• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of sleepGo To SLEEPAuthorsCurrent IssueSubscribeGo To SLEEP
Sleep. Feb 1, 2008; 31(2): 259–270.
PMCID: PMC2225576

Sleep-Stabilizing Effects of E-6199, Compared to Zopiclone, Zolpidem and THIP in Mice

Chloé Alexandre, PhD,1,2 Alberto Dordal, BSc,3 Ramon Aixendri, BSc,3 Antonio Guzman, BSc,3 Michel Hamon, PhD,1,2 and Joëlle Adrien, PhD1,2


Gamma aminobutyric acid (GABA)A receptor modulators constitute the majority of clinically used sedative-hypnotics. These compounds have the capacity to initiate and maintain sleep, but decrease REM sleep and delta activity within NREM sleep. In order to avoid such sleep adverse effects, the development of novel compounds remains of interest.

Study objectives:

The present study aimed at characterizing the acute effects of a novel putative hypnotic compound, E-6199, compared to zopiclone, zolpidem, and THIP on sleep-wakefulness patterns in mice. We also investigated whether repeated administration (daily injection during 10 days) of E-6199 was associated with tolerance and sleep disturbances at cessation of treatment.

Measurements and Results:

Polygraphic recordings were performed during 8 h after acute treatment with the various compounds. Under such conditions, E-6199 (5–20 mg/kg i.p.), zopiclone and zolpidem (2–10 mg/kg i.p.), but not THIP (2–10 mg/kg i.p.), exerted a marked sleep-promoting effect. Furthermore, E-6199 specifically increased the duration of NREM and markedly improved sleep continuity by lengthening NREM sleep episodes and reducing short awakenings and microarousal frequency. It also intensified NREM sleep by enhancing the slow wave activity within NREM at wake-NREM transitions. These effects were sustained and became even larger during chronic administration. Finally, abrupt E-6199 withdrawal did not elicit negative sleep effects.


Our findings demonstrate that E-6199 may be an effective hypnotic compound that promotes and improves NREMS, without producing EEG side effects, tolerance or withdrawal phenomena, when administered under chronic conditions.


Alexandre C; Dordal A; Aixendri R; Guzman A; Hamon M; Adrien J. Sleep-stabilizing effects of E-6199, compared to zopiclone, zolpidem and THIP in mice. SLEEP 2008;31(2):259–270.

Keywords: EEG, hypnotic, slow wave activity, sleep, chronic treatment, mice, zolpidem, THIP

INSOMNIA IS THE MOST COMMON OF ALL SLEEP DISORDERS AND CORRESPONDS TO A SUBJECTIVE COMPLAINT OF POOR SLEEP QUALITY OR INADEQUATE quantity of sleep. Insomniacs may present with one or more symptoms including difficulties in initiating and maintaining sleep, frequent nocturnal arousals or early morning awakening with inability to get back to sleep, and non-restorative sleep.1,2

Gamma amino butyric acid (GABA)A receptor modulators such as zolpidem and zopiclone are the most widely prescribed hypnotics. Their actions on sleep are well-established and include a decrease in sleep latency and awakenings frequency, associated with an increase in total sleep time.3,4 However, they may alter the physiological sleep architecture by reducing REM sleep amounts and delta wave amplitude during slow wave sleep, and/or enhancing the high frequencies activity in the electroencephalogram (EEG).3,5,6 Although zolpidem and zopiclone were originally reported to be devoid of benzodiazepine-like side-effects, subsequent studies showed that they may produce impaired daytime psychomotor performance, pharmacological tolerance, withdrawal symptoms and be endowed with abuse potential.2,4,68

Currently in clinical trials for the treatment of sleep disorders, THIP (4,5,6,7-tetrahydroisoxazolo [5,4-c] pyridin-3-ol), also named gaboxadol, is a selective extrasynaptic GABAA receptor agonist that interacts with benzodiazepine-insensitive GABAA receptors. THIP substantially improves sleep by increasing NREM time, lengthening NREM episode duration and enhancing the delta activity during NREMS (NREM sleep), without disrupting REMS (REM sleep) in humans9,10 and rodents.1113 In addition, THIP has been reported to induce neither tolerance nor withdrawal syndrome under chronic treatment conditions.12,14 Nevertheless, it produces dose-dependent alterations in EEG pattern resembling epileptiform-like activity during both waking and sleep in rats and mice,11,13 and may increase the duration of spontaneous petit mal–like seizures.15

Thus, pharmacological strategies must achieve a balance between sedative and adverse effects in the search of alternative compounds for the treatment of insomnia. Properties for an ideal hypnotic should be a rapid and reliable sleep induction and maintenance, prevention of repeated awakenings and unaltered physiological sleep architecture. In addition, such a compound should not show any decrease in efficacy upon repeated administration and no rebound insomnia should occur at treatment cessation. Finally, it should be devoid of residual sedative effects, which could impair daytime functioning and cognitive performance.

The present study aimed at characterizing the effects of a novel putative hypnotic compound, E-6199 ([4-(4-Methoxy-pyrimidin-2-yl)-piperazin-1-yl]-pyridin-2-yl-methanone; see Figure 1), compared to zolpidem, zopiclone and THIP on sleep-wakefulness patterns. For this purpose, we performed polygraphic recordings under acute and chronic conditions of treatment with these compounds, and analyzed key elements of vigilance states and sleep architecture.

Figure 1
Chemical structure of E-6199 [4-(4-Methoxy-pyrimidin-2-yl)-piperazin-1-yl]-pyridin-2-yl-methanone].


All the procedures involving animals and their care were conducted in conformity with the institutional guidelines that are in compliance with national and international laws and policies (Council directive 87–848, October 19, 1987, Ministère de l'Agriculture et de la Forêt, Service vétérinaire de la santé et de la protection animale, permissions 75–116 to M.H. and 75–125 to J.A.).


Experiments were performed on male C57BL/6J mice acclimated to the animal facilities for at least 1 week after receipt from the breeding center (Centre d'élevage R. Janvier, Le Genest St Isle, France). Mice were housed five per cage under standard conditions (12 h light/dark cycle; lights on at 07:00; 23±1°C ambient temperature; 60% relative humidity; food and water ad libitum).

