Search

We developed a rapid and sensitive radioimmunohistochemical method for the quantification of tyrosine hydroxylase (TH) at both the anatomical and cellular level. Coronal tissue sections from fresh-frozen rat brains were incubated in the presence of a TH monoclonal antibody. The reaction was revealed with a 35S-labeled secondary antibody. TH content was quantified in catecholaminergic brain areas by measuring optical density on autoradiographic films or silver grain density on autoradiographic emulsion-coated sections. Regional TH concentrations determined in the locus ceruleus (LC), substantia nigra pars compacta (SNC), and ventral tegmental area (VTA) were significantly increased by 45% after reserpine treatment in the LC but unchanged in the SNC and VTA. Microscopic examination of TH radioimmunolabeling showed a heavy accumulation of silver grains over catecholaminergic cell bodies. In the LC, grain density per cell was heterogeneous and higher in the ventral than in the dorsal part of the structure. After reserpine treatment, TH levels were significantly increased (57%) in the neurons of the LC but not in those of the SNC or VTA. The data support the validity of this radioimmunohistochemical method as a tool for quantifying TH protein at the cellular level and they confirm that TH protein content is differentially regulated in noradrenergic and dopaminergic neurons in response to reserpine.

I 2 -imidazoline binding site (I 2 BS) has been identified with a regulatory site located on a sub-population of monoamine oxidase (MAO)-A and -B. Previous studies showed a modification of MAO and I 2 BS in the elderly and in neurodegenerative processes such as Alzheimer's disease. In the present study, we studied the potential modification of I 2 binding sites and monoamine oxidases in Parkinson's disease. Putamen and cerebral cortex were collected from 17 normal subjects (79±12 yr) and 16 patients (76±9 yr) affected by Parkinson's disease. In mitochondrial preparations, radioligand binding studies with [ 3 H]idaxozan showed that putamen and frontal cortex express equivalent amount of I 2 BS. The density and affinity of I 2 BS were similar in normal subjects (putamen: B max =207±58 fmol/mg of protein, K d =10.1±3.4 nM; cerebral cortex: B max = 193 ± 54 fmol/mg of protein, K d = 12.8 ± 6.8 nM) and Parkinson's disease patients (putamen: B max =193±60 fmol/mg of protein, K d =9.8±4.6 nM; cerebral cortex: B max =199±49 fmol/mg of protein, K d = 15.9 ± 8.1 nM). The activity of total monoamine oxidase and monoamine oxidase B, measured by [ 14 C]tyramine and [ 14 C]phenylethylamine oxidation, respectively, were higher in putamen than in cerebral cortex. No differences have been detected in the enzyme activity between normal and pathological subjects. These data suggest that, although MAO and I 2 BS may play a role in the development of Parkinson's disease, they are not altered in the chronic phase of this disease.

Stereotaxic injection of a limited amount of 6-hydroxydopamine in the lateral part of the rat substantia nigra induces a partial degeneration of the nigrostriatal dopaminergic system. This animal model in which the destruction of the dopaminergic nigral cell population reaches approximately 50% could be considered as a preclinical Parkinson's model. Autoradiography of dopaminergic uptake sites performed with a specific marker ([3H]GBR 12935) allowed the precise determination of dopaminergic denervated and non-denervated areas in the striatum 1 month after partial lesion of the substantia nigra pars compacta. In both striatal areas, dopaminergic D1 and D2 receptor densities and dopaminergic D2 and preproenkephalin mRNAs levels were measured by autoradiography and in situ hybridization coupled to an image analysis system. Our results show that in the denervated striatal subregion, none of the dopaminergic targets were modified, contrary to the observations made after complete lesion of the nigrostriatal DA system at the same post-lesion delay. However, striatal Fos activation induced by amphetamine (5 mg/kg i.p., 2 h before killing) revealed that the number of Fos-positive cells detected in the denervated striatal subregion was lower than that observed in the non-denervated one. These data argue in favour of the existence of compensatory mechanisms different from the up-regulation of DA receptor densities, thereby allowing the maintenance of striatal dopaminergic transmission. Such mechanisms could contribute to the delay of the appearance of neurological symptoms (which are reported to be clinically apparent only when depletion of striatal dopamine levels reaches near 80%) in Parkinsonian patients.

A neuropathological examination was performed on a patient with parkinsonism induced by prolonged exposure to a mixture of aliphatic hydrocarbons, mainly n-hexane and halogenated compounds. The patient developed a rapid-course disease that progressed even after withdrawal from the toxic exposure. Pathological examination and immunohistochemical analysis of the brain revealed severe and widespread dopaminergic neuronal loss, associated with severe gliosis, in the substantia nigra, and almost complete loss of tyrosine hydroxylase immunostaining in the striatum. No Lewy bodies were detected. Neuronal loss was also observed in the periaqueductal gray matter, locus ceruleus, and pedunculopontine nucleus. These changes, combined with the moderate anemia due to marrow suppression, and the mild axonal neuropathy observed in vivo, are suggestive of a hydrocarbon toxic insult.

