enough to reach maximal levels.

While there is a strong argument for direct effect of hypocretins on feeding from studies specifically designed to address the involvement of wakefulness and locomotor activities in hypocretin-induced hyperphagia, negative results have also been reported, especially when the peptides were applied directly into specific brain regions (Dube et al. 1999; Ida et al. 1999; Sweet et al. 1999; Rodgers et al. 2000; Rodgers et al. 2001; Espana et al. 2001; Kotz et al. 2002). Negative results may have occurred when the peptides are delivered to areas that are not involved in sleep/wake consolidation. Our study does not exclude the possibility that the hypocretin system is involved in the direct regulation of feeding and drinking behaviours. As seen with central injection of hypocretins, stressors, diets and other environmental challenges may increase activity of hypocretin-containing neurons, which directly stimulates feeding centres in the brain. In addition, the lack of functional hypocretin system in the narcoleptic animals may be compensated by the functions of other orexigenic systems such as neuropeptide Y, melanin-concentrating hormone, and ghrelin, resulting in the observed intake dependency on wake consolidation. However, our studies caution that the phenotypes found in narcoleptic individuals or animals must be considered carefully before ascribing a direct functional role for hypocretin in regulating the behaviour.

Temperature regulation in the hypocretin-deficient mice

As reported in the preprohypocretin knockout mice (Mochizuki et al. 2006), we observed decreased instantaneous Tb fluctuations in TG mice (see Supplemental material, Fig. S3 and examples in Fig. 4A), reflecting sleep fragmentation in narcoleptic mice. The reduced instantaneous fluctuations in Tb are also reflected in the constantly low levels during the day and constantly high levels during the first half of the night in the TG mice, causing a lower average at ZT14–18 and a higher average at ZT22–4 when data were binned with 2 h intervals (Fig. 1B), compared to the WT mice. Because the rise and fall in Tb during the respective wakefulness and NREM sleep are similar in both genotypes (Fig. 6A), the blunted fluctuation in the TG mice is likely to be caused by increased sleep/wake fragmentation that prevents the Tb from dropping or rising as much as occurs in WT.

Despite the blunted fluctuation of Tb in the TG mice, the mean temperature over 24 h was similar in both WT and TG mice (Fig. 1B), indicating a normal core temperature regulation in the hypocretin-deficient mice. This result is in contrast with data reported in preprohypocretin knockout mice, where 24 h mean temperature was higher (Mochizuki et al. 2006). Preprohypocretin knockout mice also showed blunted temperature decreases after sleep onset (Mochizuki et al. 2006), a result we did not find in the TG mice (Fig. 6A). Our finding of normal temperature regulation in TG animals is in contrast to the reduced 24 h EE level. It is possible that there are greater energetic requirements when body temperature changes with large amplitudes. Therefore, to maintain a similar level of overall body temperature, the metabolic rate in the WT mice needs to be higher than that in the TG mice. Alternatively, there could be a difference in heat loss mechanisms between these mice.

Reduced energy expenditure in the hypocretin-deficient mice

We found reduced EE across 24 h in the TG mice, compared to their age- and weight-matched WT littermates (Fig. 3A). The selection of age- and weight-matched controls was made to facilitate comparison, keeping in mind that hypocretin neuron-ablated mice have slightly higher body weight, although this finding is variable, depending on genetic background, age, diet and sex (Hara et al. 2005; Fujiki et al. 2006). By selecting weight-matched subjects, we probably underestimated the difference between these animals. Indeed, we have studied a line of female preprohypocretin knockout mice that were significantly obese, and found that the metabolic rate in these mice was even more decreased than in normal weight hypocretin neuron-ablated animals; the weight-adjusted 24 h EE (expressed as the area under curve) in age-matched WT mice was 23 ± 6% more that that of the knockout mice (authors' unpublished observations).

Simultaneous monitoring of sleep/wake and EE with a high time resolution revealed that EE was instantaneously linked to sleep staging (Fig. 4). EE fluctuated less in the TG than in the WT mice, probably due to the frequent disruption of sleep/wake states. Like LMA, feeding and drinking, levels of which gradually increase upon awakening, EE also gradually increases during wake. However, unlike LMA or feeding, there exists a significant difference in the rate of change in EE between the WT and TG mice, with a steeper increase in the WT mice than in the TG. These results suggest that the change in EE is both dependent and independent of the sleep/wake fragmentation in the TG mice.

