Search

The morphological aspects of chromaffin tissue are reviewed, based mainly on our studies on the mouse adrenal gland. Particular attention was focused on the differential fixation of adrenaline and noradrenaline, and on the uptake and storage of [3H]dopa, [3H]dopamine and related substances in the adrenaline-storing (A) and noradrenaline-storing (NA) cells. Scanning electron microscopy combined with the NaOH-maceration method was useful for demonstrating the 3-dimensional organisation of nerve terminals, chromaffin cells, glial elements and vascular elements. In transmission electron microscopy, 3 types of chromaffin cell were distinguished. They were A, NA and SGC (small granule chromaffin) cells. After glutaraldehyde fixation followed by postosmication, A cell granules showed lower electron density, whereas NA cell granules were solid and dark. This difference in appearance between A and NA cells was first explained by the hypothesis that, after glutaraldehyde fixation, most of the adrenaline dissolved, whereas noradrenaline was precipitated in situ. Later, this hypothesis was supported by a series of autoradiographic and radioisotopic assay studies using [3H]dopa, [3H]dopamine and related substances; when [3H]adrenaline occurred, radioactivity in A cells mostly disappeared in the specimen, whereas that in the NA cells remained. At 15-60 min after an i.p. injection of [3H]dopa or [3H]dopamine, the concentration of radioactivity in A cells was higher than that in NA cells. However, in hypophysectomised mice, the radioactivity was low and evenly distributed in these 2 types of chromaffin cell. It was deduced that the carrier activities for extracellular dopa and dopamine were made greater in the A cells than in the NA cells by the pituitary gland. The hypophysectomy effects were restored by i.p. administration of ACTH.

The pre- and postganglionic sympathetic innervation of the guinea-pig adrenal medulla was investigated using the retrograde neuronal tracers Fast Blue and WGA-HRP. Labelled preganglionic cell bodies were located in the intermediolateral horn of spinal segments T3-L2, the majority (73.9%) were found between T6-T12 representing 70.2% of the total number of labelled cells; the segment T10 contained the largest number of labelled neurons. Labelled postganglionic cell bodies were found in the paravertebral ganglia between vertebral levels T3-T12 (representing 22.6% of the total labelled neurons), the maximum number was found at T10. In addition, labelled neurons were found in the suprarenal ganglion (representing 7.2%). No labelled cells were found in the coeliac ganglia. The labelled neurons were found ipsilateral to the site of injection into the left adrenal gland. It is concluded that the guinea-pig adrenal gland receives both a pre- and a significant postganglionic sympathetic innervation. The destination of these nerve fibres within the adrenal gland has yet to be determined.

The innervation of the adrenal medulla has been investigated in normal Wistar rats from birth to old age and ultrastructural findings compared with biochemical markers of the cholinergic innervation of the adrenal gland and catecholamine storage. Morphological evidence of the immaturity of the innervation during the first postnatal week is provided and using quantitative morphometry the innervation of chromaffin cells is shown to reach a mean total of 5.4 synapses per chromaffin cell during the period 26 days to 12 weeks of age. The variation in contents of synaptic profiles is discussed in the light of recent work that demonstrates a major sensory as well as visceral efferent innervation of the gland. Adrenal medullary neurons usually occur in closely packed groups, intimately associated with Schwann cells. Axodendritic and axosomatic synapses on these neurons are described and the likely origin of axonal processes innervating the neurons discussed. In old age the density of innervation remains the same as in young adult animals even though the medulla shows evidence of hyperplasia and hypertrophy of individual chromaffin cells.

Following the injection of 3-5 microliters of horseradish peroxidase or Fast Blue into the left adrenal medulla, labelling of neurons in the vagal sensory ganglia of rats and guinea-pigs has been demonstrated in all cases. Labelled vagal motor neurons with cell bodies in the dorsal motor nucleus of the vagi have been demonstrated in all cases in the guinea-pig and occasionally in the rat. The possible reasons for this variation are discussed. In both rat and guinea-pig the motor and sensory vagal innervation of the rat adrenal is derived from bilaterally situated cell bodies and the proportion of innervation derived from the two sides is approximately equal but with a slight ipsilateral predominance.

This study has investigated the flow of blood to the adrenal gland of the anaesthetised rat under basal conditions and after the induction of haemorrhagic hypotension. By combining the use of radioactive microspheres with subsequent serial sectioning we have been able not only to assess total adrenal blood flow but also to determine its distribution within the gland. In normal anaesthetised rats approximately 7.4% of the adrenal blood flow passes directly to the medulla, the remainder being distributed to the cortex. This proportional flow is also observed immediately after withdrawal of sufficient venous blood to cause a 32% reduction in systolic blood pressure. However, the latter procedure results in a 52% increase in the total blood flow to the gland despite the associated fall in cardiac output.

The sensory innervation of the guinea-pig adrenal medulla was studied using the fluorescent retrograde tracer Fast Blue. Labelled neurons were found in the dorsal root ganglia at segments T3-L2, the greatest contribution arising from T10 representing 15.2% of the total number of labelled cells. Labelling was ipsilateral to the site of injection of tracer into the adrenal medulla and the labelled neurons ranged in size between 6 and 36 microns. The overall size distribution fell into two groups of between 6 and 15 microns and of between 18 and 36 microns. Presumed sensory nerve endings were found in the adrenal medulla associated with chromaffin cells and vascular elements. The nerve endings were ovoid to spherical and densely packed with mitochondria. It is concluded that events occurring in the guinea-pig adrenal medulla, and possibly cortex, are monitored by sensory nerve terminals whose cell bodies lie within dorsal root ganglia. The exact distribution of the sensory nerve terminals within the adrenal gland is yet to be determined.

