The regional distribution of somatostatin and neuropeptide Y in control and Alzheimer's disease striatum

Somatostatin-immunoreactive senile plaque-like structures in the frontal cortex and nucleus accumbens of aged tree shrews and Japanese macaques

BACKGROUND

Previously, we demonstrated decreased expression of somatostatin mRNA in aged macaque brain, particularly in the prefrontal cortex. To investigate whether or not this age-dependent decrease in mRNA is related to morphological changes, we analyzed somatostatin cells in the cerebra of aged Japanese macaques and compared them with those in rats and tree shrews, the latter of which are closely related to primates.

METHODS

Brains of aged macaques, tree shrews, and rats were investigated by immunohistochemistry with special emphasis on somatostatin.

RESULTS

We observed degenerating somatostatin-immunoreactive cells in the cortices of aged macaques and tree shrews. Somatostatin-immunoreactive senile plaque-like structures were found in areas 6 and 8 and in the nucleus accumbens of macaques, as well as in the nucleus accumbens and the cortex of aged tree shrews, where amyloid accumulations were observed.

CONCLUSIONS

Somatostatin degenerations may be related to amyloid accumulations and may play roles in impairments of cognitive functions during aging.

Human striatum: chemoarchitecture of the caudate nucleus, putamen and ventral striatum in health and Alzheimer's disease

The morphology and distribution of perikarya positive for choline acetyltransferase, somatostatin, calcium binding protein (calbindin D28K) and nicotinamide adenine dinucleotide phosphate diaphorase were surveyed in the human striatum. Choline acetyltransferase and somatostatin antibodies labeled separate populations of large striatal interneurons. Somatostatin immunoreactivity and nicotinamide adenine dinucleotide phosphate diaphorase (nitric oxide synthase) activity were completely co-localized. Calbindin antibody identified two distinct groups of striatal neurons: (1) numerous medium-sized, lightly stained neurons, probably analogous to striatopallidal projection neurons in the rat, and (2) much less numerous, large, darkly stained neurons. Half of the latter group, but none of the former, were also nicotinamide adenine dinucleotide phosphate diaphorase-positive. Somatostatin-positive and medium-sized, calbindin-positive neurons were more numerous in the caudate nucleus than in the putamen or ventral striatum. By contrast, large calbindin-immunoreactive neurons were more frequently encountered in the putamen. Choline acetyltransferase-positive neurons were evenly distributed across striatal components. In aged control subjects, the size of large, darkly stained calbindin-positive neurons was reduced relative to young subjects. Aging had no effect on somatostatin-, medium-sized calbindin-, or choline acetyltransferase-positive neurons. However, in histologically confirmed cases of Alzheimer's disease, there was a selective, 75% loss of choline acetyltransferase-immunoreactive perikarya from the ventral striatum, but not from the dorsal striatum, compared to aged controls. Furthermore, the remaining cholinergic neurons in the ventral striatum of Alzheimer's disease cases were significantly smaller than similar neurons in controls. These results indicate that various striatal components which have been shown to differ in their anatomical connectivity and functional specialization, also differ in their neurochemical signatures. The specific and marked loss of choline acetyltransferase-positive neurons from the ventral striatum in Alzheimer's disease is consistent with the characteristic cholinergic and 'limbic' pathology in this disease.

Reduced cerebral cortical but elevated striatal concentration of somatostatin-like immunoreactivity in dominantly inherited olivopontocerebellar atrophy

Somatostatin-like immunoreactivity (SLI) was measured in the brains of nine patients with dominantly inherited olivopontocerebellar atrophy (OPCA), who all had a marked deficit of the cholinergic marker choline-acetyltransferase (ChAT) in the cerebral cortex and striatum. Mean concentrations of SLI in OPCA were significantly reduced by 42-58% in parietal and occipital cortices and frontal cortical eye fields, but were normal in other cortical areas, including two subdivisions of the temporal cortex which show marked depletions of both SLI and ChAT in Alzheimer's disease. This dissociation of SLI and ChAT indicates that a cortical cholinergic deficit does not invariably lead to reduction of somatostatin. In the caudate nucleus, the region of OPCA brain having the most severe ChAT deficit (-81%), SLI levels were significantly elevated by 46% and were negatively and significantly correlated with ChAT activities (r = -0.66). The SLI alterations could be due to abnormal somatostatin metabolism or release, or an increased number of somatostatin-containing neurons and could contribute to the brain dysfunction of OPCA.

