An electron microscopic study of lesion-induced synaptogenesis in the dentate gyrus of the adult rat. II. Reappearance of morphologically normal synaptic contacts

Induction and maintenance of free postsynaptic membrane thickenings in the adult superior cervical ganglion

The superior cervical ganglion (SCG) of adult rats was exposed to GABA, either by long lasting microapplication (implantation of glass bulbs for 1-24 days) or in short term experiments (external application up to 6 h). Autoradiography showed that [3H] GABA accumulated selectively in satellite cells. The GABA produced the following effects: (1) Specialized membrane thickenings--similar in fine structural appearance to those seen as postsynaptic membrane thickenings at Gray type I synapses--were formed at the extrasynaptic dendritic surface of principal ganglion cells. (2) Morphometry revealed that the surface to volume ratio of dendrites increased significantly corresponding to an enlargement of their extrasynaptic surface as a result of the formation of spine-like projections. (3) Electrophysiology confirmed that, at least after short term application, the action potentials induced by preganglionic stimulation were heavily suppressed. These results suggest that, in the course of depressed ganglionic activity, so-called free postsynaptic membrane thickenings are generated and maintained in the SCG of adult rats even in the absence of significant axonal degeneration. The discussion focuses on two points: (1) possible similarities between the conditions of neurons after denervation and under the influence of GABA; (2) a possible role of GABA and other substances with inhibitory action in synaptogenesis.

Terminal proliferation in the partially deafferented dentate gyrus: time courses for the appearance and removal of degeneration and the replacement of lost terminals

The time courses for the appearance and removal of degenerating terminals and the loss and reappearance of intact terminals were investigated in the partially denervated inner molecular layer of the dentate gyrus of the adult rat. Dense degeneration was evident in the neuropil within 26 hours following contralateral hippocampectomy. These profiles increased rapidly in number until the maximal degree was reached at two to three days postlesion, after which the degenerating terminals were quickly removed from the neuropil. A more rapid rate of removal occurred during the 3-to 5-day survival period than from 6 to 50 days postlesion. The intact terminal population dropped 35% within two days of the lesion and remained at this level until six to eight days postlesion when the number began to steadily increase. The time course for this reappearance can be divided into two phases: a period of rapid terminal addition from 6 to 15 days followed by a phase of slower acquisition. This recovery continued until the normal synaptic density was regained by 50 to 65 days postlesion. These results indicate that a substantial proportion of degenerating endings are removed well in advance of the time at which terminal proliferation begins, suggesting that certain changes other than merely the removal of competitive inputs must take place prior to growth of new terminals. Possible explanation suggested by the present results for the delay in the onset of sprouting include: (1) an absence of appropriate postsynaptic targets during the 2-to 5-day postlesion period and (2) inhibition of axonal growth by the glial cells which are phagocytizing the degenerating products. Beyond the sixth postlesion day the rate at which new terminals appear does correlate with the rate at which degeneration is removed. This suggests that once underway the time course for sprouting may be determined by the avaiabliity of postsynaptic sites.

The specificity of reactive synaptogenesis: a comparative study in the adult rat hippocampal formation

The CA1 region of the hippocampus in the mature rat is shown to possess the capacity to form new synapses following a lesion of either the commissural afferents, which removes 41% of the synaptic input to stratum radiatum, or commissural and associational afferents, which destroys 74% of the synaptic input. With both types of lesion, extensive reinnervation occurs without obvious changes in lamination of afferent fibers and without accompanying changes in the acetylcholinesterase-(AChE) staining pattern. This is in contrast to what is known to occur in the hippocampal dentate gyrus following an ipsilateral entorhinal lesion where afferent lamination is reordered and where AChE-staining intensifies. A comparison between the disparate patterns of reinnervation in these closely related structures affords the opportunity to examine some of the specific factors that may regulate synaptic readjustments in brain.

Terminal proliferation and synaptogenesis following partial deafferentation: the reinnervation of the inner molecular layer of the dentate gyrus following removal of its commissural afferents

