Sleep and brain energy levels: ATP changes during sleep

Abstract

Sleep is one of the most pervasive biological phenomena, but one whose function remains elusive. Although many theories of function, indirect evidence, and even common sense suggest sleep is needed for an increase in brain energy, brain energy levels have not been directly measured with modern technology. We here report that ATP levels, the energy currency of brain cells, show a surge in the initial hours of spontaneous sleep in wake-active but not in sleep-active brain regions of rat. The surge is dependent on sleep but not time of day, since preventing sleep by gentle handling of rats for 3 or 6 h also prevents the surge in ATP. A significant positive correlation was observed between the surge in ATP and EEG non-rapid eye movement delta activity (0.5-4.5 Hz) during spontaneous sleep. Inducing sleep and delta activity by adenosine infusion into basal forebrain during the normally active dark period also increases ATP. Together, these observations suggest that the surge in ATP occurs when the neuronal activity is reduced, as occurs during sleep. The levels of phosphorylated AMP-activated protein kinase (P-AMPK), well known for its role in cellular energy sensing and regulation, and ATP show reciprocal changes. P-AMPK levels are lower during the sleep-induced ATP surge than during wake or sleep deprivation. Together, these results suggest that sleep-induced surge in ATP and the decrease in P-AMPK levels set the stage for increased anabolic processes during sleep and provide insight into the molecular events leading to the restorative biosynthetic processes occurring during sleep.

Figure 1.

Diurnal variation in steady-state ATP levels in brain regions. Mean ATP levels (±SEM) normalized to 7:00 A.M. values in frontal cortex, basal forebrain, cingulate cortex, and hippocampus. Average ATP values during the wake (dark) period (dark background) remained stable, whereas during the sleep (lights-on) period the ATP levels showed significant alterations in values within each brain region [Kruskal–Wallis ANOVAs, n = 6 in each region; FC (H = 21.622; df = 4; p = 0.001); BF (H = 17.123; df = 4; p = 0.002); CCX (H = 15.449; df = 4; p = 0.004); HIPP (H = 14.439; df = 4; p = 0.006)] compared with waking. The levels surged significantly (n = 6; *p = 0.01) for 10:00 A.M. and 1:00 A.M. compared with other time points. Coronal anatomical plates are from the atlas of Paxinos and Watson (2007).

Figure 2.

NREM delta activity for entire lights-on (sleep) period. A, Delta activity values are calculated as percentage of mean NREM delta activity (±SEM) for each hour of NREM sleep, with values placed at midpoint of respective hour. NREM delta activity was significantly higher in the first 4 h compared with later time points (12:00 P.M. to 7:00 P.M.) (F, 2.538; df = 11; *p < 0.05). B, Correlation of the average NREM delta activity and ATP. NREM delta was averaged for 3 h bins (solid line with black diamond) and compared with the ATP levels for basal forebrain (dashed line with open triangle) calculated at the end of corresponding 3 h. Note that the increases in ATP levels closely follow the changes in 3 h average NREM delta activity.

Figure 3.

Unilateral AD perfusion into basal forebrain causes highly correlated delta and ATP increases. A, Compared with no treatment (dotted line, 100%) unilateral AD perfusion into basal forebrain (black bars) during 3 h (7:00–10:00 P.M.) in the dark period significantly increased both NREM sleep duration (240.8 ± 38.8%) and NREM delta activity (167.6 ± 27%) (within animal comparisons; n = 5; Student's t test, *p = 0.008). In contrast, aCSF perfusion values (white bars) were not significantly different from no treatment. B, ATP concentrations measured in FC, BF, CCX, and HIPP at the end of 3 h of AD or aCSF infusion into basal forebrain. AD-perfused rats showed significant increases in ATP concentrations in all brain regions when compared with aCSF-perfused controls (n = 5/region for AD and aCSF groups; Kruskal–Wallis ANOVA on ranks, H = 28.455, df = 7, p = 0.001; Student–Newman–Keul pairwise comparison for each region, *p = 0.05). C, D, Correlation analysis of ATP levels and NREM delta activity in AD- and aCSF-perfused rats. In all animals (adenosine, black diamonds; aCSF, white diamonds), a positive correlation was observed between ATP levels and NREM delta in FC (C) (n = 5/group; Spearman's ρ = 0.83; p = 0.0006) and in BF (D) (n = 5/group; Spearman's ρ = 0.64; p = 0.04). Error bars indicate SEM.

Figure 4.