Regional distribution of neuropeptide somatostatin gene expression in the human brain

The regional distribution of mRNA coding for the neuropeptide somatostatin has been studied in the human brain by in situ hybridization histochemistry using 32P-labeled oligonucleotides. We show that somatostatin mRNA-containing neurons are widely distributed in a number of nuclei and grey areas of the human brain, including neocortex, putamen, nucleus caudatus, nucleus accumbens, amygdala, midbrain, medulla oblongata, hippocampal formation, reticular nucleus of the thalamus, and posterior nucleus of the hypothalamus. No significant hybridization signal was observed in the substantia nigra, claustrum, globus pallidus, thalamus, and cerebellum. The topographic localization of neurons containing SOM mRNA in the human brain is in agreement with previous studies using immunocytochemical or radioimmunoassay techniques. These results show that in situ hybridization histochemistry with oligonucleotide probes can be used to map the distribution of neurons expressing SOM mRNA in human postmortem materials.

Distribution of somatostatin immunoreactivity in the forebrain of the squirrel monkey: Basal ganglia and amygdala

The distribution of somatostatin immunoreactivity in the basal ganglia and amygdala of the squirrel monkey (Saimiri sciureus) was studied with specific polyclonal antibodies directed against somatostatin-28 and somatostatin-28(1-12). Both antibodies gave similar results with regard to the distribution of somatostatin-immunoreactive neuronal profiles. A moderately dense and highly heterogeneous network of somatostatin-positive fibers was observed throughout the striatum. A dorsoventral gradient of increasing immunoreactivity was noted in the striatum and the caudate nucleus was found to strain generally less intensely than the putamen. The immunoreactive fibers within the striatum were mostly thin and varicose and formed patches corresponding to the striosomes, as visualized on adjacent sections immunostained for calbindin. Although some somatostatin cell bodies rimmed the striosomes, most of the positive cells were rather uniformly scattered in the striatum. These medium-sized cells were significantly smaller in the caudate nucleus (93 microns2, S.D. = 26 microns2) than in the putamen (122 microns2, S.D. = 39 microns2), but their density was significantly higher in the caudate nucleus (29.7 cells/mm2, S.D. = 8.8 cells/mm2) than in the putamen (20.5 cells/mm2, S.D. = 7.0 cells/mm2). The nucleus accumbens stained moderately and positive cell bodies were evenly dispersed throughout this structure. In contrast, the olfactory tubercle displayed a heavily stained neuropil but positive neurons were encountered only in its polymorph layer. In the sublenticular region, dense fiber plexuses appeared in register with nonreactive cell clusters of the nucleus basalis of Meynert and of the nucleus of the anterior commissure. More caudally, a dense bundle of positive fibers was observed at the level of the ansa lenticularis, the inferior thalamic peduncle, and the adjoining bed nucleus of the stria terminalis. Several fibers contributing to this bundle were of the woolly type. Woolly fibers also coursed in the substantia innominata between the ventral aspect of the globus pallidus and the optic tract, and ascended in the internal medullary lamina separating the internal and external segments of the globus pallidus. Somatostatin-immunoreactive cell bodies were uniformly scattered throughout the substantia innominata. The various nuclei of the amygdala showed a wide range of immunoreactivity. The central nucleus was lightly reactive, whereas the intercalated masses displayed a moderate staining. A dorsoventral gradient of immunostaining was noted in the ventrolateral portion of the amygdala, the lateral nucleus being moderately to densely stained and the basal nucleus very lightly to lightly immunoreactive.(ABSTRACT TRUNCATED AT 400 WORDS)

Neuropeptide Y: an overview of central distribution, functional aspects, and possible involvement in neuropsychiatric illnesses

Neuropeptide Y (NPY) was first discovered and characterized as a 36-amino-acid peptide neurotransmitter in 1982. It is widely distributed in the central nervous system, with particularly high concentrations within several limbic and cortical regions. A number of co-localizations with other neuromessengers such as noradrenaline, somatostatin, and gamma-aminobutyric acid have been demonstrated. A large number of physiological and pharmacological actions of NPY have been suggested. Recent clinical data also suggest the involvement of NPY in several neuropsychiatric illnesses, particularly in depressive and anxiety states. This article gives a comprehensive review of central distribution of NPY and its receptors, co-localizations and interactions with other neuromessengers, genetic aspects, pharmacological and physiological actions, influence on neuroendocrine functions, and possible involvement in various neuropsychiatric illnesses.

