Ultrastructural studies on the hypothalamic neurosecretory neurones of the rat. I. The paraventricular neurones of the non-treated rat

Ultrastructure of Hypothalamic Neurons and of the Median Eminence

Our light, and electron microscopic (EM) findings within the hypothalamic supraoptic (SO) and paraventricular (PV) nuclei of the normal female rabbit are in agreement with those reported earlier by other investigators for the same nuclei of the dog and rat. The neurons of these nuclei are the hypothalamic synthesis sites of the neurohypophyseal hormones.

With the exception of the arcuate nucleus, none of the hypothalamic nuclei associated with the control of adenohypohpyseal function have been studied extensively with the electron microscope. On the basis of our EM findings within the female rabbit hypothalamus, all neurons observed within the preoptic (PO) and suprachiasmatic (SCH) nuclei of the non-mated control animal were morphologically identical to the conventional neuron as described by Peters, Palay and Webster (1970). However, following coitus, castration and laparotomy, many neurons of these nuclei showed subcellular changes that have been repeatedly associated with enhanced protein synthesis. These large ‘neurosecretory’ neurons were usually located near capillaries and characterized by their well developed Rough endoplasmic reticulum (RER) and Golgi profiles, dense populations of mitochondria and lysosomes and by the presence of a homogeneous population of densecore vesicles (DCV) showing a peak distribution of 120-140 nm. Since similar neurons were not observed within the PO and SCH of the normal control rabbit it is suggested that we were observing functional states of the same type of neuron and that these ultrastructural changes occur in response to endocrine manipulation.

Two types of neurons described as ‘pale’ and ‘dark’ were observed within the arcuate nucleus of both the control and experimental female rabbit. Ultrastructurally, these neuron types were identical to those described by other investigators for the rat. It has been suggested that the ‘pale’ and ‘dark’ neurons of this hypothalamic nucleus represent functional states of the same type of cell. However, increases in the ratio of ‘dark’ to ‘pale’ neurons as observed within the arcuate nucleus of the rat following castration, were not seen in the rabbit. Similar findings were also not evident within the arcuate nucleus of the female rabbit following coitus.

As far as could be determined, all neurons of the ventromedial (VMN) nuclei of both the control and experimental rabbit were morphologically identical to the smaller, conventional type neuron. Certainly, ultrastructural changes similar to those observed within the PO and SCH nuclei of the female rabbit following coitus, castration or laparotomy, were never observed.

The basic zonation and subcellular organization of the female rabbit Median Eminence (ME) is similar to that described for other mammalian species. Our EM findings within the external layer of the rabbit ME, however, are not entirely in agreement with the earlier study of Duffy and Menefeef 1965). These investigators reported only one population of DCV within the axon terminals of the rabbit ME external layer. We feel that we have ultrastructural evidence for the presence of at least two distinct populations of DCV within this layer of the rabbit ME. Furthermore, since these vesicle populations occurred within separate axon profiles and terminals, differences in their content and origin are suggested.

Certainly, the relationship between releasing factors (RF) and the various populations of DCV observed within the external layer of the mammalian ME is not well established. The smaller (90 nm - 100 nm) DCV we have observed probably contain the catecholamines, while those of larger (120 nm - 140 nm) diameters may well represent the carriers of the RF associated with gonadotropic activity. The latter view is based primarily on our finding or numerous ‘vesicle ghosts’ within the axon terminals abutting the perivascular space (PVS) of portal capillaries of rabbits sacrificed at 10 minutes post-coitus. The mean diameters of 137±14 nm obtained for these ghosts strongly supports the suggested depletion of only the larger of the two DCV populations. Similar changes were not apparent within the axon terminals containing homogenous populations of only the smaller DCV.

Unquestionably, the precise hypothalamic synthesis sites for the RF associated with control of adenohypophyseal function, continues to provoke comment. From the results obtained from countless studies that have employed a variety of neuroendocrinilogical techniques, two main hypothalamic centers of RF synthesis have been suggested: a) the medial basal hypothalamus (MBH) or hypophysiotropic area (HTA) and b) the anterior hypothalamus. The ultrastructural studies carried out to date within this laboratoiy are in favour of the latter for the following reasons:

1) — the presence of large DCV and ‘vesicle ghosts’ within the external layer of the rabbit ME with diameters similar to those of the large (120-150 nm) DCV synthesized within the PO and SCH nuclei of the same animal in response to coitus, castration and laparotomy.