The inner one-third of the dendritic region of the dentate gyrus granule cells in adult rats receives projections primarily from the commissural fibers of the contralateral hippocampus and the associational fibers of the ipsilateral hippocampus. At two to four days following the complete removal of the contralateral hippocampus, approximately 25% of the terminals in the inner molecular layer are observed degenerating. This provides an excellent model system to investigate possible terminal proliferation induced by deafferentation since (1) the experimental lesion is easily reproducible, (2) no retrograde reactions occur in the granule cells as a direct result of the lesion, (3) no shrinkage is detected in this region following commissural deafferentation, (4) the same dendritic region can be relocated precisely in each animal, and (5) the synaptic counts are highly consistent between animals. Results from this study and from previous investigations demonstrate that the commissural projection is contained within a 0-80 mu zone directly above the granule cell layer; Complete photomontages of this zone were taken, but only the 40-80 mu zone was quantified for neuronal and glial changes in three normal, five 2- to 4-day, and five 50- to 75-day postlesion animals. The average synaptic count dropped to 64% of control values by 2 to 4 days, returned to 97% by 50- to 75 days postlesion, The number of terminals showing multiple synaptic contacts increased slightly in the long-term animals. Measurements of average terminal area showed no change between the short- and long-term survival groups. These results indicate that this dendritic region is reinnervated following partial deafferentation and that the reinnervation is due primarily to the formation of new terminals rather than the expansion of pre-existing terminals.

Reversible cerebral atrophy in recently abstinent chronic alcoholics measured by computed tomography scans

Eight chronic alcoholics received repeated computed tomography scans. Four, who maintained abstinence and functionally improved, showed partially reversible cerebral atrophy. Two nonabstinent patients and two abstinent patients who had completed functional improvement before the first scan showed no change in atrophy.

Identification of the cells of origin of a central pathway which sprouts following lesions in mature rats

Following unilateral destruction of the entorhinal cortical region of the adult rat, the denervated granule cells of the dentate gyrus are reinnervated as a result of the proliferation of a pathway from the surviving contralateral entorhinal area. The present study investigates the cells of origin of this lesion-induced pathway. Following HRP injections into the reinnervated dentate gyrus, heavily labeled cells were evident in layers II and III of the contralateral entorhinal area, in marked contrast to the pattern of labeling in normal animals, where labeled cells are restricted almost entirely to layer III. The atypically labeled cells in the operated animals were found predominantly in the dorsal half of the entorhinal area, and were concentrated in the medial most portion of layer II. These atypically labeled cells in layer II of the operated animals were an average of 16% larger than their unlabeled neighbors in the same lamina. This was not related to the loading with HRP, however, since in normal animals, cells in layer II which are labeled with HRP were no different in size than unlabeled cells. The atypically labeled cells in layer II of operated animals could also be identified at the electron microscopic level, and could be distinguished from the cells in layer III which normally project to regio superior of the contralateral hippocampal formation. While labeled cells were evident in layers II and III following injections into the reinnervated dentate gyrus, no labeled cells were found in the presubiculum or parasubiculum. In combination, these results suggest (1) the pathway which reinnervates the dentate gyrus from the contralateral entorhinal area originates predominantly, if not exclusively, from the cells in layer II, (2) these cells in layer II have the same preferential distribution within the entorhinal area as the rare lightly labeled cells which can be found contralateral to an injection in normal animals and (3) cells which participate in the reinnervation are larger than their unlabeled neighbors which presumably do not give rise to fibers which reinnervate the contralateral dentate gyrus. Since the cells in layer II which sprout following lesions can be identified at both the light and elctron microscopic level, a potentially valuable model system is available in which to analyze cellular changes during sprouting.

The effects of denervation and stimulation upon synaptic ultrastructure

Quantitative studies of synaptic ultrastructure were made in the upper layers of cat cerebral cortex. Tissues were from intact cortex and from long-term (chronic) undercut cortex with or without electrical stimulation. The synaptic effects of chronic electrical stimulation of denervated cortex are most readily understood as growth and remodeling of synaptic elements. Associated with chronic stimulation were increases in: symmetric membrane contacts; areas of round and flat vesicle containing terminals; dendritic shaft contacts; and synaptic contact lengths. Even without stimulation there were indications of synaptic plasticity in denervated cortex; compared with intact cortex, synapses having symmetric membrane contacts showed an increase in bouton area and an increase in synaptic contacts on dendritic shafts. These data are consistent with the observations of others in which axonal terminal growth occurred after differentation. But it appears that chronic electrical stimulation in the adult nervous system promotes significantly more plasticity than occurs without stimulation. In a functional sense stimulation in the present experiments produced effective inhibition which did not occur with denervation alone. Thus the plasticity observed with stimulation had both structural and functional components.

Progressive brain damage accelerates axon sprouting in the adult rat

An entorhinal cortical lesion causes undamaged fibers in the deafferented hippocampus to sprout and form new connections within 4 to 7 days after the lesion was made. When a partial lesion of the entorhinal cortex precedes a second, more complete entorhinal lesion by a few days, the rate of axon sprouting is accelerated so that the response to the second lesion occurs within only 2 days. This priming effect is present within 4 days, lasts for a few weeks, and eventually subsides. This acceleration may explain, in part, the faster recovery and reduced deficits seen in behavioral studies that have followed serial lesion paradigms.