SD postpones the sleep-induced surge in ATP. A–D, SD (hatched bar) at the 7:00 A.M. onset of light period until 10:00 A.M. significantly attenuated the ATP increase in all brain regions compared with the sleeping diurnal controls (n = 6; *p = 0. 01). When RS (open bars) was permitted after 3 h SD, a surge in ATP occurred after a 3 h lag in basal forebrain and after a 6 h lag in frontal and cingulate cortex. E–H, Three hour SD beginning at 10:00 A.M., when the ATP surge was maximal, caused a decline in ATP to levels lower than sleeping controls at 1:00 P.M. (BF, p = 0.022; FC, p = 0.006; CCX, p = 0.004; HIPP, p = 0.004). When the 3 h SD was followed by 3 h of RS (1:00 P.M. to 4:00 P.M.), ATP levels increased to levels significantly higher (*p < 0.01) than in diurnal controls (4:00 P.M.) in basal forebrain and cingulate cortex, and equal to the diurnal control levels in frontal cortex and hippocampus. I–L, Six hour SD (hatched box) markedly decreased ATP levels compared with diurnal controls in all brain regions (each *p < 0.05). In addition, a small but statistically significant decrease occurred to below the baseline (7:00 A.M.) level in all brain regions (each **p < 0.05) except for FC, where the levels matched the baseline. In all brain regions, 3 h of RS (open bar) produced an ATP increase to levels that matched those of diurnal controls at 4:00 P.M. Error bars indicate SEM.

Figure 5.

Effect of sleep deprivation on the levels of ATP in LH and VLPO. A, Coronal sections of the brain showing bregma coordinates of dissected brain regions: FC, LH, VLPO. B, Rats (N = 5/group), sleep deprived for 3 h (7:00–10:00 A.M.; open bar) showed significant decrease in ATP in FC (−53.00 ± 8.01%; **p < 0.01) and LH (−40.17 ± 19.8%; *p < 0.05), whereas no significant change was observed in VLPO (+10.69 ± 20.36%; p = 0.719) when compared with undisturbed time matched sleeping controls (black bars). Error bars indicate SEM.

Figure 6.

Changes in the phosphocreatine and creatine levels. After 3 h of sleep (10:00 A.M.), the levels of PCr (solid bars) showed a trend increase when compared with the levels observed at 7:00 A.M. in all four regions (n = 6/group), with no change in Cr (open bars) levels. The levels of PCr decreased significantly in FC (−57 ± 17%; *p < 0.001), BF (−58 ± 23%; *p < 0.01), and CCX (−45 ± 20%; *p = 0.044) after 3 h SD (7:00 A.M. to 10:00 A.M.) when compared with undisturbed time-matched sleeping controls. HIPP did not show significant change (−24 ± 27; *p = 0.228). The Cr levels showed a significant increase in HIPP (+42 ± 5%; *p = 0.002), whereas the other three regions showed trend increases. Error bars indicate SEM.

Figure 7.

Sleep–wake and SD-associated changes in P-AMPK. A, Western blots show the P-AMPK band intensity and the corresponding band for the normalizer β-actin. B, A graph of the levels of P-AMPK protein in BF show a decline with sleep, with lower values at 10:00 A.M. and 1:00 P.M. when compared with 7:00 A.M. (open triangles; solid line), whereas 3 h of SD (7:00 A.M. to 10:00 A.M.) produces significantly higher levels (n = 4; *p = 0.02) of P-AMPK when compared with the 10:00 A.M. diurnal control and rats that were allowed 3 h of RS (closed squares, dotted line). C–F, Reciprocal relationship between ATP and P-AMPK. C, Western blots show that, in BF, the levels of P-AMPK protein are higher at 7:00 A.M. and after 3 h SD when compared with 10:00 A.M. values in sleeping animals. D, A reciprocal relationship is observed between P-AMPK (open bar) and ATP (black bar) levels in basal forebrain. Compared with 10:00 A.M. values, P-AMPK levels are significantly higher (n = 4; *p = 0.02), whereas the ATP levels are significantly lower (n = 6; p = 0.01) at 7:00 A.M. Compared with 10:00 A.M. values in sleeping controls (C), 10:00 A.M. values in animals not allowed to sleep after 7:00 A.M. (SD) show that the P-AMPK levels remain high (n = 4; *p = 0.02) and ATP values remain low (n = 4; *p = 0.02). E, F, In frontal cortex, a similar reciprocal relationship between P-AMPK (trend-level, 7:00 A.M. vs 10:00 A.M., p = 0.13; 10:00 A.M. vs 3 h SD, p = 0.08) and ATP is present (*p = 0.02). Error bars indicate SEM.

Figure 8.

Model for sleep–wake-dependent changes in ATP, AMPK, and AMPK-regulated anabolic and catabolic pathways. This figure summarizes our hypotheses about regulation of AMPK-dependent anabolic and catabolic processes during sleep and wake. During spontaneous waking and sleep deprivation, both characterized by increased neuronal activity, ATP use is increased and the resulting higher AMP/ATP ratio leads to higher levels of phosphorylated AMPK (P-AMPK). This active form facilitates catabolic processes and inhibits energy requiring anabolic processes. In contrast, during the initial hours of sleep and during recovery sleep after sleep deprivation, states characterized by high NREM delta activity/low neuronal activity, ATP levels surge. This leads to a lower AMP/ATP ratio and a decrease in phosphorylated AMPK. The inactive AMPK allows anabolic processes to proceed including biosynthesis of protein, glycogen, fatty acid, etc., and thereby facilitates restorative biosynthetic processes occurring during sleep.