PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
 
J Neurosci. Author manuscript; available in PMC 2011 March 29.
Published in final edited form as:
PMCID: PMC2950004
NIHMSID: NIHMS240871

Manipulation of adenosine kinase affects sleep regulation in mice

Svitlana Palchykova and Raphaelle Winsky-Sommerer
Institute of Pharmacology and Toxicology, University of Zurich, Zurich, Switzerland

Abstract

Sleep and sleep intensity are enhanced by adenosine and its receptor agonists, while adenosine receptor antagonists induce wakefulness. Adenosine kinase (ADK) is the primary enzyme metabolizing adenosine in adult brain. To investigate whether adenosine metabolism or clearance affects sleep we recorded sleep in mice with engineered mutations in Adk. Adk-tg mice over-express a transgene encoding the cytoplasmic isoform of ADK in the brain, but lack the nuclear isoform of the enzyme. Wild-type mice and Adk+/− mice that have a 50% reduction of the cytoplasmic and the nuclear isoforms of ADK served as controls.

Adk-tg mice showed a remarkable reduction of EEG power in low frequencies in all vigilance states and in theta activity (6.25–11 Hz) in REM sleep and waking. Adk-tg mice were awake 58 min more per day than wild-type mice and spent significantly less time in REM sleep (102±3 vs 128±3 min in wild-type). After sleep deprivation slow-wave activity (0.75–4 Hz), the intensity component of NREM sleep, increased significantly less in Adk-tg mice and their slow-wave energy was reduced. In contrast, the vigilance states and EEG spectra of Adk+/− and wild-type mice did not differ. Our data suggest that over-expression of the cytoplasmic isoform of ADK is sufficient to alter sleep physiology. ADK might orchestrate neurotransmitter pathways involved in the generation of EEG oscillations and regulation of sleep.

Keywords: adenosine kinase, sleep homeostasis, EEG spectral analysis, REM sleep, sleep deprivation, mice

Introduction

The neuromodulator adenosine plays a key role in sleep regulation (Radulovacki, 1985; Benington and Heller, 1995; Basheer et al., 2004; Datta and Maclean, 2007; Huang et al., 2007; Scharf et al., 2008). Adenosine and adenosine receptor agonists enhance NREM and REM sleep (Marley and Nistico, 1972; Haulica et al., 1973; Snyder et al., 1981; Virus et al., 1983; Radulovacki et al., 1985; Sarda et al., 1986; Martin et al., 1989; Ticho and Radulovacki, 1991; Portas et al., 1997). Moreover, the adenosine agonist cyclopentyladenosine enhanced EEG slow-wave activity (SWA; power between 0.75–4 Hz; Benington et al., 1995; Schwierin et al., 1996). Adenosine receptor antagonists (caffeine, cyclopentyltheophylline) induce wakefulness (Yanik et al., 1987; Biaggioni et al., 1991; Landolt et al., 1995; Schwierin et al., 1996; Huang et al., 2005). The effects of adenosine on sleep are mediated by adenosine A1 (Basheer et al., 2001; Thakkar et al., 2003; Alanko et al., 2004; Bjorness et al., 2009) and A2A receptors (Hayaishi et al., 2004). Inhibition of gliotransmission attenuated the accumulation of sleep pressure through A1 receptors (Halassa et al., 2009).

Despite its ubiquitous presence, some studies indicated a role of adenosine in sleep in specific brain regions. The basal forebrain was suggested to gate sleep via adenosinergic inhibition of wake-promoting cholinergic neurons (Porkka-Heiskanen et al., 1997; Porkka-Heiskanen et al., 2000; Strecker et al., 2000). However, stimulation of adenosine receptors in other regions (e.g. hypothalamus) also induced sleep (Satoh et al., 1998; Satoh et al., 1999; Scammell et al., 2001; Gallopin et al., 2005; Methippara et al., 2005; Oishi et al., 2008).

A little explored avenue in sleep is the in vivo manipulation of adenosine kinase (ADK), the primary enzyme regulating adenosine metabolism in rodents (Mathews et al., 1998; Boison, 2006). Inhibition of ADK activity induced a larger adenosine increase in hippocampal and cortical slices than adenosine deaminase inhibition (Pak et al., 1994; Lloyd and Fredholm, 1995; White, 1996). Moreover, adenosine release was higher in ADK-deficient fibroblasts compared to adenosine deaminase-deficient fibroblasts (Huber et al., 2001). Sleep deprivation (SD) had no effect on ADK activity in the cortex and basal forebrain, but the diurnal variation of ADK activity (Chagoya de Sanchez et al., 1993; Alanko et al., 2003) was maximal in the cortex (Mackiewicz et al., 2003).

Here we used mice constitutively over-expressing a transgene for the cytoplasmic isoform of Adk (Adk-tg). These mutants are characterized by increased ADK enzyme activity and reduced adenosine tone in the brain (Fedele et al., 2005). Based on the reduced adenosine tone, mutants display spontaneous intrahippocampal seizures (Li et al., 2007; Li et al., 2008a; Li et al., 2008b), increased susceptibility to stroke or seizure-induced neuronal cell death (Pignataro et al., 2007) and cognitive impairment (Yee et al., 2007). We hypothesized that the transgene over-expression in Adk-tg mice should affect sleep. As controls we chose Adk+/− mice, which have reduced ADK in the liver, but normal adenosine metabolites (Boison et al., 2002) and wild-type (WT) littermates. The mice were subjected to 6-h SD to stimulate sleep regulatory mechanisms.

Materials and methods

Animals

Adult heterozygous male mice originating from a targeted disruption of the Adk-gene (Adk+/−), maintained on a mixed 129/JEms × C57Bl/6 background (see Boison et al., 2002 for characterization), and wild-type (WT) littermates were obtained by heterozygous breeding (n = 15 Adk+/−, n = 14 WT mice). Mice over-expressing the cytoplasmic isoform of ADK were derived by rescuing the lethal phenotype of Adk−/− mice by introducing a UbiAdk transgene into the Adk−/− background (Fedele et al., 2005) and maintained by homozygous breeding of Adk-tg mice (n = 11). The mice were kept individually in Macrolon cages (36 × 20 × 35 cm) with food and water available ad libitum, maintained on a 12 h light - 12 h dark cycle (~ 30 lux; light onset at 8:00) at 23°C ambient temperature. The Cantonal Veterinary Office of Zurich approved all experimental procedures.

