EFFECT OF MUSCIMOL MICROINJECTIONS INTO THE PREPOSITUS-HYPOGLOSSI ANDTHE MEDIAL VESTIBULAR NUCLEI ON CAT EYE-MOVEMENTS

1. For horizontal eye movements, previous observations led to the hypo thesis that the legendary neural integrator necessary for correct gaze holding, adequate vestibuloocular reflex (VOR), and optokinetic nysta gmus, was located in the region of the complex formed by the nucleus p repositus hypoglossi (NPH) and the medial vestibular nucleus (MVN). 2. The aim of the present study was to test the respective contributions of the NPH, of the rostral part of the MVN, which contains most secon d-order vestibular neurons, and of the central part of the MVN to the horizontal integrator.3. An injection of muscimol was used to inactiva te each of these three zones in the cat's brain. Muscimol is a gamma-a minobutyric acid (GABA) agonist. By binding to GABA(A) receptors, it i nduces a hyperpolarization of the neurons that nullifies their activit y. Muscimol was injected into the brain stem of the alert cat through a micropipette by an air pressure system. 4. The search coil technique was used to record spontaneous eye movements and the VOR induced by r otating a turntable at a constant velocity. VOR was analyzed by a new method: transient analysis of vestibular nystagmus. 5. A unilateral in jection of muscimol into the NPH induced a bilateral gaze-holding fail ure: saccades were followed by a centripetal postsaccadic drift. A ves tibular imbalance was also present but it was moderate and variable. T he VOR responses were distorted drastically. Through transient analysi s of vestibular nystagmus, that distortion was revealed to be due more to a failure of the neural integrator than to an alteration of the ve stibular input to the neural integrator. The responses to a rotation e ither toward the injected side or in the opposite direction were asymm etrical. The direction of that asymmetry was variable. 6. A unilateral injection of muscimol into the rostral part of the MVN caused a vesti bular imbalance: in complete darkness, a nystagmus appeared, whose lin ear slow phases were directed toward the side of injection. 7. A unila teral injection of muscimol into the central part of the MVN induced a syndrome where a severe bilateral gaze-holding failure was combined w ith a vestibular imbalance. In the light, saccades were followed by a bilateral centripetal postsaccadic drift. In complete darkness, a nyst agmus was observed, whose curved slow phases were directed towards the side of injection. The VOR responses were distorted drastically. Here again, that distortion was revealed by our analysis to be due more to a failure of the neural integrator than to an alteration of the vesti bular input to the neural integrator. The VOR response to a rotation t oward the side opposite to the injection side was always larger than t hat elicited by a rotation toward the injected side. 8. We conclude th at both the NPH and the adjacent central MVN are key anatomic substrat es for the horizontal neural integrator, whereas the rostral MVN is no t a key site for the horizontal neural integrator.