2) — the absence of evidence for the storage of these large DCV within the somata of PO and SCH nuclei, suggesting their immediate transport toward the ME.

3) — the absence of any ultrastructural changes within neuron somata of the rabbit arcuate nuclei which might reflect enhanced neurosecretory activity in response to coitus and/or castration.

These ultrastructural findings within the rabbit hypothalamus may, therefore, provide the first evidence of a morphological nature for the actual release of RF from their ME storage sites, as well as their synthesis within certain neurons of the anterior hypothalamus.

The magnocellular neurosecretory system of the hamster hypothalamus: an ultrastructural and morphometric study during lifetime

A quantitative study regarding the age-related changes occurring in the nucleus and the somatic organelles of neurosecretory magnocellular neurons of the hypothalamo neurohypophyseal system (HNS) was carried out in the hamster at six age-points during animal life. The magnocellular cells of both parts of the supraoptic nucleus (SON) and the paraventricular nucleus (PVN) of male Syrian hamsters between 3 and 30 months of age were examined ultrastructurally. Cells of all age groups present the same morphological ultrastructure. Standard manual morphometric techniques are used to calculate the following parameters related directly or indirectly with cellular activity: nuclear area, nucleolar area, nuclear invagination index and volumetric fractions of some intracellular structures (Golgi apparatus, mitochondria, rough endoplasmic reticulum and lipofuscin). With respect to the cell nucleus, the parameters are not modified during aging. No significant differences in the volume density of subcellular components, except lipofuscin, were detected at the age groups studies. However, there is a positive linear trend among all parameters and age except for the rough endoplasmic reticulum. Our results suggest maintenance of the synthetic activity of the magnocellular neurons in the hamster during aging but in no case an increase in their metabolic activity.

Ultrastructural characterization of the central component of the medial preoptic nucleus

The fine structure of the sexually dimorphic central part of the medial preoptic nucleus was studied in male and female rats. Neurons of this subnucleus are characterized by having convoluted nuclear membranes and fibrillar nuclear inclusions. Within the perinuclear domain there is a well developed neurosecretory apparatus, i.e., Golgi complexes, assorted types of secretory granules, and lysosome-like bodies, whereas the marginal cytosol contains stacked cisterns of rough endoplasmic reticulum, nucleolus-like inclusions, and few mitochondria. The neuropil is made up of thick dendrites, fascicles of thin unmyelinated axons, and few small myelinated axons. Additionally four types of synaptic boutons were found according to the morphology of their vesicular contents. No apparent qualitative sex differences were noted. These cytologic features which suggest secretory activity, coupled with reports that have shown the presence of several neurohormones and peptides within the central part of the medial preoptic nucleus lend further support to the neuroendocrine nature of these neurons.

The neurosecretory system of the adult Melanogryllus desertus Pall. (Orthoptera, Gryllidae). I. Ultrastructural study of the median neurosecretory cells in the brain

Based on the nature of their granules, eight prinicipal types of neurons, six of which are thought to be neurosecretory, are recognized in the median neurosecretory cell group of the brain of Melanogryllus desertus. Most of the neurosecretory cells contain granules with diameters of 200-300 nm. In a few the granules are smaller with diameters varying from 60-100 nm. Most of the cells have well developed Golgi areas and dense bodies of different sizes. Dense bodies are closely associated the neurosecretory granules. Accumulations of electron-dense granular material occur in expanded cisterns of endoplasmic reticulum, particularly in type-I cells.

Cytodifferentiation of the supraoptic nucleus correlated with vasopressin synthesis in the rat

This study demonstrates that neurons in the supraoptic nucleus attain many of the prerequisites for functional activity prior to birth. Immunoassayable vasopressin (VP) was detected in the hypothalamo-neurohypophyseal system (HNS) of the rat as early as 17 days post-coitus (dpc). Vasopressin concentrations increased 3--6-fold daily from an average of 21 pg/animal on 17 dpc to 5984 pg/animal at 21 dpc. The daily increases were highly significant (P less than 0.001). Between 21 dpc and the morning of the day of birth on 22 dpc, a further significant increase (P less than 0.05) occurred to a mean level of 9672 pg VP/animal. Birth usually occurred on the afternoon of the 22nd day. Parturition did not seem to deplete VP stores in the HNS. Differentiation of the magnocellular neurons in the supraoptic nucleus closely paralleled the appearance and increases in VP. It was first possible to dintinguish a supraopic nucleus in the 17 dpc rat and to identify dense core granules in the developing neurons of the nucleus. Cytodifferentiation of the magnocellular neurons was essentially complete by 21 dpc. Synaptic contacts could not be found on the soma and dendrites of the supraoptic neurons until 21 dpc and were extremely rare throughout the period examined.

