The effect of dark rearing on the time course of the critical period in cat visual cortex

Effects of dark-rearing on the vascularization of the developmental rat visual cortex

Cerebral vascular density corresponds to metabolic demand, which increases in highly active areas. External inputs play an important role in the modeling and development of the visual cortex. Experience-mediated development is very active during the first postnatal month, when accurate simultaneous blood supply is needed to satisfy increased demand. We studied the development of visual cortex vascularization in relation to experience, comparing rats raised in darkness with rats raised in standard conditions. The parameters measured were cortical thickness, vascular density and number of perpendicular vessels, constituting the first stage of cortical vascular development. Vessels were stained using butyryl cholinesterase histochemistry, which labels some neurons and microvascularization (vessels from 5 to 50 microns). Animals from both groups were sampled at 0, 7, 14, 21 and 60 days postnatal. Vascularization of the brain starts with vertically oriented intracortical vascular trunks whose density decreases notably after birth in rats reared in standard laboratory conditions. The most striking finding of our work is the significantly lower decrease in the number of these vessels in dark-reared rats. Our results also show that cortex thickness and vessel density are significantly lower in dark-reared rats. These results suggest that the absence of visual stimuli retards the maturation of the visual cortex including its vascular bed.

Immediate early gene expression in cat visual cortex during and after the critical period: differences between EGR-1 and Fos proteins

Immediate early gene (IEG) expression in the cat visual cortex is highly responsive to visual input and may initiate genetic mechanisms responsible for neuronal plasticity. The present study used immunohistochemical methods to address two issues regarding IEG expression in response to visual input. One was to define the differential response of distinct IEG families by comparing EGR-1 (also termed zif-268, NGFI-A, and Krox-24) and Fos proteins. The second was to determine whether IEG expression, in addition to reflecting neural activity, is related to the state of plasticity by comparing young and adult visual cortex. Immunoreactivity of the two IEG proteins was compared between 5-week-old and adult cats under three conditions of visual input: ambient light to assess basal levels of expression, 1 week of darkness to assess the effect of reduced activity, and exposure to light after 1 week of darkness to determine rapid changes in expression as a result of visual input. At both ages, there were marked differences in the expression of the two IEG proteins. EGR-1 responded to visual input with sustained changes in its level of expression. It showed high basal levels, reduced expression in darkness, and a rapid return to high constitutive levels with the introduction of light. Fos showed a markedly different profile. It had very low basal expression which was not demonstrably affected by darkness and its principal response was a marked transient induction upon exposure to light after darkness. These unique changes in expression highlight the complex response across IEGs to environmental input and suggest a genetic "on/off' signaling mechanism. There were marked differences in the laminar distribution of EGR-1 and Fos proteins between young and adult cats. In young animals, cells in all visual cortical layers showed high levels of EGR-1 and Fos proteins. In adults, immunostaining was largely specific to cells located above and below layer IV and only very faint labeling occurred within layer IV. These differences in laminar distribution between ages are inconsistent with a simple explanation of IEG expression in terms of neural activity level; rather, they suggest a relation between IEG expression and the state of plasticity in visual cortex.

Role of metabotropic glutamate receptors in the cat's visual cortex during development

We have studied the effect of metabotropic glutamate receptors on the second messenger cAMP, and how it varies with age in light- and dark-reared cats; the overall level of the metabotropic glutamate receptors mGluR1, 2/3, and 5 during development; the laminar distribution of these receptors; and how the physiological effect of the metabotropic glutamate receptor agonist ACPD varies with layer. The increase in cAMP produced by ACPD correlates well with the critical period for ocular dominance plasticity in both light- and dark-reared animals. Basal levels of cAMP also correlate well, but overall levels of mGluRs do not. Thus, the second messenger is likely to be the critical factor in plasticity, rather than the mGluRs. Both group I mGluRs (1 and 5) and group II mGluRs (2 and 3) contribute to the increase in cAMP. However, mGluR5 is affected by rearing in the dark, while mGluR1 is not. Moreover, the laminar distribution of mGluRs 2/3 and 5 changes with age, while mGluR1 does not. The laminar distribution is correlated with the functional effect of ACPD, which varies with layer. In upper layers, ACPD has a depressive effect on both visual response and spontaneous activity, while in lower layers it has a depressive effect on visual response and a facilitatory effect on spontaneous activity. These variations in functional effect with layer need to be taken into account before the role of metabotropic glutamate receptors in development will be fully understood.

