Skip to main content
NCBI home page
The Journal of Physiology logoLink to The Journal of Physiology
. 2011 Jul 4;589(Pt 17):4157–4166. doi: 10.1113/jphysiol.2011.212514

Intracellular energy status regulates activity in hypocretin/orexin neurones: a link between energy and behavioural states

Zhong-Wu Liu 1,2,3, Geliang Gan 1, Shigetomo Suyama 2, Xiao-Bing Gao 1,2,4
PMCID: PMC3180576  PMID: 21727218

Non-technical summary

A growing body of evidence has shown that energy status has a significant impact on the behavioural states in animals and that the availability of nutrients (energy state) in the brain may modulate animal behaviours. In this study, we report that the intracellular energy stores determine activity in a selective group of nerve cells (hypocretin-containing neurones) in the brain. The unique energy state of hypocretin neurones correlates with behavioural states of animals, i.e. the energy level is low during sleep and high during wakefulness. These results suggest that hypocretin neurones may act as an ‘energy gauge’ in the brain, which integrates nutritional, energetic and behavioural signals critical for the survival of animals.

Abstract

Abstract

The hypocretin/orexin (Hcrt)-containing neurones within the lateral hypothalamus integrate nutritional, energetic and behavioural cues to generate the final output in order to exert their functions. It is still not clear how Hcrt neurones monitor changes in energy status in animals. In brain slices from transgenic mice expressing green fluorescent protein (GFP) exclusively in Hcrt neurones, we examined the roles of intracellular levels of ATP ([ATP]i) in regulating activities in these cells with conventional and perforated whole-cell recording. By using ‘ATP clamp’ we demonstrated that membrane potential (Vm) correlated with the [ATP]i in Hcrt neurones. Perforated recording revealed a Vm of −46.1 ± 1.6 mV (n = 18), close to the level measured with an [ATP]i equal to 5–6 mm (–48.7 ± 1.4 mV, n = 16, 5 mm ATP), suggesting that a unique demand for energy is required to maintain normal functionality in Hcrt cells. A direct disruption of ATP production or reduction in ambient glucose levels resulted in an inhibition of activity in Hcrt neurones. The Vm was significantly depolarized in Hcrt neurones in sleep-deprived mice as compared with controls (P < 0.01, t test), which was eliminated by experimental manipulations causing the same level of [ATP]i and KATP channel opening in both groups, suggesting a decrease during sleep and an increase during sustained wakefulness in [ATP]i in Hcrt cells. In summary, these data demonstrate that a delicate control of activity by monitoring the availability of intracellular energy stores in Hcrt cells may serve as a novel mechanism regulating energy expenditure and behavioural state dependent upon the energy state in animals.

Introduction

The behavioural state of animals is determined by many internal and external environmental factors and the energy status of animals is highly correlated with their behavioural state (Shulman et al. 2003). On the one hand, the brain consumes more energy in the activated and even the so-called ‘resting state’ than in the anaesthetized state (Shulman et al. 1999); on the other hand, a low energy state leads to unconsciousness in animals (Shulman et al. 2009). It has been hypothesized that insufficient energy may lead to attenuation of wakefulness and promotion of sleep (Benington & Heller, 1995; Scharf et al. 2008). Therefore, sensing of the availability of energy in the brain and allocating energy expenditure between the brain and the rest of the animal body is not only necessary for the maintenance of the functioning of the nervous system, but also essential for survival. Evolutionarily, foraging for food (the energy source for animals) consists of a complex array of behaviours, which are governed by the brain through continuously monitoring the intake and expenditure of energy. It is not yet clear how the brain circuitry responsible for energy homeostasis is influenced by the availability of energy and how it adjusts the animal's behaviour to accommodate the energy state.

The neurones in the lateral hypothalamus (LH) that synthesize the neuropeptide hypocretin/orexin (Hcrt) play a key role in many brain functions critical to the survival of animals (de Lecea et al. 1998; Sakurai et al. 1998). It has been established that Hcrt participates in the determination of the behavioural state of animals, and that dysfunction of the Hcrt system leads to impairments to arousal and wake maintenance (Chemelli et al. 1999; Lin et al. 1999). It has also been reported that Hcrt neurones are modulated by molecules relevant to energy intake (such as leptin, insulin and glucose) in animals (Yamanaka et al. 2003; Burdakov et al. 2006; Guyon et al. 2009; Parsons & Hirasawa, 2010). However, the question remains whether the energy state of Hcrt neurones determines the activity of these cells and in turn shapes the functions governed by these neurones, particularly the behavioural state of animals. Adenosine-5′-triphosphate (ATP) is an important intracellular molecule utilized by cells to operate critical biochemical and biophysical processes, and the intracellular concentration of ATP ([ATP]i) may be an indicator of the energy state of cells (Peters et al. 2004). The sensing of ATP levels via the ATP-sensitive potassium channels in terms of modulating the excitability of excitable cells is particularly critical to neurones and other cell types that sense the energy state of animals (Nichols, 2006).

In this study, we investigated the role of [ATP]i on the excitability of Hcrt neurones. Importantly, our results suggest that the energy state of Hcrt neurones as represented by the [ATP]i might shape the response of these neurones to external stimuli and thus the behavioural state of animals.