Studies on vinblastine-induced autophagocytosis in mouse liver. II. Origin of membranes and acquisition of acid phosphatase

The origin of the membranes of autophagic vacuoles (AV) and acquisition of acid phosphatase into AV's were studied in vinblastine-induced autophagocytosis (VBL, 50 mg/kg, i.p.) in mouse hepatocytes. Using unbuffered OsO4, very intense staining was observed in the outer cisternae of the Golgi apparatus and also frequently in the cavity between the double membranes obviously destined to form AV's as well as in the cavity between the double membranes of newly formed AV's. There may occur a transformation process in the membranes limiting an AV analogous to that observed at the Golgi cisternae. The transformation of the outer AV membrane occurs independently of fusion with lysosomes. Inosine diphosphatase activity was localized within the cisternae and on the membranes of the endoplasmic recticulum and occasionally within the innermost cisterna of the Golgi apparatus. The results together with the unbuffered OsO4-staining pattern suggest that the membranes of most AV's are derived from the transformed smooth surfaced cisternae of the endoplasmic reticulum which do not have inosine diphosphatase activity. Acid phosphatase activity was localized in lysosomes, occasionally within the innermost cisternae of the Golgi apparatus, between the double membranes of a few newly formed AV's and within most older single membranes of a few newly formed AV's and within most older single membrane-limited AV's. VBL did not prevent the fusion of lysosomes with AV's.

Light microscope radioautographic study of glycoprotein secretion in the hypothalamic-neurohypophysial system of the rat, after L-fucose-3H injection

Fucose-3H was injected into the cerebral ventricle of rats that were killed at several time intervals after injection. Semi-thin sections of the supraoptic nucleus and neurohypophysis were processed for radioautography and analysed quantitatively. Silver grains indicating the site of fucoselabeled glycoproteins were first located at the perinuclear region of the secretory neurons. The highest silver-grain density in these cells was observed at 2 h after injection , declining afterwards. Silver grains over the neurohypophysis were observed from 2 h on, reached a peak at 1 day after injection and decreased in the subsequent time intervals. The distributions of the silver grains over the neurohypophysis fitted Poissonian distributions and these were shown to be heterogeneous at the several time intervals. Pituicytes were not labeled. The percentage of silver grains over the Herring bodies increased with time. In rats deprived of water after fucose-3H injection there was a great increase in the release of labeled glycoproteins from the neurohypophysis. These results indicate that the glycoproteins synthesized by the secretory neurons of the hypothalamic nuclei are secreted in the neurohypophysis.

Neuronal and non-neuronal control of the neurosecretory caudo-dorsal cells of the freshwater snail Lymnaea stagnalis (L.)

The cerebral ganglia of the freshwater snail Lymnaea stagnalis contain two clusters of neurosecretory Caudo-Dorsal Cells (CDC). These cells produce a neurohormone which stimulates ovulation. Ganglion transplantation and quantitative electron microscopy show that neuronal isolation of the cerebral ganglia complex (CCC) results in an activation of the CDC. It was, therefore, concluded that the CDC are controlled by an inhibitory neuronal input originating outside the cerebral ganglia. Ultrastructural studies on synaptic degeneration in the CCC suggest that this input reaches the CDC via a special type of synapse-like structure, the type C-SLS. Furthermore, transplantation of CCC into acceptor snails leads to a reduced release and an increased intracellular brekdown of neurohormone in the CDC of the nervous system of the acceptors. It is supposed that these phenomena are caused by the release of an (unknown) factor from the transplanted CCC. Special attention was given to the formation and degradation of a peculiar type of neurohormone granule, the large electron dense granule. The physiological significance of the neuronal and non-neuronal control mechanisms which regulate CDC activity is discussed.