Glutamate receptors and development of the visual cortex: effect of metabotropic agonists on cAMP

Glutamate receptors are more active in several respects in young animals than in adults. Here we examine the effect of metabotropic glutamate agonists on rat cortical cAMP during and after the critical period for visual cortex plasticity. Quisqualate produced a substantial increase in cAMP, which was larger during the critical period than in the adult. The increase was not affected by CNQX or APV, showing that it was not due to the action of quisqualate on ionotropic glutamate receptors. Both Type I mGluRs (mGluRs 1 and/or 5) and Type II mGluRs (mGluRs 2 and/or 3) probably contributed to the cAMP increase because (i) ACPD and L-CCG-I, which are more active on Type II mGluRs, were more effective than DHPG, which is more active on Type I mGluRs; and (ii) there was a significant difference in the effect of ACPD on the increase in cAMP, comparing mGluR1 knockout mice with control mice. Agonists which produce large stimulation of cAMP production (ACPD, L-CCG-I), as well as L-AP4, also produced small attenuations of forskolin-stimulated cAMP, but only at high concentrations. Thus, we conclude that it is the stimulation and/or potentiation of cAMP production that is significant, rather than the attenuation of forskolin-stimulated cAMP. Since this stimulation and/or potentiation is higher during the critical period than in the adult, and the cAMP second messenger system has been implicated in hippocampal plasticity, it may also play a role in visual cortex plasticity.

Dark-rearing changes dendritic microtubule-associated protein 2 (MAP2) but not subplate neurons in cat visual cortex

Sensory-dependent modification of cortical morphology is one component of the cortical plasticity that occurs during the critical period for ocular dominance changes. In this study, we used dark-rearing to examine the sensory dependency of subplate neuron death and the quantity of microtubule-associate protein 2 (MAP2)-positive dendrites. Kittens reared in total darkness until the peak of the critical period had fewer laterally extended MAP2-positive dendrites than age-matched normal kittens. This reduction was found in layer IV but not in layer V. Subsequent exposure to light for 10 days after dark-rearing was sufficient to bring the number of MAP2-positive dendrites to the normal level. Contrarily, dark-rearing did not prevent subplate neurons from dying. Exposure to light after dark-rearing did not increase the number of potential dying neurons. These results show that the quantity of MAP2-positive dendrites is sensory-dependent; however, the death of the subplate neurons is not. Therefore, the death of subplate neurons is probably not directly involved in sensory-dependent modifications of synaptic connections. The possible involvement of laterally extended MAP2-dendrites in visual plasticity is discussed.

Co-regulation of long-term potentiation and experience-dependent synaptic plasticity in visual cortex by age and experience

Long-term potentiation (LTP) is a lasting enhancement of excitatory synaptic transmission that follows specific patterns of electrical stimulation. Although the mechanism of LTP has been intensively studied, particularly in the hippocampus, its significance for normal brain function remains unproven. It has been proposed that LTP-like mechanisms may contribute to naturally occurring, experience-dependent synaptic modifications in the visual cortex. The formation of normal binocular connections within the visual cortex requires simultaneous input from both eyes during a postnatal critical period that can be delayed by rearing animals in complete darkness. To explore the role of LTP in this experience-dependent maturation process, we induced LTP in visual cortical slices taken at different ages from light-reared and dark-reared rats. Susceptibility to LTP coincides with the critical period and, like the critical period, can be prolonged by rearing animals in darkness. These findings support the hypothesis that LTP reflects a normal mechanism of experience-dependent synaptic modification in the developing mammalian brain.