Surgical Procedure

Mice (nine 10-week-old, body weight: 23–27 g) were implanted under sodium pentobarbital anaesthesia (70–75 mg/kg i.p.) with the classical set of electrodes (made of enameled nichrome wire, 150 μm in diameter) for polygraphic sleep monitoring.16 Briefly, two EEG electrodes were positioned epidurally in holes perforated into the skull, over the right cerebral cortex and the cerebellum (2 mm lateral and 2 mm caudal to the bregma suture; and at midline, 2 mm caudal to the lambda, respectively), two EOG electrodes were located subcutaneously on each side of the orbit, and two EMG electrodes were inserted into the neck muscles. All electrodes were anchored to the skull with super-bond and acrylic cement, and were soldered to a miniconnector also embedded in cement. Mice were then allowed to recover and acclimate to the recording conditions (one animal per cage: 20 × 20 × 30 cm, the home-cage being the recording cage; light-weight cable and swivel system allowing free displacements in the cage) for 9–11 days before the experiments.

Pharmacological Treatments

E-6199 (Esteve S.A., Barcelona, Spain), zopiclone (6-[5-chloro-2-pyridyl]-6,7-dihydro-7-oxo-5H-pyrrolo-[3,4-b]pyrazin-5-yl-4-methyl-piperazine-1-carboxylate, Sanofi-Aventis, Vitry, France), zolpidem (N,N,6-trimethyl-2-p-tolylimidazo[1,2-a] pyridin-3-acetamide hemitartrate, SYLACHIM, Mourenx, France), and THIP (4,5,6,7-tetrahydroisoxazolo[5,4-c] pyridin-3-ol hydrochloride, H.Lundbeck A/S, DK-2500 Valby, Denmark) were used. E-6199 was administered as its hydrochloride salt. All drugs were dissolved in pyrogen-free saline (0.9% NaCl), except zopiclone, which was suspended in 1% hydroxyethylcellulose (HEC) solution. In all cases, a volume of 0.1 mL was injected intraperitoneally (i.p.) at 10:00 a.m. For baseline data, mice were injected with vehicle (i.p.). Three to four days prior to initiation of treatments, mice received the vehicle (i.p.) each day in order to habituate to the experimental conditions.

Acute Treatments

E-6199 was injected at 5, 10, and 20 mg/kg, while zopiclone, zolpidem, and THIP were injected at the doses of 2, 5, and 10 mg/kg.13,17,18 Drug studies were carried out according to a standard crossover design, each animal receiving control (vehicle) and one drug at all doses tested. Accordingly, 4 groups (of 7–9 mice each) were used for this study (one group for each compound). A washout period of 4 to 7 days was allowed between two consecutive injections.

Chronic Treatments

A second series of experiments were performed for investigating the effects of E-6199 compared to zolpidem under chronic treatment conditions. The doses of 20 mg/kg of E-6199 and 5 mg/kg of zolpidem were chosen based upon the effects observed after acute treatment. For each compound, a group of 8 mice were used, and the experiments lasted 13 consecutive days. Mice received injections (i.p.) of the following: on the first day, pyrogen-free saline for baseline data (BSL), on each of the following 10 days (T1 to T10), either 20 mg/kg E-6199 or 5 mg/kg zolpidem, and on the last 2 days, pyrogen-free saline again (withdrawal period: W1 and W2).

Polygraphic Recordings

Recording was started at 10:00 a.m., i.e., immediately after drug injection, and was continued for 8 hours. For chronic conditions, recordings were performed on days BSL, T1, T5, and T10 of treatment, and W1 and W2 of withdrawal.

The polygraphic signals were amplified, sampled at 100 Hz (EMG and EOG) or 200 Hz (EEG), and digitalized by an AddLife A/D Module and the Somnologica software (Medcare, Reykjavik, Iceland).

Data Analyses


Polygraphic recordings were visually scored every 5-s epoch as wakefulness (W), NREMS, or REMS following classical criteria16,19 using the Somnologica software (Medcare, Reykjavik, Iceland). After THIP injection at dose of 5 and 10 mg/kg, some cases occurred where W and NREMS were indistinguishable on the EEG during the first 90 minutes.13,20 For these particular pharmacological conditions, the EMG activity and behavioral observations of mice (directly or with simultaneous video recordings) were used to determine the vigilance states.

Sleep Parameters

NREMS latency after treatment was defined as time elapsed between the injection and the first NREMS bout lasting at least 30 s. In order to assess the effect of drug treatments on REMS latency, we determined the time elapsed between the beginning of the first NREMS bout and the occurrence of the first REMS bout lasting at least 15 s.

The effects of each dose of each drug tested for each state of vigilance were analyzed over 8 h after injection. For the four drugs tested, the effects on vigilance states dissipated after 4 h post-injection, i.e., no modifications occurred during the 4–6 h and 6–8 h intervals. Therefore, these data are not presented in this study. Vigilance states amounts for each animal were computed within the 0–2 and 2–4 h intervals. Furthermore, sleep architecture was assessed by calculating the mean duration and frequency of vigilance states bouts (a bout could be as short as one epoch). To examine the maintenance of NREMS in detail, we analyzed the distribution of NREMS bouts as a function of bout length. We also scored single events (<15 s) in NREMS corresponding to a transient reduction in EEG power, accompanied by a slight EMG activation. These events were scored as microarousals (MAs).19 They are distinct from single episodes in NREMS, in which the EEG spectral content increases in the theta band with an EMG activation and which were thus scored as “short awakenings” and counted as W (see discussion in19). The fragmentation of NREMS by short awakenings and MAs was quantified by the number of events per minute of NREMS. NREMS bouts duration ignored MAs.19 Analyses were done with custom Matlab (MathWorks, Natick, MA) routines.19

Power Spectra Analysis

The EEG signal was processed for power spectrum analysis.19 Spectral analysis was performed on E-6199 and zolpidem, as both displayed remarkable hypnotic profile in mice.