Surgery: EEG/EMG electrode implantation

Mice were implanted epidurally with gold-plated miniature screws (0.9 mm diameter) for EEG recording under isoflurane anesthesia (right hemisphere: above the frontal (FRO), 1.5 mm anterior to bregma, 2 mm lateral to the midline, and the parietal cortex (PAR), 2 mm posterior to bregma, 3 mm lateral to midline; reference: above the cerebellum, 2 mm posterior to lambda, on the midline). Two gold wires (0.2 mm diameter) were inserted bilaterally into the neck muscles for electromyogram (EMG) recording. The electrodes were connected to stainless steel wires fixed to the skull with dental acrylic cement. The mice were connected by a fine cable to a swivel, and remained connected throughout the experiment. At least 3 weeks were allotted for recovery from surgery and a minimum of 3 days for habituation to the recording chambers before data collection.

Experimental protocols

Recordings were performed during a 24-h baseline followed by 6 h SD starting at light onset and 18 h recovery. SD was performed by directly observing the mice and introducing new material into the cage (e.g. nesting material, wood or cardboard pieces) whenever a mouse appeared to be drowsy. If necessary, acoustic stimulation was used by gentle tapping on the cage (Tobler et al., 1997). Special care was taken not to interfere with eating and drinking behavior. Data were collected from frontal and parietal EEG derivations, because slow waves and spindles in NREM sleep are more prominent in the frontal EEG, whereas theta activity in waking and REM sleep is prominent in the parietal EEG.

A separate batch of Adk-tg (n = 12) and WT mice (n = 11) were placed in single cages equipped with an infra-red sensor mounted above the cage for undisturbed, continuous motor activity recordings for 10 consecutive days (Chronobiology Kit, Stanford Software Systems, Stanford, CA).

Data acquisition and analysis

The EEG and EMG signals were amplified (amplification factor approx. 2000), conditioned by analogue filters (high-pass filter: −3 dB at 0.016 Hz; low-pass filter: −3 dB at 40 Hz, less than −35 dB at 128 Hz) sampled with 512 Hz, digitally filtered (EEG: low-pass FIR filter 25 Hz; EMG: band-pass FIR filter 20–50 Hz) and stored with a resolution of 128 Hz. The EEG power spectra were computed for 4-s epochs by a Fast Fourier Transform (FFT) routine. Adjacent 0.25-Hz bins were averaged into 0.5-Hz (0.25–5 Hz) and 1.0-Hz (5.25–25 Hz) bins. The EMG was full-wave rectified and integrated over 4-s epochs; ambient temperature within the animal cage was sampled at 4-s intervals. Before each recording the EEG and EMG channels were calibrated with a 10-Hz, 300 μV peak-to-peak sine wave.

The three vigilance states NREM sleep, REM sleep and waking were determined off-line by visual inspection of the parietal and frontal EEG, EMG records and EEG power in the slow-wave range (0.75–4 Hz) for 4-s epochs. Epochs containing EEG artifacts in one derivation were excluded from spectral analyses of both EEG derivations (6.3 ± 0.6 % of recording time, which occurred mainly during active waking). Vigilance states could always be determined. The frequency and duration of vigilance state episodes was computed using the methods of Deboer et al. (1994). Sleep continuity was assessed by determining the number of brief awakenings (episodes of waking <16 sec, Deboer et al., 1994). The time course of slow-wave activity (SWA; 0.75–4 Hz), spindle frequency activity (SFA; 10.25–15 Hz) and theta activity (6.25–9 Hz) for 2 min before and after the transition from waking or REM sleep to NREM sleep or from NREM sleep to REM sleep was computed as previously (Franken et al., 1994). Data analysis was performed using the MATLAB software package (The Math Works, Inc., Natick, MA, USA).

Western blot analysis

We determined protein levels of ADK in brains of Adk-tg, Adk+/− and WT mice which had been used for sleep recordings (n=6 mice per genotype). Brain tissue from the right hemispheres was processed for aqueous protein extracts as previously described (Gouder et al., 2004). Cell extracts were resuspended, each containing 40 μg proteins, and electrophoresed in a 10% Tris-glycine gel. After blotting, the membranes were incubated with polyclonal rabbit antiserum against ADK (1:5,000) and processed as described (Gouder et al., 2004). The membranes were then developed by incubation with peroxidase-conjugated anti-rabbit antibody (#7074, 1:8,000, Cell Signaling, Danvers, MA). Immunoblots were quantified using a Kodak Scientific Imaging System (v3.6.5.k2, Kodak). To normalize ADK immunoreactivity, a mouse monoclonal anti-α-Tubulin antibody (# sc-8035, 1:5,000; Santa Cruz, CA) was reprobed on the same blot and the OD ratio of ADK/ α-Tubulin was calculated.

Statistical analysis

Genotype differences were evaluated by linear mixed model analysis of variance (ANOVA), corrected for multiple comparison when appropriate, or by a repeated measures ANOVA (SAS software, SAS Institute, Cary, NC, USA). Post-hoc paired and unpaired t-test or Tukey-Kramer, Bonferroni and Kruskal-Wallis tests were performed if the results of the ANOVA reached statistical significance (p < 0.05). All reported values are means over 6 – 15 mice ± SEM.

Results

Over-expression of Adk affects the EEG and impairs sleep regulation

We previously demonstrated reduced adenosinergic tone in Adk-tg mice (Fedele et al., 2005) that was subsequently validated in several experimental paradigms (Li et al., 2007; Pignataro et al., 2007; Yee et al., 2007; Li et al., 2008a; Li et al., 2008b). We therefore hypothesized that this genetic modification would affect sleep.

EEG power spectra

A statistically significant reduction of EEG power density was observed in the frequencies below 4 Hz in all vigilance states in Adk-tg mice compared to WT mice (Figs. 1 and and2).2). In addition, in REM sleep and waking, EEG power was significantly lower in the Adk-tg mice in the frequencies encompassing the theta band (6.25–11 Hz; Fig. 1).

Figure 1
EEG power density in microvolt square per Hertz (0.75–25 Hz) in NREM sleep, REM sleep and waking in wild-type (WT; black dots; n = 14) and Adk-tg mice (white dots; n = 11) during the 24-h baseline. Mean values of the parietal (PAR) and frontal ...
Figure 2
Representative 5-s raw parietal EEG and EMG traces during NREM sleep, REM sleep and waking in wild-type (WT), Adk+/− and Adk-tg mice.

Baseline sleep

Adk-tg mice slept approximately 60 min less than WT mice in the dark period (Table 1; Fig. 3 left). The duration of their waking and NREM sleep episodes was significantly longer than in WT mice, while their total episode number was lower, indicating more consolidated sleep in Adk-tg mice (Table 2). The relative number of longer NREM sleep episodes (>128 sec) was higher in Adk-tg mice compared to WT mice at the cost of shorter (36–64 sec) episodes (Fig. 4 left; p < 0.005, Tukey-Kramer test). Sleep fragmentation (defined as the occurrence of waking episodes <16 sec per hour of sleep), differed between the genotypes (Adk-tg, 20.1 ± 1.3; WT, 24.7 ± 1.5; p = 0.039, unpaired t-test).