DNA fragments from a genomic library were used to establish the partial structure of the human dopamine D{sub 3} receptor gene (DRD3). Its coding sequence contains 6 exons and stretches over 40,000 base pairs. The complete DRD3 transcript and three shorter variants, in which the second and/or third exon are deleted, were detected in similar proportions in brains from four controls and three psychiatric patients. The Msp I polymorphism was localized in the fifth intron of the gene, 40,000 base pairs downstream the Bal I polymorphism and a PCR-based method was developed for genotyping this polymorphism. The distributions of the Msp I and Bal I genotypes were not independent in 297 individuals ({chi}{sup 2} = 10.5, df = 4, P = 0.03), but only a weak association was found between allele 1 of the Bal I polymorphism and allele 2 of the Msp I polymorphism ({chi}{sup 2} = 3.99, df = 1, P = 0.04). The previously reported association between homozygosity at both alleles of the Bal I polymorphism and schizophrenia was presently maintained in an extended sample, comprising 119 DSM-III-R chronic schizophrenics and 85 controls ({chi}{sup 2}= 5.3, df = 1, P = 0.02) and found more important inmore » males than in females. The presence of the Bal I allele 2 is associated with an early age at onset, particularly in males (df = 35, t value = 2.6, P = 0.014). In the same sample, allelic frequencies, genotype counts, and proportion of homozygotes for the Msp I polymorphism did not differ between schizophrenics and controls ({chi}{sup 2}= 0.06, df = 1, P = 0.80, {chi}{sup 2} = 0.22, df = 1, P = 0.90 and {chi}{sup 2} = 0.16, df = 1, P = 0.69, respectively). The large distance of the Msp I polymorphism from the Bal I polymorphism and its localization in the 3{prime} part of the gene may explain the discrepant results obtained with the two polymorphisms. 36 refs., 2 figs., 4 tabs.« less

The cellular expression of the genes encoding the neuropeptides enkephalin and substance P were examined in the caudate nucleus and putamen of parkinsonian 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP)-treated cynomolgus monkeys by in situ hybridization using radioactive antisense oligonucleotides coupled with computer-assisted image analysis. Behavioural evaluation of the animals revealed two levels of motor impairment; one group moderately impaired and the other severely disabled. A marked increase in the cellular content of preproenkephalin A messenger RNA was observed in medium-sized (106 +/- 9 microns2) cells in the caudate-putamen of all MPTP animals when compared with controls, the increase being greatest in the most severely impaired animals. By contrast, a marked reduction in the cellular abundance of preprotachykinin gene expression was detected in striatal cells (101 +/- 16 microns2) of these same MPTP animals. These changes in neuropeptide gene expression were not associated with a change in the density (approximately 10 cells per mm2) of messenger RNA-expressing cells. L-DOPA treatment of two of the severely-impaired MPTP monkeys resulted in a dissociation of expression of these two genes: the cellular abundance of preproenkephalin A remained elevated whilst preprotachykinin levels were normalized and comparable with controls. No change in the cellular abundance of preprotachykinin messenger RNA was observed in cells of the insular cortex or a small discrete population of large cells (208 +/- 27 microns2) in the ventral putamen. These results demonstrate that MPTP treatment of primates results in a marked potentiation in preproenkephalin messenger RNA coupled with a attenuation in preprotachykinin messenger RNA in the dopamine-denervated caudate-putamen. L-DOPA therapy given on an intermittent schedule reverses the decrease in preprotachykinin messenger RNA, but fails to reverse the increase in preproenkephalin messenger RNA in the same animal. These observations suggest that a dissociation of the activity of these two neuropeptide systems may underlie the improvement in motor skill that accompanies dopamine replacement therapy and that this dissociation may be instrumental in the long-term complications associated with L-DOPA therapy.

Degeneration of the nigro-striatal dopaminergic neurons occurring in Parkinson's disease is associated with an increase in iron concentrations in the substantia nigra. As this metal catalyzes the production of free radicals, and oxidative stress may participate in the cascade of events ending in cell death, this increase in iron content may be involved in dopaminergic neuronal death. The localization and number of receptors for transferrin were investigated postmortem by quantitative autoradiography of iodinated-ferrotransferrin binding in the basal ganglia from controls, parkinsonian patients and MPTP-lesioned monkeys. In human controls, specific [125i] ferrotransferrin binding-site density was highest in the putamen and the caudate nucleus, and lowest in the globus pallidus. In parkinsonian patients, it was increased in the putamen and caudate nucleus, while in MPTP-intoxicated monkeys, there was a tendency for levels to decrease in these two regions. An inverse relationship between binding density and iron content reported in the studied regions supports the assumption of a possible capture of iron at the level of dopaminergic terminals and, in parkinsonian patients, on other cellular elements as well. These results suggest an involvement of transferrin receptors in iron uptake in the striatum of patients with Parkinson's disease, with differences between the 'acute' nigro-striatal MPTP-induced degeneration syndrome and the chronic long lasting human disease.

Clinical symptoms in Parkinson's disease do not appear until almost total depletion of dopamine has occurred in the striatum, suggesting the existence of compensatory mechanisms to offset the loss of nigrostriatal dopaminergic neurons. This compensation has been attributed mainly to an increased turnover of dopamine in the remaining dopaminergic neurons. Besides this biochemical phenomenon intrinsic to dopaminergic neurons, we tested whether morphological changes in the nerve afferents to the dopaminergic neurons could participate in these compensatory mechanisms. The afferents to the dendrites of dopaminergic neurons were analyzed ultrastructurally in the substantia nigra of parkinsonian patients and matched controls, using simultaneous histochemical detection of acetylcholine-like cation and tyrosine hydroxylase. The size of acetylcholine-like cation-containing terminals in contact with dopaminergic dendrites increased significantly by 38% in the substantia nigra of parkinsonian patients; whereas their number per section of dopaminergic dendrite showed an increase of 60%, although not reaching statistical significance. The number of the terminals devoid of acetylcholine-like cation per section of dopaminergic dendrite decreased significantly by 52% in the substantia nigra of parkinsonian patients. These results suggest (1) a plasticity of excitatory cholinergic neurons targeting nigral dopaminergic neurons and (2) an involution of noncholinergic nerve terminals, mostly originating from inhibitory nigral, pallidal, and striatal GABAergic neurons. The findings provide evidence of a capacity for neuronal plasticity in the elderly human brain, even in the presence of neurodegenerative disorders.