Variations in absolute power density in each animal within NREMS were taken for calculations, as follows. Consecutive 5-s epochs were subjected to a fast Fourier transformation routine (FFT), yielding power spectra between 0.4 to 25 Hz, with a 0.2 Hz frequency and a 5-s time resolution. For each vigilance state, a mean EEG power spectrum was obtained by averaging the spectra of all 5-s epochs of a state in the first 2-h interval, after exclusion of epochs at the transition between two different states. Under our conditions, C57BL/6J mice exhibited a similar EEG density as that previously reported for this mouse strain.21 Absolute power densities are not suitable for visualization of drugs effects because the interindividual variation is considerable. Therefore, for each animal, the power values of the drug condition were expressed relative to control: EEG power density of each frequency bin within NREMS was computed for the first 2-h interval and expressed as fractional change of the corresponding bin (and interval) after vehicle treatment.

To analyze changes in the dynamics of slow wave activity (SWA; 0.75–4.5 Hz) during NREMS bouts, all NREMS bouts that were preceded at least by three 5-s epochs (15 s) of W and that lasted at least eight 5-s epochs (40 s) were selected from the first 2-h interval post-injection. For normalization, all 5-s epochs were expressed relative to the mean power of the three 5-s epochs scored as W before the W-NREMS transition.22


All results are expressed as means ± SEM. For a given treatment, each animal was referred to its own baseline represented by the data obtained under the vehicle/baseline condition. The effects of the different drugs tested upon each vigilance state parameter were analyzed for 2-h intervals. For sleep latencies, vigilance states amounts, number and duration of bouts, episode length distribution, as well as MAs and occurrence of short awakenings, significance was tested by performing one-way ANOVA with repeated measure (rANOVA) over dose or day according to the treatment condition (acute or chronic). For the evaluation of dynamics of SWA at W-NREMS transition, a two-way rANOVA was performed over dose or day and over 5-s epochs. In case of significance (P <0.05), the F test was followed by a two-tailed paired Student's t-test.

EEG spectra values after 20 mg/kg of E-6199 or 5 mg/kg of zolpidem were compared to vehicle by a two-tailed paired Student's t-test.


Behavioral Observations

No behavioral alterations were observed at any doses tested for E-6199, zopiclone, or zolpidem. In contrast, mice treated with THIP at 5 and 10 mg/kg exhibited for 60 to 90 minutes following injection motionless behavior, abnormal posture, and myclonus, while the eyes remained open. This state was scored as W.13

Acute Treatments

Effects of Hypnotics on Sleep Latency

Sleep latencies were similar in all groups after vehicle injection (Figure 2). After injection of all compounds except THIP, NREMS latency was dose-dependently reduced compared to vehicle (E-6199: F3,18 = 4.26, P <0.05; zopiclone: F3,21 = 3.75, P <0.05; zolpidem F3,15 = 5.94, P <0.01). These effects reached significance for the highest dose of E-6199 (20 mg/kg) and zopiclone (10 mg/kg), and for 5 and 10 mg/kg of zolpidem. In contrast, THIP markedly prolonged the NREMS latency (F3,21 = 27.34, P <0.0001) at 5 and 10 mg/kg.

Figure 2
NREMS and REMS latencies after acute treatment with E-6199, zopiclone (ZPC), zolpidem (ZPD) and THIP at various doses. Data (mean ± SEM of 7 to 9 animals per group; see Table 1) are expressed as minutes. * P < 0.05, significantly different ...

In parallel, the REMS latency was not significantly modified after E-6199 or zopiclone, but was lengthened after the two highest doses (5 and 10 mg/kg) of zolpidem and THIP (F3,15 = 5.98, P <0.05 and F3,21 = 20.54, P <0.0001, respectively; Figure 2).

Effects of Hypnotics on Time Spent in Each Vigilance State

Compared to vehicle, E-6199 (5–20 mg/kg i.p.) increased NREMS amounts for 2 h post-injection (dose F3,18 = 3.63, P <0.05) and this effect reached significance at 10 and 20 mg/kg (Table 1). In parallel, the amounts of W decreased during the same period (dose F3,18 = 3.79, P <0.05). In contrast, no consistent modification of REMS was observed.

Table 1
Amounts of the Three Vigilance States (W, NREMS, and REMS) During 4 h After Acute Treatment with E-6199, Zopiclone, Zolpidem, or THIP at Various Doses

Zopiclone (2–10 mg/kg i.p.) failed to significantly change the amounts of any vigilance state during the first 2 h after its injection compared to vehicle. However, this drug increased the time spent in NREMS during the 2–4 h interval post-injection (F3,21 = 4.85, P <0.05; Table 1).

Zolpidem (2–10 mg/kg i.p.) dose-dependently enhanced the amounts of NREMS for the first 2 h following injection compared to vehicle (dose F3,15 = 7.93, P <0.01) and reduced those of W and REMS (dose F3,15 = 7.58, P <0.01; dose F3,15 = 5.58, P <0.05, respectively). The decrease in REMS reached significance at 10 mg/kg, while the effects on W and NREMS were significant at all doses. Moreover, at the dose of 10 mg/kg, these latter effects persisted during the 2–4 h interval post-injection (Table 1).

In contrast to the other three drugs tested, THIP at 5 and 10 mg/kg i.p. induced a dramatic increase in W during the first 2 h post-injection (dose F3,21 = 21.90, P <0.0001; Table 1). However, this effect was associated with behavioral alterations and did not correspond to a physiological waking state.13,20 In parallel, NREMS and REMS amounts were drastically reduced compared to vehicle (dose F3,21 = 19.4, P <0.0001 and F3,21 = 17.9, P <0.0001, respectively; Table 1). This strong reduction of NREMS and REMS during the first 2 h following THIP injection actually resulted from a delayed initiation of sleep (see Figure 2). The dose of 2 mg/kg did not elicit the same effects on vigilance states total duration: it decreased W amounts and enhanced both NREMS and REMS for 2 h after injection. During the following hours, only the decrease in time spent in W persisted at the highest dose of THIP.