Methods

Animals

Male Hcrt-GFP mice (14 to 21 days old, on a C57BL6 genetic background) were group housed and maintained on a 12–12 h light–dark cycle with food and water available ad libitum. Sleep deprivation by gentle handling was performed as described previously (Rao et al. 2007). Briefly, mice in the SD group were sleep deprived (SD) for 4 hours (08.00 h to 12.00 h) by being gently touched with a small soft paintbrush on the back upon eye closure by an experimenter, while mice in the control group were allowed to undergo normal sleep. Mice were habituated to the experimenter and the paintbrush for 3 days before the SD experiment. All animal procedures were performed in strict accordance with NIH Care and Use of Laboratory Animals Guidelines and were approved by the Yale University Animal Care and Use Committee.

Electrophysiology

The preparation of hypothalamic slices containing Hcrt neurones was performed as described previously (Rao et al. 2007). Briefly, after mice were anaesthetized with isoflurane and decapitated, the brains were rapidly removed and immersed in cold (4°C) oxygenated high-sucrose solution containing (mm): sucrose 220, KCl 2.5, NaH2PO4 1.23, NaHCO3 26, CaCl21, MgCl2 6 and glucose 10, pH 7.3 with NaOH. After being trimmed to a small tissue block containing the hypothalamus, coronal slices (350 μm thick) were cut on a vibratome and maintained in a holding chamber with artificial cerebrospinal fluid (ACSF, bubbled with 5% CO2 and 95% O2) containing (in mm): NaCl 124, KCl 3, CaCl2 2, MgCl2 2, NaH2PO4 1.23, NaHCO3 26, glucose 2.5, pH 7.4 with NaOH. After a one-hour recovery period, slices were transferred to a recording chamber and were constantly perfused with ACSF (33°C) at a rate of 2 ml min−1.

Both conventional and perforated whole-cell patch clamp recording were performed in Hcrt-GFP neurones under voltage- and current clamp as reported previously (Rao et al. 2007). The micropipettes were made of borosilicate glass (World Precision Instruments) with a Sutter micropipette puller (P-97) and back filled with a pipette solution containing (mm): potassium gluconate 135, MgCl2 2, Hepes 10, EGTA 1.1, Mg-ATP 1–6 (according to our experimental design), Na2-phosphocreatin 10 and Na2-GTP 0.3, pH 7.3 with KOH. In the perforated whole-cell recording, amphotericin B (Sigma, final concentration 3%) was added to a modified pipette solution with 0 ATP and GTP. Both input resistance and series resistance were monitored throughout the experiments and the former was partially compensated. Only those recordings with stable series resistance and input resistance were accepted. The membrane and spontaneous action potential were recording in Hcrt neurones under current clamp, while whole-cell KATP current was recorded under voltage clamp. Tolbutamide, diazoxide and NaN3 were from Sigma and were applied to the recording chamber via ACSF. All data were sampled at 3–10 kHz, filtered at 1–3 kHz and analysed with an Apple Macintosh computer using AxoGraph X (AxoGraph Scientific). Statistics and plotting were performed with KaleidaGraph (Synergy software) and Igor Pro (WaveMetrics).

Results

[ATP]i modulates the excitability via KATP channels in Hcrt neurones

ATP level is an essential parameter representing energy availability and serves as the link between the energy state and excitability in neurones. Although Hcrt neurones are critical in sensing glucose and lactate levels in the brain of animals (Yamanaka et al. 2003; Burdakov et al. 2006; Parsons & Hirasawa, 2010), it is not yet clear whether the [ATP]i has any effects on neuronal activity in these cells. In Hcrt neurones under current clamp recorded with a pipette solution containing 2 mm ATP (a typical concentration in patch pipette solution), a few minutes after the establishment of whole-cell recording configuration, the spontaneous action potentials (APs) gradually disappeared and membrane potential (Vm) hyperpolarized significantly from −58.3 ± 2.2 mV (n = 20) at the start of current clamp recording to −67.5 ± 1.5 mV (n = 13) and maintained this steady-state level throughout the experiment (P < 0.05, t test, Fig. 1A). Next, we performed conventional whole-cell recording in Hcrt neurones with a fixed ATP concentration ranging from 1 to 6 mm in the pipette solution (‘ATP clamp’) in the absence or presence of tetrodotoxin (TTX, 0.5 μm), which eliminates the influence of APs on the Vm. As the Vm measured in the presence and absence of TTX at each ATP concentration was not significantly different, the results obtained under these two conditions were combined. The Vm of Hcrt neurones underwent a similar hyperpolarization and reached a plateau 5 min later when [ATP]i was lower than 5 mm (Fig. 1B). The final stabilized Vm was as follows: −69.6 ± 1.9 mV (n = 14, 1 mm ATP), −65.1 ± 1.4 mV (n = 29, 2 mm ATP), −61.9 ± 2.7 mV (n = 13, 3 mm ATP), −59.1 ± 2.3 mV (n = 14, 4 mm ATP), −48.7 ± 1.4 mV (n = 16, 5 mm ATP) and −49.7 ± 2.2 mV (n = 16, 6 mm ATP). When Hcrt neurones were examined with perforated recording with amphotericin B to preserve intracellular content (including [ATP]i), the Vm and APs remained at a stable level throughout the experiment and Vm (–46.1 ± 1.6 mV, n = 18) was close to that measured with 5–6 mm ATP in the pipette solution (Fig. 1A). Our further experiments verified that the KATP channels mediated the effects of [ATP]i on Hcrt neurones as seen in other types of cells (Nichols, 2006). A selective KATP channel blocker tolbutamide (100 μm) applied in the ACSF significantly reversed hyperpolarization in Hcrt neurones in conventional recording mode with 2 mm ATP in the pipette solution (control: −71.9 ± 1.7 mV, n = 11; tolbutamide: −53.7 ± 2.6 mV, n = 11; washout: −66.2 ± 2.6 mV, n = 11; P < 0.01, ANOVA, Fig. 1C). In contrast, Vm was unchanged in the presence of tolbutamide in perforated mode (control: −47.6 ± 2.0 mV, n = 5; tolbutamide: −48.0 ± 1.7 mV, n = 5; washout: −49.0 ± 1.9 mV, n = 5; P > 0.05, ANOVA, Fig. 1C). Furthermore, the KATP channel opener diazoxide (500 μm) applied in ACSF significantly hyperpolarized Hcrt neurones (control: −50.5 ± 2.1 mV, n = 12; diazoxide: to −69.1 ± 1.7 mV, n = 12; washout: −53.2 ± 1.6 mV, n = 11; P < 0.01, ANOVA, Fig. 1E and F). Together, these data indicate that the physiological concentration of ATP may be at 5–6 mm in Hcrt neurones and that the alteration in [ATP]i leads to changes in Vm via the opening of KATP channels in these cells.