Parkinson's disease is characterized by the degeneration of melanized dopaminergic neurons of the substantia nigra. The functional capacity of the surviving dopaminergic neurons is affected, as suggested by the subnormal levels of tyrosine hydroxylase messenger RNA and protein found in the remaining cells. The reduced expression of tyrosine hydroxylase may be due to either the evolving neurodegenerative process or its downregulation, possibly secondary to chronic levodopa treatment. The cellular content of tyrosine hydroxylase was determined in the mesencephalon from 16 Macaca fascicularis monkeys, using a semiquantitative immunocytochemical method. Thirteen monkeys were rendered parkinsonian by weekly intravenous injections of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) for 2 (subacute treatment) or 20 (chronic treatment) weeks. Three of the monkeys received levodopa and 3 others received GM1 ganglioside. The loss of dopaminergic neurons in the mesencephalon of the MPTP-intoxicated monkeys was severe in the substantia nigra, intermediate in cell groups A8 and A10, and almost undetectable in the central gray substance. After both subacute and chronic treatment, the cellular content of tyrosine hydroxylase was reduced by 40% in the surviving neurons of the lesioned substantia nigra, but by less in the other mesencephalic dopaminergic regions. Neuronal survival and tyrosine hydroxylase content in monkeys that had received levodopa were not significantly different. The cellular content of tyrosine hydroxylase was increased in the substantia nigra of the monkeys that received GM1 ganglioside injections. The results show that the decreased expression of tyrosine hydroxylase found in nigral dopaminergic neurons after partial degeneration of the mesostriatal dopaminergic system is not influenced by levodopa treatment and is partially reversed by GM1 ganglioside administration.

The distribution of manganese-dependent superoxide dismutase, an enzyme that transforms superoxide radicals into hydrogen peroxide, was studied in the human brain post mortem using a sheep polyclonal antiserum raised against the enzyme from liver mitochondria. One band, corresponding to a protein of 22,000 mol. wt was detected in the human brain by western blot analysis. At the light-microscopy level, a punctate immunostaining was observed in the neuropil and in some but not all, glial and neuronal cell bodies. Electron-microscopy revealed that the staining was exclusively confined to the inner mitochondrial membrane. A heterogeneous distribution of manganese-dependent superoxide dismutase was observed in the human brain. In the forebrain, numerous immunostained neurons were detected in the striatum, thalamus, pallidal complex and the nucleus basalis of Meynert. In the cerebellum, only granular and Purkinje cells were immunostained. Various nuclei from the brainstem displayed superoxide dismutase immunoreactivity, including the cranial nerve nuclei, the nucleus supratrochlearis, the red nucleus, the substantia nigra, the nucleus cuneiformis and subcuneiformis, the nucleus parabigeminal, the nucleus centralis superior, the nucleus supraspinalis, the nucleus of the medullae oblongata and the gigantocellularis nucleus. Large pyramidal neurons containing superoxide dismutase were detected in the CA subsectors, the hilus of the hippocampus and the cerebral cortex. Smaller immunostained neurons were also observed in layers I, IV and VI of all cortical regions studied. The distribution of immunostained glial cells was more limited, and restricted to the internal and external capsules, the hypothalamus, the red nucleus, the pyramidal white matter and surrounding areas, the cerebral cortex and the sub-ependymal layer, the alveus and the stratum oriens of the hippocampus. This heterogeneous but not ubiquitous distribution of cells expressing manganese-dependent superoxide dismutase suggests that not all cells in the human brain are protected to the same extent against the deleterious effects of superoxide.

Activated glial cells observed in the substantia nigra in Parkinson's disease may participate in the mechanism of nerve cell death by producing toxic substances such as cytokines. Among these compounds, tumor necrosis factor-α (TNF) is of interest because it can provoke cell death. We detected TNF-immunoreactive glial cells in the substantia nigra of parkinsonian patients but not in those of control subjects. Immunoreactivity for TNF receptors was found in cell bodies and processes of most dopaminergic neurons of control and parkinsonian subjects, suggesting that nigral dopaminergic neurons might be sensitive to TNF produced in Parkinson's disease. These results suggest that TNF may participate in the degenerative processes occurring in Parkinson's disease, at least after a primary insult inducing a reactive gliosis.

In human brain, [3H]glibenclamide binds with high affinity (KD about 3.5 nM) to sulfonylurea binding sites which are associated with ATP-sensitive potassium (KATP) channels. Regarding to the important neuromodulatory action of KATP channels in some neuronal populations, sulfonylurea binding sites were measured in several cortical areas (frontal and temporal cortex, hippocampus) and striatum (caudate nucleus and putamen) in controls and patients with Parkinson's disease or progressive supranuclear palsy. There was no modification of [3H]glibenclamide specific binding in the cerebral regions studied in both pathologies. These results indicate that KATP channels do not seem to be involved in the pathophysiology of these degenerative processes. Brain samples from five patients with Huntington's disease were studied. A small decrease in sulfonylurea binding sites was measured in the frontal cortex, caudate nucleus and putamen which could be due to the loss of either neurons or nerve endings. This low decrease contrasts with the dramatic diminution of many other markers associated with the profound striatal degeneration occurring in Huntington's disease.

Parkinson's disease is characterized by a loss of dopaminergic neurons in the substantia nigra and, in the most severe cases, by degeneration of mesopontine cholinergic neurons. In a monkey model of Parkinson's disease induced by 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine we report that, despite a severe loss of dopaminergic neurons, in the mesopontine tegmentum cholinergic neurons are preserved in the same region. This suggests that the loss of mesopontine cholinergic neurons in parkinsonian patients may represent an end-stage degenerative process, the cause of which may be independent of the mechanism of dopaminergic cell death.