Effects of Hypnotics on Sleep Architecture and Continuity

We investigated the sleep architecture after each treatment by evaluating the mean duration, frequency and the distribution of bouts frequency as a function of bout length, with particular attention to NREMS. Within this state, short awakenings and MAs (<15 s; see Materials & Methods) were assessed in order to analyze sleep fragmentation. Focus was brought to the first 2 h after injection, because it was when compounds had the most of their effects.

Administration of E-6199 did not affect the frequency, but induced a dose-dependent increase in the mean duration of NREMS bouts (dose F3,18 = 4.45, P <0.05; Figure 3a), that was accounted for by a reduction in short (5–75 s) and a marked increase in long (≥320 s) bouts (see supplementary data, Figure S10). In parallel, the occurrence of short awakenings and MAs was strikingly reduced (dose F3,18 = 3.33, P <0.05 and F3,18 = 10.20, P <0.001, respectively; Figure 3b). A slight, but non-significant, decrease in the number of W was also noted, while the duration and frequency of REMS bouts remained globally unchanged after treatment with E-6199 (see supplementary data, Table 2).

Figure 3
Architecture of NREMS bouts after acute treatment with E-6199, zopiclone, zolpidem or THIP at various doses. a) Number and mean duration (an index of continuity) of all NREMS episodes. b) Occurrence of short awakenings and microarousals (MAs) within NREMS, ...

Zopiclone at 2–10 mg/kg did not modify the mean duration of W, NREMS, or REMS bouts and their global frequency, nor the number of short awakenings in a significant manner (Figure 3 and supplementary data, Table 2). However, at 5 and 10 mg/kg, it slightly enhanced the number of long NREMS bouts (lasting at least 320 s) and tended to reduce those of shorter duration (20–75 s; see supplementary data, Figure S10). In contrast, it also induced an increase in the number of MAs occurring during NREMS (Figure 3b).

Zolpidem did not change significantly the frequency or the mean duration of NREMS bouts, nor the occurence of short awakenings, compared to vehicle (Figure 3). Analysis of the time-weighted frequency histogram only showed a reduction in the number of very short NREMS bouts (5 s) for all doses of zolpidem (see supplementary data, Figure S10). In contrast, it induced an enhancement in the number of MAs (dose F3,15 = 3.25, P <0.05; Figure 3b), which reached significance at 2 and 10 mg/kg. Zolpidem also reduced the mean duration of W bouts (dose F3,15 = 3.36, P <0.05) and the number of REM bouts (dose F3,15 = 3.83, P <0.05; see supplementary data, Table 2).

Administration of THIP lessened the frequency of all state bouts compared to vehicle (dose W: F3,21 = 9.52, P <0.001; NREMS: F3,21 = 9.55, P <0.001 and REMS: F3,21 = 6.07 P <0.01; Figure 3a and supplementary data, Table 2), and these effects reached significance at 5 and 10 mg/kg. THIP also lengthened NREMS bouts (dose F3,21 = 3.68, P <0.05) at 2 and 5 mg/kg. In parallel, the frequency of long bouts (≥160 s) of NREMS was increased compared to vehicle (see supplementary data, Figure S10). Furthermore, NREMS was less fragmented after administration of THIP, as the occurrence of short awakenings and MAs was decreased (dose F3,21 = 3.91, P <0.05 and F3,21 = 2.94, P <0.05, respectively; Figure 3b). These latter effects were significant only at 10 mg/kg. Finally, THIP also prolonged the mean duration of W bouts for 2 h post-injection (dose F3,21 = 12.07, P <0.0001; see supplementary data, Table 2), which was accounted for by the long sleep latency at the highest dose (see Figure 2).

Effects of Hypnotics on SWA and EEG Power Spectra within NREMS

After injection of vehicle, the power density of SWA (0.75–4.5 Hz) was higher during NREMS than during W as illustrated by the analysis of its dynamics at the transition between W and NREMS (epoch F11,165 = 104.49, P <0.0001 and F11,165 = 63.19, P <0.0001, for E-6199 and zolpidem groups, respectively; Figure 4). E-6199 induced, following the same time course, an increase of SWA during NREMS (dose F3,15 = 3.69, P <0.05; Figure 4). On the contrary, zolpidem induced a decrease in SWA, that was significant at 5 and 10 mg/kg (dose F3,15 = 3.29, P <0.05; Figure 4).

Figure 4
Dynamics of slow wave activity at W-NREMS transitions during the first 2 h after acute treatment with E-6199 or zolpidem at various doses. SWA (mean EEG power density in the 0.75–4.5 Hz band) was calculated in absolute values and is expressed ...

We also analyzed the absolute EEG power values during NREMS after vehicle injection and after 20 mg/kg of E-6199 and 5 mg/kg of zolpidem performed on the following day (Figure 5). E-6199 elicited a slight increase in the 2–3.5 Hz frequencies, whereas zolpidem reduced all frequencies above 2 Hz, so that the averaged SWA (0.75–4.5 Hz) tended to be enhanced after E-6199 administration, and was significantly reduced after zolpidem (Figure 5).

Figure 5
EEG power density and slow wave activity (SWA) in NREMS during the first 2 h after acute treatment with E-6199 (20 mg/kg) and zolpidem (5 mg/kg). Curves connect mean values (mean ± SEM of 6 animals per group) of power density of each frequency ...

Chronic Treatments

Based on the effects observed under acute conditions, the dose of 20 mg/kg i.p. was selected for daily treatment with E-6199. Zolpidem was taken as reference compound because it demonstrated under acute treatment a better hypnotic profile than zopiclone or THIP. It was used at the daily dose of 5 mg/kg, because its effects lasted during 2 h post-injection, like the dose of 20 mg/kg of E-6199 (see Table 1).

