Rapid synaptic changes and bundling in the supraoptic dendritic zone of the perfused rat brain.

In the dendritic region of the supraoptic nucleus of young adult rats we found ultrastructural evidence of rapid (< 30 min) morphological reorganization following transcardial perfusion of the anesthetized animals with high (325 mOsm), but not with normal, osmolality (302 mOsm) balanced salt solutions. The changes observed included significant increases in dendritic bundling and increased numbers of multiple axodendritic synapses (where one presynaptic terminal contacted two or more postsynaptic dendrites). In our initial experiments the perfusion medium contained tannic acid, but when the tannic acid was omitted in a separate set of experiments, the results were both similar and more striking. These studies indicate that the plasticity previously seen in this region under physiological conditions can occur rapidly and probably involves astrocytic process withdrawal from between adjacent dendrites. This change in the distribution of astrocytic processes allows previously single synapses access to additional postsynaptic dendritic membrane. Together, increased dendritic bundling and multiple synapses may facilitate the synchronization of cell firing and, thus, hormone release.

Reevaluation of the plasticity in the rat supraoptic nucleus after chronic dehydration using immunogold for oxytocin and vasopressin at the ultrastructural level.

It has been shown that during physiological stimuli, such as dehydration, supraoptic nucleus (SON) neurons undergo profound morphological changes. However, little is known about how much each type of cell, oxytocin (OT) or vasopressin (VP), contributes to this plasticity during dehydration. Using postembedding immunogold cytochemistry for both OT and VP hormones at the electron microscopic level, we address this question. Rats were chronically dehydrated (given 2% saline to drink for 10 days) and their SON neurons were studied morphologically. The results were compared to control animals with free access to water. Both VP and OT somata showed an enlargement in size in dehydrated animals. Percentage of somasomatic/dendritic membrane contact increased significantly in both VP and OT neurons, with no significant changes in percentage of coverage of the cells by astrocytic membrane. Only the VP cells had a lesser amount of axosomatic membrane contact after dehydration, possibly due to an increase in cell size rather than a decrease in synaptic contact. Multiple synapses (MSs) (i.e., terminals that form more than one synapse with adjacent somata and or dendrites) occurred only between positively labeled cells and between negatively labeled cells, but not between positively and negatively labeled cells. The number of MSs per 100 microns OT somatic membrane or per 100 OT cells was significantly higher in dehydrated rats but was unchanged with regard to VP neurons. These findings indicate that both VP and OT neurons undergo morphological changes during chronic dehydration and, thus, that plasticity is not limited to OT cells as some earlier reports have suggested.

Neurosecretory endings in the rat neurohypophysis are en passant.

A combination of Golgi, Phaseolus vulgaris leucoagglutinin, and transmission electron microscopic (EM) techniques was used to investigate the morphology of neurosecretory axons and their endings in the neurohypophysis of the rat. Light microscopy indicated that the neurosecretory processes are very tortuous, varicose, and branched, often running in close association with blood vessels. EM, as well as reconstruction from serial thin sections, demonstrated that the combination of synaptoid membrane specializations, dense core vesicles, and accumulations of microvesicles, thought to indicate areas of preferential hormone release, can occur anywhere that the axons contact the basal lamina (BL) lining the perivascular space. Usually, but not always, this was accompanied by some degree of axonal dilation. Individual neurosecretory axons frequently entwined around, or ran adjacent to, blood vessels, a short length of axon forming multiple "endings." Thin glial (pituicyte) processes were interposed between these endings. Axonal processes were also seen to end blindly as end-bulbs suspended in the perivascular space where they were often surrounded by pituicyte processes. The morphology observed suggests that reductions in the contact length of individual nerve terminals could be mediated through increased pituicyte coverage along BL. Conversely, a reduction of pituicyte coverage along the BL could lead to more areas of axonal contact with the BL inducing or allowing the formation of more endings, such as has been reported to occur during lactation or prolonged elevation of plasma testosterone levels.

Ultrastructural changes in the rat neurohypophysis following castration and testosterone replacement.