Figure 3
Time course of the vigilance states NREM sleep, REM sleep, waking and REM sleep as percentage of total sleep time (REMS/TST) in WT and Adk-tg mice during the 24-h baseline (BL; left panels) and 18-h recovery (Rec.; right panels) after 6 h sleep deprivation ...
Figure 4
Left: Number of NREM sleep episodes in WT (n = 14) and Adk-tg mice (n = 11) during the 24-h baseline. Mean values are expressed as percentage of the total number of NREM sleep episodes. Numbers below the x-axis are duration range in seconds. Triangles: ...
Table 1
Amount of vigilance states
Table 2
Frequency and duration of vigilance state episodes

The amount of REM sleep as well as REM sleep per total sleep time was reduced in Adk-tg mice (Table 1; Fig. 3 left), due to a decreased REM sleep episode frequency (Table 2, p = 0.0041, unpaired t-test).

Motor activity data obtained in a separate group of Adk-tg and WT mice corroborated the sleep findings. Adk-tg mice spent more time active in the dark period (12-h dark: 328.8 ± 11.3 vs 268.5 ± 9.1 min, p = 0.0017, unpaired t-test). The intensity of their movements (activity counts per minute) was also significantly higher compared to WT mice (Fig. 4).

In summary, during baseline, sleep and EEG power in frequencies encompassing SWA in NREM sleep and theta activity in waking and REM sleep were reduced in Adk-tg mice.

Effects of enhanced sleep pressure by 6 h SD

During recovery after SD, Adk-tg and WT mice showed the typical compensatory increase of NREM sleep with a concomitant reduction in waking (Table 1; Fig. 3). However, the NREM sleep increase was significantly larger in the Adk-tg mice in hours 7–12 (Table 1). The REM sleep rebound was delayed to the second 6-h recovery interval in all mice (Table 1). Although the baseline genotype differences in REM sleep persisted during recovery (Fig. 3), the response to SD was similar in Adk-tg and WT mice (recovery-baseline, 6-h intervals 1 – 3: −1.9 ± 1.2 vs −1.5 ± 1.1 min, p = 0.81; 8.9 ± 1.5 vs 9.2 ± 1.4 min, p = 0.88; 1.8 ± 1.5 vs 3.2 ± 2.0 min, p = 0.56, unpaired t-test, in WT vs Adk-tg mice, respectively).

SWA and slow wave energy

We hypothesized that following SD, mice over-expressing ADK would display a reduced increase in SWA. Accordingly, sleep intensity was significantly lower in Adk-tg mice compared to WT mice (Fig. 5A). The magnitude of the typical SD-induced initial increase in SWA was not only lower by 9–12 % in Adk-tg mice (first 2-h interval: p < 0.01, unpaired t-test, for PAR and FRO EEG), but SWA remained lower in Adk-tg mice throughout the entire recovery period in both EEG derivations (Fig. 5A). While baseline levels were attained by all mice after 4–6 h recovery, only Adk-tg mice showed a subsequent negative rebound below baseline. The genotype SWA difference in NREM sleep was also evident at the transitions from waking to NREM sleep during the first 6-h recovery interval (Fig. 5B). The increase (i.e. build-up) of SWA in Adk-tg mice was significantly lower compared to WT mice immediately after the transitions and remained below the WT increase throughout the 2-min NREM sleep episodes. During the corresponding baseline interval, the dynamics of SWA build-up had been similar in the two genotypes (Fig. 5B). All genotype differences dissipated in the second 6-h recovery interval.

Figure 5
A: Time course of slow-wave activity (SWA) in NREM sleep in the parietal (PAR) and frontal (FRO) EEG in WT and Adk-tg mice during 18-h recovery after 6 h SD (grey area). Mean 2-h values are expressed as percentage of the mean 24-h baseline SWA in NREM ...

We computed slow-wave energy (SWE; 0.75–4 Hz) which takes into account the difference in the amount of NREM sleep between genotypes. Despite the larger amount of NREM sleep, Adk-tg mice reached a significantly lower SWE level after 18-h recovery compared to WT mice in the parietal EEG (82.1 ± 1.3 % vs 87.5 ± 1.8 %; p = 0.034, unpaired t-test). Significance was not attained in the frontal EEG due to larger inter-individual variability (82.2 ± 1.9 vs 87.9 ± 2.2 %; p = 0.072).

Spindle frequency activity (SFA)

Slow waves and spindles are EEG hallmarks of NREM sleep, the latter being most prominent at the transition from NREM sleep to REM sleep (Franken et al., 1994; Vyazovskiy et al., 2004). Spindle density and SWA in NREM sleep show inter-related dynamics in rats and humans (Aeschbach et al., 1997; Vyazovskiy et al., 2004). In addition to the reduced levels of SWA attained by Adk-tg mice, the typical surge of SFA immediately before the NREM-REM sleep transitions was significantly reduced in Adk-tg mice during baseline (Fig. 6) and recovery (not shown) compared to WT mice.

Figure 6
Time course of specific frequency bands during the 2 min before and after transitions from NREM to REM sleep in WT and Adk-tg mice. Spindle frequency activity (SFA; 10.25–15 Hz) in the frontal EEG (left panel) and theta frequency activity (7.25–11 ...

Overall, Adk-tg mice showed a reduced homeostatic response to enhanced sleep pressure compared to WT mice.

Theta peak frequency in REM sleep was lower in Adk-tg mice

Oscillations in the theta frequency range are a distinct feature of REM sleep in rodents (Figs. 1 and and2;2; Huber et al., 2000). REM sleep EEG spectra were significantly lower in the theta frequency range in Adk-tg mice compared to WT mice. Based on the significant bins in Fig. 1 (7.25–11 Hz) we investigated whether EEG power would be affected at the transitions from NREM to REM sleep. Adk-tg mice lacked the typical, prominent increase in theta activity after the transitions to REM sleep (Fig. 6). Their theta activity levels were significantly lower than those of WT mice 20 s before the transitions, and remained at a consistently low level after the transitions to REM sleep (the result was similar when the time course of the traditional theta frequency band, 6.25–9 Hz, was computed).

Adenosine is directly linked to energy metabolism in the cell (Scharf et al., 2008) and the peak frequency of theta in REM sleep was affected by a deficiency in energy metabolism (Tafti et al., 2003). We determined the peak frequency within the theta frequency band for individual mice. In Adk-tg mice the peak was at a significantly lower frequency than in WT mice (7 ± 0.1 Hz vs 8 ± 0.1 Hz, p < 0.0001, unpaired t-test).