Figure 1. Intracellular ATP levels determine membrane potential (Vm) by modulating KATP channels in Hcrt neurones.

Figure 1

Open in a new tab

A, ‘run down’ of Vm in Hcrt neurones after the dialysis of pipette solution containing 2 mm ATP into Hcrt neurones. Top and middle panels, raw traces of Vm and APs recorded in Hcrt neurones under current clamp in perforated (grey) and conventional (with 2 mm ATP in pipette solution, black) whole-cell modes. Bottom, time courses of Vm in Hcrt neurones examined under current clamp in perforated (grey, n = 6) and conventional (black, n = 29) whole-cell modes. Scale bar for the inset: 20 mV, 100 ms. B, the effects of [ATP]i on Vm in Hcrt neurones. Top, sample traces of Vm recorded with ATP levels at 1, 3 and 5 mm in the pipette solution. Bottom, pooled results from all cells tested at various ATP levels (1–6 mm) and undisturbed ATP levels (P, perforated recording). C and D, the time courses and pooled results show effects of the KATP channel blocker tolbutamide on Vm in Hcrt neurones at a low (2 mm, filled circles/bars) and an intact (open circles/bars) [ATP]i. E and F, time course and pooled results show that diazoxide significantly induces a hyperpolarization of Vm in Hcrt neurones with undisturbed [ATP]i. G, pooled results show that changes in intracellular ADP levels at the ATP/ADP ratios within the physiological range do not significantly alter Vm in Hcrt neurones. ATP (concentration was fixed at 2 mm) and ADP (at different concentrations according to the ATP/ADP ratios) were included in the pipette solution when conventional whole-cell recording was performed in Hcrt neurones. H, time courses of Vm in ARC neurones examined under current clamp in perforated (circles) and conventional whole-cell modes with 0 (diamonds), 0.5 (triangles) and 2 mm ATP (squares) in the pipette solution. *P < 0.05, t test.

Since the ATP metabolite ADP has also been shown to modulate KATP channels and the ratio of ATP/ADP triggers the energy-sensing mechanisms in neurones (Vesce et al. 2005; Nichols, 2006), we verified whether the ratio of ATP/ADP participated in the modulation of Vm in Hcrt neurones. To achieve this goal, we performed conventional whole-cell recording with both ATP and ADP in the pipette solution in Hcrt neurones. Specifically, we examined the effects of the ATP/ADP ratio found under physiological conditions on Vm in Hcrt neurones (Vesce et al. 2005). When the ATP/ADP ratio was adjusted to levels at 10:1, 5:1 and 2:1, the Vm was not significantly different from that without ADP in the pipette solution (P > 0.05, ANOVA) (Fig. 1G). Vm was −70.1 ± 2.1 mV (n = 12, 2 mm ATP only), −68.3 ± 1.7 mV (n = 14, ATP/ADP at 10:1), −72.3 ± 0.6 mV (n = 11, ATP/ADP at 5:1) and −67.6 ± 1.8 mV (n = 12, ATP/ADP at 2:1). These results suggest that the increase in ADP levels resulting from the use of ATP does not induce significant changes in Vm as long as the ATP levels are maintained stable through cellular metabolism in neurones and that the intracellular levels of ATP dominate the effects of adenosine nucleosides on Vm in Hcrt neurones. Lastly, to verify the specificity and validity of examining the effects of [ATP]i on Vm in Hcrt neurones, we used the same approach to examine the correlation of [ATP]i with Vm in neurones in the arcuate nucleus (ARC), since it was reported that [ATP]i was close to 0.9 mm in a previous report (Ainscow et al. 2002). As shown in Fig. 1H, Vm maintained stable in ARC neurones under perforated (initial value: −49.6 ± 2.3 mV, n = 7; 6 min later: −53.1 ± 2.7 mV, n = 7; P > 0.05, t test; Fig. 1H, circles) and conventional whole-cell recording with 2 mm ATP (initial value: −49.2 ± 3.4 mV, n = 6; 6 min later: −50.3 ± 3.8 mV, n = 6; P > 0.05, t test; Fig. 1H, squares) and 0.5 mm ATP (initial value: −48.3 ± 3.4 mV, n = 8; 6 min later: −48.5 ± 2.5 mV, n = 8; P > 0.05, t test; Fig. 1H, triangles) in the pipette solution. The Vm was significantly hyperpolarized 5 min after the establishment of whole-cell configuration in ARC neurones when a pipette solution with 0 mm ATP was used (initial value: −48.2 ± 2.6 mV, n = 7; 5 min later: −57.2 ± 3.1 mV, n = 7; P < 0.05, t test; Fig. 1H, diamonds). These results suggest that the [ATP]i required to maintain a normal Vm was no more than 0.5 mm, which is close to the [ATP]i measured with the luciferase–luciferin approach (Ainscow et al. 2002).

