Unrelated course of subthalamic nucleus and globus pallidus neuronal activities across vigilance states in the rat Nadia Urbain, 1 Damien Gervasoni, 2 Fabienne Soulie Áre, 1 Letõ Âcia Lobo, 1 Nicolas Rente Âro, 1 Franc Ëois Windels, 3 Bernadette Astier, 1 Marc Savasta, 3 Patrice Fort, 2 Bernard Renaud, 1 Pierre-Herve  Luppi 2 and Guy Chouvet 1 1 Laboratoire de Neuropharmacologie et Neurochimie, INSERM U512 and 2 Laboratoire de Neurobiologie des Etats de Sommeils et d'Eveil, INSERM U480, Universite  Claude Bernard-Lyon 1, 8 avenue Rockefeller, 69373 Lyon cedex 08, France 3 Laboratoire de Neurobiologie Pre Âclinique, INSERM U318, Centre Hospitalier Universitaire de Grenoble, Universite  Joseph Fourier, 38043 Grenoble cedex 09, France Keywords: anaesthetic-free preparation, basal ganglia, EEG and EMG polygraphy, extracellular single-unit recordings, ®ring pattern analysis Abstract The pallido-subthalamic pathway powerfully controls the output of the basal ganglia circuitry and has been implicated in movement disorders observed in Parkinson's disease (PD). To investigate the normal functioning of this pathway across the sleep±wake cycle, single-unit activities of subthalamic nucleus (STN) and globus pallidus (GP) neurons were examined, together with cortical electroencephalogram and nuchal muscular activity, in non-anaesthetized head-restrained rats. STN neurons shifted from a random discharge in wakefulness (W) to a bursting pattern in slow wave sleep (SWS), without any change in their mean ®ring rate. This burst discharge occurred in the 1±2 Hz range, but was not correlated with cortical slow wave activity. In contrast, GP neurons, with a mean ®ring rate higher in W than in SWS, exhibited a relatively regular discharge whatever the state of vigilance. During paradoxical sleep, both STN and GP neurons increased markedly their mean ®ring rate relative to W and SWS. Our results are not in agreement with the classical `direct/indirect' model of the basal ganglia organization, as an inverse relationship between STN and GP activities is not observed under normal physiological conditions. Actually, because the STN discharge pattern appears dependent on coincident cortical activity, this nucleus can hardly be viewed as a relay along the indirect pathway, but might rather be considered as an input stage conveying corticothalamic information to the basal ganglia. Introduction Basal ganglia are currently viewed as a network that receives, processes and transmits back to the cortex, the signals necessary for the appropriate execution of movements (Albin etal., 1989; Alexander & Crutcher, 1990; DeLong, 1990). In the rat, the major entrance of this circuit, the striatum, receives excitatory afferents from the cortex and is connected to output nuclei by a direct and an indirect pathway (Albin etal., 1989; Gerfen & Wilson, 1996; Smith etal., 1998). The indirect pathway consists of striatal g-aminobutyric acid (GABA)ergic projections to the globus pallidus (GP), which in turn sends GABAergic efferents to the subthalamic nucleus (STN); the latter provides a glutamatergic input to the basal ganglia output neurons of the substantia nigra pars reticulata and the entopeduncular nucleus (Bevan etal., 1994a, b). It has been suggested that most of the motor symptoms observed in Parkinson's disease (PD) could result from an imbalance between the direct and indirect pathways, leading to an overactivity of the STN, itself consecutive to a decreased activity of GP neurons (DeLong, 1990). Accordingly, in monkeys treated with the dopaminergic neurotoxin MPTP (1-methyl- 4-phenyl-1,2,3,6-tetrahydropyridine), movement disorders are alle- viated by chemical lesion or high frequency stimulation of the STN (Bergman etal., 1990; Benazzouz etal., 1993, 1996). The latter treatment has been successfully applied to patients with PD (Limousin etal., 1995a,b, 1998; Pollak etal., 1996). However, the conceptual view of STN hyperactivity attributed to GP hypoactivity is presently challenged (Chesselet & Delfs, 1996; Fe Âger, 1997; Levy etal., 1997; Parent & Cicchetti, 1998), and it appears that other mechanisms might be involved (Hassani etal., 1996; Kreiss etal., 1997). Although several experimental factors may account for such a controversy, the in¯uence of anaesthesia has been, until now, relatively neglected. Indeed, general anaesthesia may interfere with a number of neurophysiological variables (Nicoll & Madison, 1982; Davies etal., 1997; Mihic & Harris, 1997). In order to avoid such an interference, some studies of the pallido-subthalamic pathway (Bergstrom etal., 1984; Pan & Walters, 1988; Kelland etal., 1995; Kreiss etal., 1996, 1997) have been performed on the awake, paralysed, locally anaesthetized and arti®cially ventilated rat. However, this preparation has been shown to be associated with profound metabolic disturbances (Saunier etal., 1991, 1993) that may alter the normal functioning of the indirect pathway. It was therefore of interest to investigate its normal functioning in rats free of any drug, and Correspondence: Dr N. Urbain, as above. E-mail: urbain@rockefeller.univ-lyon1.fr Received 29 December 1999, revised 22 May 2000, accepted 25 May 2000 European Journal of Neuroscience, Vol. 12, pp. 3361±3374, 2000 ã Federation of European Neuroscience Societies