Retinal projections to the subcortical nuclei in the Japanese field vole (Microtus montebelli)

Retinofugal projections in the mouse

The laboratory mouse is increasingly a subject for visual system investigation, but there has been no comprehensive evaluation of this species' visual projections. Here, projections were visualized and mapped following intraocular injection of cholera toxin B subunit. Tissue was processed using standard procedures applied to 30 μm free-floating sections with diaminobenzidine as the chromogen. The mouse retina projects to ~46 brain regions, including 14 not previously described in this species. These include two amygdaloid nuclei, the horizontal limb of the diagonal band, the paraventricular hypothalamic nucleus, several visual thalamic nuclei, the paranigral nucleus, several pretectal nuclei, and the dorsal cortex of the inferior colliculus. Dense retinal patches were also observed in a narrow portion of the ipsilateral intermediate layer of the superior colliculus. The superior fasciculus of the accessory optic tract, which innervates the medial terminal nucleus, was also determined to be a terminal zone throughout its length. The results are compared with previous descriptions of projections from mouse intrinsically photoreceptive retinal ganglion cells, and with data from the hamster, Nile grass rat, and laboratory rat. The retinal projection patterns are similar in all four species, although there are many differences with respect to the details. The specific visual functions of most retinorecipient areas are unknown, but there is substantial convergence of retinal projections onto regions concerned with olfaction and audition.

Aberrant retinal projections in congenitally deaf mice: how are phenotypic characteristics specified in development and evolution?

The contribution of sensory input to the formation of sensory system-specific (sensoritopic) connections of the thalamus and midbrain was investigated using mice lacking the Na+-K+-2Cl- cotransporter (NKCC1) or the plasma membrane Ca2+-ATPase isoform2 (PMCA2). Because these mice are congenitally deaf, the developing nervous system has no exposure to sensory-driven neural activity from the auditory system. Here we compared the retinofugal pathway in normal and congenitally deaf mice using intraocular injections of neuroanatomical tracers into each eye, and relating tracer patterns to identified thalamic nuclei and superior colliculus layers. We demonstrate that loss of such activity results in aberrant projections of the retina into nonvisual auditory structures such as the medial geniculate nucleus and the intermediate layers of the superior colliculus. These findings indicate that activity from peripheral sensory receptor arrays is necessary not only for the refinement of developing connections within a unimodal structure, but for the establishment of sensoritopic or sensory-specific connections of unimodal and multimodal structures. We hypothesize that specification of such connections may occur through the modulation of spatial expression patterns of molecules known to be involved in the development of topography of connections between brain structures, such as the ephrins, via activity-dependent, CRE-mediated gene expression.

Retinofugal projections in the short-tailed opossum (Monodelphis domestica)

In the current investigation, retinofugal projections to midbrain and thalamic nuclei of Monodelphis domestica were investigated using wheat-germ agglutinin conjugated to horseradish peroxidase (WGA-HRP). Large intraocular injections of WGA-HRP were placed into the eye, and patterns of labeled axon terminals were related to nuclear boundaries in tissue that was stained for Nissl or reacted for cytochrome oxidase (CO). Our results demonstrate that the major projection from the retina is to the contralateral dorsal lateral geniculate nucleus (LGNd) and the superior colliculus (SC). Connections were also observed with the contralateral pretectal nucleus (PRT), the lateral posterior nucleus (LP), and the ventral division of the lateral geniculate nucleus (LGNv). Ipsilateral connections were with the LGNv and LGNd. These findings are consistent with reports in other marsupials as well as with studies in a number of eutherian mammals. Thus, there appears to be a common pattern of retinofugal projections that all mammals share, probably due to retention from a common ancestor. However, some features such as a lack of ipsilateral input to the SC (which are absent only in certain species like Monodelphis, platypus, and echidnas) may represent a primitive state retained from a common ancestor. When comparisons of retinofugal connections and LGNd organization are made across taxa, three types of organization are observed: a homogenous LGNd with a high degree of binocular overlap of projections; a partially differentiated LGNd with some segregation of eye-specific inputs; and a fully segregated structure with a large degree of segregation of eye-specific inputs. We discuss the factors that contribute to the organization observed in extant mammals and conclude that phylogeny and lifestyle appear to be the underlying factors contributing to the organization of the LGNd.