Synaptic replacement in the dentate gyrus after unilateral entorhinal lesion: electron microscopic analysis of the extent of replacement of synapses by the remaining entorhinal cortex

In response to a unilateral entorhinal lesion the input from the contralateral entorhinal cortex to the dentate gyrus appears to increase. We have studied this crossed projection by electron microscopy in normal animals and in animals one year or more after a unilateral entorhinal lesion. In normal animals few degenerating boutons are found after a contralateral entorhinal lesion. However, when the contralateral lesion was made one year after an ipsilateral entorhinal lesion, degenerating boutons were readily identified. The boutons were relatively few in number, but formed an abnormally large number of synaptic contacts. These results support the previous conclusion that fibres from the contralateral entorhinal cortex form additional synapses when their ipsilateral homologues are removed. However, these new cortical synapses probably account for only a small portion of those formed in response to the lesion. Thus an anatomically homologous input does not, in this case, selectively capture most of the newly available synaptic sites.

Histochemical evidence of altered development of cholinergic fibers in the rat dentate gyrus following lesions. I. Time course after complete unilateral entorhinal lesion at various ages

The entorhinal cortex of rats was removed at various times during development, and the reaction of the cholinergic septohippocampal input to the dentate gyrus was examined by use of acetylcholinesterase histochemistry. When the ipsilateral entorhinal cortex is completely removed, the outer 70-75% of the molecular layer of the dentate gyrus is almost completely denervated. After such a lesion at 5 to 33 days of age, the acetylcholinesterase staining initially intensified throughout the denervated area, indicating that the septohippocampal fibers branched or elongated. This reaction could be detected within one day after a lesion at 11 days of age and within three or five days after lesions at earlier or later times. Whereas the initial response of the septohippocampal fibers was independent of the age at which the lesion was made, their final localization depended on the developmental state of the animal. After lesions at the age of 5 or 11 days, the reactive septohippocampal fibers became restricted to the outer one-sixth to one-third of the molecular layer within two days after appearance of their initial reaction. A similar concentration of reactive fibers was demonstrable after lesions at 16, 18 or 21 days of age, but some reaction persisted in the middle third of the molecular layer. Finally, after lesions at 26 or 33 days of age the proliferating cholinergic fibers ultimately were uniformly distributed throughout the outer 60% of the molecular layer. These results suggest that septohippocampal fibers initially extend or sprout throughout the denervated area to replace the lost perforant path fibers. However, the reactive fiber population becomes restricted to the outer edge of the molecular layer if the entorhinal lesion is made before the period of cholinergic synaptogenesis and concentrates in this same zone if it is made while cholinergic synapses are forming. We suggest that either the proliferative reaction continues in the outer part of the molecular layer and subsides in other parts of the denervated area or septohippocampal fibers move outward through the molecular layer to assume a more superficial location. After entorhinal lesions at 16 days of age or later the pale-staining zone (containing fibers that originate in hippocampus regio inferior) immediately deep to the denervated area widened. If the lesion was made earlier, this zone never developed at most septotemporal levels of the dentate gyrus. These results are probably related to the extension of regio inferior fibers into the denervated area.

An electron microscopic study of lesion-induced synaptogenesis in the dentate gyrus of the adult rat. I. Magnitude and time course of degeneration

Synapses in the rat dentate gyrus are rapidly lost after removal of the primary input from the entorhinal cortex. In this paper we describe the extent and time course of degeneration and in the subsequent paper the nature of the reinnervation processes. They synapses of entorhinal afferents are remarkably concentrated in their zone of termination. Unilateral removal of the rat entorhinal cortex results in the loss of about 86% of all synapses in the outer three-fourths of the molecular layer of the epsilateral dentate gyrus. Entorhinal synapses are all asymmetric (Gray type I) and terminate on dendritic spines. Analysis of the degeneration reaction provides a means to examine the characteristics of the loss of a relatively homogeneous afferent on a single cell type. The morphological characteristics of the the degenerating terminals showed some heterogeneity; both the electron lucent and electron dense types of degenerating terminals were identified. The electron lucent type was observed only at short survival times. The time course of the loss of degenerating terminals was resolvable into two components, each of which followed first order decay kinetics. Thus degenerating entorhinal terminals behaved as a population which disappeared randomly at a rate dependent on the fraction of terminals present at any time. The loss of degenerating terminals was accompanied by the loss of postsynaptic sites. At short survival times the majority of postsynaptic sites (defined by the presence of a postsynaptic density) had disappeared. There was also a loss of complex spines and some shrinkage of the molecular layer.