To further verify the effects of [ATP]i on Vm via KATP channels, we examined the opening of KATP channels at different [ATP]i in Hcrt neurones. The whole-cell KATP current was measured when a voltage ramp from −120 mV to −40 mV was applied to Hcrt neurones under voltage clamp with a pipette solution containing 1 mm ATP (in our case, the Hcrt neurones could not tolerate a long period of recording with no ATP in the pipette solution). The KATP current responses in the same Hcrt neurones were recorded in control, tolbutamide- and diazoxide-containing ACSF sequentially (Fig. 2, bottom right panel). The net KATP current at 1 mm ATP and the total KATP current were obtained by subtracting current responses recorded in control and diazoxide-containing ACSF from that recorded in the presence of tolbutamide (non-KATP membrane currents) (Fig. 2A). The total whole-cell conductance of KATP currents (calculated as the slope of the tolbutamide-sensitive current–voltage curve between −70 mV and −95 mV in the presence of diazoxide) was 53.68 ± 38.18 pS pF−1 (n = 7) in Hcrt neurones. The KATP current responses at ATP levels of 3 and 5 mm were examined by the same approach (Fig. 2B and C). At ATP levels of 1, 3 and 5 mm in Hcrt neurones, the open probability of KATP channels (p), defined as p = (Icontrol–Itolbutamide)/(Idiazoxide–Itolbutamide) × 100%, was 54.39 ± 13.01% (n = 7), 17.25 ± 3.95% (n = 6) and 4.52 ± 3.26% (n = 6), respectively (Fig. 2E). In perforated mode, the open probability of KATP channels measured with the approach described above was 0.3 ± 0.36% (n = 4), suggesting that the KATP channels are in the closed state in Hcrt neurones with intact intracellular ATP levels.

Figure 2. The state of KATP channels at different ATP levels in Hcrt neurones in LH slices.

Figure 2

Open in a new tab

Lower bottom right panel, sample traces of whole-cell KATP current responses to a voltage ramp from −120 mV to −40 mV in the presence of control, diazoxide- and tolbutamide-containing ACSF, respectively. A–D, whole-cell KATP current I–V curves at various ATP levels in Hcrt neurones (black traces). The grey traces are I–V curves of total KATP currents measured under each condition. E, pooled results of open probability of KATP channels at different [ATP]i in Hcrt neurones.

In summary, these results suggest that [ATP]i has a profound effect on the opening of KATP channels in Hcrt neurones, which may significantly dictate the Vm in these cells.

[ATP]i shapes the activity in Hcrt neurones

Next, we tested whether the interruption of ATP production by blocking oxidative phosphorylation or decreasing extracellular glucose has any effects on Vm in Hcrt neurones. In perforated mode, the application of NaN3 (1 mm, an inhibitor of oxidative phosphorylation) through ACSF (Matsumoto et al. 2002) significantly hyperpolarized Hcrt neurones a few minutes later (control: −50.0 ± 1.4 mV, n = 9; NaN3: −63.7 ± 1.2 mV, n = 9; washout: −48.9 ± 1.4 mV, n = 9; P < 0.01, ANOVA, Fig. 3A). Tolbutamide (100 μm) applied in the presence of NaN3 significantly reversed the effect of NaN3 on Vm in Hcrt neurones (control: −50.3 ± 1.6 mV, n = 6; NaN3: −65.5 ± 0.8 mV, n = 6; NaN3 plus tolbutamide: −44.5 ± 2.1 mV, n = 6; P < 0.01, ANOVA) (Fig. 3B). The same experiment was performed in Hcrt neurones under conventional whole-cell recording with an ATP level of 6 mm in the pipette solution. The application of NaN3 did not induce any changes in Vm in this protocol (control: −47.4 ± 2.0 mV, n = 5; NaN3: −47.7 ± 2.1 mV, n = 5; washout: −45.3 ± 1.7 mV, n = 5, P > 0.05, ANOVA, Fig. 3E and F). In a parallel set of experiments, we lowered the extracellular glucose concentration from 2.5 mm to 0.1 mm in the ACSF. The Vm of Hcrt neurones did not exhibit an acute change in response to the initial decrease in extracellular glucose concentration, but rather significantly hyperpolarized about 20 min later (control: −48.41 ± 1.66 mV, n = 7; low glucose: −53.46 ± 1.73 mV, n = 7, washout: −45.07 ± 1.81 mV, P < 0.01, ANOVA, Fig. 3G), which could be reversed by the application of tolbutamide (100 μm) (control: −51.64 ± 0.80 mV, n = 7; low glucose: −58.06 ± 1.93 mV, n = 7; low glucose + tolbutamide: −46.63 ± 1.44 mV, n = 7, P < 0.01, ANOVA, Fig. 3H). Together, these results suggest that the lowered [ATP]i resulting from the altered cellular metabolism induces the opening of KATP channels and hyperpolarization of Vm in recorded neurones. Replenishing the exogenous ATP could reverse this process.

