Local differences in the amount of early cell death in neocortex predict adult local specializations.

The amount of early cell loss in five neocortical areas was inversely related to adult numbers of neurons in those areas. Differential cell death predicted particularly the thickness of the upper cortical laminae; it was not related to neuron numbers in the lower laminae. Cell loss thus determines some features of local neocortical differentiation.

Linked regularities in the development and evolution of mammalian brains.

Analysis of data collected on 131 species of primates, bats, and insectivores showed that the sizes of brain components, from medulla to forebrain, are highly predictable from absolute brain size by a nonlinear function. The order of neurogenesis was found to be highly conserved across a wide range of mammals and to correlate with the relative enlargement of structures as brain size increases, with disproportionately large growth occurring in late-generated structures. Because the order of neurogenesis is conserved, the most likely brain alteration resulting from selection for any behavioral ability may be a coordinated enlargement of the entire nonolfactory brain.

Early removal of one eye reduces normally occurring cell death in the remaining eye.

During normal development of the hamster eye, there is a substantial loss of cells from the retinal ganglion cell layer in the first two postnatal weeks. If one eye is lost at birth, this cell death is reduced in the remaining eye. This may account for the increased ipsilateral projection from this eye to the thalamus and midbrain observed in these animals.

The limbic system in Mammalian brain evolution.

Previous accounts of mammalian brain allometry have relied largely on data from primates, insectivores and bats. Here we examine scaling of brain structures in carnivores, ungulates, xenarthrans and sirenians, taxa chosen to maximize potential olfactory and limbic system variability. The data were compared to known scaling of the same structures in bats, insectivores and primates. Fundamental patterns in brain scaling were similar across all taxa. Marine mammals with reduced olfactory bulbs also had reduced limbic systems overall, particularly in those structures receiving direct olfactory input. In all species, a limbic factor with olfactory and non-olfactory components was observed. Primates, insectivores, ungulate and marine mammals collectively demonstrate an inverse relationship between isocortex and limbic volumes, but terrestrial carnivores have high relative volumes of both, and bats low relative volumes of both. We discuss developmental processes that may provide the mechanistic bases for understanding these findings.

Changes in synaptic density after developmental compression or expansion of retinal input to the superior colliculus.

The retinal projection to the superior colliculus can be made abnormally dense by inducing a "compressed" retinal projection into a subnormal tectal volume, or abnormally sparse by monocular enucleation early in development. Any or all of the features of cell number, axonal arbor, dendritic arbor, and synaptic density could potentially be adjusted to compensate for such variations in the convergence of one cell population on another. We have examined the consequences of neonatal partial tectal ablation or monocular enucleation for synaptic length, density, and relative numbers of synapse classes in the superficial gray layer of the hamster superior colliculus. Monocular enucleation resulted in a reduction of synaptic density in the superficial gray layer of the colliculus ipsilateral to the remaining eye. This decrease in density was entirely accounted for by a reduction of the number of synapses with round vesicles, large asymmetric terminal specializations, and pale mitochondria characteristic of retinocollicular terminals (RLP synapses). There was no compensatory increase in any other synaptic class. RLP synapses were larger in monocular enucleates. Partial tectal ablation had no effect on synaptic density, nor on the relative proportions of different synaptic types. Synapses of the RLP class were slightly smaller than normal. These results suggest that synaptic density is normally at a maximum that cannot be altered by increases in potential input. However, density may be reduced by decreasing the number of inputs. Terminal classes do not appear to compete with each other within the collicular volume, suggesting that postsynaptic cells controls both the classes and numbers of their potential inputs.

Cell death in the mammalian visual system during normal development : I. Retinal ganglion cells.

Degenerating cells may be observed with light microscopy in the hamster retinal ganglion cell layer during early postnatal development. On the first postnatal day, degenerating cell profiles were found at a rate of 2.7 per 1,000 live cells. This rate increased to a peak of 14.7 degenerating cells per 1,000 live on postnatal day 5 and then slowed to 4.2 per 1,000 live by postnatal day 10. These rates of cell death correspond to a 49% reduction in cell number in the ganglion cell layer. Examination of the spatial pattern of cell death revealed that although on visual inspection degenerating cells appear to occur in clumps, statistical analysis demonstrated a random distribution within renal areas. Across the retina, cell death rates were higher in peripheral retina than in central retina. The timing and pattern observed correspond well with that of cell degeneration observed in the superficial layers of the superior colliculus, the major target of the retinal projection.

Developmental changes in the distribution of retinal catecholaminergic neurones in hamsters and gerbils.