Diurnal variation in EE is shown in Fig. 3A. EE levels peak at the beginning and the end of the dark period, and remain low during the light period. This pattern does not resemble diurnal variation in hypocretin release, which peaks at the end of the dark period and reaches the lowest at the second half of the light period (Yoshida et al. 2001). This indicates that hypocretin is not the major contributing factor of overall energy metabolism. The diurnal variation of the difference in EE between the two genotypes was also evaluated by subtracting the EE values in the TG mice from those in the WT mice (from Fig. 3A) at corresponding time points (data not shown). The difference was highest at lights-on, gradually decreasing during the light period until ZT8. The difference remained low until the second half of the dark period, when it started to rise again. This pattern resembles diurnal fluctuations in hypocretin levels as measured in rat cerebrospinal fluid and extracellular spaces of the lateral hypothalamus (Fujiki et al. 2001; Yoshida et al. 2001; Zhang et al. 2004). Thus, hypocretin may directly and/or indirectly influence metabolic rate in mice. The mechanisms by which hypocretin interacts with EE are unknown.

Because the expression of ataxin-3 in the hypocretin neurons also eliminates or reduces the levels of other factors that coexpress with the hypocretin peptides (e.g. dynorphins, neuronal activity-regulated pentraxin, and precursor-protein convertase 1) (Chou et al. 2001; Reti et al. 2002; Nilaweera et al. 2003), it is possible that some of the observed differences between the WT and TG mice are caused by the changes of these other factors (Hara et al. 2005). However, effects of coexpressing factors may be subtle and could be under the influence of genetic background, sex and environmental conditions. Our recent findings using orexin-ataxin-3 transgenics and preprohypocretin knockout mice reveal a similar obesity tendency in the female mutants of both narcolepsy models (Fujiki et al. 2006), suggesting that hypocretins, rather than their coexpressing factors, play a critical role in downstream effects on energy balance. Regardless of this argument, our results are informative for considering changes in metabolism associated with sleep abnormalities in human narcolepsy.

In conclusion, we demonstrate that regulating sleep/wake and maintaining state stability are the central downstream effects of the hypocretin system. The lack of functional hypocretin transmission leads to severe sleep/wake fragmentation, which secondarily reduces locomotion, feeding, drinking and energy metabolism. We also discovered an intrinsic function of the hypocretin system in metabolic regulation that is independent of sleep/wake, locomotion and food intake.