Ultrastructural changes in male rat neurohypophyses were studied 8 or 30 days after castration, with or without testosterone (T) replacement via capsule implants. Morphometric analyses determined the: (a) amount of neural contact at the basal lamina (BL) of the neurovascular contact zone, (b) average length of individual terminal contacts with the BL, (c) number of neurosecretory terminals per 100 micrograms of BL, and (d) mean number of enclosed axonal processes per pituicyte. Eight days after castration there was decreased neural/BL contact and increased pituicyte enclosure of neurosecretory processes, conditions associated with decreased hormone release. In contrast, T replacement resulted in increased individual nerve terminal length, a measure associated with increased hormone demand. This observation may indicate a stimulatory effect of continuous high-normal circulating levels of testosterone from the capsule implants. There were no differences from control in 30-day castrate rats, but the 30-day rats with T replacement showed morphological indications of increased hormone release. These consisted of increased neural contact with the BL apparently through a significant increase in the number of neurosecretory terminals per unit length of BL. These findings support studies showing a complex feedback interaction between circulating levels of testosterone and vasopressin release.

Morphological adaptability at neurosecretory axonal endings on the neurovascular contact zone of the rat neurohypophysis.

To compare the effects of a variety of acute and chronic stimuli that bring about or terminate hormone release the ultrastructure of nerve terminal contact at the basal lamina of the neurohypophysial neurovascular contact zone was examined quantitatively in young adult rats of the following treatment groups: untreated virgin females, untreated male rats, prepartum (day 21 of gestation), postpartum (on the day of parturition), lactating (14 days of suckling), mothers 10 days after their pups were weaned, 48 h water-deprived males, males given 2% saline solution (dehydrated) for 10 days, males given 2% saline as described then given tap water to rehydrate for 2 or 5 weeks. Morphometric analysis of electron micrographs revealed that all stimuli leading to increased hormone release were accompanied by both increased occupation of the basal lamina by nerve terminals as well as decreased enclosure of neurosecretory processes by pituicyte cytoplasm. Neural occupation of the basal lamina remained significantly elevated 10 days post-weaning and at 2 weeks (but not 5 weeks) of rehydration following 10 days of dehydration. Pituicyte enclosure of neurosecretory axons had returned to control values in the postweaning and 5 week (but not 2 week) rehydrated animals. The mean length of individual nerve terminal contact with the basal lamina was found to increase under some, but not all, conditions associated with increased hormone release (i.e. parturition, acute and chronic dehydration, but not during lactation) and to decrease below control values in prepartum females and after 5 weeks of rehydration.(ABSTRACT TRUNCATED AT 250 WORDS)

Vacant postsynaptic densities on supraoptic dendrites of adult rats diminish in number with chronic stimuli.

In earlier ultrastructural studies of the supraoptic nucleus in adult rats we noted "free" and incompletely covered postsynaptic densities (collectively referred to here as vacant postsynaptic densities) on dendritic shafts. Free postsynaptic densities have been reported in other parts of the central nervous system of normal rodents. We investigated the possibility that physiological activation of the supraoptic cells, which produces changes in many aspects of their morphology, would alter the incidence of the free or incompletely covered postsynaptic densities on dendrites in the supraoptic basal dendritic zone. The cells of the supraoptic nucleus are activated to increase cell firing and secretion of oxytocin and/or vasopressin in response to dehydration, gestation, and lactation. We have examined: untreated virgin females; untreated males; 24 h water-deprived males; prepartum (21st day of gestation) females; postpartum females (on the day of parturition); lactating females (14 days of suckling); mothers 10 days after weaning their pups; females given 2% saline to drink (dehydrated) for 10 days; and females or males given 2% saline to drink for 10 days, then given tap water for 2 or 5 weeks to allow rehydration. Only long-term activation of the supraoptic nucleus by lactation or by drinking saline for 10 days brought about significant decreases in the percentage of dendrites with vacant postsynaptic densities. These densities did not reappear in saline treated rats which had been rehydrated for 2 weeks, but did return in both the 5-week rehydration and the 10-day postweaning groups.(ABSTRACT TRUNCATED AT 250 WORDS)

Neuronal/glial plasticity in the supraoptic dendritic zone in response to acute and chronic dehydration.