Parkinson's disease is characterized by massive degeneration of the melanized dopaminergic neurons in the substantia nigra. The functional capacity of the surviving nigral neurons is affected, as indicated by the subnormal levels of tyrosine hydroxylase (TH) mRNA in these neurons and the presence in the parkinsonian mesencephalon of melanized neurons lacking TH immunoreactivity. This is apparently in contraction with the known overactivity of dopamine synthesis and release that occurs in the remaining dopaminergic terminals. To test the ability of the surviving neurons to express TH protein, a semiquantitative immunocytochemical method was developed. The relative amounts of TH were estimated with a computer-assisted image analysis system in the dopaminergic neurons of representative mesencephalic sections of control and parkinsonian brains and for comparison in brains from patients with Alzheimer's disease. In control brains, the mean TH content per neuron differed from one subject to another and between the different dopaminergic cell groups of the mesencephalon in the same subject. Within a given dopaminergic region, the level of TH was variable among neurons. In patients with Parkinson's disease, the ratio of TH protein content per neuron in the substantia nigra by reference to that of the central gray substance was reduced. In patients with Alzheimer's disease, the amount of TH was selectively reduced in the remaining dopaminergic neurons of the ventral tegmental area, a region characterized by a loss in dopaminergic neurons. The decrease in cellular TH content might therefore be related to the presence of the neurodegenerative process in the area considered. In patients with Parkinson's disease, the incapacity of the surviving neurons to express normal TH levels may reduce the efficiency of the hyperactivity mechanisms that develop in the remaining striatal dopaminergic terminals.

The question has been raised as to whether neuromelanin, a by-product of catecholamine metabolism which accumulates during aging in primate midbrain neurons, contributes to the selective vulnerability of subgroups of dopaminergic neurons in Parkinson's disease. 1-Methyl-4-phenylpyridinium (MPP+) a metabolite of 1-methyl, 4-phenyl, 1,2,3,6-tetrahydropyridine (MPTP) is toxic to dopaminergic neurons, particularly in primates, producing a motor syndrome similar to that observed in Parkinson's disease. To test whether this neurotoxin preferentially affects melanized neurons, the survival of melanized and non-melanized catecholaminergic neurons was analysed after MPTP intoxication in the midbrain of the cynomolgus monkey (Macaca fascicularis). Experiments were performed on six animals chronically treated with MPTP (two were severely disabled, four moderately affected) and two age-matched control monkeys. Two populations of neurons were examined on regularly spaced sections throughout the midbrain: catecholaminergic neurons, identified by tyrosine hydroxylase immunohistochemistry and neuromelanincontaining neurons, visualized by Masson's method. The total number of neurons of each type was estimated in the different midbrain catecholaminergic cell groups using computer assisted image analysis. In the midbrains of control animals not all catecholaminergic neurons contained neuromelanin. The percentage of melanized neurons compared to the total population of tyrosine hydroxylase-positive neurons was high in the substantia nigra pars compacta (81.5%) and in the locus coeruleus (98%), intermediate in the substantia nigra pars lateralis (70%), in the catecholaminergic cell group A8 (50%), and in the ventral tegmental area (41.5%) and almost nil in the central gray substance. In MPTP-treated monkeys, the severity of the loss of catecholaminergic neurons was variable within the different midbrain cell groups, though of similar intensity in severely and mildly disabled monkeys. A relationship was found between the loss of dopaminergic neurons in the different mesencephalic cell groups of MPTP-intoxicated animals and the percentage of melanized neurons they normally contain (r = 0.98; P = 0.04). The percentage loss of catecholaminergic neurons in the locus coeruleus, the only noradrenergic cell group studied, was lower than expected from the correlation curve obtained for dopaminergic cell groups. Altogether, these findings indicate: (i) that dopaminergic neurons are more vulnerable to MPTP-toxicity than noradrenergic neurons; and (ii) that among dopaminergic neurons, those containing neuromelanin are more susceptible, indicating a possible role of neuromelanin in MPTP-toxicity. Nevertheless, neuromelanin may only represent one of the factors involved in the neurotoxin-induced neuronal death, since some nigral melanized neurons also survive MPTP-intoxication.

The copper-zinc-dependent superoxide dismutase messenger RNA expression was studied at cellular level by in situ hybridization, using a 35S-labelled complementary DNA probe homologous to human copper-zinc-dependent superoxide dismutase messenger RNA, in the dopaminergic neuron-containing areas of the human mesencephalon (the substantia nigra pars compacta, ventral tegmental area, central gray substance and peri- and retrorubral region corresponding to catecholaminergic cell group A8). The autoradiographic labelling signal was localized in neurons. No detectable hybridization signal could be found in the glial cells. Copper-zinc-dependent superoxide dismutase messenger RNA was detected in melanin-containing neurons as well as in non-melanized neurons. Quantification at cellular level, taking the autoradiographic silver grain density as an index of the abundance of copper-zinc-dependent superoxide dismutase messenger RNA, indicated that hybridization level was higher in the melanized than in the non-melanized neurons within a region. Among melanized neurons, cellular copper-zinc-dependent superoxide dismutase messenger RNA content was lowest in the neurons of the substantia nigra. No significant difference in levels of transcripts was evidenced between the groups of non-melanized neurons. The data suggest that the abundance of copper-zinc-dependent superoxide dismutase messenger RNA is higher in the mesencephalic neurons containing neuromelanin compared to other neurons. Thus, the melanized neurons have a particular defence system against oxygen toxicity, which may represent a basis for their preferential vulnerability to Parkinson's disease.

