Distinct Effect of Dual Orexin 1/2 Receptor Antagonists versus Zolpidem on the Sleep-Wake Profile and EEG Power Spectrum of Rats

LH Jacobson1, L Perrot2, M Fendt3, D Behnke2, S Cotesta2, G Laue2, S Ofner2, E Legangneux2, S Hintermann2, C Betschart2, CE Gee4, D Hoyer5. 1The Florey Institute of Neuroscience and Mental Health, University of Melbourne, Parkville, Victoria 3010, Australia, 2Novartis Institutes for BioMedical Research, Basel, Switzerland, 3Institute for Pharmacology and Toxicology, Otto-von-Guericke University Magdeburg, D-39120 Magdeburg, Germany, 4Institute for Synaptic Physiology, Center for Molecular Neurobiology Hamburg, 20251 Hamburg, Germany, 5Department of Pharmacology & Therapeutics, University of Melbourne, Parkville, Victoria 3010, Australia

Orexin peptide-producing neurons are clustered in the lateral hypothalamus, project widely across the brain and play a major role in sleep-wake regulation. Orexin A and B act via two GPCRs, Orexin 1 and Orexin 2 (Ox1R, Ox2R) which have become the targets of drug discovery programs for sleep disorders. Four dual Ox1R/Ox2R antagonists (DORAs) have been clinically validated for sleep modulation in primary insomnia. Advantages of DORAs compared to marketed GABAA receptor-enhancing hypnotics may include reduced next-day effects, lack of potentiation of ethanol effects, and reduced tolerance and withdrawal [1]. Some debate, however, remains as to the effect of DORAs on sleep architecture. While GABAA receptor-mediated drugs increase NREM sleep at the expense of REM, DORAs are reported to either increase sleep without affecting sleep architecture, or to enhance REM sleep, while very few studies have examined EEG power spectra. We therefore sought to compare the effects of the clinically validated DORAs SB-649868 (50 mg/kg), almorexant (50, 150 mg/kg), MK-6096 (filorexant; 50,100 mg/kg) and MK-4305 (suvorexant; 30 mg/kg) with that of the GABAA receptor-mediated hypnotic zolpidem (10 mg/kg), on the sleep-wake profile and EEG power spectrum in the rat.

Male Sprague-Dawley rats were surgically implanted with an EEG hub with 4 electrocorticogram electrodes positioned on the cerebellum, visual and frontal cortices, the frontal electrode serving as an earth. Piezoelectric sensors were placed on the bottom of the cages to detect activity and aid in the definition of vigilance states [2]. Animals acted as their own controls in 3 day experiments (n = 5-8 per experiment), with dosing at the start of the dark phase with water on day 1 (habituation), vehicle on day 2, and compound on day 3. EEG signals were amplified, filtered and subsequently analysed using Somnologica. Post- hoc correction of epoch vigilance states was performed using the piezoelectric signal. Relative spectral power (Hanning window) was analysed in REM epochs and in selected continuous wake and NREM epochs for each day and spectral ratios plotted (Habitation / Vehicle and Compound / Vehicle).

All compounds tested induced sleep. DORAs had similar effects on the sleep-wake profile, increasing REM sleep and enhancing REM as a proportion of total sleep. In contrast, zolpidem reduced REM sleep, as expected. SB-649868, almorexant (150 mg/kg), filorexant (50 mg/kg) and zolpidem increased NREM sleep, while filorexant (100 mg/kg) and suvorexant had no effect on NREM sleep despite potent REM-enhancing effects. Wake and NREM spectral power were unaffected by DORAs, while zolpidem increased 5Hz, beta, gamma and decreased theta power during wake, relative to vehicle. During NREM, zolpidem enhanced delta while reducing theta power and higher frequencies. Interestingly, both filorexant and suvorexant enhanced theta power within REM sleep. In summary, the four clinically validated DORAs preferentially enhance REM sleep, suggestive of a class action of the DORA modality. Further studies are needed to investigate the effects of DORAs on enhanced REM sleep theta power and its consequences.

1. Winrow CJ & Renger JJ, Br J Pharmacol in press, 2013

2. Flores AE et al, IEEE Trans Biomed Eng 54:225, 2007