Although Syrian hamsters and Mongolian gerbils are closely related, they have quite different patterns of retinal ganglion cell distribution and different patterns of retinal growth that produce their distributions. We have examined the morphology and distribution of catecholaminergic (CA) neurones in adult and developing retinae of these species in order to gain a more general understanding of the mechanisms producing cellular topographies in the retina. CA neurones were identified with an antibody to tyrosine hydroxylase (TH), the rate limiting enzyme in the production of catecholamines. In adult retinae of both hamsters and gerbils, most CA somata were located in the inner part of the inner nuclear layer (INL) and CA dendrites spread in a outer stratum of the inner plexiform layer (IPL). Their somata varied with retinal position, being largest in temporal and smallest in central retina. In hamsters, but not gerbils, a small number of CA interplexiform cells was also observed. In development, CA somata of hamster retinae were observed first in the middle and/or scleral regions of the cytoblast layer (CBL) at P (postnatal day) 8. By P12, CA somata were commonly located in the inner part of the INL and their dendrites spread into the outer region of the IPL. In developing gerbil retinae, CA somata were first observed at P6 in the middle of the CBL. Over subsequent days, they migrated into the inner part of the INL and spread their dendrites into the outer strata of the IPL. In both hamsters and gerbils, CA cells were initially concentrated in the superior temporal margin of the retina. In hamsters, this supero-temporal concentration persisted until adulthood, whereas in adult gerbils, the greatest density of CA cells was found just superior to the visual streak. These distributions were distinct from those of the ganglion cells in adult and developing retinae of each species. We discuss the role of maturational expression of TH, cell death, and retinal growth in the generation of the distinct distribution of the CA cells.

The early development of thalamocortical and corticothalamic projections.

The early development of thalamocortical and corticothalamic projections in hamsters was studied to compare the specificity and maturation of these pathways, and to identify potential sources of information for specification of cortical areas. The cells that constitute these projections are both generated prenatally in hamsters and they make reciprocal connections. Fluorescent dyes (DiI and DiA) were injected into the visual cortex or lateral geniculate nucleus in fixed brains of fetal and postnatal pups. Several issues in axonal development were examined, including timing of axon outgrowth and target invasion, projection specificity, the spatial relationship between the two pathways, and the connections of subplate cells. Thalamic projections arrive in the visual cortex 2 days before birth and begin to invade the developing cortical plate by the next day. Few processes invade inappropriate cortical regions. By postnatal day 7 their laminar position is similar to mature animals. By contrast, visual cortical axons from subplate and layer 6 cells reach posterior thalamus at 1 day after birth in small numbers. By 3 days after birth many layer 5 cell projections reach the posterior thalamus. On postnatal day 7, there is a sudden increase in the number of layer 6 projections to the thalamus. Surprisingly, these layer 6 cells are precisely topographically mapped with colabeled thalamic afferents on their first appearance. Subplate cells constitute a very small component of the corticothalamic projection at all ages. Double injections of DiI and DiA show that the corticofugal and thalamocortical pathways are physically separate during development. Corticofugal axons travel deep in the intermediate zone to the thalamic axons and are separate through much of the internal capsule. Their tangential distribution is also distinct. The early appearance of the thalamocortical pathway is consistent with an organizational role in the specification of some features of cortical cytoarchitecture. The specific initial projection of thalamocortical axons strongly suggests the recognition of particular cortical regions. The physical separation of these two pathways limits the possibility for exchange of information between these systems except at their respective targets.

Duration of retinogenesis: its relationship to retinal organization in two cricetine rodents.

The Mongolian gerbil (Meriones unguiculatus) has a prolonged period of development relative to other muroid rodents. We have explored the consequences of this relatively long period of maturation on retinal cell number and topography by comparing the duration and topography of neurogenesis in the gerbil retina with that of a closely related species which develops rapidly, the Syrian hamster (Mesocricetus auratus) (Sengelaub et al.: J. Comp. Neurol. 246:527-543, 1986). An analysis of thymidine-labeled retinas indicate that cells destined for the gerbil retinal ganglion cell layer are generated for at least 12 embryonic days, twice the duration in the hamster. The period of cell loss in the gerbil retinal ganglion cell layer extends for at least 14 postnatal days, more than twice as long as in the hamster. The gerbil retina is generated in a center-to-periphery gradient for both retinal ganglion cells and displaced amacrine cells, while no such gradients are evident in the hamster retina. We conclude that the longer developmental period of the gerbil is associated with 1) a longer period of neurogenesis resulting in greater retinal cell number, 2) the expression of spatial gradients in neurogenesis, and 3) a larger eye at maturity. The last two factors, in part, may be related to the development of a highly differentiated area centralis and visual streak in the retina of this rodent. Unrelated to duration of growth, early differences in retinal shape between these two species contributes to the development of retinal topography. The gerbil, but not the hamster retina, is initially asymmetric, longer in its nasotemporal than its dorsoventral dimension. The gerbil retina then grows asymmetrically, producing a spherical retina, and coincident in time, a nasotemporally extended visual streak.