Cholinergic neurons were studied by immunohistochemistry, with an antiserum against choline acetyltransferase (ChAT), in the basal forebrain (Ch1 to Ch4) of four patients with Alzheimer's disease (AD) and four control subjects. ChAT-positive cell bodies were mapped and counted in Ch1 (medial septal nucleus), Ch2 (vertical nucleus of the diagonal band), Ch3 (horizontal nucleus of the diagonal band) and Ch4 (nucleus basalis of Meynert). Compared to controls, the number of cholinergic neurons in AD patients was reduced by 50% on average. The interindividual variations in cholinergic cell loss were high, neuronal loss ranging from moderate (27%) to severe (63%). Despite the small number of brains studied, a significant correlation was found between the cholinergic cell loss and the degree of intellectual impairment. To determine the selectivity of cholinergic neuronal loss in the basal forebrain of AD patients, NPY-immunoreactive neurons were also investigated. The number of NPY-positive cell bodies was the same in controls and AD patients. The results (1) confirm cholinergic neuron degeneration in the basal forebrain in AD and the relative sparing of these neurons in some patients, (2) indicate that degeneration of cholinergic neurons in the basal forebrain contributes to intellectual decline, and (3) show that, in AD, such cholinergic cell loss is selective, since NPY-positive neurons are preserved in the basal forebrain.

Hyperoxidation phenomena are suspected to be involved in dopaminergic cell death in Parkinson's disease, which affects preferentially the neuromelanin-containing dopaminergic neurons of the substantia nigra. Glutathione peroxidase is the major protective enzyme against hydrogen peroxide toxicity. The distribution of glutathione peroxidase-containing cells was investigated by immunohistochemistry in the midbrain of four control subjects and four patients with Parkinson's disease. (1) Glutathione peroxidase-like immunoreactivity was detected exclusively in glial cells. (2) In control brains, the density of glutathione peroxidase-positive cells was higher in the vicinity of the dopaminergic cell groups known to be resistant to the pathological process of Parkinson's disease. (3) In Parkinson's disease, an increased density of glutathione peroxidase-immunostained cells was observed, surrounding the surviving dopaminergic neurons. The increase in glutathione peroxidase-containing cells was correlated with the severity in dopaminergic cell loss in the respective cell groups. The data suggest that in control brains, a low density of glutathione peroxidase-positive cells surround the dopaminergic neurons the most vulnerable to Parkinson's disease, and that in parkinsonian brains, the increased number of glutathione peroxidase-positive cells may contribute to protect neurons against pathological death. Thus, the amount of glutathione peroxidase protein-containing cells may be critical for a protective effect against oxidative stress, although it cannot be excluded that the level of the enzyme activity remains the crucial factor.

The subnormal choline acetyltransferase (ChoAcTase) activity in the cerebral cortex of patients with Alzheimer disease (AD) is thought to originate from the loss of cholinergic neurons in the nucleus basalis of Meynert (nbM). To examine possible changes in the functional activity of the remaining cholinergic neurons in the nbM of patients with AD, the level of expression of ChoAcTase mRNA was evaluated. A procedure for double-labeling cholinergic neurons to detect ChoAcTase mRNA and the corresponding protein in the same cell was developed, taking advantage of an anti-ChoAcTase antibody and the recently isolated cDNA complementary to a sequence of the human ChoAcTase mRNA. In the study of three controls and four patients with AD, the presence of both ChoAcTase mRNA and protein was observed in the same large neurons in both nbM and putamen. Specificity of in situ hybridization was further supported by the absence of neuronal staining with a sense probe. In AD patients a subnormal level of expression of ChoAcTase mRNA per cholinergic cell was detected in the nbM but not in the putamen. Our data support the hypothesis that expression of ChoAcTase mRNA might be down-regulated in the surviving cholinergic neurons in the nbM of patients with AD, raising the possibility of functional restoration by stimulating ChoAcTase synthesis.

The somatostatinergic innervation, and by comparison that of neuropeptide Y, was examined by immunocytochemistry, in the hippocampal formation of 6 patients with clinically and neuropathologically confirmed diagnosis of Alzheimer's disease (AD) and 4 matched controls. The quantitative assessment of the somatostatin and neuropeptide Y immunoreactive cell perikarya was conducted on regularly spaced sections covering the span of the whole hippocampus and parahippocampal cortex. The estimated total number of somatostatin or neuropeptide Y containing neurons was similar in the hippocampal formation of patients with AD and controls. By contrast, the density of the somatostatinergic fibers and terminals as estimated by the optical density of the somatostatin neuropil, significantly decreased by 30-45% in all the hippocampal subregions and the parahippocampal cortex, whereas the density of the neuropeptide Y immunoreactive neuropil significantly decreased (45%) only in the parahippocampal cortex. Numerous enlarged fibers containing somatostatin or neuropeptide Y, some of which were observed within senile plaques, were detected in the brain of patients with AD. These data suggest that in AD, somatostatin and neuropeptide Y containing cell bodies in the hippocampus and parahippocampal gyrus are anatomically preserved, in spite of an altered terminal field.

The somatostatinergic innervation, and by comparison that of neuropeptide Y, was examined by immunocytochemistry, in the hippocampal formation of 6 patients with clinically and neuropathologically confirmed diagnosis of Alzheimer's disease (AD) and 4 matched controls. The quantitative assessment of the somatostatin and neuropeptide Y immunoreactive cell perikarya was conducted on regularly spaced sections covering the span of the whole hippocampus and parahippocampal cortex. The estimated total number of somatostatin or neuropeptide Y containing neurons was similar in the hippocampal formation of patients with AD and controls. By contrast, the density of the somatostatinergic fibers and terminals as estimated by the optical density of the somatostatin neuropil, significantly decreased by 30-45% in all the hippocampal subregions and the parahippocampal cortex, whereas the density of the neuropeptide Y immunoreactive neuropil significantly decreased (45%) only in the parahippocampal cortex. Numerous enlarged fibers containing somatostatin or neuropeptide Y, some of which were observed within senile plaques, were detected in the brain of patients with AD. These data suggest that in AD, somatostatin and neuropeptide Y containing cell bodies in the hippocampus and parahippocampal gyrus are anatomically preserved, in spite of an altered terminal field.