References

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. References
  7. Appendix
  • Akiyama M, Yuasa T, Hayasaka N, Horikawa K, Sakurai T & Shibata S (2004). Reduced food anticipatory activity in genetically orexin (hypocretin) neuron-ablated mice. Eur J Neurosci 20, 30543062.
  • Arnulf I, Lin L, Zhang J, Russell IJ, Ripley B, Einen M, Nevsimalova S, Bassetti C, Bourgin P, Nishino S & Mignot E (2006). CSF versus serum leptin in narcolepsy: is there an effect of hypocretin deficiency? Sleep 29, 10171024.
  • Asakawa A, Inui A, Goto K, Yuzuriha H, Takimoto Y, Inui T, Katsuura G, Fujino MA, Meguid MM & Kasuga M (2002). Effects of agouti-related protein, orexin and melanin-concentrating hormone on oxygen consumption in mice. Int J Mol Med 10, 523525.
  • Beuckmann CT, Sinton CM, Williams SC, Richardson JA, Hammer RE, Sakurai T & Yanagisawa M (2004). Expression of a poly-glutamine-ataxin-3 transgene in orexin neurons induces narcolepsy-cataplexy in the rat. J Neurosci 24, 44694477.
  • Beuckmann CT & Yanagisawa M (2002). Orexins: from neuropeptides to energy homeostasis and sleep/wake regulation. J Mol Med 80, 329342.
  • Burdakov D, Jensen LT, Alexopoulos H, Williams RH, Fearon IM, O'Kelly I, Gerasimenko O, Fugger L & Verkhratsky A (2006). Tandem-pore K+ channels mediate inhibition of orexin neurons by glucose. Neuron 50, 711722.
  • Cai XJ, Evans ML, Lister CA, Leslie RA, Arch JR, Wilson S & Williams G (2001). Hypoglycemia activates orexin neurons and selectively increases hypothalamic orexin-B levels: responses inhibited by feeding and possibly mediated by the nucleus of the solitary tract. Diabetes 50, 105112.
  • Chemelli RM, Willie JT, Sinton CM, Elmquist JK, Scammell T, Lee C, Richardson JA, Williams SC, Xiong Y, Kisanuki Y, Fitch TE, Nakazato M, Hammer RE, Saper CB & Yanagisawa M (1999). Narcolepsy in orexin knockout mice: molecular genetics of sleep regulation. Cell 98, 437451.
  • Chou TC, Lee CE, Lu J, Elmquist JK, Hara J, Willie JT, Beuckmann CT, Chemelli RM, Sakurai T, Yanagisawa M, Saper CB & Scammell TE (2001). Orexin (hypocretin) neurons contain dynorphin. J Neurosci 21, RC168.
  • Dahmen N, Bierbrauer J & Kasten M (2001). Increased prevalence of obesity in narcoleptic patients and relatives. Eur Arch Psychiatry Clin Neurosci 251, 8589.
  • De Lecea L, Kilduff TS, Peyron C, Gao X, Foye PE, Danielson PE, Fukuhara C, Battenberg EL, Gautvik VT, Bartlett FS, Frankel WN, Van Den Pol AN, Bloom FE, Gautvik KM & Sutcliffe JG (1998). The hypocretins: hypothalamus-specific peptides with neuroexcitatory activity. Proc Natl Acad Sci U S A 95, 322327.
  • Diano S, Horvath B, Urbanski HF, Sotonyi P & Horvath TL (2003). Fasting activates the nonhuman primate hypocretin (orexin) system and its postsynaptic targets. Endocrinology 144, 37743778.
  • Dube MG, Kalra SP & Kalra PS (1999). Food intake elicited by central administration of orexins/hypocretins: identification of hypothalamic sites of action. Brain Res 842, 473477.
  • Edwards CM, Abusnana S, Sunter D, Murphy KG, Ghatei MA & Bloom SR (1999). The effect of the orexins on food intake: comparison with neuropeptide Y, melanin-concentrating hormone and galanin. J Endocrinol 160, R7R12.
  • Espana RA, Baldo BA, Kelley AE & Berridge CW (2001). Wake-promoting and sleep-suppressing actions of hypocretin (orexin): basal forebrain sites of action. Neuroscience 106, 699715.
  • Fujiki N, Yoshida Y, Ripley B, Honda K, Mignot E & Nishino S (2001). Changes in CSF hypocretin-1 (orexin A) levels in rats across 24 hours and in response to food deprivation. Neuroreport 12, 993997.
  • Fujiki N, Yoshida Y, Zhang S, Sakurai T, Yanagisawa M & Nishino S (2006). Sex difference in body weight gain and leptin signaling in hypocretin/orexin deficient mouse models. Peptides 27, 23262331.