Effects of Hypnotics on Sleep Latency

Daily administration of E-6199 (20 mg/kg) or zolpidem (5 mg/kg) markedly shortened NREMS latency compared to baseline (BSL) (day F5,35 = 5.59, P <0.0001 and F5,35 = 11.60, P <0.0001, for E-6199 and zolpidem, respectively; Figure 6). These effects were observed on the first (T1) and persisted until the last (T10) day of treatment. Latencies values returned to baseline at withdrawal (W1 and W2). For both compounds, REMS latency was increased during the whole treatment (day F5,35 = 6.57, P <0.001 and F5,35 = 4.52, P <0.01 for E-6199 and zolpidem, respectively), and were normalized at withdrawal (Figure 6).

Figure 6
NREMS and REMS latencies after repeated treatment with E-6199 (20 mg/kg) or zolpidem (5 mg/kg). Data (mean ± SEM of 8 animals per group) are expressed as minutes for the baseline (BSL), treatment (T1, T5 and T10) and drug withdrawal (W1 and W2) ...

Effects of Hypnotics on Time Spent in Each Vigilance State

These effects were observed essentially during the first 2 h after injection. Both E-6199 and zolpidem increased the total time spent in NREMS compared to BSL (day F5,35 = 5.60, P <0.001 and F5,35 = 12.37, P <0.0001, respectively; Figure 7 and see supplementary data, Table 3). This increase was significant on the first day and persisted during the whole treatment, while values returned to baseline at withdrawal. Both compounds also caused a concomitant decrease in W amounts (day F5,35 = 3.38, P <0.05 and F5,35 = 7.44, P <0.0001, respectively), that was sustained during the whole treatment for E-6199, (Figure 7 and see supplementary data, Table 3). In parallel, REMS was reduced compared to BSL (day F5,35 = 6.77, P <0.001 and F5,35 = 4.51, P <0.01, for E-6199 and zolpidem respectively; Figure 7 and see supplementary data, Table 3), and this effect was sustained over day of treatment for only zolpidem. All vigilance states were normalized at cessation of treatment (Figure 7).

Figure 7
Amounts of the three vigilance states (W, NREMS and REMS) during the first 2 h after repeated treatment with E-6199 or zolpidem. The data (mean ± SEM of 8 animals per group) are expressed relative to recording time for the baseline (BSL), treatment ...

Effects of Hypnotics on Sleep Architecture and Continuity

The increased mean duration of NREMS bouts after treatment with 20 mg/kg of E-6199 (as under acute treatment with the same dose) was found again on the first day of chronic treatment. This effect persisted throughout the whole treatment (day F5,35 = 11.71, P <0.0001; Figure 8a) and also, but less, at withdrawal. It was accounted for, during treatment days, by a reduction in the number of short (5–75 s), and a marked increase in that of long (≥160 s) NREMS bouts (see supplementary data, Figure S11). In parallel, the total number of NREMS bouts was reduced (day F5,35 = 3.39, P <0.05; Figure 8a). These modifications were associated with a marked decrease in the frequency of short awakenings and MAs (day F5,35 = 3.44, P <0.05 and day F5,35 = 6,7, P <0.001, respectively; Figure 8b), and in the number of W and REMS bouts (day F5,35 = 3.38, P <0.05 and F5,35 = 9.13, P <0.0001, respectively; see supplementary data, Table 4).

Figure 8
Architecture of NREMS bouts after repeated treatment with E-6199 or zolpidem. a) Number and mean duration of all NREMS episodes. b) Occurrence of short awakenings and microarousals (MAs) within NREMS, as a measure of fragmentation. All data are plotted ...

Zolpidem administration (5 mg/kg) reduced the frequency of NREMS bouts (Figure 8a) compared to BSL (day F5,35 = 2.35, P <0.05), notably at T5 and at withdrawal. A concomitant decrease in W bouts was also observed (day F5,35 = 3.35, P <0.05; Table 4). In parallel, the mean duration of NREMS bouts was increased compared to BSL (day F5,35 = 4,24, P <0.05; figure 8a), an effect that persisted on the first day of withdrawal. No changes were noted in the distribution of NREMS bouts (see supplementary data, Figure S11), nor in the occurrence of short awakenings (Figure 8b), but the number of MAs was enhanced (day F5,35 = 2,88, P <0.05). In contrast, the number of REMS bouts was dramatically reduced during the whole treatment (day F5,35 = 6.79, P <0.001; Table 4). Baseline values were restored at withdrawal.

Effects of Hypnotics on SWA within NREMS

The increased SWA observed on the first day as after acute treatment with 20 mg/kg of E-6199 was found again and persisted throughout the whole treatment (day F5,25 = 4.61, P <0.05; Figure 9). Baseline values were restored at withdrawal. The decreased SWA after zolpidem was attenuated in the course of treatment, but remained significant (day F5,25 = 2.97, P <0.05; Figure 9).

Figure 9
Dynamics of slow wave activity at W-NREMS transitions during the first 2 h after repeated treatment with E-6199 or zolpidem. SWA (mean EEG power density in the 0.75–4.5 Hz band) was calculated in absolute values and is expressed relatively to ...


This study provides an exhaustive characterization of the effects on sleep-wake patterns of a novel compound, E-6199, compared to the widely used zopiclone and zolpidem, and to THIP (gaboxadol), a potential hypnotic currently under clinical development. After acute administration, E-6199 exhibited clear-cut sleep-promoting effects in mice, including shortened NREMS latency, increased NREMS time and improved sleep continuity and efficiency. These effects were sustained during chronic dosing (daily administration during 10 days), indicating that no rapid development of tolerance to the hypnotic action of E-6199 took place. Finally, abrupt drug withdrawal did not elicit negative sleep rebound phenomena.

