Department of Psychiatry
Brockton VAMC and Harvard Medical School
Brockton, MA 02401, USA.
ver the past decade, converging lines of evidence have linked the occurrence of the Rapid Eye Movement (REM) phase of sleep to the activity of mesopontine cholinergic neurons. A subpopulation of these neurons selectively discharges just before and during REM sleep, and a variety of in vivo and in vitro experiments indicate they excite effector neurons in the pontine reticular neurons to produce the brainstem-generated signs of REM sleep. Attention has now turned to determining what modulates the activity of these cholinergic neurons. In Vitro studies in our lab have demonstrated that serotonin inhibits identified cholinergic neurons via an inwardly rectifying potassium current. We recently used In Vivo microdialysis application of 8-OH-DPAT to inhibit serotonergic neuronal activity in the dorsal raphe, a major source of serotonergic innervation of the cholinergic neurons; this inhibition, measured as a reduction in serotonin output, produced a 300% increase in REM sleep, consistent with a disinhibition of cholinergic REM-active neurons.
It is also known that other cholinergic neurons profoundly modulate the EEG activation seen in waking and REM sleep. These waking- and REM-active cholinergic neuronal populations are located in both the mesopontine and the basal forebrain (BF) cholinergic zones. In evaluating what might modulate these populations, especially in production of the increased sleep propensity following prolonged wakefulness, a metabolically produced increased concentration of adenosine seems one likely candidate. Our In Vitro study has demonstrated that both mesopontine and basal forebrain neurons identified as cholinergic are under a powerful tonic inhibitory control by endogenous adenosine. We are now in the process of evaluating adenosine influence using In Vivo microdialysis. Adenosine (300 mM) perfused into either the cholinergic BF or the mesopontine laterodorsal tegmental nucleus in the cat produced a dramatic decrease in waking, to about 50% of the baseline level. Furthermore, BF perfusion of S-(4-nitrobenzyl)-6-thioinosine (NBTI; 1mM), a potent adenosine uptake inhibitor also decreased wakefulness and increased non-REM (NREM) sleep. However, perfusion of NBTI in ventrolateral and ventroanterior thalamic nuclei did not affect behavioral state, although local adenosine concentration was increased to the same extent as with BF perfusion. This suggests that adenosine may have specific effects on cholinergic zones important in arousal. We have initial data indicating that spontaneous levels of extracellular adenosine in BF parallel metabolic activity levels, NREM=85% of waking=REM values, tend to increase following mild sleep deprivation and show a slow decline in recovery sleep. The REM sleep increase following BF (substantia inominata) perfusion of adenosine and NBTI suggests the presence of BF REM-suppressing activity that is inhibited by adenosine.