Altered development of visual subcortical projections following neonatal thalamic ablation in the hamster.

Previous research has demonstrated that precise patterns of axonal connectivity often develop during a series of stages characterized by pathfinding, target recognition, and address selection. This last stage involves the focusing of projections to a precisely defined region within the target. Because thalamic projections begin to innervate cortex before the latter stages are reached, these projections may be important in the establishment of adult-like patterns of cortical connectivity. To address this issue, we examined the mature corticopontine and corticospinal projections of visual cortex deprived of early thalamic input by visual thalamic ablation. Although ablations on the day of birth in hamsters did not disrupt the targeting of appropriate subcortical structures by visual cortical axons, they did alter the organization of projections within the basilar pons and spinal cord. The density and spread of visual corticopontine connections in lesioned animals was greatly increased relative to unlesioned animals, suggesting that thalamic afferents are required during address selection, when the topographic specificity of projections is established. To determine whether early visual thalamic ablation increases connectivity by stabilizing an exuberant developmental projection, we examined the normal development of visual corticopontine connections in hamsters ages postnatal days 1-17 (P1-P17). From the earliest ages, visual cortical axons innervate the pontine nucleus in regions specific to their adult projection zones and show progressive growth within these zones. At no time during development do projections exist that are equivalent to the projections found after thalamic ablation, suggesting that removal of thalamic input does not simply stabilize a developmental projection.

Cell generation, death, and retinal growth in the development of the hamster retinal ganglion cell layer.

During the early postnatal period in the hamster, the retinal ganglion cell layer grows, establishes its central connections, and undergoes substantial cell loss. In this study, we describe the development of the retinal ganglion cell layer with particular attention to the creation of local specializations in cell density. Changes in the number and spatial distribution of cells identified by a single 3H thymidine injection were examined through the period of maximal cell loss (postnatal days 4-10) and at adulthood. The cells of the retinal ganglion cell layer are generated from embryonic day 10 to postnatal day 3. Overall, cell number in the ganglion cell layer increases by approximately 108,000 cells (223%) from postnatal day 1 to 5, because of continued migration of cells generated prenatally. Cell number decreases from postnatal day 5 to 10 (25%), coincident with the presence of degenerating cells. Cell type is correlated with day of generation: the largest cells, all having retinal ganglion cell morphology, are generated on embryonic days 10 and 11; intermediate-sized cells predominantly of ganglion cell morphology on embryonic day 12; and smaller cells of displaced amacrine or glial cell morphology thereafter. At adulthood, the hamster retina shows a streaklike elevation of cell density through central retina. However, at the time of maximal cell number (postnatal day 5), cell density is uniform across the retina. During the period of cell degeneration, cells are lost in greater relative numbers from the retinal periphery. This cell loss occurs principally from the first-generated cells (embryonic days 10 and 11), as shown by both changes in the distribution of labeled cells and by the spatial pattern of labeled degenerating cells. From postnatal day 10 to adulthood, relative cell density continues to decline in the periphery of the retina, thus suggesting that differential growth completes the production of the adult cell density distribution.

Regulation of retinal ganglion cell axon arbor size by target availability: mechanisms of compression and expansion of the retinotectal projection.

The ability of pre- and postsynaptic populations to achieve the proper convergence ratios during development is especially critical in topographically mapped systems such as the retinotectal system. The ratio of retinal ganglion cells to their target cells in the optic tectum can be altered experimentally either by early partial tectal ablation, which results in an orderly compression of near-normal numbers of retinal projections into a smaller tectal area, or by early monocular enucleation, which results in the expansion of a reduced number of axons in a near-normal tectal volume. Our previous studies showed that changes in cell death and synaptic density consequent to these manipulations can account for only a minor component of this compensation for the population mismatch. In this study, we examine other mechanisms of population matching in the hamster retinotectal system. We used an in vitro horseradish peroxidase labeling method to trace individual retinal ganglion cell axons in superior colliculi partially ablated on the day of birth, as well as in colliculi contralateral to a monocular enucleation. We found that individual axon arbors within the partially lesioned tectum occupy a smaller area, with fewer branches and fewer terminal boutons, but preserve a normal bouton density. In contrast, ipsilaterally projecting axon arbors in monocularly enucleated animals occupy a greater area than in the normal condition, with a much larger arbor length and greater number of boutons and branches compared with normal ipsilaterally projecting cells. Alteration of axonal arborization of retinal ganglion cells is the main factor responsible for matching the retinal and tectal cell populations within the tectum. This process conserves normal electrophysiological function over a wide range of convergence ratios and may occur through strict selectivity of tectal cells for their normal number of inputs.

