The role of GABAergic modulation of mesopontine cholinergic neurotransmission in rapid eye movement (REM) sleep

The complexity of the pontine neuronal mechanisms that control and modulate expression of rapid eye movement (REM) sleep has been only recently recognized. In this review of some of these mechanisms, results from studies in the cat and rat are summarized and compared, and the differences between neuronal populations that might be effective, modulatory, or executive for REM sleep control are considered. One of the principal pontine areas that has been found to be important for the induction of REM sleep is located in the nucleus pontis oralis. Another is located in a more dorsocaudal area. The latter, designated the peri-locus coeruleus α in the cat, corresponds most closely with the sublaterodorsal (SLD) nucleus in the rat. In both areas in the cat, cholinergic mechanisms are important for REM sleep induction, whereas in the rat, only the pontis oralis area is cholinergically sensitive. In contrast, the SLD is responsive to glutamatergic processes. The emerging recognition of the importance of GABAergic modulation of these sites in both species is reviewed, and from this work, some preliminary conclusions can be drawn. In the caudal pontis oralis of rat, GABAergic mechanisms can affect the release of acetylcholine, inferring that transmitter release in a terminal region can be decoupled from impulse flow. This is important in light of the concept of changes in the discharge rate of reciprocally interacting populations of neurons being correlated with differences in vigilance state. A second conclusion, based on the work in the SLD, is that network interactions are critical for REM sleep control. The latter can be extended to conclude that activation of neurons that are effective for at least two of the signs of REM sleep may be sufficient to induce the state. Distributed networks would obviate the need for a single locus of executive control for REM sleep. But another possibility derived from the SLD studies is that executive control of REM sleep may actually represent the coordination of effective mechanisms, explaining the difficulty that has been encountered in finding a REM sleep generator.