The magnocellular neurosecretory cells of the supraoptic nucleus increase production and secretion of oxytocin and/or vasopressin in response to dehydration, gestation and lactation. Dynamic neuronal/glial interactions have also been shown to occur in response to these stimuli, resulting in a reversible increase in soma-somatic direct membrane apposition at these times. Chronic (lactation, 10 days of saline drinking) but not acute stimuli (4-24 h water deprivation) are further accompanied by the reversible formation of axo-somatic double synapses (one presynaptic terminal contacting two postsynaptic elements), which are virtually absent in control animals. The dendrites of these cells course ventrolaterally toward the ventral glial lamina, and have also been shown to be involved in this plasticity: dendro-dendritic direct membrane apposition and axo-somatic double synapses significantly vary with gestation and parturition. The present study investigated the dendritic zone response to both chronic and acute dehydration and rehydration. Increased dendro-dendritic membrane contacts resulted from both stimuli. Rehydration following acute dehydration resulted in a dose-dependent return to control levels, while rehydrated chronic dehydrates did not show such a return until 35 days of rehydration. The percentage of dendrites contacted by double synapses did not vary with treatment, and there were no sex differences. The recalcitrance on the part of the dendrites to return to normal following chronic dehydration may reflect a readiness to respond to renewed hormone demand.

Horseradish peroxidase evidence for a spinal projection from the preoptic area of the goldfish, a light and electron microscopic study.

Horseradish peroxidase (HRP) was applied to the transected spinal cord of goldfish labeled neurons in the preoptic area. Since leakage of HRP into the blood could produce the labeling of neurosecretory cells, intraperitoneal (i.p.) injections of HRP were made with a wide range of dosages in order to intentionally label preoptic neurosecretory cells. The distribution of preoptic neurons labeled after spinal HRP application was far more restricted than the labeling via uptake of HRP from the blood, even when cells in the spinal cord-transected fish were intensely labeled. Furthermore, in HRP electron microscopic material, morphological differences were observed between neurons labeled by the two procedures. Large numbers of dense core vesicles and well-developed stacks of rough endoplasmic reticulum, features typical of cells projecting to the pituitary, were not observed in cells labeled via the spinal cord. These findings indicate that in goldfish a direct projection exists from the preoptic area to the spinal cord which could be homologous to one arising from the paraventricular nucleus of mammals. Both i.p. injection and spinal transection also produced labeling of more caudal periventricular diencephalic cells which resemble preoptic cells in efferent projections as well as ultrastructural features.

Neuronal/glial plasticity in the supraoptic dendritic zone: dendritic bundling and double synapse formation at parturition.

The magnocellular neurosecretory cells of the supraoptic nucleus increase production and release of oxytocin and/or vasopressin under such conditions as parturition, lactation and dehydration. These stimuli have been shown to result in increased direct apposition of neuronal membranes and the formation of double synapses (one presynaptic terminal contacting two postsynaptic elements) within the supraoptic nucleus at the level of the cell bodies. These morphological changes are due to the retraction of the thin glial processes which are normally interposed between adjacent neurons. The present study was undertaken to ascertain whether, and to what extent, neuronal/glial plasticity occurs in the dendritic zone (i.e. the ventral glial laminar area) of the supraoptic nucleus. The instances of two or more dendrites with membrane in direct apposition (dendritic bundles), the number of dendrites per bundle, the amount of dendritic membrane in direct apposition and the percentage of dendrites contacted by double synapses were quantified at the ultrastructural level in virgin female, prepartum (21 days of gestation), postpartum (day of parturition) and lactating rats. All parameters measured varied significantly with the hormone demand states created by pregnancy and lactation, apparently due to glial retraction. Moreover, in the 2-24 h period between pre- and postpartum there was a significant increase in the number of dendrites per bundle, dendritic membrane in direct apposition and the percentage of dendrites contacted by double synapses. This time course corresponds to the known increased release of oxytocin and vasopressin at parturition. These findings constitute the first demonstration that dendritic bundles and double synapses occur in the ventral glial lamina/dendritic zone of the supraoptic nucleus and vary under the physiological conditions of pregnancy, parturition and lactation.

Synapse formation and disappearance in adult rat supraoptic nucleus during different hydration states.

Activation of the adult rat supraoptic nucleus by a chronic stimulus (10 days of drinking 2% NaCl instead of tap water) brought about the appearance of newly formed specialized synapses onto the magnocellular neurosecretory cells. The increase in these synapses was reversed when the animals were allowed to rehydrate by drinking tap water. The apparent retraction and reinsertion of the thin glial processes from between the neurosecretory cell somata, which results in significant changes in soma-somatic direct membrane appositions, is most likely involved in the formation and elimination of these synapses.

Plasticity in the in vitro neurohypophysis: effects of osmotic changes on pituicytes.