The motor areas of the cerebral cortex contain dense dopaminergic and noradrenergic innervation in humans. We looked for changes of these innervations in cases with Parkinson's disease (PD). The density of fibers immunolabeled with tyrosine hydroxylase or dopamine-beta-hydroxylase was evaluated in the primary motor, premotor, and prefrontal cortical regions in 6 cases with PD and 7 control cases. Reductions of both noradrenergic and dopaminergic cortical innervations were observed, with similar magnitudes of reduction found in the motor and prefrontal regions of the cortex. Depletion of noradrenergic innervation was diffuse, involving all cortical laminae. Depletion of dopaminergic innervation was laminar specific, with the most significant reductions in layers I and II; reductions in layers V and VI were either less marked (prefrontal cortex) or not detectable (primary motor). The results suggest the existence of two separate mesocortical dopaminergic systems in humans, with the one distributing to upper cortical layers being preferentially involved in PD.

The levels of different elements were studied by x-ray microanalysis in the substantia nigra and the central gray substance of patients with Parkinson's disease, progressive supranuclear palsy, and matched controls. In control brains, only iron, potassium, silicium, sodium, sulfur, and zinc were within the limit of detection of the technique. The abundance of each element was different, but their respective concentrations in the two brain regions were similar, except for sulfur levels which were higher on neuromelanin aggregates in the substantia nigra than in nigral regions lacking neuromelanin, and in the central gray substance. In Parkinson's disease, but not in progressive supranuclear palsy, nigral iron levels increased in regions devoid of neuromelanin and decreased on neuromelanin aggregates, but were unchanged in the central gray substance, when compared to control values. Concentrations of the other elements in the central gray substance and substantia nigra were not different from controls in brains from patients with Parkinson's disease and progressive supranuclear palsy. Analysis of Lewy bodies in the parkinsonian substantia nigra revealed high levels of iron and the presence of aluminum. Metal abundance was not affected in progressive supranuclear palsy, in spite of the nigral cell death. This suggests that the increased iron levels and the detection of aluminum observed in Parkinson's disease are not solely the consequence of the neuronal degeneration.

The distribution and the density of 125I-Tyr0-DTrp8-somatostatin (125I-SRIF) binding sites were compared by quantitative radioautography in the hippocampi and entorhinal cortices from normal humans and patients with senile dementia of the Alzheimer type (SDAT). In SDAT, entorhinal cortex concentrations of choline acetyl transferase and SRIF were significantly decreased as compared to controls. The neuroanatomical distribution of I25I-SRIF binding sites was similar in control and SDAT hippocampus and entorhinal cortex, being high in molecular layers of the dentate gyrus and stratum oriens of the CA1 and deeper cortical layers (V-VI), intermediate in superficial (II—III) cortical layers and low in subiculum, hilus of the dentate gyrus and cortical layer I and IV. Competition curves with SMS 201995 to displace 125I-SRIF binding allowed to differentiate between high (SS-A)-and low (SS-B)-affinity somatostatin binding sites. In control brains, SS A (Kd A = 0.06 ± 0.02 nM) represented 23–33% of the binding while SS B (KdB = 23.5 ± 2.8 nM) accounted for 67–77% of total specific binding, depending on the regions. In dentate gyrus and CA1, the proportion of SS-A was significantly reduced in SDAT as compared to controls. The reduction did not reach significance in the entorhinal cortex, and inhibition of adenylate cyclase activity by SRIF was equivalent in SDAT and controls. It is concluded from these data that the number of SS-A decreases in SDAT hippocampus but that SS-B and negative coupling to adenylate cyclase are preserved.

The aim of this study was to compare brain glial fibrillary acidic protein (GFAP) levels to the modifications of cognitive functions (Blessed test score [BTS]), the density of the main neuropathological lesions (senile plaques [SP] and neurofibrillary tangles [NFT]), and the density of the two main subtypes of beta A4 deposits (classic plaques and diffuse deposits) in a series of patients with normal aging and senile dementia of the Alzheimer type of various degrees of severity. GFAP levels (enzyme-linked immunosorbent assay [ELISA] technique) and the densities of changes were measured in the temporal lobe of 12 women over 75 years of age. Under these conditions, the ELISA assay could determine GFAP in brain homogenates (aqueous-Triton buffer soluble extract) in a range from 2.5 ng to 600 ng per assay. Least affected patients (with a BTS of 19 and over) all ranged below 60 micrograms/mg protein. Most affected patients (with a BTS under 6) ranged above 150 micrograms/mg protein. However, interindividual variations were wide. A significant correlation between the BTS and the amount of GFAP could be found only when using the non parametric test of Spearman. There was a significant positive correlation between the amount of GFAP and the density of 1) SP, 2) NFT both revealed by Bodian's silver stain, and 3) classic beta A4 plaques shown by immunocytochemistry. On the contrary, no correlation was observed with diffuse beta A4 deposits. One case with very large amounts of diffuse beta A4 deposits without SP or NFT showed no associated GFAP reactivity. This suggests that GFAP production is a critical event in the formation of classic SP.

