Abstracts

Rationale: The supramammillary region (SuM) is known to demonstrate theta frequency activity, innervates the hippocampus and is part of a circuit responsible for hippocampal theta activity. Posterior hypothalamic lesions result in somnolence, and we hypothesized that the SuM region, well-placed to activate the forebrain, with terminals in the lateral hypothalamus, basal forebrain, septal area and cerebral cortex, might be wake-promoting, as well exerting control of hippocampal theta activity. Methods: We manipulated activity in the SuM area with a combination of pharmacogenetic, optogenetic and conditional deletion experiments. For in vivo experiments, mice were injected with a vector transducing to an excitatory desginer receptor (human modified muscarinic 3, hM3) in an anatomically discrete and Cre-dependent manner, in vgat-Cre or vgltu2-Cre mice. A Cre-lox technique was then used to selectively delete either GABA or glutamate release, while also inserting hM3. We made prolonged recordings of epidural EEG, hippocampal field potentials, EMG and video to examine the effects of the otherwise inert hM3 agonist (clozapine-N-oxide, CNO), on behavior, state of arousal, EEG and hippocampal activity. In further experiments, after vector injection and transduction of channelrhodopsin in the SuM, we recorded dentate gyrus pyramidal cells, in vitro, to examine transmitter release from SuM neurons. Results: Pharmacogenetic activation of glutamate- or GABA-containing neurons of the SuM similarly and potently promoted very prolonged normal wakeful behavior (time to NREM sleep in vglut2-Cre: NaCl 22m ± 3m49s v. CNO 8h54m ± 53m56s, n = 9, p < 0.01; vgat-Cre: NaCl 20m ± 2m53s v CNO 5h40m ± 18m9s, n = 5, p = 0.014), gamma band cortical activity and hippocampal theta rhythm (spectral peak ~ 9Hz). Given the same effect of activation of both putative neuronal types and previous anatomical studies¹, we hypothesized, and demonstrated, in vitro, that a subpopulation of paramedian supramammillary neurons release both neurotransmitters monosynaptically. Activation of SuM neurons after the deletion of glutamate release resulted in marked attenuation of the wake-promoting effect (CNO sleep onset 1h0m ± 5m5s), while one spontaneous seizure during recording and delayed mortality (4 of 10 mice), consistent with the time course of GABA-release deletion, suggest that this manipulation may result in seizures. Conclusions: The SuM not only modulates hippocampal theta, but potently promotes wakefulness, with activation of the cerebral cortex and hippocampus. This region is reciprocally connected with other key components of the arousal system and the frontal cortex, suggesting that this may be a key node that is susceptible to both the top-down and bottom-up activation by cognitive networks as well as other components of the arousal system. Interestingly, a subpopulation of these neurons release both glutamate and GABA, with glutamate likely responsible for wake-promoting effects. Deletion of GABA from these neurons may result in seizures.