Early isolation from conspecific song does not affect the normal developmental decline of N-methyl-D-aspartate receptor binding in an avian song nucleus

Early effects of experience on synaptic reorganization and behavior often involve activation of N-methyl-D-aspartate (NMDA) receptors. We have begun to explore the role of this glutamate-receptor subtype in the development of learned birdsong. Song learning in zebra finches occurs during a restricted period that coincides with extensive synaptic reorganization within neural regions controlling song behavior. In one brain region necessary for song learning, the lateral magnocellular nucleus of the anterior neostriatum (IMAN), NMDA receptor binding is twice as high at the onset of song learning as in adulthood. In the present study, we used quantitative autoradiography with the noncompetitive NMDA antagonist [3H]MK-801 to examine more closely the developmental decline in NMDA receptor binding within IMAN and found that it occurred gradually over the period of song learning and was not associated with a particular stage of the learning process. In addition, early isolation from conspecific song did not affect [3H]MK-801 binding in IMAN at 30, 60, or 80 days. Since behavioral studies confirmed that our isolate rearing conditions extended the sensitive period for song learning, we conclude that the normal developmental decline in overall NMDA receptor binding within IMAN does not terminate the capacity for song learning. Finally, early deafening, which prevents both stages of song learning, also did not affect [3H]MK-801 binding in IMAN at 80 days, indicating that the decline in NMDA receptor binding occurs in the absence of auditory experiences associated with song development.

The origin of synaptic neuroplasticity: crucial molecules or a dynamical cascade?

What is neuroplasticity and what are its origins? These questions have been the subject of intense theoretical and experimental research in the neurosciences for decades. Basically, the term neuroplasticity refers to the ability of neurons to alter some functional property in response to alterations in input. Traditional definitions, however, are often imprecise and restricted to particular 'model' neural systems. In the present article we will consider several of the most widely studied models of synaptic-level neuroplasticity including alterations in response properties of two types of invertebrate sensory neurons, long-term potentiation (LTP) in mammalian hippocampus and cortex, and ocular dominance shifts in cat visual cortex. While many other forms of neuroplasticity have been studied, these examples typify the diversity of the subject, as well as illustrate our contention that no unitary model of the phenomena is possible for all conditions. This last point is of particular importance for the mammalian literature, since many hypotheses concerning the mechanism(s) underlying the initiation of neuroplasticity have proposed a single crucial molecular element as the primary causal agent. A closer examination of these various hypotheses, in concert to several examples from the invertebrate literature, leads, however, to the conclusion that synaptic neuroplasticity must arise from a series of inter-related molecular events of a particular form, a cascade, in which individual elements may differ radically from system to system. We next provide an overview of our studies of age-dependent regulation of excitatory and inhibitory ionotropic neurotransmitter receptor populations in cortex in response to agonist and depolarizing stimulation. We provide evidence that such regulation for ionotropic receptors is under the control of ionically driven receptor kinase and phosphatase activity which is also age-dependent in function. These data provide the basis for a cascade model of receptor regulation. Based on this qualitative model, we describe a quantitative computer simulation of certain age-dependent stages in the receptor regulatory cascade which may interact to produce LTP-like effects. While such a model is not exclusive, it nevertheless provides a demonstration that elements in the proposed cascade may comprise the necessary and sufficient conditions for some forms of neuroplasticity. We also propose some of the principles underlying our model as a means of unifying much of the diverse phenomenology reported in the literature. Finally, we make a series of explicit predictions which are testable with current experimental techniques.(ABSTRACT TRUNCATED AT 250 WORDS)

Visual deprivation decreases long-term potentiation in rat visual cortical slices

