European Journal zyxwvutsrqp of Neuroscience, Vol. 6, pp. 173-186 zyxwvut 0 I994 European Neuroscience Association z Lamellar Organization of Pontocerebellar Neuronal Populations. A Multi-tracer and zyx 3-D Computer Reconstruction Study in the Cat Alfeo M. Nikundiwe’, Jan G. Bjaalie and Per Brodal Department of Anatomy, Institute of Basic Medical Sciences, University of Oslo, P.O. Box 1105 Blindern, N-0317 Oslo, Norway ’Permanent address: Department of Zoology and Marine Biology, University of Dar es Salaam, P.O. Box 35063, Dar-es-Salaam, Tanzania zyxwvutsrqpo Key words: pontine nuclei, cerebellum, fluorescent tracers, three-dimensional, neural network Abstract This study deals with the three-dimensional arrangement of populations of pontocerebellar cell bodies projecting to the parafloccular complex. The fluorescent tracers rhodamine B isothiocyanate, fluoro-gold and fast blue were injected in either adjacent or separated cerebellar folia. A set of coordinates zyxw (x, y, z) was assigned to each retrogradely labelled cell and the total distribution reconstructed and displayed on a graphics workstation. At a large scale, we found that the majority of the cells of each labelled population (all projecting to the same folium) were confined to a lamella-shaped tissue volume. Each lamella extended from medial to lateral, and accordingly followed the curving of the pontine grey around the corticospinal and corticobulbar fibre tracts. At a smaller scale, i.e. within each lamellar subspace, ihe neurons belonging to one labelled population were distributed in aggregates of various shapes. To enable further analysis of the shapes of the intralaminar aggregates, we developed a computer program for unfolding of the lamellae, based on cubic B-spline approximation. The flattened reconstructions were three-dimensional polygonal windows, circumscribing the large majority of the labelled cell swarm (usually 70-800/0 of the total number of labelled cells in one population). The present findings, taken together with previous data on a gradual, rather than disjunctive, shift of pontocerebellar neuronal position in relation to a gradual shift of target region (Bjaalie et a/., Anat. Rec., 231, 510 - 523, 1991), suggest that the cerebropontocerebellar system may be organized according to a set of fairly simple topographic rules. Introduction zyxwvutsrqp The pontine nuclei constitute the major source of mossy fibres to the cerebellar hemispheres. In cat there are - 400 OOO pontine neurons on each side (Bjaalie, in preparation), and only - 1 % of these appear to be interneurons (Brodal ef af., 1988). The only known efferent target of the pontine nuclei is the cerebellum. zyxwvuts Until the introduction of tracers for retrograde axonal transport, the organization of the pontocerebellar projection was known only in very broad outlines (for a discussion of the older literature see Brodal and Jansen, 1946). Thus, retrograde degenerative changes after lesions of the cerebellum appeared to be diffusely distributed, and only detectable after large cerebellar lesions (very unlike the distinct and circumscribed changes occurring in the inferior olive). Experiments with retrograde transport of horseradish peroxidase (and later other tracers) explained why the retrograde degeneration experiments only gave evidence of a diffuse pattern within the pontocerebellar projection (for references see Gould, 1980; Brodal, 1982; Brodal and Bjaalie, 1992). Thus, the pontocerebellar projection is strikingly convergent; each small part of the cerebellar cortex receives afferents from cells in extensive parts of the pontine nuclei. The distribution of retrogradely labelled cells in sections of the pontine nuclei even after a small cerebellar tracer injection appears to be very complex, and the three-dimensional arrangement of pontocerebellar cell groups is not clearly understood. This is witnessed by the use of various descriptive terms, such as ‘columns’, ‘clusters’, ‘bands’, ‘slabs’, ‘patches’, ‘mosaic’ etc. In spite of the wide distribution of pontine neurons projecting to a small part of the cerebellum, there must exist a relatively high degree of spatial order within the pontocerebellar projection since different cerebellar subdivisions receive afferents from differently placed pontine cell groups (Hoddevik et al., 1977; Brodal and Walberg, 1977; Eisenman, 1980; Eisenman and Noback, 1980; Mihailoff et al., 1980, 1981; Rosinaand F’rovini, 1981, 1987; Robinson et al., 1984; Aas and Brodal 1989; Brodal, 1979, 1982, 1987; Brodal and Bjaalie, 1992). General rules determining the mutual arrangement of cell groups with different cerebellar targets have so far not been formulated. Yet, it seems inconceivable that the position of each minor pontocerebellar cell group is individually specified with regard to its position in the pontine nuclei. Most likely, a limited set of general rules specify the position of the pontocerebellar neurons according to the position of their target. The main aim of the present study is to look Correspondence to: zyxwvutsrqponm Jan G. Bjaalie, as above Received 4 May, zyxwvutsrqponm 1993, revised 3 August 1993. accepted 17 September 1993