Kathrin Herrmann

Ultrastructural Evidence for Synaptic Interactions between Thalamocortical Axons and Subplate Neurons

Thalamic axons are known to accumulate in the subplate for a protracted period prior to invading the cortical plate and contacting their ultimate targets, the neurons of layer 4. We have examined the synaptic contacts made by visual and somatosensory thalamic axons during the transition period in which axons begin to leave the subplate and invade the cortical plate in the ferret. We first determined when geniculocortical axons leave the subplate and begin to grow into layer 4 of the visual cortex by injecting 1,1′‐dioctadecyl‐3, 3, 3′, 3′‐tetramethyl indocarbocyanine (Dil) into the lateral geniculate nucleus (LGN). By birth most LGN axons are still confined to the subplate. Over the next 10 days LGN axons grow into layer 4, but many axons retain axonal branches within the subplate. To establish whether thalamic axons make synaptic contacts within the subplate, the anterograde tracer PHA‐L was injected into thalamic nuclei of neonatal ferrets between postnatal day 3 and 12 to label thalamic axons at the electron microscope level. The analysis of the PHA‐L injections confirmed the Dil data regarding the timing of ingrowth of thalamic axons into the cortical plate. At the electron microscope level, PHA‐L‐labelled axons were found to form synaptic contacts in the subplate. The thalamic axon terminals were presynaptic primarily to dendritic shafts and dendritic spines. Between postnatal days 12 and 20 labelled synapses were also observed within layer 4 of the cortex. The ultrastructural appearance of the synapses did not differ significantly in the subplate and cortical plate, with regard to type of postsynaptic profiles, length of postsynaptic density or presynaptic terminal size. These observations provide direct evidence that thalamocortical axons make synaptic contacts with subplate neurons, the only cell type within the subplate possessing mature dendrites and dendritic spines; they also suggest that functional interactions between thalamic axons and subplate neurons could play a role in the establishment of appropriate thalamocortical connections.

DIFFERENTIAL DISTRIBUTION OF AMPA RECEPTORS AND GLUTAMATE DURING PRE- AND POSTNATAL DEVELOPMENT IN THE VISUAL CORTEX OF FERRETS

Immunohistochemical methods were used to study the distribution and time‐course of appearance of cells expressing glutamate and α‐amino‐3‐hydroxy‐5‐methyl‐4‐isoaxazole propionic acid (AMPA)‐type glutamate receptors (GluR1 and GluR2/3) during development of the ferret visual cortex. Glutamate is present in many neurons in the ventricular zone, intermediate zone, developing cortical plate, and marginal zone as early as embryonic day (E) 34 (birth is at E41 in ferrets). Glutamate attains its adult distribution coincident with the completion of cellular migration. By contrast, GluR1 immunoreactivity emerges more slowly. By birth, GluR1 immunoreactivity is present only in a few neurons in the marginal zone and ventricular zone but is abundant in the marginal zone and subplate, where synaptogenesis commences. The number and staining intensity of GluR1‐positive cells increases dramatically during the first two postnatal weeks and is maximal between the second and third week, before slowly declining to adult levels. Cortical cells immunopositive for GluR2/3 follow a similar pattern, although their distribution differs: GluR2/3‐positive cells are mainly pyramidal cells. During the first postnatal week, GluR2/3 is also transiently present in fibers in the intermediate zone, which at this stage contains many thalamocortical and callosal and corticofugal axons. The abundance of glutamate at fetal stages, especially in the ventricular zone, is consistent with the previously proposed role of glutamate in mediating trophic effects in vivo, as previously demonstrated in vitro. The expression of AMPA receptors, as well as their transient overexpression, confirms the results of in situ hybridization studies and may imply a developmental role in neuronal differentiation for these receptors, in addition to their mature role in mediating cortical transmission.

Effects of monocular deprivation in the nucleus rotundus of zebra finches: a Nissl and deoxyglucose study

SummaryWe evaluated in zebra finches the effects of monocular deprivation on morphological and physiological features of the nucleus rotundus, the thalamic relay station of the tectofugal pathway. In a first series of experiments neuron size and total volume were estimated in animals deprived for 20, 40 and at least 100 days and compared to values obtained from normally reared birds. Monocular closure for more than 40 days causes a marked hypertrophy in cells receiving their main input from the open eye, whereas the deprived cells are normal in size. However, with only 20 days of monocular deprivation both deprived and non-deprived rotundal neurons are larger than normal. This indicates that monocular closure has a biphasic effect: firstly, an unselective hypertrophy of deprived and non-deprived neurons, and secondly, a subsequent period of shrinkage of the deprived cells to normal values, while cells driven by the open eye remain hypertrophied. The total volume of the deprived n. rotundus turns out to be smaller in all age groups. In a second series of experiments the activity of the n. rotundus of animals monocularly deprived from birth for 100 days was investigated with the 2-deoxyglucose-method (Sokoloff et al. 1977). With binocular stimulation the activity of the deprived n. rotundus was reduced by about 40%. Depriving adult animals for 100 days does not result in asymmetric labeling of the n. rotundus. We interpretate the 2-DG data as evidence for the existence of a sensitive period for the effects of monocular deprivation. The anatomical data suggest, however, that the effects of monocular deprivation in birds are different from those observed in mammals.

