BACKGROUND: Neurons expressing gamma-aminobutyric acid (GABA) play an important role in the regulation of wakefulness to sleep, as well as the maintenance of sleep. However, the role of GABAergic neurons in the tuberomammillary nucleus (TMn), with regard to the sleep-wakefulness cycle, is poorly understood.OBJECTIVE: To investigate the effects of GABAergic neurons in the TMn on the sleep-wakefulness cycle.DESIGN, TIME AND SETTING: Randomized controlled study, performed at the Laboratory of Neurobiology, Department of Anatomy, Histology and Embryology, School of Basic Medical Sciences, Lanzhou University from July 2007 to February 2008.MATERIALS: Fifteen healthy, adult, male, Sprague Dawley rats were randomly divided into three groups(n = 5): control, ventrolateral preoptic area (VLPO) lesion, and VLPO lesion plus GABAA receptor antagonist-treated. Ibotenic acid and bicuculline were provided by Sigma (St. Louis, USA). METHODS: Four electroencephalogram screw electrodes were implanted into the skull at a frontal region (two) and parietal bones (two) on each side. Three flexible electromyogram wire electrodes were placed into the nuchal muscles. On day 8, a fine glass micropipette (10-20 mm tip diameter) containing ibotenic acid solution (10 nmol/L) was injected into the VLPO in both hemispheres following bone wax removal under anesthesia. One week after the second surgery, sleep-wakefulness states were recorded in rats from the VLPO lesion group. On day 10 after VLPO lesion, bicuculline (10 nmol/L), a GABAA-receptor antagonist, was microinjected into the TMn and sleep-wakefulness states were recorded for 24 hours.MAIN OUTCOME MEASURES: Duration of the sleep-wakefulness cycle in each group using a Data acquisition unit (Micro1 401 mk2) and Data collection software (Spike Ⅱ). RESULTS: VLPO lesion induced an increased duration of wakefulness (W, 13.17%) and light slow-wave sleep (SWS1, 28.9%), respectively. Deep slow-wave sleep (SWS2, 43.74%) and paradoxical sleep (PS, 44.07%) were respectively decreased for 24 hours at day 9 post-lesion, compared with pre-lesion (P< 0.01). Microinjection of bicuculline into the TMn following VLPO lesion at 10:00 am on the 10th day elicited a wake state for 40-55 minutes, with a latency of 15 minutes. However, 24-hour sleep-wake states demonstrated that the ratio of W and SWS1 were increased by 12.61% (P< 0.01) and 50.97% (P< 0.01), respectively. In addition, SWS2 and PS were decreased by 68.08% (P< 0.01) and 39.92% (P< 0.05), respectively, compared with prior to VLPO lesion. CONCLUSION: The evidence of decreased deep slow-wave sleep, which was induced by VLPO lesion, suggested that GABAergic neurons in the VLPO play an important role in maintaining sleep. Bicuculline microinjection into the TMn, following VLPO lesion, elicited wakefulness and sleep depression for 50 minutes, with contrary increased light slow-wave sleep for 24 hours, which suggested that GABAergic neurons in the TMn play a role in sleep drive (sleepiness) via local circuit to directly inhibit histaminergic neurons.