The cholinergic innervation of the mediodorsal (MD) nucleus of the thalamus was visualized immunohistochemically in human brain postmortem, using an antibody against human choline acetyltransferase (ChAT). The ChAT staining of the MD nucleus was more intense than in the surrounding thalamic nuclei but weaker than that of the striatum. No ChAT-positive cell bodies were detected. The ChAT-positive neuropil was unevenly distributed, with patches of dense immunoreactivity contrasting with a weaker surrounding matrix. In adjoining sections stained for ChAT immunoreactivity and for acetylcholinesterase (AChE) activity, the zones enriched in ChAT-immunostained neuropil corresponded to AChE-rich regions. The three-dimensional reconstruction of the richest zone in AChE/ChAT activity evidenced a cylindrical organization throughout the rostrocaudal axis of the MD nucleus. Counts of ChAT-positive varicosities confirmed an inhomogeneous distribution; the density of varicosities was 30% higher in ChAT-rich regions than in surrounding matrix. These findings suggest that the activity of intrinsic neurons within the nucleus may be differentially regulated by cholinergic systems.

The distribution of both tyrosine hydroxylase (TH) antigen and messenger RNA was investigated in the ventral mesencephalon from human brain using a 35S-labeled human TH cDNA probe and a polyclonal TH antibody. A hybridization signal was detected not only in the perikaryon but also in the neurites, most probably dendrites, of immunoreactive neurons. In cell bodies, the level of TH mRNA varies, revealing a regional heterogeneity. In the substantia nigra of patients with Parkinson's disease the number of labeled cells was markedly reduced as compared to controls.

The autoradiographic distribution of D1 dopaminergic binding sites was studied in the human ventral mesencephalon using the D1 antagonist [3H]SCH 23390. [3H]SCH 23390 binding was characterized by a single class of sites with a Kd of 2.5 nM and a Bmax of 31 fmol/mg of tissue. The density of [3H]SCH 23390 binding sites was high in the substantia nigra, moderate in the ventral tegmental area and low in the peri- and retrorubral field (catecholaminergic region A8). Binding densities were similar in pars compacta and pars reticulata of the substantia nigra, except for a peak value of high [3H]SCH 23390 in the pars reticulata, at a level just ventral to a zone of hyperdensity of melanized dopaminergic neurons in the pars compacta. The anatomical organization of the human ventral mesencephalon was analysed on adjacent sections stained for acetylcholinesterase histochemistry and tyrosine hydroxylase, substance P, dynorphin B, somatostatin and methionine-enkephalin immunohistochemistry, respectively. The similarity in distribution of [3H]SCH 23390 binding sites and substance P or dynorphin B immunoreactivity suggests that D1 binding sites are mainly located on the striatonigral projections. In accordance with these results: (1) the density of [3H]SCH 23390 binding sites was reduced in the substantia nigra of a patient with Huntington's chorea, a disease associated with a degeneration of striatonigral neurons; (2) the density of [3H]SCH 23390 binding sites was unaffected in the substantia nigra of a patient with Parkinson's disease, a disorder characterized by a marked loss in nigral tyrosine hydroxylase-positive neurons. [3H]SCH 23390 binding sites showed a characteristic, heterogeneous distribution within the human ventral mesencephalon, confirming data obtained in other species. The preferential localization of D1 dopamine receptors on striatonigral projections in human brain suggests that pharmacological manipulation of these receptors modulates the activity of striatonigral pathways, thereby affecting the various outputs of the nigral complex.

Summary Nineteen Macaca fascicularis monkeys were divided into four different groups: Group A ( n = 3), control; Group B ( n = 3), monkeys treated with I-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP); Group C ( n = 8), animals treated with MPTP in which the subthalamic nucleus (STN) was unilaterally lesioned by kainic acid injection; in Group D ( n = 5), the STN was lesioned prior to MPTP administration. Subthalamotomy resulted in a bilateral improvement of tremor, spontaneous activity, bradykinesia (evaluated by a manual motor test) and freezing in Group C. All these monkeys developed hemichorea contralateral to the lesion. The improvement was maintained and the hemichorea continued until death. The monkeys in group D showed severe hemiballism which persisted throughout MPTP administration and developed parkinsonian signs mainly on the side ipsilateral to the lesion. Analysis of the in situ hybridization of the mRNA coding for glutamic acid decarboxylase (GAD) of MPTP monkeys showed a significant increase in the mean density of silver grains over every labelled neuron in the globus pallidum lateralis (56.8% over control) as well as the globus pallidus medialis (GPM) (45.7% over control) and the substantia nigra reticulata (SNR) (35.8% over control). No significant change was observed in the thalamic nucleus reticularis. Subthalamotomy (Groups C and D) produced a significant reduction in mRNA GAD expression on the side of the lesion in the GPM and the SNR (34% and 42.3%, respectively) with respect to the ipsilateral (non-lesioned) side and also when compared with parkinsonian monkeys. These results confirm and expand, at the cellular level, the paramount role of STN hyperactivity in the pathophysiology of parkinsonism. The therapeutic consequences of these findings for surgical treatment of Parkinson's disease are discussed.

To examine the consequences of nigrostriatal denervation and L-dopa treatment on the basal ganglia output system, we analyzed, by quantitative in situ hybridization, the messenger RNA coding for glutamic acid decarboxylase (Mr 67,000) (GAD67 mRNA) in pallidal cells from patients with Parkinson's disease (PD), monkeys rendered parkinsonian by 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) receiving or not receiving L-dopa, and their respective control subjects. In MPTP-treated monkeys, the expression of GAD67 mRNA was increased in cells from the internal pallidum, and this effect was abolished by L-dopa treatment. There were no differences in the levels of GAD67 mRNA between patients with PD, who were all treated with L-dopa, and control subjects. These results indicate that the level of GAD67 mRNA is increased in the cells of the internal pallidum after nigrostriatal dopaminergic denervation and that this increase can be reversed by L-dopa therapy.NEUROLOGY 1996;47: 219-224

Although the Huntington's disease (HD) gene defect has been identified, the structure and function of the abnormal gene product and the pathogenetic mechanisms involved in producing death of selective neuronal populations are not understood. Indirect evidence from several sources indicates that a defect of energy metabolism and consequent excitotoxicity are involved in HD. Toxin models of HD may be induced by 3-nitropropionic acid or malonate, both inhibitors of succinate dehydrogenase, complex II of the mitochondrial respiratory chain. We analyzed mitochondrial respiratory chain function in the caudate nucleus (n = 10) and platelets (n = 11) from patients with HD. In the caudate nucleus, severe defects of complexes II and III (53-59%, p < 0.0005) and a 32-38% (p < 0.01) deficiency of complex IV activity were demonstrated. No deficiencies were found in platelet mitochondrial function. The mitochondrial defect identified in HD caudate parallels that induced by HD neurotoxin models and further supports the role of abnormal energy metabolism in HD. The relationship of the mitochondrial defect to the role of huntingtin is not known.