Effects of Adk over-expression on EEG power during induced wakefulness

To investigate whether the genotype differences in the response to SD were due to behavioral differences during SD, we analyzed the corresponding EEG spectra (Fig. 7). The typical increase in EEG power in the course of SD was observed for a large frequency range (0.75–8 Hz) in both genotypes and in higher frequencies (13–15 Hz; p < 0.05, paired t-test) in Adk-tg mice. The rise rate of EEG power (interval 3 as % of interval 1) differed between the two genotypes in frequencies within the SWA band, theta activity and SFA (Fig. 7). The increase in the 2.25–3.5 Hz band was apparent in all mice (interval 1 vs 3: p < 0.0005), but the rise rate was significantly lower in the Adk-tg mice (p = 0.0487, Kruskal-Wallis test). Theta activity was significantly higher in Adk-tg mice than in WT mice throughout SD (rise rate, p = 0.0004, unpaired t-test). SFA increased in the course of SD in Adk-tg mice only (p = 0.044, paired t-test), but was significantly lower during the first two 2-h intervals compared to WT mice (rise rate, p = 0.0455, unpaired t-test).

Figure 7
Time course of EEG power in waking in three frequency bands in the parietal EEG of WT and Adk-tg mice during the 6 h SD (grey area) and first 6-h recovery. Mean 2-h values are expressed as percentage of the mean 24-h baseline power in the corresponding ...

Reduced expression of Adk has no major effect on sleep

Our prior data suggest that adenosine metabolism is not significantly altered in Adk+/− mice (normal AMP, ADP, ATP, SAM, SAH levels and normal physiology; Boison et al., 2002). We therefore hypothesized that sleep would not be altered under conditions of reduced ADK, but normal adenosine metabolites. The amount of sleep (Table 1; Fig. 8), as well as the average duration and frequency of vigilance state episodes (data not shown), did not differ between Adk+/− and WT mice. Moreover, the 24-h baseline time course of SWA in NREM sleep was similar in the two genotypes (Fig. 8). When sleep pressure was enhanced by SD, the typical increase in NREM sleep, REM sleep and SWA in NREM sleep was evident, but none of these variables differed significantly between Adk+/− and WT mice (Table 1). We conclude that the changes in ADK expression per se in the absence of any changes in adenosine metabolites do not affect sleep.

Figure 8
Time course of NREM sleep (NREMS), REM sleep (REMS) and slow-wave activity (SWA; EEG power between 0.75–4 Hz) in NREM sleep in the parietal (PAR) and frontal (FRO) EEG in WT (black dots; n = 14) and Adk+/− mice (white dots; n = 15) during ...

ADK protein expression level

To validate our findings, we retrospectively quantified ADK protein expression levels in brain extracts from representative experimental animals of each genotype after conclusion of the experiments (n=6 per genotype). Adk-tg mutants expressed significantly higher ADK levels in the brain (142 % vs. WT mice, p < 0.001, unpaired t-test) of the short ADK isoform (Fig. 9), while ADK expression levels of the short and long ADK isoforms in Adk+/− mice were significantly lower (65 % of WT mice, p < 0.001, unpaired t-test).

Figure 9
Top: Western blot of ADK protein levels from the right hemisphere of adult WT, Adk+/− and Adk-tg mice (numbers 1–6 correspond to individual mice). Bottom: Corresponding quantitative analysis of the ADK levels of n = 6 mice per genotype ...

Discussion

Adenosine levels in adult brain are largely regulated by an astrocyte-based adenosine cycle (Boison, 2008). Astrocytes can release adenosine precursor ATP either by vesicular transport (Pascual et al., 2005) or via hemichannels (Kang et al., 2008). Once outside the cell ATP is rapidly degraded by a cascade of ectonucleotidases (Zimmermann, 2000). In contrast to conventional neurotransmitters, there is no regulated transporter-mediated reuptake system for adenosine. On the contrary, astrocytes express two types of equilibrative nucleoside transporters (Baldwin et al., 2004). Inside the cell adenosine is removed by phosphorylation into AMP by ADK. Thus, ADK drives the influx of adenosine into the astrocytes and regulates the abundance of extracellular adenosine.

Role of ADK in sleep homeostasis

In this study we show for the first time that increasing ADK, the key enzyme controlling adenosine levels, robustly altered vigilance states and specific EEG oscillations which are the hallmarks of NREM sleep, REM sleep and wakefulness. In particular, the increases in brain ADK protein and ADK activity in Adk-tg mice resulted in a remarkable reduction of EEG power below 4 Hz and in the theta frequency range (Fig. 1), pointing to a deficiency in mechanisms generating EEG slow waves and theta oscillations. Moreover, ADK over-expression led to a reduction of sleep, which was partially compensated for by an increase in sleep consolidation. Under enhanced sleep pressure a deficiency in the homeostatic regulation of sleep became apparent in Adk-tg mice. The increase of SWA in NREM sleep and its build-up at the waking-NREM sleep transitions were lower in Adk-tg mice (Fig. 5), indicating a reduced capacity to intensify sleep. This finding is remarkable since Adk-tg mice had slept less than WT mice during the dark period preceding SD. Therefore, they were under higher sleep pressure at the beginning of SD. The deficiency in sleep regulatory mechanisms was further demonstrated by the lower SWE accumulated after 18 h recovery in Adk-tg mice.

ADK is highly conserved in evolution and ubiquitously expressed (Boison, 2006). Its importance for survival is emphasized by the lethality of the disruption of the Adk gene (Boison et al., 2002). Two different isoforms of ADK exist (Cui et al., 2009), a cytoplasmic isoform thought to regulate ambient levels of adenosine, while a nuclear isoform might be implicated in epigenetic mechanisms. This dual functionality of the enzyme has been postulated previously (Studer et al., 2006). Heterozygous Adk mice have no obvious phenotype (Boison et al., 2002), which we here confirmed for sleep and sleep regulation. We reported earlier that Adk+/− mice display alterations in ADK expression in the absence of changes in adenosine metabolism. These findings suggest that the nuclear isoform of ADK is likely not involved in sleep regulation. In line with this notion, Adk-tg mice that selectively over-express the cytoplasmic isoform of ADK, but lack the nuclear isoform of ADK show a profound sleep phenotype. Together these data suggest that the cytoplasmic, but not the nuclear isoform of ADK is involved in sleep regulation.

Sleep deprivation enhanced extracellular adenosine which thereafter decreased during recovery sleep (Porkka-Heiskanen et al., 1997; Porkka-Heiskanen et al., 2000; Basheer et al., 2004). Benington and Heller (1995) proposed that adenosine continues to be released during recovery from SD as a product of a putative restorative process occurring during sleep. The astrocytic and neuronal ADK over-expression in Adk-tg mice should lead to a faster adenosine clearance from the extracellular space resulting in a slower accumulation of sleep need during prolonged wakefulness, and in a reduced response to SD. Thus, we had hypothesized that the overall reduction in brain adenosine levels in Adk-tg mice might not only modify sleep under baseline conditions but also attenuate the rise in sleep pressure during SD, and impair sleep restorative processes during recovery.