Figure 3. The production of ATP determines the excitability in Hcrt neurones.

Figure 3

Open in a new tab

A and C, the inhibition of ATP production by NaN3 induces hyperpolarization in Hcrt neurones under perforated whole-cell recording. A sample trace is shown in A while the pooled results are presented in C. *P < 0.01, ANOVA. B and D, the effect of NaN3 on Vm in Hcrt neurones is reversed by the application of tolbutamide. *P < 0.01, ANOVA. E and F, the time course (E) and pooled results from all tested neurones (F) indicate that the effect of NaN3 on Vm is eliminated in Hcrt neurones under conventional whole-cell recording with 6 mm ATP in the pipette solution. G and H, the time course (top) and pooled results from all tested neurones (bottom) indicate that the depletion of extracellular glucose significantly hyperpolarizes Hcrt neurones (G, *P < 0.01, ANOVA), which is reversed by tolbutamide (H, *P < 0.01, ANOVA).

[ATP]i is relevant to the activity state of animals

It has been shown that the cerebral consumption of glucose and oxygen falls during sleep in humans, and that glucose uptake is high during wake and low during sleep in animals (Boyle et al. 1994; Vyazovskiy et al. 2008). It is not clear, however, whether [ATP]i fluctuates in central neurones during the sleep–wake cycle. We examined whether [ATP]i altered in Hcrt neurones dependent upon the behavioural state of animals in control (normal sleep) and sleep-deprived (SD, from 08.00 h to 12.00 h) mice (Rao et al. 2007). As shown in Fig. 4A, in perforated mode the Vm was significantly depolarized in SD mice versus controls (control: −52.0 ± 0.5 mV, n = 19; SD: −47.6 ± 0.8 mV, n = 17; P < 0.01, t test). When conventional recording with 2 mm ATP in the pipette solution was performed in control and SD mice, the Vm was −60.3 ± 1.6 mV (n = 8, control) and −53.4 ± 2.9 mV (n = 7, SD) initially, declined gradually and reached almost the same level (control: −69.8 ± 1.8 mV, n = 8; SD: −69.3 ± 1.5 mV, n = 7; P > 0.05, t test, Fig. 4B) once Vm stabilized (after the completion of exchange between the pipette solution and intracellular content). This result suggests that the initial difference in Vm between control and SD mice may be due to the different levels of ATP in Hcrt neurones. To further verify the effects of intracellular ATP on Vm measured in control and SD mice, we examined the effects of diazoxide (a KATP channel opener) on Hcrt neurones in these mice (Fig. 4C). In perforated recording mode, the Vm was −52.4 ± 0.8 mV (n = 11, control) and −48.0 ± 1.3 mV (n = 9, SD) initially (P < 0.05, t test), then hyperpolarized to −66.0 ± 1.9 mV (n = 11, control) and −66.3 ± 2.8 mV (n = 9, SD) during the application of diazoxide (P > 0.05, t test) and recovered upon removal of diazoxide. Note that we also recorded Vm in Hcrt neurones in conventional recording mode with 5 mm ATP in the pipette solution in control and SD mice. In Hcrt neurones in control mice, Vm modestly depolarized from the initial value of −55.1 ± 3.9 mV (n = 7) to −49.6 ± 2.7 mV (n = 7, P > 0.05, t test) 5 min after the establishment of whole-cell configuration, while it was unchanged in Hcrt neurones in SD mice (initial value: −49.8 ± 4.6 mV, n = 7; 5 min later: −49.2 ± 2.3 mV, n = 7). This result suggests that a concentration of 5 mm may be lower than (or close to) the level of ATP in Hcrt neurones in SD mice but higher than that in control mice. In a parallel experiment we measured Vm in ARC neurones from control and SD mice under perforated whole-cell recording. The Vm was not significantly different in ARC neurones in control and SD mice (control, −54.3 ± 2.7 mV, n = 10; SD, −55.6 ± 3.9 mV, n = 6, P > 0.05, t test). Together, these results suggest that Vm is significantly depolarized in Hcrt neurones in SD mice as compared to controls and that a higher level of ATP in Hcrt neurones in SD mice may contribute to this difference.