Monocular deprivation affects neuron size in the ectostriatum of the zebra finch brain

The effects of different periods of monocular deprivation on cell sizes in the ectostriatum, the telencephalic relay of the tectofugal pathway in zebra finches, were evaluated. Following 20 days of monocular closure, neurons in the deprived and undeprived hemisphere show an unselective hypertrophy of 10%. Extending the deprivation period results in a shrinkage of neurons of the deprived side to values of adult normally reared birds, whereas the non-deprived neurons maintain their hypertrophied size.

The sensitive period for the morphological effects of monocular deprivation in two nuclei of the tectofugal pathway of zebra finches

Previous experiments with 2-deoxyglucose (2-DG) suggested the existence of a critical period for the effects of monocular deprivation in the nucleus rotundus of zebra finches. The present study concerns the time course of this sensitive period for the morphological effects of monocular deprivation in two areas of the tectofugal visual pathway of zebra finches, the nucleus rotundus of the thalamus and the telencephalic ectostriatum. Cell size and volume changes were measured in birds subjected to 40 days of unilateral eye closure starting at ages spaced regularly throughout the first 70 days of life. The results show that monocular deprivation markedly affects cell size in both areas if the treatment starts at one or 10 days posthatch. The differences between deprived and non-deprived neurons decline monotonically with increasing visual experience prior to deprivation. However, deprivation onset at day 40 again causes as severe effects as early monocular closure. Deprivation as from day 50 or later no longer leads to abnormalities. The measurements of the volume of the nucleus rotundus parallel the cell size measurements, with the exception that the second increase in sensitivity occurs with deprivation onset at day 50 instead of day 40. These data indicate that the time course of the sensitive period for the effects of monocular deprivation may be double-peaked: the sensitivity for external stimuli declines from hatch until day 30, but has another peak at 40-50 days of life. The definite end of the sensitive period, as determined with this method, can therefore be assumed to be at around day 50-60.

Effects of Intraocular Activity Blockade on the Morphology of Developing LGN Neurons in the Cat

Spontaneous neuronal activity is known to play a major role in the construction and final functioning of neuronal circuits, especially those that are formed through interactions of competing inputs. The requirement for neuronal activity in the formation of highly specific connections found in the adult has been well documented in the development of the cat visual system. Shatz and Stryker (1988) and Sretavan et al. (1988) showed that prenatal intracranial infusion of the sodium channel blocker tetrodotoxin, TTX, blocks the segregation of retinogeniculate afferents into eye-specific layers in the lateral geniculate nucleus, LGN, that normally form before birth in the cat. In addition, Stryker and Harris (1986) demonstrated that the formation of the ocular dominance columns in layer 4 of the cat visual cortex can be prevented if retinal ganglion cell discharges are eliminated by intraocular injections of TTX within the critical period postnatally. These studies have clearly shown that action potential activity is required for the elimination of excessive axonal branches and the remodeling of the axonal arbor. Little, however, is known about the role of activity on dendritic development. In this paper we wished to determine whether action potential activity from the retina also influences the morphological development of the dendritic arbors of LGN neurons.

Development and Plasticity of the Tectofugal Visual Pathway in the Zebra Finch

In the recent years the development of the visual system of mammals and its alterability by environmental influences has been addressed by hundreds of investigations. Detailed knowledge on the development and plasticity of the visual cortex, for example, has been accumulated for a variety of mammals, in particular cats and monkeys (rev. Fregnac and Imbert 1984). In contrast, very few studies were concerned with the development of higher stations of the visual system of other vertebrates. In birds, for example, the retinotectal projection of the chick tectofugal pathway is one of the best investigated paradigms for the development of the specificity of neuronal connections and axonal pathfinding (see Thanos, this volume); however, there were almost no developmental studies of other stations of the tectofugal pathway until recently. In one of the earliest studies Pettigrew and Konishi (1976a,b) showed that the visual wulst has physiological properties very similar to those of the visual cortex in mammals, and that monocular deprivation in these animals induces the same shift in ocular dominance of wulst neurons as it was observed in area 17 of the cat or the monkey.