The deficiency in the Adk-tg mice was not restricted to EEG slow waves, but also comprised spindle activity in NREM sleep and theta activity in waking and REM sleep. Spindle and slow-wave oscillations originate in the thalamocortical network and depend on the hyperpolarization of thalamocortical neurons (Steriade et al., 1993). Notably, Adk-tg mice showed both reduced SWA levels in NREM sleep and a 24% lower surge of SFA at NREM-REM sleep transitions compared to WT mice (Fig. 6). During SD-induced wakefulness, Adk-tg mice showed a larger increase of SFA and a smaller increase of SWA (Fig. 7). Adenosine A1 receptors, which are predominantly expressed in the cortex and in several thalamic nuclei, including the thalamic reticular nucleus (Rosin et al., 1998; Ochiishi et al., 1999), may mediate these genotype differences. Robust, but opposing pre- and postsynaptic effects of adenosine via A1 receptors on thalamocortical oscillations were reported in ex-vivo thalamic slices (Pape, 1992; Ulrich and Huguenard, 1995; Fontanez and Porter, 2006). Our results are in agreement with a potent role of adenosine in mediating the hyperpolarization of thalamocortical neurons, which de-inactivates the low-threshold calcium channels in thalamic relay nuclei underlying the occurrence of slow waves (reviewed in McCormick, 1992). The reduced adenosinergic tone in Adk-tg mice (Fedele et al., 2005) may also alter ATP-regulated phosphorylation of T-type Ca2+ channels (Leresche et al., 2004). In addition, adenosine modulates the release of several neurotransmitters (Fredholm et al., 2005), including cholinergic, monoaminergic and histaminergic inputs to the thalamocortical network which, thereby, may alter the firing pattern of thalamocortical neurons.

The waking EEG is affected by Adk over-expression

Both the amount and quality of wakefulness were affected by ADK over-expression (Table 1). Adk-tg mice were more active during their waking bouts compared to WT mice (Fig. 4), confirming previous results (Fedele et al., 2005; Yee et al., 2007). These findings mimic the stimulant properties of the adenosine receptor antagonist caffeine (Schwierin et al., 1996; Fredholm et al., 2005; Huang et al., 2005). It is likely that the increase in wakefulness and corresponding reduction in sleep in Adk-tg mice is a consequence of reduced adenosinergic tone throughout the brain, leading to increased activity of the ascending monoaminergic system on the thalamus.

It is well-established from human and animal studies that EEG power in the low frequencies increases in the course of prolonged wakefulness (e.g. Torsvall and Akerstedt, 1987; Franken et al., 1993; Cajochen et al., 1995; Aeschbach et al., 1999; e.g. Achermann and Borbely, 2003; Vyazovskiy and Tobler, 2005). We found a robust increase in EEG power of frequencies below 8 Hz in both genotypes in the course of SD. Interestingly, initial levels and rise rates of power in specific frequencies were different in Adk-tg mice (Fig. 7). These differences may result from the behavioral alterations mentioned above. The mechanisms underlying these alterations remain unclear.

Role of Adk in REM sleep

Strikingly, Adk-tg mice showed an overall reduction of REM sleep (Table 1). Several lines of evidence support a role of the adenosinergic system in the onset and maintenance of REM sleep. Microinjections of the A1 receptor agonist cyclopentyladenosine in the pontine reticular formation, region implicated in the modulation of REM sleep (Monti and Monti, 2000), enhanced REM sleep duration in rats (Marks and Birabil, 2000; Marks et al., 2003). This effect was counteracted by pretreatment with the adenosine receptor antagonist cyclopentyltheophylline (Marks and Birabil, 1998). Consistently, microinjections of an A2A receptor agonist in the same region also enhanced REM sleep and increased acetylcholine release in rodents (Marks et al., 2003; Coleman et al., 2006). Initiation of REM sleep may be deficient in Adk-tg mice due to alterations in glutamatergic excitation and GABAergic inhibition of REM sleep active neurons in the sublaterodorsal nucleus (Luppi et al., 2006). Once REM sleep is initiated, it is maintained normally as indicated by the similar duration of REM sleep episodes in WT and Adk-tg mice. Under increased sleep pressure the REM sleep rebound was similar in the two genotypes, indicating that basic REM sleep regulatory mechanisms are intact in Adk-tg mice.

In the present study, reduced theta activity after transitions to REM sleep in Adk-tg mice implicates a role for adenosine in the modulation of theta oscillations. Moreover, theta peak frequency in REM sleep was significantly lower in Adk-tg mice compared to WT mice, while waking theta peak frequency was similar in the two genotypes (p = 0.18, unpaired t-test). Changes in adenosinergic tone due to ADK over-expression may modulate the frequency and amplitude of theta oscillations during REM sleep by changing the excitatory input from the brainstem reticular formation to the septum (Vinogradova, 1995).

Conclusions

Our data suggest that ADK over-expression profoundly alters the sleep phenotype, likely by lowering adenosine tone in the brain. It is tempting to speculate that ADK, by controlling adenosine levels and its receptor mediated effects, may tune and integrate other neurotransmitter pathways involved in sleep-wake regulation and the generation of related EEG oscillations. Overall, our results are in accordance with a global rather than a region-specific role of adenosine in sleep regulation (Heller, 2006).

Acknowledgements

We thank Dr. P. Achermann for advice in data analysis and Drs. A. Borbély, V. Vyazovskiy and L. Tarokh for critical comments on the manuscript. This work was supported by the Swiss National Science Foundation grant nr. 3100A0-112528/1, EU LSHM-CT-2005-518189, EU MCRTN-CT-2004-512362, and the National Institutes of Mental Health (R01-MH83973).