Sleep Effects after Acute Treatment

In mice, acute administration of E-6199, zopiclone and zolpidem strikingly shortened NREMS latency, thus demonstrating a rapid onset of hypnotic action for these three compounds, as previously described for zolpidem and zopiclone in humans and rats.3 E-6199 was efficient at a fourfold higher dosage than zolpidem, and both compounds substantially increased total time spent in NREMS for 2 h, indicating a short duration of action consistent with their short half-lives in rodents (E-6199: 0.6 h, R.A., unpublished data; zolpidem: 1.5 h7), but the modalities of this increase differ between the two drugs. Indeed, E-6199 improved NREMS maintenance, by markedly increasing NREMS bout duration. In addition, it diminished the frequency of both short awakenings and MAs, which disrupt NREMS and correspond to transient events underlying alertness.19,23 It is widely accepted that the SWA (EEG power in the delta range: 0.75–4.5 Hz) during NREMS is an index for sleep intensity,24 spindles (12–16 Hz25) being markers of light sleep. Along this line, we observed that E-6199 decreased EEG power in the spindle range, while it enhanced SWA. Therefore, E-6199 concomitantly decreased sleep fragmentation, lengthened NREMS episodes, and enhanced SWA, all features that characterize increased sleep quality, illustrated by the negative correlation between sleep fragmentation and SWA observed after prolonged wakefulness.21,22,25 Furthermore, such positive effects occurred without altering consistently REMS and W parameters. Altogether, these findings indicate that E-6199 substantially enhanced sleep continuity and efficiency in the mouse.

On the contrary, the sleep-promoting effect of zolpidem took place at the expense of W, in concordance with most findings in humans and rodents,3,18,26,27 but was neither associated with a reduction in NREMS fragmentation nor with an increase in NREMS bout duration. Moreover, the number of MAs was increased. These results are opposite to findings in humans,3 where zolpidem reduced the amounts of briefs awakenings. Such discrepancy might be related to species differences, since benzodiazepines like diazepam also diminished the number of briefs awakenings in humans, but not in mice.25 In addition, zolpidem decreased SWA during NREMS, an effect that was observed, but not significantly, in another study.18 It also elicited an overall reduction of EEG frequencies over 5 Hz, in agreement with the latter study.18 Finally, zolpidem altered REMS latency, total duration and bouts frequency, as previously reported in rats5 and mice.18 Altogether, these data confirm in the mouse the powerful sleep-promoting properties of zolpidem, and shed light on its moderate negative effects on sleep stability and EEG synchronization.

The effects of zopiclone on sleep have not been reported previously in mice, and surprisingly, we found only a minor effect on sleep amounts. Although well documented in humans28,29 and in rats,30,31 the ability of zopiclone to enhance NREMS is therefore much less obvious in mice, as it elicited a delayed increase in NREMS (only during the 2–4 h interval after injection). However, zopiclone promoted long NREMS bouts at the expense of shorter ones, and reduced NREMS latency, as also reported in humans, whether insomniac or good sleepers.3 REMS was not altered, in contrast to the clear-cut reduction of this sleep state observed in rats.5,30

Finally, THIP appeared unable to induce and/or promote sleep, confirming previous data in mice.13,20 This lack of sleep-inducing effect is consistent with earlier studies in rats using THIP and other GABAA agonists such as muscimol,3,32 and further supports the idea that GABAA agonists do no affect the initiation of sleep. In addition, THIP exhibited dualistic features according to its dosing, since high doses dramatically delayed NREMS onset and produced marked abnormalities in the EEG during W as well as NREMS,13,20 whereas the lowest dose improved NREMS stability and elicited a surge in the delta activity within NREMS (data not shown). Such NREMS-consolidating and -improving effects support previous reports in humans9 and rats.11 In contrast, higher dosing of THIP induced dramatic EEG alterations unrelated to any specific vigilance state during 60–120 minutes after injection, in concordance with the short half-life of this drug (<2 h in mice33). These results are consistent with previous findings in rats and mice where THIP was found to induce abnormal EEG pattern 11,13, and to enhance the duration of spontaneous petit-mal seizure.15

Sustained NREMS-promoting effects under chronic treatment

Chronic drug administration can lead to a decrease in drug efficacy or potency (pharmacological tolerance) or the appearance of withdrawal or abstinence signs when treatment terminates (physiological dependence). Indeed, there is evidence that pharmacological tolerance and physiological dependence to hypnotic/sedative effects of at least some benzodiazepines4 can develop, especially at high doses. In particular, abrupt withdrawal of benzodiazepines often produces a transient deterioration of sleep (rebound insomnia), with a prolongation of sleep onset latency, an increase in intermittent wakefulness and/or a decrease in total sleep time, which may lead to continued use.34 Thus, it is of interest to investigate whether the development of tolerance and/or dependence occurs under repeated administration of a putative hypnotic compound.

The effects on NREMS of E-6199 and zolpidem under chronic treatment conditions were globally similar to those observed after acute treatment at the same dosage. All E-6199-induced changes, including the lengthening of NREMS bouts, the reduction of short awakenings and MAs occurrence, as well as the increase in SWA levels in the course of NREMS bouts, persisted during the whole treatment, and became even larger as chronic treatment proceeded. These effects, added to the reduction in the number of W bouts, suggest that E-6199 under chronic conditions effectively enhances NREMS continuity and intensity. However, and in contrast to acute treatment, E-6199 induced on day 1, a reduction in REMS amounts, that was accounted for by an increase in REMS latency and a decrease in bout frequency. Because this discrepancy was a concern, we used a third group of naïve mice (n = 6) to re-investigate the effects of E-6199 at the dose of 20 mg/kg. REMS characteristics were intermediate between those of acute and chronic treatment, with no significant differences from these two conditions. Thus, the effects of E-6199 on REMS were highly variable, suggesting a rather weak/inconsistent action on this sleep state. With regard to zolpidem, enhanced NREMS and reduced REMS amounts, as well as less prominent effects on the number and duration of NREMS bouts, persisted during the whole treatment. However, the decrease of SWA did not persist, suggesting that the mechanisms underlying the NREMS rise after zolpidem are more labile than those after E-6199.

Finally, the abrupt cessation of treatment with E-6199 or zolpidem did not induce sleep disturbances, indicating the lack of withdrawal and/or rebound insomnia. In the case of zolpidem these findings are consistent with previous studies in humans, which reported no or only weak rebound insomnia after cessation of treatment for two weeks or longer4.