The ultrastructure of rat neurohypophyses incubated in vitro was studied to investigate secretion-related changes that may be intrinsic to the isolated neurohypophysis and to establish the morphological integrity of this preparation. Neurohypophyses were incubated for 2 h in an in vitro chamber in medium of low (290 mosM/kg), normal (310 mosM/kg), or high (340 mosM/kg) osmolality. Subsequent morphometric analyses at the ultrastructural level revealed that the number of axons completely enclosed by pituicyte cytoplasm was inversely related and the amount of neuro-vascular contact was directly related to the osmolality of the medium. These results mimic those found in preparations from in vivo experiments with hydrated and dehydrated rats. The number of pituicyte liposomes did not vary consistently with osmolality. We conclude that: dynamic interactions between pituicytes and neurosecretory axons may be stimulated, at least in part, by the immediately surrounding milieu; pituicytes may have an active role as modulators of hormone release; and the in vitro neurohypophysis provides a suitable model to study events intrinsic to this area.

Dynamic neuronal-glial interactions in hypothalamus and pituitary: implications for control of hormone synthesis and release.

Various lines of evidence have suggested that astrocytes play a dynamic role in control of hormone synthesis and release from the CNS. The model system most studied has been the rat hypothalamo-neurohypophysial system, consisting chiefly of the supraoptic and paraventricular nuclei and their axonal terminals. Neurons of this system manufacture and secrete oxytocin and vasopressin. Electron microscopic studies have shown that certain physiological conditions (e.g., dehydration, lactation) produce increases in direct apposition among these neurosecretory cells, an effect due to withdrawal of glial processes from between the neurons. Neurohypophysial astrocytes (pituicytes) show dynamic interactions with the neurons at the level of the terminals, by engulfing them and interposing processes between the terminals and the basement membrane when hormone demand is low. Pituicyte processes retract from both areas when hormone demand is high, allowing the neuronal terminals direct access to the perivascular space. Recently, osmotic manipulations (in the physiological range) have shown that these changes can be produced in vitro in neurohypophysial explants without stimulated hormone release. Experiments on cultured adult rat pituicytes have revealed similar morphological changes in response to noradrenaline. These changes were reversed or blocked by propranolol. The increase in direct soma-somatic apposition (7-9 nm separation) of magnocellular neurons could produce a tonic rise in (K+)o which would increase protein synthesis and contribute to the raised excitability of these neurons. Also, the removal of interposed glia could allow the formation of gap junctions and specialised synapses which are known to occur between these neurons. These in turn may participate in producing the coordinated firing that maximizes hormone release. The interactions of pituicytes with the terminals in the neurohypophysis suggests that these astrocytes are also a part of the mechanism of control of hormone release.

Maturation of the goldfish preoptic area: ultrastructural correlates of position and maturation of neurosecretory cells.

A gradation of small rostral to larger caudal neurosecretory cells is found in the preoptic area of fish. Regional variations in ultrastructural features were assessed in this region of fish ranging widely in size (1.7-283.1 g), some of which had received intraperitoneal injections of horseradish peroxidase in order to label neurons projecting beyond the blood-brain barrier (i.e. neurosecretory cells). In small (less than or equal to 6.6 g) fish, neurosecretory cells of the caudal preoptic area were 10-15 microns in diameter and contained dense core vesicles. More rostral smaller cells had few or no dense core vesicles, but were labeled following intraperitoneal horseradish peroxidase injections. Cells in both areas were progressively larger, with larger nuclei and more granular reticulum in larger fish. In large fish, neurons that contained dense core vesicles ranged from 10 microns in diameter rostrally to 70 microns in diameter caudally. Some intermediate and large cells were extensively vacuolated and contained extremely convoluted and possibly multiple nuclei. Degenerating material was seen in apparently normal large fish. A few neurosecretory cells were extensively surrounded by perineuronal electron-dense glial cells. Most neurosecretory cells were involved in extensive soma-somatic apposition, which was especially pronounced in fish which were removed from cold water in late fall. Gap junctions were also clearly present on the somata of preoptic neurosecretory cells of these fall fish. These data imply that increased capacity for hormone secretion in large fish may be accomplished in part by neuronal hypertrophy. Maturational and seasonal variation in structure suggest that physiological characteristics are also variable in this nucleus.

Magnocellular neuropeptidergic neurons in hypothalamus: increases in membrane apposition and number of specialized synapses from pregnancy to lactation.