References

  • Achermann P, Borbely AA. Mathematical models of sleep regulation. Front Biosci. 2003;8:s683–693. [PubMed]
  • Aeschbach D, Dijk DJ, Borbely AA. Dynamics of EEG spindle frequency activity during extended sleep in humans: relationship to slow-wave activity and time of day. Brain Res. 1997;748:131–136. [PubMed]
  • Aeschbach D, Matthews JR, Postolache TT, Jackson MA, Giesen HA, Wehr TA. Two circadian rhythms in the human electroencephalogram during wakefulness. Am J Physiol. 1999;277:R1771–1779. [PubMed]
  • Alanko L, Heiskanen S, Stenberg D, Porkka-Heiskanen T. Adenosine kinase and 5'-nucleotidase activity after prolonged wakefulness in the cortex and the basal forebrain of rat. Neurochem Int. 2003;42:449–454. [PubMed]
  • Alanko LO, Laitinen JT, Stenberg D, Porkka-Heiskanen T. Adenosine A1 receptor-dependent G-protein activity in the rat brain during prolonged wakefulness. Neuroreport. 2004;15:2133–2137. [PubMed]
  • Baldwin SA, Beal PR, Yao SY, King AE, Cass CE, Young JD. The equilibrative nucleoside transporter family, SLC29. Pflugers Arch. 2004;447:735–743. [PubMed]
  • Basheer R, Strecker RE, Thakkar MM, McCarley RW. Adenosine and sleep-wake regulation. Prog Neurobiol. 2004;73:379–396. [PubMed]
  • Basheer R, Halldner L, Alanko L, McCarley RW, Fredholm BB, Porkka-Heiskanen T. Opposite changes in adenosine A1 and A2A receptor mRNA in the rat following sleep deprivation. Neuroreport. 2001;12:1577–1580. [PubMed]
  • Benington JH, Heller HC. Restoration of brain energy metabolism as the function of sleep. Prog Neurobiol. 1995;45:347–360. [PubMed]
  • Benington JH, Kodali SK, Heller HC. Stimulation of A1 adenosine receptors mimics the electroencephalographic effects of sleep deprivation. Brain Res. 1995;692:79–85. [PubMed]
  • Biaggioni I, Paul S, Puckett A, Arzubiaga C. Caffeine and theophylline as adenosine receptor antagonists in humans. J Pharmacol Exp Ther. 1991;258:588–593. [PubMed]
  • Bjorness TE, Kelly CL, Gao T, Poffenberger V, Greene RW. Control and function of the homeostatic sleep response by adenosine A1 receptors. J Neurosci. 2009;29:1267–1276. [PMC free article] [PubMed]
  • Boison D. Adenosine kinase, epilepsy and stroke: mechanisms and therapies. Trends Pharmacol Sci. 2006;27:652–658. [PubMed]
  • Boison D. Adenosine as a neuromodulator in neurological diseases. Curr Opin Pharmacol. 2008;8:2–7. [PMC free article] [PubMed]
  • Boison D, Scheurer L, Zumsteg V, Rulicke T, Litynski P, Fowler B, Brandner S, Mohler H. Neonatal hepatic steatosis by disruption of the adenosine kinase gene. Proc Natl Acad Sci U S A. 2002;99:6985–6990. [PubMed]
  • Cajochen C, Brunner DP, Krauchi K, Graw P, Wirz-Justice A. Power density in theta/alpha frequencies of the waking EEG progressively increases during sustained wakefulness. Sleep. 1995;18:890–894. [PubMed]
  • Chagoya de Sanchez V, Hernandez Munoz R, Suarez J, Vidrio S, Yanez L, Diaz Munoz M. Day-night variations of adenosine and its metabolizing enzymes in the brain cortex of the rat--possible physiological significance for the energetic homeostasis and the sleep-wake cycle. Brain Res. 1993;612:115–121. [PubMed]
  • Coleman CG, Baghdoyan HA, Lydic R. Dialysis delivery of an adenosine A2A agonist into the pontine reticular formation of C57BL/6J mouse increases pontine acetylcholine release and sleep. J Neurochem. 2006;96:1750–1759. [PubMed]
  • Cui XA, Singh B, Park J, Gupta RS. Subcellular localization of adenosine kinase in mammalian cells: The long isoform of AdK is localized in the nucleus. Biochem Biophys Res Commun. 2009;388:46–50. [PubMed]
  • Datta S, Maclean RR. Neurobiological mechanisms for the regulation of mammalian sleep-wake behavior: reinterpretation of historical evidence and inclusion of contemporary cellular and molecular evidence. Neurosci Biobehav Rev. 2007;31:775–824. [PMC free article] [PubMed]
  • Deboer T, Franken P, Tobler I. Sleep and cortical temperature in the Djungarian hamster under baseline conditions and after sleep deprivation. J Comp Physiol A. 1994;174:145–155. [PubMed]
  • Fedele DE, Gouder N, Guttinger M, Gabernet L, Scheurer L, Rulicke T, Crestani F, Boison D. Astrogliosis in epilepsy leads to overexpression of adenosine kinase, resulting in seizure aggravation. Brain. 2005;128:2383–2395. [PubMed]
  • Fontanez DE, Porter JT. Adenosine A1 receptors decrease thalamic excitation of inhibitory and excitatory neurons in the barrel cortex. Neuroscience. 2006;137:1177–1184. [PubMed]
  • Franken P, Tobler I, Borbely AA. Effects of 12-h sleep deprivation and of 12-h cold exposure on sleep regulation and cortical temperature in the rat. Physiol Behav. 1993;54:885–894. [PubMed]
  • Franken P, Dijk DJ, Tobler I, Borbely AA. High-frequency components of the rat electrocorticogram are modulated by the vigilance states. Neurosci Lett. 1994;167:89–92. [PubMed]
  • Fredholm BB, Chen JF, Masino SA, Vaugeois JM. Actions of adenosine at its receptors in the CNS: insights from knockouts and drugs. Annu Rev Pharmacol Toxicol. 2005;45:385–412. [PubMed]
  • Gallopin T, Luppi PH, Cauli B, Urade Y, Rossier J, Hayaishi O, Lambolez B, Fort P. The endogenous somnogen adenosine excites a subset of sleep-promoting neurons via A2A receptors in the ventrolateral preoptic nucleus. Neuroscience. 2005;134:1377–1390. [PubMed]
  • Gouder N, Scheurer L, Fritschy JM, Boison D. Overexpression of adenosine kinase in epileptic hippocampus contributes to epileptogenesis. J Neurosci. 2004;24:692–701. [PubMed]
  • Halassa MM, Florian C, Fellin T, Munoz JR, Lee SY, Abel T, Haydon PG, Frank MG. Astrocytic modulation of sleep homeostasis and cognitive consequences of sleep loss. Neuron. 2009;61:213–219. [PMC free article] [PubMed]
  • Haulica I, Ababei L, Branisteanu D, Topoliceanu F. Letter: Preliminary data on the possible hypnogenic role of adenosine. J Neurochem. 1973;21:1019–1020. [PubMed]