Overall, these findings indicate that E-6199 did not induce tolerance phenomena during chronic treatment nor adverse effects at withdrawal, and that its sleep-stabilizing action became more effective as the treatment proceeded.

Mechanisms Possibly Involved

The mechanisms of action and the molecular/cellular target(s) of E-6199 in the central nervous system are presently unknown, but this compound does not seem to potentiate or mimic the GABA response at GABAA receptors, for which it has only a low affinity (Ki> 10 μM, for all GABAA receptors sites, unpublished data). In contrast to the other three compounds tested in parallel in our study, E-6199 exhibited sleep-inducing and -promoting properties by deeply improving NREMS stability, an effect also induced by low doses of THIP. Furthermore, E-6119 enhanced NREMS mostly at the cost of single events disrupting NREMS, and exerted only weak effects on W and REMS. These latter observations suggest that E-6199 does not interfere with ascending arousal systems and that this compound selectively affects NREMS processes. In contrast, GABAA receptor allosteric modulators such as zolpidem, zopiclone or benzodiazepines globally increase NREMS at the expense of W and REMS in humans and rodents,3,4,25 as in our study where mice treated with zolpidem displayed a marked decrease in W and REMS amounts.

Regarding EEG activity, E-6199 increased SWA during NREMS without dramatically altering EEG power densities in frequencies over 5 Hz, and elicited a reduction in power of the spindles range. This is in contrast with the effects of zopiclone and zolpidem on the EEG spectrum, which closely resemble those of other positive GABAA receptor allosteric modulators like benzodiazepines. In humans, these compounds induce a typical benzodiazepine-like EEG “fingerprint”, consisting of a marked reduction in delta oscillations and an increase in the spindle range activity within NREMS.3 Such alterations are also found in rodents after diazepam and midazolam.3,25,35 However, zolpidem did not induce all of these benzodiazepine-like EEG changes in mice, in agreement with other results,18 which possibly reflects species differences in the regional distribution of GABAA receptors subunits in the brain.36 Overall, these results may indicate that the hypnotic effects of E-6199 do not directly involve GABAA receptors.

At the present time, the possible mechanisms underlying the action of E-6199 on sleep remain highly speculative, but as our understanding of the mechanisms responsible for sleep regulation improves, and the pharmacology of existing hypnotic agents becomes clearer, it is possible that novel strategies, unrelated to presently characterized receptor modulation, will be identified and open new ways of interpretation.


Epidemiological studies on the general population have explored the different clinical expressions of insomnia complaints: difficulty initiating sleep, repetitive arousals, early morning awakening, diminished total sleep time, and/or nonrestorative sleep.2 In primary insomnia, the decrease of sleep time is mostly related to a curtailment of NREMS, indicating a sustained difficulty maintaining consolidated sleep.1,37 Aging is also associated with a deterioration of sleep, that consists of a progressive decrease in the amount of deep sleep coupled with a frequent sleep fragmentation by nocturnal awakenings. The present data with E-6199 in mice suggest that this drug might be of interest for reducing sleep complaints, especially the poor sleep quality in insomniac patients and elderly people.


This research was supported by Institut National de la Santé et de la Recherche Médicale, Université Pierre et Marie Curie, and Esteve S.A. We are very grateful to Dr. C Lena for his generous help in Matlab routines writing.


Disclosure Statement

This research was supported by Institut National de la Santé et de la Recherche Médicale, Université Pierre et Marie Curie, and Esteve S.A. Alberto Dordal, Ramon Aixendri, and Antonio Guzman are employees of Esteve S.A. The other authors have indicated no financial conflicts of interest.