Neural development in metatherian and eutherian mammals: variation and constraint.

A model for predicting the timing of neurogenesis in mammals (Finlay and Darlington [1995] Science 268:1578-1584) is here extended to an additional five metatherian species and to a variety of other events in neural development. The timing of both the outgrowth of axonal processes and the establishment and segregation of connections proves to be as highly predictable as neurogenesis. Expressed on a logarithmic scale, late developmental events are as predictable as early ones. The fundamental order of events is the same in eutherian and metatherian animals, but there is a curvilinear relation between the event scales of the two; for metatherians, later events are slowed relative to earlier events. Furthermore, in metatherians, the timing of developmental events is more variable than in eutherians. The slowing of late developmental events in metatherians is associated with their considerably longer time to weaning compared with eutherians.

Anomalous ipsilateral retinotectal projections in Syrian hamsters with early lesions: topography and functional capacity.

Retinotectal topography, response properties of neurons in superior colliculus, and visual orienting behavior were studied in hamsters whose superior colliculi were innervated by one or the other of two types of anomalous ipsilateral projections. For the first type, an abnormally large uncrossed projection was created by monocular enucleation on the day of birth. This projection extended over the superficial part of the rostral half of the colliculus. The upper visual field was represented medially, and the lower visual field laterally, which corresponds to a normal projection. The rostrocaudal axis was disordered, but showed a slight tendency for nasal visual field to be represented rostrally and temporal field caudally; this tendency corresponds to an inversion of the normal ipsilateral projection, fitting instead the pattern of a contralateral projection. For the second type of anomalous ipsilateral projection, an abnormal intertectal decussation of optic tract fibers was created by neonatal ablation of the superficial layers of one superior colliculus and removal of the ipsilateral eye (Schneider, '73). Retinotectal topography observed in this recrossing projection was predominantly mirror-symmetric to the normal contralateral projection; however, some distortions in retinotopic order were observed, including misplaced fields and local inversions of the mirror-symmetric topography, and distortions of local magnification factor. Response properties of single units found medially in the left colliculus were similar to those found in normal colliculus. Units found more laterally were underresponsive, showing response decrements with repeated stimulation which is abnormal for units in the superficial gray, and many had abnormally large receptive fields. This physiological pattern was reflected in the pattern of errors made in visual orienting to small targets. It was concluded that polarity cues exist in the tectum sufficient to order the terminals of the retinotectal projection independent of the direction of fiber arrival or order in the optic tract as it enters the tectum. In addition, the functional competence of the abnormal recrossing retinotectal projection has been demonstrated by both electrophysiological and behavioral methods.

Cell death in the mammalian visual system during normal development: II. Superior colliculus.

Degenerating cells may be observed with light microscopy in the hamster superior colliculus during early postnatal development. In the superficial gray layer and stratum opticum, 1.8 degenerating cells for each 1,000 live cells could be seen on the first postnatal day. This rate increased to 5.6 degenerating cells per 1,000 live cells by postnatal day 8. The rate of cell degeneration was consistently elevated at the medial, lateral, and caudal margins of the superficial gray layer relative to the center. In the intermediate and deep gray layers, the rate of cell death was consistently higher, starting at three degenerating cells per 1,000 on postnatal day 5, and declining to 4.7 per 1,000 by postnatal day 8. In contrast to the superficial gray layer, the number of degenerating cells in the central versus peripheral segments of the intermediate and deep gray layers was quite similar. Although the rate of observable degeneration is low, the likely rapid clearance of degenerating cell debris indicates a substantial loss of cells from the midbrain tectum in early development. The time course of observable degeneration, the amount, and the distribution of degenerating cells are quite similar in the tectum, and its major innervating structure, the retina.

Translating developmental time across mammalian species.