  • Hayaishi O, Urade Y, Eguchi N, Huang ZL. Genes for prostaglandin d synthase and receptor as well as adenosine A2A receptor are involved in the homeostatic regulation of nrem sleep. Arch Ital Biol. 2004;142:533–539. [PubMed]
  • Heller HC. A global rather than local role for adenosine in sleep homeostasis. Sleep. 2006;29:1382–1383. discussion 1387–1389. [PubMed]
  • Huang ZL, Urade Y, Hayaishi O. Prostaglandins and adenosine in the regulation of sleep and wakefulness. Curr Opin Pharmacol. 2007;7:33–38. [PubMed]
  • Huang ZL, Qu WM, Eguchi N, Chen JF, Schwarzschild MA, Fredholm BB, Urade Y, Hayaishi O. Adenosine A2A, but not A1, receptors mediate the arousal effect of caffeine. Nat Neurosci. 2005;8:858–859. [PubMed]
  • Huber A, Padrun V, Deglon N, Aebischer P, Mohler H, Boison D. Grafts of adenosine-releasing cells suppress seizures in kindling epilepsy. Proc Natl Acad Sci U S A. 2001;98:7611–7616. [PubMed]
  • Huber R, Deboer T, Tobler I. Topography of EEG dynamics after sleep deprivation in mice. J Neurophysiol. 2000;84:1888–1893. [PubMed]
  • Kang J, Kang N, Lovatt D, Torres A, Zhao Z, Lin J, Nedergaard M. Connexin 43 hemichannels are permeable to ATP. J Neurosci. 2008;28:4702–4711. [PubMed]
  • Landolt HP, Werth E, Borbely AA, Dijk DJ. Caffeine intake (200 mg) in the morning affects human sleep and EEG power spectra at night. Brain Res. 1995;675:67–74. [PubMed]
  • Leresche N, Hering J, Lambert RC. Paradoxical potentiation of neuronal T-type Ca2+ current by ATP at resting membrane potential. J Neurosci. 2004;24:5592–5602. [PubMed]
  • Li T, Lan JQ, Boison D. Uncoupling of astrogliosis from epileptogenesis in adenosine kinase (ADK) transgenic mice. Neuron Glia Biol. 2008a;4:91–99. [PMC free article] [PubMed]
  • Li T, Quan Lan J, Fredholm BB, Simon RP, Boison D. Adenosine dysfunction in astrogliosis: cause for seizure generation? Neuron Glia Biol. 2007;3:353–366. [PMC free article] [PubMed]
  • Li T, Ren G, Lusardi T, Wilz A, Lan JQ, Iwasato T, Itohara S, Simon RP, Boison D. Adenosine kinase is a target for the prediction and prevention of epileptogenesis in mice. J Clin Invest. 2008b;118:571–582. [PMC free article] [PubMed]
  • Lloyd HG, Fredholm BB. Involvement of adenosine deaminase and adenosine kinase in regulating extracellular adenosine concentration in rat hippocampal slices. Neurochem Int. 1995;26:387–395. [PubMed]
  • Luppi PH, Gervasoni D, Verret L, Goutagny R, Peyron C, Salvert D, Leger L, Fort P. Paradoxical (REM) sleep genesis: the switch from an aminergic-cholinergic to a GABAergic-glutamatergic hypothesis. J Physiol Paris. 2006;100:271–283. [PubMed]
  • Mackiewicz M, Nikonova EV, Zimmerman JE, Galante RJ, Zhang L, Cater JR, Geiger JD, Pack AI. Enzymes of adenosine metabolism in the brain: diurnal rhythm and the effect of sleep deprivation. J Neurochem. 2003;85:348–357. [PubMed]
  • Marks GA, Birabil CG. Enhancement of rapid eye movement sleep in the rat by cholinergic and adenosinergic agonists infused into the pontine reticular formation. Neuroscience. 1998;86:29–37. [PubMed]
  • Marks GA, Birabil CG. Infusion of adenylyl cyclase inhibitor SQ22,536 into the medial pontine reticular formation of rats enhances rapid eye movement sleep. Neuroscience. 2000;98:311–315. [PubMed]
  • Marks GA, Shaffery JP, Speciale SG, Birabil CG. Enhancement of rapid eye movement sleep in the rat by actions at A1 and A2a adenosine receptor subtypes with a differential sensitivity to atropine. Neuroscience. 2003;116:913–920. [PubMed]
  • Marley E, Nistico G. Effects of catecholamines and adenosine derivatives given into the brain of fowls. Br J Pharmacol. 1972;46:619–636. [PubMed]
  • Martin JV, Berman KF, Skolnick P, Mendelson WB. Behavioral and electroencephalographic effects of the adenosine1 agonist, L-PIA. Pharmacol Biochem Behav. 1989;34:507–510. [PubMed]
  • Mathews, Erion MD, Ealick SE. Structure of human adenosine kinase at 1.5 A resolution. Biochemistry. 1998;37:15607–15620. [PubMed]
  • McCormick DA. Neurotransmitter actions in the thalamus and cerebral cortex and their role in neuromodulation of thalamocortical activity. Prog Neurobiol. 1992;39:337–388. [PubMed]
  • Methippara MM, Kumar S, Alam MN, Szymusiak R, McGinty D. Effects on sleep of microdialysis of adenosine A1 and A2a receptor analogs into the lateral preoptic area of rats. Am J Physiol Regul Integr Comp Physiol. 2005;289:R1715–1723. [PubMed]
  • Monti JM, Monti D. Role of dorsal raphe nucleus serotonin 5-HT1A receptor in the regulation of REM sleep. Life Sci. 2000;66:1999–2012. [PubMed]
  • Ochiishi T, Chen L, Yukawa A, Saitoh Y, Sekino Y, Arai T, Nakata H, Miyamoto H. Cellular localization of adenosine A1 receptors in rat forebrain: immunohistochemical analysis using adenosine A1 receptor-specific monoclonal antibody. J Comp Neurol. 1999;411:301–316. [PubMed]
  • Oishi Y, Huang ZL, Fredholm BB, Urade Y, Hayaishi O. Adenosine in the tuberomammillary nucleus inhibits the histaminergic system via A1 receptors and promotes non-rapid eye movement sleep. Proc Natl Acad Sci U S A. 2008;105:19992–19997. [PubMed]
  • Pak MA, Haas HL, Decking UK, Schrader J. Inhibition of adenosine kinase increases endogenous adenosine and depresses neuronal activity in hippocampal slices. Neuropharmacology. 1994;33:1049–1053. [PubMed]
  • Pape HC. Adenosine promotes burst activity in guinea-pig geniculocortical neurones through two different ionic mechanisms. J Physiol. 1992;447:729–753. [PubMed]
  • Pascual O, Casper KB, Kubera C, Zhang J, Revilla-Sanchez R, Sul JY, Takano H, Moss SJ, McCarthy K, Haydon PG. Astrocytic purinergic signaling coordinates synaptic networks. Science. 2005;310:113–116. [PubMed]