The sympathetic innervation of the rat adrenal medulla was studied using the fluorescent tracer Fast Blue. Labelled preganglionic cell bodies were located in the intermediolateral horn of the spinal cord at segments T1 and L1; the greatest number was found in T9. The ILp nucleus contained 95%, the ILf nucleus 4%, and the IC nucleus 1% of the total number of labelled preganglionic cells. Labelled postganglionic cell bodies were found in the sympathetic ganglia at levels T4 and T12; the maximum number were located in ganglia at T9 and T10. In addition, labelled cells were found in the suprarenal ganglion. No labelled cells were found in the ganglia of the coeliac plexus. The number of labelled preganglionic cells in the spinal cord accounted for 88.6%, the labelled cells in the sympathetic chain for 8.7% and those in the suprarenal ganglia for 2.7% of the total number of labelled cells found. The detailed distribution within the gland of postganglionic axons has yet to be determined, but it is thought that some are destined for the adrenal cortex.

The localisation of chromogranins A and B, met-enkephalin-arg6-gly7-leu8 (met-enk 8) and protein gene product 9.5 (PGP 9.5) in the adrenal medulla and extra-adrenal chromaffin tissue has been studied in the developing rat by immunogold-silver staining. In the adult rat adrenal the cytoplasm of all medullary chromaffin cells showed a positive response with chromogranin A and B; in each case occasional groups of cells with a low reactivity that may have been NA cells were seen. Chromogranin A was first detected in adrenal medullary and extra-adrenal chromaffin cells at 18 days of gestation whilst chromogranin B was not detected in animals younger than 7 days. In 15 days old animals the adrenal medullary response to A and B was of the same intensity as that seen in the adult. Less than 1% of adult medullary chromaffin cells were responsive to met-enk 8 staining and medullary cells were unreactive in the fetus, with only extra-adrenal chromaffin tissue responding prenatally. During the first postnatal week immunoreactive cells appeared in the adrenal medulla in considerably greater proportions than in the adult gland. In contrast, positively stained nerve terminals associated with chromaffin cells and abundant in the adult adrenal were not detected during the first week of life. Immunoreactive nerve terminals were first seen early in the second week of life at a time when positive chromaffin cells were becoming less common. PGP 9.5 was located in all chromaffin cells of the adult adrenal and was readily detected in chromaffin cells in the adrenal and in extra-adrenal locations of the earliest stage examined (E16). Our findings suggest that the ontogenesis of the chromogranin-like immunostaining reflects the maturation of chromaffin granules and the PGP 9.5 immunostaining detected a protein common to cells of neuronal origin and expressed at an early stage of differentiation. The reciprocal relationship between the presence of enkephalins in chromaffin cells and in their presynaptic terminals merits further investigation.

Adrenal medullary arteries have been followed from their origins in the capsular or subcapsular plexus to the medullary plexus in the mouse, rat, hamster, cat, guinea-pig, rabbit and bovine. Adrenal medullary arteries have a zonal distribution to the medullary capillary plexus and the number of arteries present in species variable and proportional to the size of the medulla. Cortical veins drain through distinct venous channels between groups of chromaffin cells, and may run adjacent to either adrenaline-storing (A) or noradrenaline-storing (NA) cells. In the normal unstimulated gland both A and NA cells lie adjacent to capillaries of the medullary plexus which derive their blood supply from adrenal medullary arteries. In the normal unstimulated gland no evidence of selective supply of arterial or venous blood to either A or NA cells has been obtained.

Following intravenous or intraperitoneal injections of 3H-l-dopa radioactivity in adrenal medullary chromaffin cells of the mouse has been followed by autoradiographic and radioisotopic assay methods. For autoradiographic purposes tissues were fixed in either phosphate buffered glutaraldehyde or in 1-5% potassium dichromate in glutaraldehyde pH 7-2. A high initial radioactivity is demonstrated by both autoradiography and radioisotopic assay. During the first hour dopamine is the predominant labelled amine. Between 1 and 4 hours after the injection noradrenaline is in highest concentration and from 1 to 24 hours labelled adrenaline appears and becomes the predominant labelled amine (about 70%) after 24 hours. Autoradiographs of tissues fixed in glutaraldehyde alone confirm that adrenaline is lost from sections by this technique. When 1-5% potassium dichromate is added to glutaraldehyde then adrenline is fixed in A cells and these show only a slightly decreased grain count as compared to NA cells at 5 days; this slight decrease is probably due to more rapid turnover of amines by A cells.

Administration of hydrocortisone in vivo to pregnant mice between the eighth and sixteenth days of gestation leads to highly significant increases in both the volume and the labelling index of extra-adrenal chromaffin tissue (as represented by the para-aortic body) of 16 days fetal mice. The labelling index of intra-adrenal chromaffin cells increases slightly after hydrocortisone administration, but the volume of the adrenal chromaffin tissue was not assessed since the groups of chromaffin cells had not in all cases aggregated to form a distinct medulla. Hydrocortisone administration to mice for the last week of gestation also leads to a highly significant increase in the size and the labelling index of the perinatal para-aortic body. Although an increase in the labelling index of perinatal intra-adrenal chromaffin cells is brought about, this is much less marked than that of the extra-adrenal tissue and neither is it reflected by any increase in the volume of the now discernible adrenal medulla. These increases in size and labelling index of the para-aortic body constitute a hyperplastic response rather than a hypertrophic response. Various possible mechanisms and implications are discussed in the light of the development of chromaffin tissue and the effects of the cortex and its secretions on the medulla. Associated effects of hydrocortisone noted in this work are the resultant marked diminution of adrenocortical volume in the perinatal gland, and a slight fall in the labelling index of perinatal intra-adrenal haemopoietic cells.