Conservation of the order in which events occur in developing mammalian brains permits use of regression theory to model the timing of neural development. Following a small adjustment to account for a systematic variability in primate cortical and limbic systems, the model is used to generate a 95-event/nine-species matrix that predicts aspects of neurogenesis and axonal outgrowth in the brains of developing mice, hamsters, rats, spiny mice, rabbits, ferrets, cats, monkeys, and humans. Although data are compiled from species in which the timing of birth and the rate of maturation vary widely, the model proves statistically accurate, with practical implications for improving estimation of milestones of neural development, particularly for humans. Using the three-factor model (species, neural events, and primate adjustments), we produce predictions for the timing of 493 neural occurrences in developing mammalian brains that either have not yet been, or cannot be, empirically derived. We also relate the timing of neural events across the nine species in the form of a reference table calibrated to the development of laboratory rats. This 'translation' table will assist in attempts to equate the neurodevelopmental literature across species with either large or small differences in gestation and maturation, and also permit studies done in a variety of mammals to be applied to better understand human development. The comparative data indicate that humans, although conventionally considered an altricial species, are neurally advanced at birth relative to the other species studied.

What do developmental mapping rules optimize?

Convergence ratios between pre- and postsynaptic cells in the visual system vary widely between cell classes, areas of the visual field, between individuals and between species. Proper stabilization of the convergence and divergence of single visual neurons is critical for visual integration generally, and for specific functions such as those of rod and cone pathways, or the center and peripheral regions of the visual field. In early development, retinal ganglion cells, target cells and all their processes are produced in excess and stabilize at certain mature values. The intent of the investigations described here is to determine what features of cell connectivity are stabilized over normal variability by these developmental processes and how such stabilization is accomplished, using the developing mammalian retinotectal system as an example. Orderly compression of the retinotopic map into a half tectum was induced by a partial tectal ablation at birth in hamsters, increasing the ratio of retinal ganglion cells to superior colliculus target cells. The convergence problem is solved in this case by undersampling the spatial array with respect to normal, preserving local spatial resolution, but potentially reducing sensitivity or introducing aliasing artifacts. Receptive field sizes of single neurons are indistinguishable from normal, and reduction of branching of presynaptic axon arbors is the mechanism of the remapping. Behaviorally, though the entire visual field is still represented in the remaining colliculus, the solution has a cost in decreased probability and increased latency to orient to visual stimuli, particularly in the peripheral visual field. The generality of this solution for retinal and other central convergence regulation problems is evaluated.

Toward a neuroethology of mammalian vision: ecology and anatomy of rodent visuomotor behavior.

The great diversity of the niches inhabited by rodents, and the variety of the visual demands of these niches, present an excellent prospect for a comprehensive neuroethological analysis of rodent visuomotor behavior. To this end, rodent taxonomy is reviewed, with special attention to the multiple independent invasions of arboreal, terrestrial, fossorial and aquatic niches by distantly related rodent species. Current work on rat, gerbil and hamster is reviewed with emphasis on visual contributions to naturalistic behaviors such as exploration, foraging, predator detection and conspecific recognition.

A neuroethological approach to hamster vision.

The contributions of the midbrain optic tectum to visuomotor behaviors likely to be important to hamsters in the wild were studied, including aperture detection, insect catching, and barrier avoidance. Following tectal undercuts, hamsters ceased to make direct approaches to apertures in the posterior 180 degrees of the visual field; this appeared to be mediated by a loss of exploratory or scanning head movements. Reorientation to and pursuit of crickets jumping out of grasp into the visual periphery was impaired, though initial approach to them was not. Barrier avoidance was unaffected by tectal undercuts. This pattern is similar to the contribution of the frog and toad optic tectum to analogous visuomotor tasks. The contribution of the tectum to searching and scanning in the hamster is an extension of the basic orienting capabilities dependent on optic tectum in anurans.

Control of cell number in the developing neocortex. II. Effects of corpus callosum section.

To determine if cell death participates in the regulation of cell number between interconnecting populations of the neocortex, we sectioned the corpus callosum of neonatal hamsters, thus depriving callosally projecting cells of their normal targets and callosally-recipient cells of their normal afference. The numbers of neurons per unit column in two areas of the cortex which have heavy callosal projections (the 17-18a border and area 6) and one area that is relatively acallosal (area 3) were compared in animals with early corpus callosum sections and controls. No differences were found, either for a 'unit cortical column,' or for the callosally-projecting layers (II-III and V). Mean soma sizes in layers II-III and V of all three areas were likewise unchanged. In area 6 and part of area 3, however, the distribution of soma sizes in callosally projecting and recipient laminae was significantly altered. The change in size distribution without change in mean soma area suggests that the cortex responds to the elimination of the callosal pathway in more than one way. Since no role for cell death in removal of diffuse connectivity or in target regulation of neuron number has yet been found, a new hypothesis for the function of cell death in local cytoarchitectural differentiation of the cortex is proposed.