  • Pignataro G, Simon RP, Boison D. Transgenic overexpression of adenosine kinase aggravates cell death in ischemia. J Cereb Blood Flow Metab. 2007;27:1–5. [PubMed]
  • Porkka-Heiskanen T, Strecker RE, McCarley RW. Brain site-specificity of extracellular adenosine concentration changes during sleep deprivation and spontaneous sleep: an in vivo microdialysis study. Neuroscience. 2000;99:507–517. [PubMed]
  • Porkka-Heiskanen T, Strecker RE, Thakkar M, Bjorkum AA, Greene RW, McCarley RW. Adenosine: a mediator of the sleep-inducing effects of prolonged wakefulness. Science. 1997;276:1265–1268. [PubMed]
  • Portas CM, Thakkar M, Rainnie DG, Greene RW, McCarley RW. Role of adenosine in behavioral state modulation: a microdialysis study in the freely moving cat. Neuroscience. 1997;79:225–235. [PubMed]
  • Radulovacki M. Role of adenosine in sleep in rats. Rev Clin Basic Pharm. 1985;5:327–339. [PubMed]
  • Radulovacki M, Virus RM, Rapoza D, Crane RA. A comparison of the dose response effects of pyrimidine ribonucleosides and adenosine on sleep in rats. Psychopharmacology (Berl) 1985;87:136–140. [PubMed]
  • Rosin DL, Robeva A, Woodard RL, Guyenet PG, Linden J. Immunohistochemical localization of adenosine A2A receptors in the rat central nervous system. J Comp Neurol. 1998;401:163–186. [PubMed]
  • Sarda N, Gharib A, Coindet J, Pacheco H, Jouvet M. Possible involvement of the S-adenosyl-L-homocysteine metabolites, adenosine and L-homocysteine, in sleep in rats. Neurosci Lett. 1986;66:287–292. [PubMed]
  • Satoh S, Matsumura H, Hayaishi O. Involvement of adenosine A2A receptor in sleep promotion. Eur J Pharmacol. 1998;351:155–162. [PubMed]
  • Satoh S, Matsumura H, Koike N, Tokunaga Y, Maeda T, Hayaishi O. Region-dependent difference in the sleep-promoting potency of an adenosine A2A receptor agonist. Eur J Neurosci. 1999;11:1587–1597. [PubMed]
  • Scammell TE, Gerashchenko DY, Mochizuki T, McCarthy MT, Estabrooke IV, Sears CA, Saper CB, Urade Y, Hayaishi O. An adenosine A2a agonist increases sleep and induces Fos in ventrolateral preoptic neurons. Neuroscience. 2001;107:653–663. [PubMed]
  • Scharf MT, Naidoo N, Zimmerman JE, Pack AI. The energy hypothesis of sleep revisited. Prog Neurobiol. 2008;86:264–280. [PMC free article] [PubMed]
  • Schwierin B, Borbely AA, Tobler I. Effects of N6-cyclopentyladenosine and caffeine on sleep regulation in the rat. Eur J Pharmacol. 1996;300:163–171. [PubMed]
  • Snyder SH, Katims JJ, Annau Z, Bruns RF, Daly JW. Adenosine receptors and behavioral actions of methylxanthines. Proc Natl Acad Sci U S A. 1981;78:3260–3264. [PubMed]
  • Steriade M, McCormick DA, Sejnowski TJ. Thalamocortical oscillations in the sleeping and aroused brain. Science. 1993;262:679–685. [PubMed]
  • Strecker RE, Morairty S, Thakkar MM, Porkka-Heiskanen T, Basheer R, Dauphin LJ, Rainnie DG, Portas CM, Greene RW, McCarley RW. Adenosinergic modulation of basal forebrain and preoptic/anterior hypothalamic neuronal activity in the control of behavioral state. Behav Brain Res. 2000;115:183–204. [PubMed]
  • Studer FE, Fedele DE, Marowsky A, Schwerdel C, Wernli K, Vogt K, Fritschy JM, Boison D. Shift of adenosine kinase expression from neurons to astrocytes during postnatal development suggests dual functionality of the enzyme. Neuroscience. 2006;142:125–137. [PubMed]
  • Tafti M, Petit B, Chollet D, Neidhart E, de Bilbao F, Kiss JZ, Wood PA, Franken P. Deficiency in short-chain fatty acid beta-oxidation affects theta oscillations during sleep. Nat Genet. 2003;34:320–325. [PubMed]
  • Thakkar MM, Winston S, McCarley RW. A1 receptor and adenosinergic homeostatic regulation of sleep-wakefulness: effects of antisense to the A1 receptor in the cholinergic basal forebrain. J Neurosci. 2003;23:4278–4287. [PMC free article] [PubMed]
  • Ticho SR, Radulovacki M. Role of adenosine in sleep and temperature regulation in the preoptic area of rats. Pharmacol Biochem Behav. 1991;40:33–40. [PubMed]
  • Tobler I, Deboer T, Fischer M. Sleep and sleep regulation in normal and prion protein-deficient mice. J Neurosci. 1997;17:1869–1879. [PubMed]
  • Torsvall L, Akerstedt T. Sleepiness on the job: continuously measured EEG changes in train drivers. Electroencephalogr Clin Neurophysiol. 1987;66:502–511. [PubMed]
  • Ulrich D, Huguenard JR. Purinergic inhibition of GABA and glutamate release in the thalamus: implications for thalamic network activity. Neuron. 1995;15:909–918. [PubMed]
  • Vinogradova OS. Expression, control, and probable functional significance of the neuronal theta-rhythm. Prog Neurobiol. 1995;45:523–583. [PubMed]
  • Virus RM, Djuricic-Nedelson M, Radulovacki M, Green RD. The effects of adenosine and 2'-deoxycoformycin on sleep and wakefulness in rats. Neuropharmacology. 1983;22:1401–1404. [PubMed]
  • Vyazovskiy VV, Tobler I. Theta activity in the waking EEG is a marker of sleep propensity in the rat. Brain Res. 2005;1050:64–71. [PubMed]
  • Vyazovskiy VV, Achermann P, Borbely AA, Tobler I. The dynamics of spindles and EEG slow-wave activity in NREM sleep in mice. Arch Ital Biol. 2004;142:511–523. [PubMed]
  • White TD. Potentiation of excitatory amino acid-evoked adenosine release from rat cortex by inhibitors of adenosine kinase and adenosine deaminase and by acadesine. Eur J Pharmacol. 1996;303:27–38. [PubMed]
  • Yanik G, Glaum S, Radulovacki M. The dose-response effects of caffeine on sleep in rats. Brain Res. 1987;403:177–180. [PubMed]
  • Yee BK, Singer P, Chen JF, Feldon J, Boison D. Transgenic overexpression of adenosine kinase in brain leads to multiple learning impairments and altered sensitivity to psychomimetic drugs. Eur J Neurosci. 2007;26:3237–3252. [PubMed]
  • Zimmermann H. Extracellular metabolism of ATP and other nucleotides. Naunyn Schmiedebergs Arch Pharmacol. 2000;362:299–309. [PubMed]