Development of thalamocortical pathways

When Torston Wiesel presented his Nobel address, he pointed out that his and David Hubel's pioneering work on the effects of sensory deprivation on the development of the functional architecture of the visual cortex had been inspired by early eighteenth century speculation that congenital blindness resulted in defective visual perception, presumably as a result of malformation of the brain (Von Senden, 1960). Earlier, Hubel and Wiesel had demonstrated that adjacent territories of area 17 receive input alternatively from the right and left eye. These so-called ocular dominance columns had been explored both by single-unit recording in the cortex, as well as by injections of anterograde tracers in the eye, followed by transsynaptic transport of the tracer to area 17 (Hubel and Wiesel, 1962, 1977). To investigate the role of the sensory periphery, they carried out similar experiments in normal kittens as well as kittens having one eye, closed throughout development. These experiments suggested that, during normal development, the ocular dominance columns emerge from an immature stage in which input from the two eyes overlaps. If one eye is closed during development, then only the experienced eye drives most neurons in the adult (Wiesel and Hubel, 1963). Transsynaptic labeling has shown that, following monocular deprivation, there is an expansion of the columns receiving input from the open eye and a reduction of the columns receiving input from the deprived eye (LeVay et al., 1980). These experiments gave rise to the concept that, during normal development, the geniculate afferents conveying responses to the right and left eye initially are extensive and, therefore, overlap. Because binocular deprivation allowed segregation of the two sets of inputs while segregation was prevented by blockade of activity (Stryker and Harris, 1986), it was thought that the formation of ocular dominance columns depended on competitive interactions between inputs from both eyes during a so-called critical period. These experiments emphasized the role of sensory experience in shaping neuronal connections and defined a time window during which sensory experience has a profound effect on the development of the brain.

Although experiments involving monocular deprivation clearly show that the brain is susceptible to deprivation, they fail to elucidate the role of experience during normal development. Further, they have failed to provide insight into the normal process of development at the cellular level. It was assumed for many years that the segregation of ocular dominance columns was attributable to pruning of the thalamic axonal arbor in layer 4 of the primary visual cortex. However, few experiments have actually attempted to examine this issue. When single geniculostriate axonal arbors have been examined at different developmental stages, the main event that is documented is the progressive increase in total axon length and complexity of axonal arbors (Antonini and Stryker, 1993; Friedlander and Martin, 1989). Pruning of extended branches either does not happen, or happens very rarely.

The role of visual experience in constructing ocular dominance columns was challenged by a study involving newborn monkeys. After prolonged in utero development and, therefore, in the absence of stimulation of the retina, the left and right eye inputs to the cortex were found to be well segregated into ocular dominance columns (Horton and Hocking, 1996; Rakic, 1976). This led to the suggestion that spontaneous activity, possibly originating in the periodic waves of excitation in the immature retina, drove segregation (Galli and Maffei, 1988; Meister et al., 1991; Mooney et al., 1993; Wong et al., 1993). However, a number of studies have suggested that the development of ocular dominance columns can actually proceed relatively normally in the total absence of the retina. Cytochrome oxidase-rich blobs in layers 2/3 in the monkey relate to the ocular dominance columns (Hendrickson, 1985; Hendry and Yoshioka, 1994) and have been shown to develop in the absence of the retina (Dehay et al., 1989; Kennedy et al., 1990; Kuljis and Rakic, 1990) and to show a normal periodicity (Kennedy et al., 1990). More direct evidence for the formation of ocular dominance columns in the absence of the retina has recently been obtained from studies of neonatal enucleated ferrets in which tracer injection in the lateral geniculate nucleus (LGN) revealed alternating stripes of label in the striate cortex (Crowley and Katz, 1999). Not only did ocular dominance columns appear to develop independently of the retina, but they were shown to appear well before the critical period and the period in which geniculostriate axons had been thought to undergo refinement and retraction (Crowley and Katz, 2000). Further, geniculate axons formed ocular dominance columns shortly after innervation of layer 4, and at this early stage, imbalance caused by removal of one eye fails to change the periodicity of the input to the cortex (Crowley and Katz, 2000). Although these results do not exclude a role for the sensory periphery in central development, they suggest that molecular cues play a major role in the formation of ocular dominance columns.

The influence of visual experience on the development of orientation columns has a long and controversial history. Unlike ocular dominance columns, where experimental manipulations quickly establish developmental plasticity, the effects of deprivation on the development of orientation columns has been a highly disputed issue (Blakemore and Cooper, 1970; Hirsch and Spinelli, 1970; Stryker and Sherk,

1975). The issue has recently been reexamined using optical imaging of orientation columns during development in normal and deprived animals. This work shows that, in agreement with earlier observations, orientation selectivity develops independently of visual experience, but neuronal activity is required to fine-tune and maintain the orientation map into adulthood (Crair et al., 1998; Sengpiel et al., 1999; White et al., 2001; Wiesel and Hubel, 1974). Although activity may not be necessary for the development of orientation columns, cross-modal rewiring experiments strongly suggest that visual afferents can create orientation columns in the auditory cortex (Sharma et al., 2000).

The emerging consensus on the development of the functional architecture of the cortex is that the construction of both orientation and ocular dominance columns is largely activity independent, but that the fine-tuning and maintenance of columns during the critical period depends on activity. These findings imply that early development of the columnar organization relies on molecular markers that may tag columns in the early phase of thalamic innervation of the cortex and that may share certain features with molecules known to be expressed in the embryonic cortex (Donoghue and Rakic, 1999; Flanagan and Vanderhaeghen, 1998).

0 0

Post a comment