Morphological changes which have been hypothesized to accompany functional alterations in magnocellular neuropeptidergic cells (MNCs) were studied in female rats. Direct soma-somatic appositions between the MNC profiles of two nuclear groups, the supraoptic nucleus (SON) and nucleus circularis (NC) were investigated at the ultrastructural level in 4 groups of animals: virgin females, immediately pre-partum pregnant rats, post-partum and 14-day lactating animals. The percentage of SON MNC profiles in soma-somatic apposition and the amount of membrane in direct contact significantly increased over control levels by the last day of pregnancy. Further significant increases in these measures were observed in lactating rats. MNCs in NC showed steady gradual increases on these measures with significant differences from controls occurring in the post-partum group. The percentage of SON cell profiles with double synapses (i.e., presynaptic terminals making synaptic contact with two postsynaptic neurons) was significantly elevated in lactating rats (approximately 10%) over the next highest group (approximately 1% for post-partum rats). In NC, approximately 10% of the cell profiles sampled had such synapses but no differences among treatments occurred. The changes during late pregnancy suggest that close appositions may serve to enhance the metabolic activity of MNCs at a time when there is a build-up of stored oxytocin. Further increases in cell-cell contact and the addition of double synapses on possibly electrotonically coupled MNCs during lactation may serve a synchronizing function, particularly in the oxytocin cells participating in the milk ejection reflex.

Magnocellular neuropeptidergic terminals in neurohypophysis: rapid glial release of enclosed axons during parturition.

The ultrastructure of pituicytes as well as their relationship with neurosecretory axons were measured from the following groups of adult rats: virgin females; immediately pre-partum; post-partum; and following 14 days of lactation. Both parturition and lactation brought about a significant decrease in the number of pituicytes with neurosecretory axons completely surrounded by their cytoplasm. Since these two physiological states are each characterized by increased neurohypophysial hormone release, the data suggest that pituicyte enclosure of neurosecretory axons occurs mainly during conditions of low hormone release. Examination of pituicyte size and number of lipid inclusions, and the number of synaptoid contacts from neurosecretory axons onto pituicytes showed no differences among groups.

Changes in lipid levels of three skeletal muscles following denervation.

Nonpolar and polar lipids extracted from denervated rat gastrocnemius, plantaris, and soleus muscles were measured 7-9 days after unilateral sciatic nerve transection. The contralateral muscle (CCON) was used to obtain control lipid levels. After denervation changes in lipid concentrations were found in all three muscles. These alterations in lipid levels were generally in same direction but not to the same extent. The change in total nonpolar lipids (NL) was an increase in soleus greater than gastrocnemius greater than plantaris concentration. This change in lipid concentration was more apparent than real since the wet weight of muscle was decreased after denervation. Since polar lipid (PL) concentrations were not increased under these conditions of muscle weight loss, an actual decrease of polar lipids after denervation may be inferred. In contrast to the other two muscles, a marked difference was noted for polar lipids of denervated gastrocnemius muscle. An unidentified spot near the origin was detected. This area is the location of a nerve sprouting factor(s). The compound(s) was not detectable for the other two muscles. When the gastrocnemius from an unoperated animal rather than a CCON muscle was used as a benchmark, slight increases were found for total nonpolar, polar, and plasmalogen fractions following denervation. The changes for individual lipid fractions were less definable, except for the significant increase for the unknown polar compound near the origin. This spot was noted in extracts from CCON and DEN muscles but not in untouched control muscle. The CCON gastrocnemius muscle is therefore a poor control for determining effects of denervation on lipid levels and perhaps other biochemical parameters as well.

Cholinesterase in muscle of dystrophic hamsters (Bio-40.54).

Isozyme patterns of cholinesterase (ChE) from heart, tongue, and skeletal muscle of normal and dystrophic hamsters are presented. Two principal bands, bands 1 and 2, were evaluated. Band 1 migrates faster towards the anode than does band 2. While bands 1 and 2 stain for AChE and were found in control muscles, only band 2 was stained by a pseudocholinesterase (BuChE) and was decreased in samples from dystrophic hamsters. The decrease in BuChE was most pronounced in dystrophic heart muscle. The low level of BuChE measured for dystrophic animal tissue was similar to isozyme patterns found in embryonic tissue and in denervated muscle. BuChE obtained by acrylamide gel electrophoresis along with 16S AchE appears to be a useful biochemical marker of nerve-muscle interactions.