1. Roth T, Drake C. Evolution of insomnia: current status and future direction. Sleep Med. 2004;5(Suppl 1):S23–30. [PubMed]
2. Sateia MJ, Nowell PD. Insomnia. Lancet. 2004;364:1959–73. [PubMed]
3. Lancel M. Role of GABAA receptors in the regulation of sleep: initial sleep responses to peripherally administered modulators and agonists. Sleep. 1999;22:33–42. [PubMed]
4. Wagner J, Wagner ML. Non-benzodiazepines for the treatment of insomnia. Sleep Med Rev. 2000;4:551–81. [PubMed]
5. Gottesmann C, Gandolfo G, Arnaud C, Gauthier P. The intermediate stage and paradoxical sleep in the rat: influence of three generations of hypnotics. Eur J Neurosci. 1998;10:409–14. [PubMed]
6. Landolt HP, Finelli LA, Roth C, Buck A, Achermann P, Borbely AA. Zolpidem and sleep deprivation: different effect on EEG power spectra. J Sleep Res. 2000;9:175–83. [PubMed]
7. Wagner J, Wagner ML, Hening WA. Beyond benzodiazepines: alternative pharmacologic agents for the treatment of insomnia. Ann Pharmacother. 1998;32:680–91. [PubMed]
8. Gericke CA, Ludolph AC. Chronic abuse of zolpidem. Jama. 1994;272:1721–2. [PubMed]
9. Faulhaber J, Steiger A, Lancel M. The GABAA agonist THIP produces slow wave sleep and reduces spindling activity in NREM sleep in humans. Psychopharmacology (Berl) 1997;130:285–91. [PubMed]
10. Mathias S, Steiger A, Lancel M. The GABA(A) agonist gaboxadol improves the quality of post-nap sleep. Psychopharmacology (Berl) 2001;157:299–304. [PubMed]
11. Lancel M. The GABA(A) agonist THIP increases non-REM sleep and enhances non-REM sleep-specific delta activity in the rat during the dark period. Sleep. 1997;20:1099–104. [PubMed]
12. Lancel M, Langebartels A. gamma-aminobutyric Acid(A) (GABA(A)) agonist 4,5,6, 7-tetrahydroisoxazolo[4,5-c]pyridin-3-ol persistently increases sleep maintenance and intensity during chronic administration to rats. J Pharmacol Exp Ther. 2000;293:1084–90. [PubMed]
13. Vyazovskiy VV, Kopp C, Bosch G, Tobler I. The GABAA receptor agonist THIP alters the EEG in waking and sleep of mice. Neuropharmacology. 2005;48:617–26. [PubMed]
14. Mathias S, Zihl J, Steiger A, Lancel M. Effect of repeated gaboxadol administration on night sleep and next-day performance in healthy elderly subjects. Neuropsychopharmacology. 2005;30:833–41. [PubMed]
15. Vergnes M, Marescaux C, Micheletti G, Depaulis A, Rumbach L, Warter JM. Enhancement of spike and wave discharges by GABAmimetic drugs in rats with spontaneous petit-mal-like epilepsy. Neurosci Lett. 1984;44:91–4. [PubMed]
16. Boutrel B, Franc B, Hen R, Hamon M, Adrien J. Key role of 5-HT1B receptors in the regulation of paradoxical sleep as evidenced in 5-HT1B knock-out mice. J Neurosci. 1999;19:3204–12. [PubMed]
17. Tsuboi M, Tanaka Y, Himori N. Changes in mouse hippocampal EEG characteristics after oral administration of Ro 41-3696, nitrazepam, or zopiclone alone and in combination with ethanol. Pharmacology. 1994;49:278–85. [PubMed]
18. Kopp C, Rudolph U, Tobler I. Sleep EEG changes after zolpidem in mice. Neuroreport. 2004;15:2299–302. [PubMed]
19. Lena C, Popa D, Grailhe R, Escourrou P, Changeux JP, Adrien J. Beta2-containing nicotinic receptors contribute to the organization of sleep and regulate putative micro-arousals in mice. J Neurosci. 2004;24:5711–8. [PubMed]
20. Winsky-Sommerer R, Vyazovskiy VV, Homanics GE, Tobler I. The EEG effects of THIP (Gaboxadol) on sleep and waking are mediated by the GABA(A)delta-subunit-containing receptors. Eur J Neurosci. 2007;25:1893–9. [PubMed]
21. Franken P, Malafosse A, Tafti M. Genetic variation in EEG activity during sleep in inbred mice. Am J Physiol. 1998;275:R1127–37. [PubMed]
22. Franken P, Malafosse A, Tafti M. Genetic determinants of sleep regulation in inbred mice. Sleep. 1999;22:155–69. [PubMed]
23. Halasz P. Hierarchy of micro-arousals and the microstructure of sleep. Neurophysiol Clin. 1998;28:461–75. [PubMed]
24. Borbely AA, Baumann F, Brandeis D, Strauch I, Lehmann D. Sleep deprivation: effect on sleep stages and EEG power density in man. Electroencephalogr Clin Neurophysiol. 1981;51:483–95. [PubMed]
25. Tobler I, Kopp C, Deboer T, Rudolph U. Diazepam-induced changes in sleep: role of the alpha 1 GABA(A) receptor subtype. Proc Natl Acad Sci U S A. 2001;98:6464–9. [PMC free article] [PubMed]
26. Renger JJ, Dunn SL, Motzel SL, Johnson C, Koblan KS. Sub-chronic administration of zolpidem affects modifications to rat sleep architecture. Brain Res. 2004;1010:45–54. [PubMed]
27. Sanger DJ. The pharmacology and mechanisms of action of new generation, non-benzodiazepine hypnotic agents. CNS Drugs. 2004;18(Suppl 1):9–15. discussion 41, 43–5. [PubMed]
28. Tiberge M, Calvet U, Khayi N, Delahaye C, Arbus L. [Comparison of the effects of zopiclone and triazolam on the sleep of normal subjects] Encephale. 1988;14:319–24. [PubMed]
29. Mann K, Bauer H, Hiemke C, Roschke J, Wetzel H, Benkert O. Acute, subchronic and discontinuation effects of zopiclone on sleep EEG and nocturnal melatonin secretion. Eur Neuropsychopharmacol. 1996;6:163–8. [PubMed]
30. Yoshimoto M, Higuchi H, Kamata M, Yoshida K, Shimizu T, Hishikawa Y. The effects of benzodiazepine (triazolam), cyclopyrrolone (zopiclone) and imidazopyridine (zolpidem) hypnotics on the frequency of hippocampal theta activity and sleep structure in rats. Eur Neuropsychopharmacol. 1999;9:29–35. [PubMed]
31. Shinomiya K, Shigemoto Y, Okuma C, Mio M, Kamei C. Effects of short-acting hypnotics on sleep latency in rats placed on grid suspended over water. Eur J Pharmacol. 2003;460:139–44. [PubMed]
32. Lancel M, Faulhaber J. The GABAA agonist THIP (gaboxadol) increases non-REM sleep and enhances delta activity in the rat. Neuroreport. 1996;7:2241–5. [PubMed]
33. Schultz B, Aaes-Jorgensen T, Bogeso KP, Jorgensen A. Preliminary studies on the absorption, distribution, metabolism, and excretion of THIP in animal and man using 14C-labelled compound. Acta Pharmacol Toxicol (Copenh) 1981;49:116–24. [PubMed]
34. Bateson AN. Basic pharmacologic mechanisms involved in benzodiazepine tolerance and withdrawal. Curr Pharm Des. 2002;8:5–21. [PubMed]
35. Lancel M, Cronlein TA, Faulhaber J. Role of GABAA receptors in sleep regulation. Differential effects of muscimol and midazolam on sleep in rats. Neuropsychopharmacology. 1996;15:63–74. [PubMed]
36. Mohler H, Fritschy JM, Vogt K, Crestani F, Rudolph U. Handb Exp Pharmacol. 2005. Pathophysiology and pharmacology of GABA(A) receptors; pp. 225–47. [PubMed]
37. Terzano MG, Rossi M, Palomba V, Smerieri A, Parrino L. New drugs for insomnia: comparative tolerability of zopiclone, zolpidem and zaleplon. Drug Saf. 2003;26:261–82. [PubMed]

Articles from Sleep are provided here courtesy of Associated Professional Sleep Societies, LLC


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


  • Compound
    PubChem Compound links
  • MedGen
    Related information in MedGen
  • PubMed
    PubMed citations for these articles
  • Substance
    PubChem Substance links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...