A major finding in the visual plasticity literature is that visual deprivation is effective only during an early 'sensitive' period, which is lengthened by dark rearing. Unresolved is whether the visual cortex is in a normally plastic state prior to light stimulation. This cannot be addressed using paradigms employing light exposure to assess plasticity. Several developmental studies have investigated a plastic phenomenon termed long-term potentiation (LTP) in slices from cat (J. Neurophysiol., 59 (1988) 124-141) and rat (Brain Res., 439 (1988) 222-229) visual cortex. Susceptibility to the induction of LTP parallels the period of sensitivity to visual deprivation. This suggests that slices can be used to assay visual cortical plasticity, avoiding light exposure. In the present study, field potentials were recorded from slices of the primary visual cortices of dark-reared (DR) and control (CONT) Long Evans hooded rats (17 to 21 days). Field potential profiles recorded before and 90 min following tetanic electrical stimulation were subjected to current source density analysis, yielding extracellular current sink amplitudes. Tetanus resulted in LTP in both CONT and DR slices, but DR slices were significantly less potentiated. These results indicate that the primary visual cortex of DR animals is not fully plastic, indicating a role for light stimulation in inducing visual cortical plasticity.

Developmental and environmental changes in GAP-43 gene expression in cat visual cortex

Northern/slot blot analysis was used to determine postnatal developmental and environmentally induced changes in the level of expression of GAP-43 mRNA in visual and frontal cortex. Both structures showed a precipitous decline during the first 5 weeks and a slight further decline to adult levels. Dark rearing resulted in a significant elevation of GAP-43 mRNA which was eliminated by brief visual experience. This effect was specific to visual cortex and did not occur in frontal cortex. The effect also did not occur in normal adult cats placed in prolonged darkness, indicating that GAP-43 mRNA levels are not simply activity dependent and are altered by visual input only during early postnatal life. These results are consistent with a role for GAP-43 in the state of visual cortical plasticity.

Do NMDA receptors have a critical function in visual cortical plasticity?

The theoretical framework, by which we understand the function of NMDA receptors, is derived, in large part, from work conducted on the hippocampal slice preparation, where NMDA receptors are crucial for a form of synaptic plasticity known as long-term potentiation (LTP). Establishing their role in plasticity mechanisms in the neocortex is proving to be far more difficult than originally envisaged, in part due to the fact that the operation of NMDA receptors is different in the intact animal than in vitro. For example, NMDA receptors are activated at low levels of sensory input in intact animals but only by high levels of input in slice preparations. Recent results suggest that a re-evaluation of the role of NMDA receptors in neocortical plasticity is required. Here we discuss some of the issues and introduce four criteria by which any factor supposedly involved in plasticity can be judged. NMDA receptors fulfill more of these criteria than any of the other factors so far investigated in the visual cortex, but maybe this is because they have been studied more intensively.

Castration and antisteroid treatment impair vocal learning in male zebra finches

Both song behavior and its neural substrate are hormone sensitive: castrated adult male zebra finches need replacement of gonadal steroids in order to restore normal levels of song production, and sex steroids are necessary to establish male-typical neural song-control circuits during early development. This pattern of results suggests that hormones may be required for normal development of learned song behavior, but evidence that steroids are necessary for normal neural and behavioral development during song learning has been lacking. We addressed this question by attempting to eliminate the effects of gonadal steroids in juvenile male zebra finches between the time of initial song production and adulthood. Males were castrated at 20 days of age and received systemic implants of either an antiandrogen (flutamide), an antiestrogen (tamoxifen), or both drugs. The songs of both flutamide- and tamoxifen-treated birds were extremely disrupted relative to normal controls in terms of the stereotypy and acoustic quality of individual note production, as well as stereotypy of the temporal structure of the song phrase. We did not discern any differences in the pattern of behavioral disruption between birds that were treated with either flutamide, tamoxifen, or a combination of both drugs. Flutamide treatment resulted in a reduced size of two forebrain nuclei that are known to play some role unique to early phases of song learning [lateral magnocellular nucleus of the anterior neostriatum (IMAN) and area X (X)], but did not affect the size of two song-control nuclei that are necessary for normal song production in adult birds [caudal nucleus of the ventral hyperstriatum (HVc) and robust nucleus of the archistriatum (RA)]. In contrast, treatment with tamoxifen did not result in any changes in the size of song-control nuclei relative to normal controls, and it blocked the effects of flutamide on the neural song-control system in birds that were treated with both drugs. Castration and antisteroid treatment exerted no deleterious effects on the quality of song behavior in adult birds, indicating that gonadal hormones are necessary for the development of normal song behavior during a sensitive period.