Figure 4. [ATP]i is elevated in Hcrt neurones in mice exposed to sleep deprivation (SD).

Figure 4

Open in a new tab

A, Vm of Hcrt neurones in brain slices recorded through perforated whole-cell recording in control and SD mice. *P < 0.01, t test. B, the time course of changes in Vm in Hcrt neurones shows that the substitution of [ATP]i with the same low ATP level (2 mm) eliminates the difference in Vm in Hcrt neurones in control and SD mice. The top panel shows representative traces of Vm recorded in Hcrt neurones in control and SD mice. C, the opening of KATP channels to the same extent with diazoxide eliminates the difference in Vm in Hcrt neurones in control and SD mice.

Discussion

In this study, by correlating [ATP]i with KATP current/membrane potential we demonstrate for the first time that Hcrt neurones in the LH rely on a high [ATP]i to maintain Vm and generate APs, and that the cellular metabolism may shape the activity in Hcrt neurones via changes in [ATP]i in these cells. Consistent with these results, our data that the [ATP]i was significantly higher in Hcrt neurones in mice with sustained wakefulness than in controls suggest a novel mechanism underlying the correlation between behavioural state and energy state in animals. We suspect that the difference in [ATP]i in Hcrt neurones between control and SD mice may be greater than what we report here, since the homogenized experimental condition in which brain slices were prepared and maintained, may lessen the difference in [ATP]i in Hcrt neurones between control and SD animals.

ATP is an essential commodity for biochemical and biophysical processes to occur in all cells in animals and humans. The determination of [ATP]i in intact neurones is critical to our understanding of energy metabolism in cells and eventually in whole animals. In this study we used an ‘ATP clamp’ approach to determine [ATP]i in neurones in brain slices. Compared to the luciferase–luciferin method (Ainscow et al. 2002), our approach can be used in neurones in vitro (acutely dissociated or cultured neurones and those in brain slices) and in vivo, as long as conventional and perforated whole-cell recording can be performed. In addition, our approach does not involve the expression of exogenous proteins in neurones, which adds an additional hurdle to overcome if using the luciferase-luciferin protocol. The resting cytosolic ATP concentration was estimated to be about 0.9 mm in cultured arcuate neurones and 0.6 mm in cultured cerebellar neurones (Ainscow et al. 2002). Consistent with these results, we showed that Vm and AP maintained intact with 0.5 mm ATP in the pipette solution in neurones in the ARC in hypothalamic slices in this study. The AP frequency and Vm remain intact for a long period of time (>10 min) under conventional whole-cell recording with 2 mm or less intracellular ATP in many central neurones, including substantia nigra, midbrain dopamine and cerebellar neurones (Jiang et al. 1994; Roper & Ashcroft, 1995; Stanford & Lacey, 1995). However, a concentration at this level seems insufficient to maintain normal Vm and APs in Hcrt neurones, in which an [ATP]i as high as 5–6 mm is required. It is noteworthy that basal forebrain cholinergic neurones may also require a high [ATP]i to maintain their normal function (Allen & Brown, 2004).

Previous results suggest that Hcrt neurones are inhibited by an acute elevation in the extracellular concentration of glucose, suggesting that Hcrt neurones act as an energy sensor to prevent excessive energy intake by animals (Yamanaka et al. 2003; Burdakov et al. 2006; Guyon et al. 2009). Our results that insufficient [ATP]i resulting from deficiency in ATP production or energy sources (low ACSF glucose supply) leads to inactivation of Hcrt neurones demonstrate heterogeneous mechanisms of energy sensing in these neurones, consistent with a recent report (Parsons & Hirasawa, 2010). This mode of energy sensing has additional physiological significance. On the one hand, the high [ATP]i in Hcrt neurones may reflect an ample supply of energy in the brain and probably in the entire body. The closure of KATP channels depolarizes the Vm of Hcrt neurones enough to facilitate AP generation triggered by various cues, leading to sufficient arousal and more importantly the mobilization of energy expenditure to support animal behaviours. This is consistent with a report on the glucose utilization in skeletal muscle through the activation of the sympathetic nervous system by Hcrt (Shiuchi et al. 2009). On the other hand, the requirement of a high [ATP]i in Hcrt neurones to maintain a depolarized Vm and AP firing may lead to a rapid reduction in energy expenditure in whole animals in addition to a neuroprotective role in neurones under hypoglycaemia or anoxia (Ballanyi, 2004). When the energy source is scarce in the brain leading to a lowered [ATP]i inside neurones, the Hcrt cells would be less active due to the opening of KATP channels resulting from the fall in [ATP]i before the same change occurs in other types of neurones, in which lower ATP levels are required to maintain their excitability (Jiang et al. 1994; Roper & Ashcroft, 1995; Stanford & Lacey, 1995). Thus, the decline in the excitability of Hcrt neurones would, in turn, tune down the activity in its target systems, leading to reduced arousal and energy expenditure in animals to conserve energy. Therefore, the Hcrt neurones may provide an early alert to the shortage of energy in the brain.