  • Hara J, Beuckmann CT, Nambu T, Willie JT, Chemelli RM, Sinton CM, Sugiyama F, Yagami K, Goto K, Yanagisawa M & Sakurai T (2001). Genetic ablation of orexin neurons in mice results in narcolepsy, hypophagia, and obesity. Neuron 30, 345354.
  • Hara J, Yanagisawa M & Sakurai T (2005). Difference in obesity phenotype between orexin-knockout mice and orexin neuron-deficient mice with same genetic background and environmental conditions. Neurosci Lett 380, 239242.
  • Haynes AC, Jackson B, Overend P, Buckingham RE, Wilson S, Tadayyon M & Arch JR (1999). Effects of single and chronic intracerebroventricular administration of the orexins on feeding in the rat. Peptides 20, 10991105.
  • Ida T, Nakahara K, Katayama T, Murakami N & Nakazato M (1999). Effect of lateral cerebroventricular injection of the appetite-stimulating neuropeptide, orexin and neuropeptide Y, on the various behavioral activities of rats. Brain Res 821, 526529.
  • Jobst EE, Enriori PJ & Cowley MA (2004). The electrophysiology of feeding circuits. Trends Endocrinol Metab 15, 488499.
  • Kaitin KI, Kilduff TS & Dement WC (1986). Sleep fragmentation in canine narcolepsy. Sleep 9, 116119.
  • Kiwaki K, Kotz CM, Wang C, Lanningham-Foster L & Levine JA (2004). Orexin A (hypocretin 1) injected into hypothalamic paraventricular nucleus and spontaneous physical activity in rats. Am J Physiol Endocrinol Metab 286, E551E559.
  • Kotz CM, Teske JA, Levine JA & Wang C (2002). Feeding and activity induced by orexin A in the lateral hypothalamus in rats. Regul Pept 104, 2732.
  • Kunii K, Yamanaka A, Nambu T, Matsuzaki I, Goto K & Sakurai T (1999). Orexins/hypocretins regulate drinking behaviour. Brain Res 842, 256261.
  • Lammers GJ, Pijl H, Iestra J, Langius JA, Buunk G & Meinders AE (1996). Spontaneous food choice in narcolepsy. Sleep 19, 7576.
  • Lin L, Faraco J, Li R, Kadotani H, Rogers W, Lin X, Qiu X, De Jong PJ, Nishino S & Mignot E (1999). The sleep disorder canine narcolepsy is caused by a mutation in the hypocretin (orexin) receptor 2 gene. Cell 98, 365376.
  • Lubkin M & Stricker-Krongrad A (1998). Independent feeding and metabolic actions of orexins in mice. Biochem Biophys Res Commun 253, 241245.
  • Marcus JN, Aschkenasi CJ, Lee CE, Chemelli RM, Saper CB, Yanagisawa M & Elmquist JK (2001). Differential expression of orexin receptors 1 and 2 in the rat brain. J Comp Neurol 435, 625.
  • Mieda M, Williams SC, Sinton CM, Richardson JA, Sakurai T & Yanagisawa M (2004). Orexin neurons function in an efferent pathway of a food-entrainable circadian oscillator in eliciting food-anticipatory activity and wakefulness. J Neurosci 24, 1049310501.
  • Mochizuki T, Crocker A, McCormack S, Yanagisawa M, Sakurai T & Scammell TE (2004). Behavioral state instability in orexin knock-out mice. J Neurosci 24, 62916300.
  • Mochizuki T, Klerman EB, Sakurai T & Scammell TE (2006). Elevated body temperature during sleep in orexin knockout mice. Am J Physiol Regul Integr Comp Physiol 291, R533540.
  • Monda M, Viggiano A & De Mondola P & V (2001). Inhibition of prostaglandin synthesis reduces hyperthermic reactions induced by hypocretin-1/orexin A. Brain Res 909, 6874.
  • Moriguchi T, Sakurai T, Nambu T, Yanagisawa M & Goto K (1999). Neurons containing orexin in the lateral hypothalamic area of the adult rat brain are activated by insulin-induced acute hypoglycemia. Neurosci Lett 264, 101104.
  • Nilaweera KN, Barrett P, Mercer JG & Morgan PJ (2003). Precursor-protein convertase 1 gene expression in the mouse hypothalamus: differential regulation by ob gene mutation, energy deficit and administration of leptin, and coexpression with prepro-orexin. Neuroscience 119, 713720.
  • Nishino S, Ripley B, Overeem S, Nevsimalova S, Lammers GJ, Vankova J, Okun M, Rogers W, Brooks S & Mignot E (2001). Low cerebrospinal fluid hypocretin (Orexin) and altered energy homeostasis in human narcolepsy. Ann Neurol 50, 381388.
