The organization of the thalamostriatal projection from the lateral posterior thalamic nuclear complex (LP) in the pigmented rat.

We investigated the thalamostriatal projection of the rat using biotinylated dextran amine (BDA) and wheat-germ agglutinin horseradish peroxidase (WGA-HRP). To obtain the patch/matrix compartments of the striatum (ST), we used mu-opioid receptor (MOR) immunoreaction labeling. Thus, an MOR-positive 'patch' was indicated by a darkly stained spot, while the MOR-negative 'matrix' was displayed as a non-immunoreactive region. A small injection of BDA was made in a subregion of the lateral posterior thalamic nucleus (LP). The LP-ST fibers originated in all subregions of LP and terminated in the dorsocaudal portion of ST, where the corticostriatal fibers from the visual cortex terminate (Serizawa et al. 1994). These LP-ST fibers and terminals were concentrated in the MOR-negative matrix compartment. Electron microscopic observations showed that the LP-ST terminals made asymmetrical synaptic contacts mainly (70%, n = 30) with the dendritic spines of the presumptive ST-output neurons, and fewer (30%) contacted dendritic shafts. The present results provide anatomical support for the contention that ST-output spiny neurons of the matrix that project to the pars reticulata of the substantia nigra or globus pallidus, may be influenced directly by the LP-ST projection.

Extrinsic and intrinsic connections of the cat's lateral suprasylvian visual area.

The lateral suprasylvian visual area (LS) is known to have numerous interconnections with visual cortical areas as well as with subcortical structures implicated in visually-guided behaviors. In contrast, little data is available regarding connections within the LS itself. In order to obtain information about intra-areal connections and to re-investigate LS connectivity with various cortical and subcortical areas, the traces (biocytin or WGA-HRP) was injected into various loci along the medial and lateral banks of the LS. The anterograde tracer, biocytin injections into both medial and lateral bank produced label contained within the respective bank that extended rostrally and caudally from the infection site. In addition, following medical bank injections, considerable label was distributed throughout the fundus and, to a lesser extent, in the lateral bank. In contrast, no label could be detected in the medial bank after lateral bank injections, and, although label was observed in the fundus, it was restricted to the most lateral aspects. Moderate labeling could be observed in the medial bank following the tracer injection into the most rostral aspect of the lateral bank. It is likely that input derived from various visual cortical areas which project to the medial bank of the LS has access to this intra-areal circuitry. This may provide a route by which visual cortical information can be relayed to other cortical and subcortical structures involved in visually-guided behaviors such as the anterior ectosylvian visual cortex, striatum, and the deep layers of the superior colliculus, despite the fact that these structures themselves do not receive substantial direct projections from the visual cortical areas that are associated with the medial bank. Examination of the laminar location of the cells-of-origin of striate and extrastriate projections to LS using retrograde trace, WGA-HRP, revealed that the supragranular laminae of areas 17, 18 and 19 were the source of LS afferents whereas afferents from the other cortical areas (e.g., 20a, 20b, 21a, 21b, 7 and anterior ectosylvian visual area) were from both supra- and infragranular laminae. In addition, all LS subregions received intra-areal afferent projections from all LS cortical laminae. Thus, although rather clear hierarchical relationship between LS and visual cortical areas appears to exist, the interconnections among LS subregions provide no clear evidence of simple hierarchical relationships between regions LS or may have feed-forward and feed-back pathways.