In the light of changes in [ATP]i in Hcrt neurones between spontaneous sleep and prolonged wakefulness, our results are consistent with reports of elevated activity in ATP production machinery in the cortex during prolonged wakefulness and the fluctuation in ATP levels in several brain areas across the sleep–wake cycle in animals (Dworak et al. 2010; Nikonova et al. 2010). However, due to the technical limitations the measurements were performed in brain tissues but not neurones in both studies (Dworak et al. 2010; Nikonova et al. 2010). Therefore, our results provide a piece of direct evidence of changes in ATP levels in neurones according to the behavioural state of animals. Moreover, the correlation between the activity and [ATP]i in Hcrt neurones may serve as one mechanism underlying the energy hypothesis of sleep, in which the depletion of ATP may limit arousal or wake and promote sleep in animals (Benington & Heller, 1995; Scharf et al. 2008). In light of most recent results that the cerebral consumption of glucose and oxygen and glucose uptake fluctuates during the sleep–wake cycle in humans and animals (Boyle et al. 1994; Vyazovskiy et al. 2008), it is intriguing to examine whether a lowered energy state in neurones during sleep is a ubiquitous adaptation in all neurons. Also, our observations reported in this study may be applicable to other brain areas.

In conclusion, we demonstrate that the unique requirement of a high energy state in Hcrt neurones provides a novel mechanism for sensing the availability of energy in the brain and shape the behavioural state dependent on the energy status in animals.

Acknowledgments

This work is supported by NIH grant DK070723. The authors would like to thank Mss Marya Shanabrough and Susan Andranovich for assistance with the manuscript.

Glossary

Abbreviations

ARC

arcuate nucleus

GFP

green fluorescent protein

Hcrt

hypocretin/orexin

LH

lateral hypothalamus

Vm

membrane potential

Author contributions

Z.-W.L., G.G. and X.-B.G. designed the experiments. Z.-W.L., G.G. and S.S. performed experiments and analysed results. Z.-W.L., G.G. and X.-B.G. wrote the manuscript. All authors approved the final version.