Neuropeptide Y receptor binding sites in human brain. Possible alteration in Alzheimer's disease

Neuropeptide Y (NPY) and peptide YY (PYY) receptor sites were studied in human brain using saturation binding experiments and receptor autoradiography. Additionally, the affinities and densities of [3H]NPY binding sites were compared in the temporal cortex, hippocampus and putamen of patients dying from Alzheimer's disease (AD) and aged matched controls. High densities of [3H]NPY binding sites were found in the putamen (192 +/- 32 fmol/mg protein), followed by the hippocampus (165 +/- 42 fmol/mg protein) and temporal cortex (118 +/- 19 fmol/mg protein). Receptor autoradiography revealed that these sites were especially concentrated in certain layers of the hippocampus, laminae I and IV-V of the temporal cortex and the amygdalo-hippocampal area. No significant changes in [3H]NPY binding affinities were seen between the AD and aged-matched groups (Kd ranges: 2.5-6.8 nM). However, significant decreases in [3H]NPY receptor densities (Bmax) were found in temporal cortex (-43%) and hippocampus (-49%) in AD brains. No significant change in [3H]NPY Bmax values was found in the putamen. It is therefore possible that decreases in [3H]NPY receptor densities may be associated to the degenerative process taking place in certain brain regions in AD, although further work will be necessary to confirm this hypothesis. Part of this work was presented at the 17th Annual Meeting of the Society for Neuroscience.

Neuropeptide levels in Alzheimer's disease and dementia with frontotemporal degeneration

The CSF levels of somatostatin-LI (SLI), neuropeptide Y (NPY-LI) and Delta Sleep Inducing Peptide (DSIP-LI) have been measured in patients with dementia of Alzheimer type (DAT) and dementia with frontotemporal degeneration of non-Alzheimer type (FTD). The distribution pattern of cortical degeneration differs between these two types of dementia. DAT shows degeneration of mainly temporo-parietal and temporo-limbic structures, whereas FTD discloses its main degeneration in the frontotemporal regions (Brun, 1987). The somatostatin-LI was significantly reduced both in DAT and FTD. NPY-LI showed a significant reduction in DAT but not in FTD. A tendency to a reduction with duration of the disease was observed in DAT whereas the contrary was noted in FTD. The DSIP-LI levels were reduced in DAT and slightly increased in FTD. The study provides an evidence of neurochemical differences between the two primary degenerative dementias.

Regional distribution of neuropeptide Y mRNA in postmortem human brain

The distribution of messenger RNA encoding neuropeptide Y (NPY) was studied in 11 different postmortem human brain regions using in situ hybridization histochemistry, and RNA blot analysis. In situ hybridization data revealed that the highest numerical density of labeled cells corresponded to neurons in accumbens area, caudate nucleus, putamen, and substantia innominata. Significantly fewer NPY mRNA-containing neurons were found in frontal and parietal cortex, amygdaloid body and dentate gyrus. No NPY mRNA-containing cells were found in substantia nigra. NPY mRNA-positive neurons from all regions studied showed relatively similar labeling, as revealed by computerized image analysis. Blot analysis showed an approximately 0.8 kb NPY mRNA in all brain regions studied, except in substantia nigra and cerebellum. Densitometric scanning of the autoradiograms revealed levels of NPY mRNA in the following order: putamen greater than caudate nucleus greater than frontal cortex (Brodmann areas 4 and 6) greater than temporal cortex (Brodmann area 38) greater than parietal cortex (Brodmann areas 5 and 7) greater than frontal cortex (Brodmann area 11). Hence, although NPY mRNA is widely distributed in neurons of the human brain large regional variation exists, with the highest expression in accumbens area and parts of the basal ganglia.

Selective loss of cholinergic neurons in the ventral striatum of patients with Alzheimer disease

Cholinergic neurons were studied by immunohistochemistry with an antiserum against human choline acetyltransferase in the caudate nucleus, putamen, and ventral striatum (including the nucleus accumbens) of three patients with Alzheimer disease and three control subjects. Immunoreactive cell bodies were mapped and counted. In the ventral striatum of patients with Alzheimer disease, a 60% decrease in the number of cholinergic neurons was observed, whereas in the caudate nucleus and putamen values for control subjects and patients were similar. To determine whether all neurons in the ventral striatum were affected, neuropeptide Y-containing neurons were also immunostained, mapped, and counted. The number of these neurons was the same in control subjects and patients with Alzheimer disease, indicating that neuronal loss is not generalized in the ventral striatum and may be specific to the cholinergic population.