Reticulo-spinal neurons participating in the control of synergic eye and head movements during orienting in the cat

Summary

Previously we described physiological properties of pontine reticulo-spinal neurons which generate bursts and decaying tonic discharges related to eye movements and neck muscle activity during ipsiversive gaze shifts (Grantyn and Berthoz 1987). Two of these “eye-neck reticulo-spinal neurons” (EN-RSN) were labeled by intra-axonal injections of HRP. The present report provides a detailed description of their morphology with an emphasis on the topography of axon collaterals, bouton numbers, and the structure of preterminal ramifications in different target areas. The cell bodies of labeled EN-RSNs were located rostro-ventrally to the abducens nucleus. Their descending axons issued 8 and 13 collaterals (left and right EN-RSN, respectively) at different rostro-caudal levels, between the abducens nucleus and the pyramidal decussation. On the basis of the size of their cell bodies, the isodendritic type of dendritic branching and their multiple collateralization, EN-RSNs correspond to the class of “generalized” reticular neurons, often referred to as The Scheibels' neurons. Collaterals of EN-RSNs terminated in the following structures: the abducens and facial nuclei, the medial and lateral vestibular nuclei, the nn. prepositus and intercalatus, and the bulbar reticular formation. As judged from bouton numbers, the strongest connection of both neurons was with the abducens nuclei. Terminations in the rostral part of the medial vestibular and prepositus nuclei indicate that EN-RSNs may also influence oculomotor output activity through these indirect routes. In the facial nucleus, a majority of terminations was found in its medial subdivision containing motoneurons of ear muscles. However, other subdivisions were also contacted by EN-RSNs. Most terminations in the rostral bulbar reticular formation are distributed to the dorsal, gigantocellular field. Within this field, there is a substantial contribution to the zone characterized by the highest density of reticulo-spinal neurons projecting directly to neck motoneurons. Other target areas which may participate in the modulation of spinal cord activity by EN-RSNs are the ventral reticular nucleus in the caudal medulla and the lateral vestibular nucleus. EN-RSNs also establish connections with precerebellar structures: the prepositus and the paramedian reticular nuclei. The numbers of boutons on collaterals issued within 6 mm of the injection site varied between 37 and 469. The occurrence of presumed axo-somatic contacts was low (0–8.2%) and not characteristic for any particular target area. Local accumulations of boutons in the form of small and large field clusters was a common observation. The variations of tendency to clustering appeared to reflect both the properties of target areas and of individual collaterals. Bouton counts on transsynaptically labeled neurons prove that connections of EN-RSNs are not uniform. They may be exceptionally strong with some selected target neurons, a fact which must be taken into account in models of reticular networks.