The growth of cat cerebral cortex in postnatal life: a magnetic resonance imaging study.

To follow up the development of an individual brain over time and to measure its growth we have analysed the brains of individual cats from postnatal day 12 to adulthood using magnetic resonance imaging. From the anatomical images, four parameters were calculated: anteroposterior extent of the telencephalon, brain volume, neocortical surface area and neocortical volume. The development of the anteroposterior extent was similar in all cats. It increased between the 3rd and 6th postnatal week from 33 to 37.5 mm ending up approximately 40 mm in adulthood. The brain volume showed greater variability. On average, the volume increased from 11.5 to 16.5 cm3 in the same period. Adult values were approximately 19 cm3. Considerable interindividual variability was observed in neocortical surface area. In one cat, it expanded from 12.5 to 26 cm2 between days 14 and 41. In another cat, this area expanded from 16 to 24.5 cm2 between days 12 and 40. On average, the surface area expanded by 34% between the 3rd and 6th week. Adult values ranged from 27 to 30 cm2. Neocortical volume increased from 2.9 to 4.1 cm3 between the 3rd and 6th postnatal week and to 4.5-5.2 cm3 in adulthood. The asymmetry between the hemispheres in both neocortical surface area and volume was < 3% in all animals for most of the observation period. Comparison of the neocortical surface measurements with data on postnatal growth of cat primary visual cortex obtained by 2-deoxyglucose autoradiography indicates that the primary visual cortex grows at the same speed and amounts to approximately 15% of the entire neocortical surface area throughout development.

Neuroanatomical and neurophysiological consequences of strabismus: changes in the structural and functional organization of the primary visual cortex in cats with alternating fixation and strabismic amblyopia.

In recent years, evidence has accumulated indicating that long-ranging neuronal connections within the primary visual cortex (area 17) mediate the influences of context and experience, possibly also those of expectation. After early onset strabismus, the layout of these connections is massively modified: in strabismic but not in normally raised cats, horizontal connections extend primarily between neurons activated by the same eye. As a possible consequence of the modified circuitry, neuronal synchronization between different ocular dominance domains is also massively reduced. Thus, the inability of strabismics to combine the signals arriving from the two eyes into a single percept may be caused by these structural and functional changes. Strabismic amblyopia is also accompanied by significant modifications of intracortical associational interactions: corresponding to the psychophysical deficits, neurons driven by the normal eye displayed stronger synchronization of their responses than neurons dominated by the amblyopic eye.(1) These data demonstrated for the first time a clear neurophysiological correlate of strabismic amblyopia in area 17. They suggest that - similar to our observations in divergent squinters - at least some of the perceptual deficits of amblyopic patients are due to experience-dependent changes in intracortical circuitry. We analyze this question by combining optical imaging of intrinsic signals with 3-D reconstructions of neuronal circuitry.

Optical imaging of orientation and ocular dominance maps in area 17 of cats with convergent strabismus.

Strabismus (or squint) is both a well-established model for developmental plasticity of the brain and a frequent clinical symptom. While the layout and topographic relationship of functional domains in area 17 of divergently squinting cats has been analyzed extensively in recent years (e.g. Löwel et al., 1998), functional maps in convergently squinting animals have so far not been visualized with comparable detail. We have therefore investigated the functional organization of area 17 in adult cats with a surgically induced convergent squint angle. In these animals, visual acuity was determined by both behavioral tests and recordings of visual evoked potentials, and animals with comparable acuities in both eyes were selected for further experiments. The functional layout of area 17 was visualized using optical imaging of intrinsic signals. Monocular iso-orientation domains had a patchy appearance and their layout was different for left and right eye stimulation, so that segregated ocular dominance domains could be visualized. Iso-orientation domains exhibited a pinwheel-like organization, as previously described for normal and divergently squinting cats. Mean pinwheel density was the same in the experimental and control animals (3.4 pinwheel centers per mm2 cortical surface), but significantly (P < 0.00001) higher than that reported previously for normal and divergently squinting cats (2.7/mm2). A comparison of orientation with ocular dominance maps revealed that iso-orientation domains were continuous across the borders of ocular dominance domains and tended to intersect these borders at steep angles. However, in contrast to previous reports in normally raised cats, orientation pinwheel centers showed no consistent topographical relationship to the peaks of ocular dominance domains. Taken together, these observations indicate an overall similarity between the functional layout of orientation and ocular dominance maps in area 17 of convergently and divergently squinting cats. The higher pinwheel densities compared with previous reports suggest that animals from different gene pools might generally differ in this parameter and therefore also in the space constants of their cortical orientation maps.

Partial regeneration and long-term survival of rat retinal ganglion cells after optic nerve crush is accompanied by altered expression, phosphorylation and distribution of cytoskeletal proteins.

In a screen to identify genes that are expressed differentially in the retina after partial optic nerve crush, we identified MAP1B as an up-regulated transcript. Western blot analysis of inner retina protein preparations confirmed changes in the protein composition of the microtubule-associated cytoskeleton of crushed vs. uncrushed nerve. MAP1B immunoreactivity and transcript levels were elevated for two weeks after crush. Immunostaining and Western blots with monoclonal antibodies directed against developmentally regulated phosphorylation sites on MAP1B revealed a gradient of MAP1B phosphorylation from the proximal optic nerve stump to the soma of retinal ganglion cells. Most interestingly, using antibodies directed against developmentally regulated phosphorylation sites on MAP1B, we observed that a significant number of crushed optic nerve axons develop MAP1B-immunopositive growth cones, which cross the crush site and migrate along the distal nerve fragment. In parallel, an abnormal distribution of highly phosphorylated neurofilament protein (pNF-H) in the cell soma and dendrites of presumably axotomized retinal ganglion cells was observed following partial nerve crush. This redistribution is present for the period between day 7 and 28 postcrush and is not seen in cells that stay connected to the superior colliculus. Axotomized ganglion cells, which contain pNF-H in soma and dendrites appear to have been disconnected from the colliculus at an early stage but survive axonal trauma for long periods.

GABA-inactivation attenuates colinear facilitation in cat primary visual cortex.

Neurons in primary visual cortex (V1) respond preferentially to stimuli of a particular orientation falling within a circumscribed region of visual space known as their receptive field (RF). However, the response to an optimally oriented stimulus presented within the RF can be enhanced by the simultaneous presentation of co-oriented, co-linearly aligned flank stimuli falling outside the RF which, when presented alone, fail to activate the cell. This type of contextual effect, termed colinear facilitation, presumably forms the physiological substrate for the integration of the line elements of a contour and the perceptual saliency of a contour in a complex environment. Here we show that colinear facilitation in single cells of cat area V1 can be substantially reduced or abolished by focal inactivation of laterally remote cells in the same area which respond strongly to the co-oriented, colinear flank stimulus inducing the facilitatory effect. The results provide evidence that horizontal intrinsic connections between cells with co-oriented and co-linearly aligned RFs make a major contribution to colinear facilitation in V1. They imply that the neuronal circuitry underlying contour integration and saliency is already present at the earliest stage of visual cortical information processing.