References

  1. Ainscow EK, Mirshamsi S, Tang T, Ashford ML, Rutter GA. Dynamic imaging of free cytosolic ATP concentration during fuel sensing by rat hypothalamic neurones: evidence for ATP-independent control of ATP-sensitive K+ channels. J Physiol. 2002;544:429–445. doi: 10.1113/jphysiol.2002.022434. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Allen TG, Brown DA. Modulation of the excitability of cholinergic basal forebrain neurones by KATP channels. J Physiol. 2004;554:353–370. doi: 10.1113/jphysiol.2003.055889. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Ballanyi K. Protective role of neuronal KATP channels in brain hypoxia. J Exp Biol. 2004;207:3201–3212. doi: 10.1242/jeb.01106. [DOI] [PubMed] [Google Scholar]
  4. Benington JH, Heller HC. Restoration of brain energy metabolism as the function of sleep. Prog Neurobiol. 1995;45:347–360. doi: 10.1016/0301-0082(94)00057-o. [DOI] [PubMed] [Google Scholar]
  5. Boyle PJ, Scott JC, Krentz AJ, Nagy RJ, Comstock E, Hoffman C. Diminished brain glucose metabolism is a significant determinant for falling rates of systemic glucose utilization during sleep in normal humans. J Clin Invest. 1994;93:529–535. doi: 10.1172/JCI117003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Burdakov D, Jensen LT, Alexopoulos H, Williams RH, Fearon IM, O'Kelly I, et al. Tandem-pore K+ channels mediate inhibition of orexin neurons by glucose. Neuron. 2006;50:711–722. doi: 10.1016/j.neuron.2006.04.032. [DOI] [PubMed] [Google Scholar]
  7. Chemelli RM, Willie JT, Sinton CM, Elmquist JK, Scammell T, Lee C, et al. Narcolepsy in orexin knockout mice: molecular genetics of sleep regulation. Cell. 1999;98:437–451. doi: 10.1016/s0092-8674(00)81973-x. [DOI] [PubMed] [Google Scholar]
  8. de Lecea L, Kilduff TS, Peyron C, Gao XB, Foye PE, Danielson PE, et al. The hypocretins: hypothalamus-specific peptides with neuroexcitatory activity. Proc Natl Acad Sci U S A. 1998;95:322–327. doi: 10.1073/pnas.95.1.322. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Dworak M, McCarley RW, Kim T, Kalinchuk AV, Basheer R. Sleep and brain energy levels: ATP changes during sleep. J Neurosci. 2010;30:9007–9016. doi: 10.1523/JNEUROSCI.1423-10.2010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Guyon A, Tardy MP, Rovère C, Nahon JL, Barhanin J, Lesage F. Glucose inhibition persists in hypothalamic neurons lacking tandem-pore K+ channels. J Neurosci. 2009;29:2528–2533. doi: 10.1523/JNEUROSCI.5764-08.2009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Jiang C, Sigworth FJ, Haddad GG. Oxygen deprivation activates an ATP-inhibitable K+ channel in substantia nigra neurons. J Neurosci. 1994;14:5590–5602. doi: 10.1523/JNEUROSCI.14-09-05590.1994. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Lin L, Faraco J, Li R, Kadotani H, Rogers W, Lin X, et al. The sleep disorder canine narcolepsy is caused by a mutation in the hypocretin (orexin) receptor 2 gene. Cell. 1999;98:365–376. doi: 10.1016/s0092-8674(00)81965-0. [DOI] [PubMed] [Google Scholar]
  13. Matsumoto N, Komiyama S, Akaike N. Pre- and postsynaptic ATP-sensitive potassium channels during metabolic inhibition of rat hippocampal CA1 neurons. J Physiol. 2002;541:511–520. doi: 10.1113/jphysiol.2002.018267. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Nichols CG. KATP channels as molecular sensors of cellular metabolism. Nature. 2006;440:470–476. doi: 10.1038/nature04711. [DOI] [PubMed] [Google Scholar]
  15. Nikonova EV, Naidoo N, Zhang L, Romer M, Cater JR, Scharf MT, et al. Changes in components of energy regulation in mouse cortex with increases in wakefulness. Sleep. 2010;33:889–900. doi: 10.1093/sleep/33.7.889. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Parsons MP, Hirasawa M. ATP-sensitive potassium channel-mediated lactate effect on orexin neurons: implications for brain energetics during arousal. J Neurosci. 2010;30:8061–8070. doi: 10.1523/JNEUROSCI.5741-09.2010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Peters A, Schweiger U, Pellerin L, Hubold C, Oltmanns KM, Conrad M, et al. The selfish brain: competition for energy resources. Neurosci Biobehav Rev. 2004;28:143–180. doi: 10.1016/j.neubiorev.2004.03.002. [DOI] [PubMed] [Google Scholar]
  18. Rao Y, Liu ZW, Borok E, Rabenstein RL, Shanabrough M, Lu M, et al. Prolonged wakefulness induces experience-dependent synaptic plasticity in mouse hypocretin/orexin neurons. J Clin Invest. 2007;117:4022–4033. doi: 10.1172/JCI32829. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Roper J, Ashcroft FM. Metabolic inhibition and low internal ATP activate K-ATP channels in rat dopaminergic substantia nigra neurones. Pflugers Arch. 1995;430:44–54. doi: 10.1007/BF00373838. [DOI] [PubMed] [Google Scholar]
  20. Sakurai T, Amemiya A, Ishii M, Matsuzaki I, Chemelli RM, Tanaka H, et al. Orexins and orexin receptors: a family of hypothalamic neuropeptides and G protein-coupled receptors that regulate feeding behavior. Cell. 1998;92:573–585. doi: 10.1016/s0092-8674(00)80949-6. [DOI] [PubMed] [Google Scholar]
  21. Scharf MT, Naidoo N, Zimmerman JE, Pack AI. The energy hypothesis of sleep revisited. Prog Neurobiol. 2008;86:264–280. doi: 10.1016/j.pneurobio.2008.08.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Shiuchi T, Haque MS, Okamoto S, Inoue T, Kageyama H, Lee S, et al. Hypothalamic orexin stimulates feeding-associated glucose utilization in skeletal muscle via sympathetic nervous system. Cell Metab. 2009;10:466–480. doi: 10.1016/j.cmet.2009.09.013. [DOI] [PubMed] [Google Scholar]
  23. Shulman RG, Hyder F, Rothman DL. Cerebral metabolism and consciousness. C R Biol. 2003;326:253–273. doi: 10.1016/s1631-0691(03)00071-4. [DOI] [PubMed] [Google Scholar]
  24. Shulman RG, Hyder F, Rothman DL. Baseline brain energy supports the state of consciousness. Proc Natl Acad Sci U S A. 2009;106:11096–11101. doi: 10.1073/pnas.0903941106. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Shulman RG, Rothman DL, Hyder F. Stimulated changes in localized cerebral energy consumption under anesthesia. Proc Natl Acad Sci U S A. 1999;96:3245–3250. doi: 10.1073/pnas.96.6.3245. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Stanford IM, Lacey MG. Regulation of a potassium conductance in rat midbrain dopamine neurons by intracellular adenosine triphosphate (ATP) and the sulfonylureas tolbutamide and glibenclamide. J Neurosci. 1995;15:4651–4657. doi: 10.1523/JNEUROSCI.15-06-04651.1995. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Vesce S, Jekabsons MB, Johnson-Cadwell LI, Nicholls DG. Acute glutathione depletion restricts mitochondrial ATP export in cerebellar granule neurons. J Biol Chem. 2005;280:38720–38728. doi: 10.1074/jbc.M506575200. [DOI] [PubMed] [Google Scholar]
  28. Vyazovskiy VV, Cirelli C, Tononi G, Tobler I. Cortical metabolic rates as measured by 2-deoxyglucose-uptake are increased after waking and decreased after sleep in mice. Brain Res Bull. 2008;75:591–597. doi: 10.1016/j.brainresbull.2007.10.040. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Yamanaka A, Beuckmann CT, Willie JT, Hara J, Tsujino N, Mieda M, et al. Hypothalamic orexin neurons regulate arousal according to energy balance in mice. Neuron. 2003;38:701–713. doi: 10.1016/s0896-6273(03)00331-3. [DOI] [PubMed] [Google Scholar]

Articles from The Journal of Physiology are provided here courtesy of The Physiological Society

RESOURCES