  • Olszewski PK, Li D, Grace MK, Billington CJ, Kotz CM & Levine AS (2003). Neural basis of orexigenic effects of ghrelin acting within lateral hypothalamus. Peptides 24, 597602.
  • Peyron C, Faraco J, Rogers W, Ripley B, Overeem S, Charnay Y, Nevsimalova S, Aldrich M, Reynolds D, Albin R, Li R, Hungs M, Pedrazzoli M, Padigaru M, Kucherlapati M, Fan J, Maki R, Lammers GJ, Bouras C, Kucherlapati R, Nishino S & Mignot E (2000). A mutation in a case of early onset narcolepsy and a generalized absence of hypocretin peptides in human narcoleptic brains. Nat Med 6, 991997.
  • Reti IM, Reddy R, Worley PF & Baraban JM (2002). Selective expression of Narp, a secreted neuronal pentraxin, in orexin neurons. J Neurochem 82, 15611565.
  • Rodgers RJ, Halford JC, Nunes de Souza RL, Canto de Souza AL, Piper DC, Arch JR & Blundell JE (2000). Dose–response effects of orexin-A on food intake and the behavioural satiety sequence in rats. Regul Pept 96, 7184.
  • Rodgers RJ, Halford JC, Nunes de Souza RL, Canto de Souza AL, Piper DC, Arch JR, Upton N, Porter RA, Johns A & Blundell JE (2001). SB-334867, a selective orexin-1 receptor antagonist, enhances behavioural satiety and blocks the hyperphagic effect of orexin-A in rats. Eur J Neurosci 13, 14441452.
  • Sakurai T (2006). Roles of orexins and orexin receptors in central regulation of feeding behavior and energy homeostasis. CNS Neurol Disord Drug Targets 5, 313325.
  • Sakurai T, Amemiya A, Ishii M, Matsuzaki I, Chemelli RM, Tanaka H, Williams SC, Richardson JA, Kozlowski GP, Wilson S, Arch JR, Buckingham RE, Haynes AC, Carr SA, Annan RS, McNulty DE, Liu WS, Terrett JA, Elshourbagy NA, Bergsma DJ & Yanagisawa M (1998). Orexins and orexin receptors: a family of hypothalamic neuropeptides and G protein-coupled receptors that regulate feeding behavior. Cell 92, 573585.
  • Schuld A, Hebebrand J, Geller F & Pollmacher T (2000). Increased body-mass index in patients with narcolepsy. Lancet 355, 12741275.
  • Smart D, Haynes AC, Williams G & Arch JR (2002). Orexins and the treatment of obesity. Eur J Pharmacol 440, 199212.
  • Sweet DC, Levine AS, Billington CJ & Kotz CM (1999). Feeding response to central orexins. Brain Res 821, 535538.
  • Taheri S, Zeitzer JM & Mignot E (2002). The role of hypocretins (orexins) in sleep regulation and narcolepsy. Annu Rev Neurosci 25, 283313.
  • Tang X & Sanford LD (2002). Telemetric recording of sleep and home cage activity in mice. Sleep 25, 691699.
  • Thannickal TC, Moore RY, Nienhuis R, Ramanathan L, Gulyani S, Aldrich M, Cornford M & Siegel JM (2000). Reduced number of hypocretin neurons in human narcolepsy. Neuron 27, 469474.
  • Wang J, Osaka T & Inoue S (2001). Energy expenditure by intracerebroventricular administration of orexin to anesthetized rats. Neurosci Lett 315, 4952.
  • Willie JT, Chemelli RM, Sinton CM & Yanagisawa M (2001). To eat or to sleep? Orexin in the regulation of feeding and wakefulness. Annu Rev Neurosci 24, 429458.
  • Yamanaka A, Beuckmann CT, Willie JT, Hara J, Tsujino N, Mieda M, Tominaga M, Yagami K, Sugiyama F, Goto K, Yanagisawa M & Sakurai T (2003). Hypothalamic orexin neurons regulate arousal according to energy balance in mice. Neuron 38, 701713.
  • Yoshida Y, Fujiki N, Nakajima T, Ripley B, Matsumura H, Yoneda H, Mignot E & Nishino S (2001). Fluctuation of extracellular hypocretin-1 (orexin A) levels in the rat in relation to the light-dark cycle and sleep-wake activities. Eur J Neurosci 14, 10751081.
  • Zhang S, Zeitzer JM, Yoshida Y, Wisor JP, Nishino S, Edgar DM & Mignot E (2004). Lesions of the suprachiasmatic nucleus eliminate the daily rhythm of hypocretin-1 release. Sleep 27, 619627.

Appendix

  1. Top of page
  2. Abstract
  3. Methods
  4. Results
  5. Discussion
  6. References
  7. Appendix

Acknowledgements

We thank Dr Giovanni Cizza (NIDDK) for inputs and discussions. This work was supported by the Howard Hughes Medical Institute and National Institutes of Health (MH073435).