DNA fragments from a genomic library were used to establish the partial structure of the human dopamine D3 receptor gene (DRD3). Its coding sequence contains 6 exons and stretches over 40,000 base pairs. The complete DRD3 transcript and three shorter variants, in which the second and/or third exon are deleted, were detected in similar proportions in brains from four controls and three psychiatric patients. The Msp I polymorphism was localized in the fifth intron of the gene, 40,000 base pairs downstream the Bal I polymorphism and a PCR-based method was developed for genotyping this polymorphism. The distribution of the Msp I and Bal I genotypes were not independent in 297 individuals (chi 2 = 10.5, df = 4, P = 0.03), but only a weak association was found between allele 1 of the Bal I polymorphism and allele 2 of the Msp I polymorphism (chi 2 = 3.99, df = 1, P = 0.04). The previously reported association between homozygosity at both alleles of the Bal I polymorphism and schizophrenia was presently maintained in an extended sample, comprising 119 DSM-III-R chronic schizophrenics and 85 controls (chi 2 = 5.3, df = 1, P = 0.02) and found more important in mal than in females. The presence of the Bal I allele 2 is associated with an early age at onset, particularly in males (df = 35, t value = 2.6, P = 0.014). In the same sample, allelic frequencies, genotype counts, and proportion of homozygotes for the Msp I polymorphism did not differ between schizophrenics and controls (chi 2 = 0.06, df = 1, P = 0.80, chi 2 = 0.22, df = 1, P = 0.90 and chi 2 = 0.16, df = 1, P = 0.69, respectively). The large distance of the Msp I polymorphism from the Bal I polymorphism and its localization in the 3' part of the gene may explain the discrepant results obtained with the two polymorphisms.

The effects of nigrostriatal denervation and L-dopa therapy on GABAergic neurons were analysed in patients with Parkinson's disease and in monkeys rendered parkinsonian by MPTP intoxication. The expression of the messenger RNA coding for the 67 kDa isoform of glutamic acid decarboxylase (GAD67 mRNA), studied by quantitative in situ hybridization, was used as an index of the GABAergic activity of the striatal neurons. A significant increase in GAD67 mRNA expression, generalized to all GABAergic neurons, was observed in MPTP-treated monkeys compared to control monkeys in the putamen and caudate nucleus (+44 and +67% respectively), but not in the ventral striatum. L-Dopa therapy significantly reduced GAD67 mRNA expression in the putamen and caudate nucleus to levels similar to those found in control monkeys. However, the return to normal of GAD67 mRNA expression was not homogeneous across all neurons since it was followed by an increase of labelling in one subpopulation of GABAergic neurons and a decrease in another. These data suggest that in MPTP-treated monkeys the degeneration of nigrostriatal dopaminergic neurons results in a generalized increase in GABAergic activity in all the GABAergic neurons of the striatum, which is partially reversed by L-dopa therapy. As the expression of GAD67 mRNA is less intense in the ventral than in the dorsal striatum, this increase in striatal GABAergic activity may be related to the severity of nigrostriatal denervation. In parkinsonian patients who had been chronically treated with L-dopa, GAD67 mRNA expression was significantly decreased in all GABAergic neurons, in the caudate nucleus (by 44%), putamen (by 43.5%) and ventral striatum (by 26%). The opposite variation of GAD67 mRNA in patients with Parkinson's disease, compared with MPTP-treated monkeys, might be explained by the combination of chronic nigrostriatal denervation and long-term L-dopa therapy.

Article abstract-We analyzed postmortem GABAergic neurons in the basal ganglia of three patients with progressive supranuclear palsy (PSP) and four matched controls by means of glutamic acid decarboxylase (Mr 67,000 (GAD67)) mRNA in situ hybridization. In PSP, we found a 50 to 60% decrease in the number of neurons expressing GAD67 mRNA in the caudate nucleus, ventral striatum, and the external and internal pallidum. The expression of GAD67 mRNA per neuron was reduced in the caudate nucleus and putamen (-43%), the ventral striatum (-55%), and the external and internal pallidum (-59% and -68%). Our data indicate that striatal and pallidal GABAergic neurotransmission is markedly reduced in PSP and we suggest that this alteration may account for the motor and cognitive symptoms observed in PSP. Furthermore, the destruction of the basal ganglia output systems may explain the lack of responsiveness to L-dopa therapy of PSP patients.NEUROLOGY 1995;45: 127-134

Reduced glutathione (GSH) and oxidized glutathione (GSSG) levels were measured in various brain areas (substantia nigra, putamen, caudate nucleus, globus pallidus, and cerebral cortex) from patients dying with Parkinson's disease, progressive supranuclear palsy, multiple-system atrophy, and Huntington's disease and from control subjects with no neuropathological changes in substantia nigra. GSH levels were reduced in substantia nigra in Parkinson's disease patients (40% compared to control subjects) and GSSG levels were marginally (29%) but insignificantly elevated; there were no changes in other brain areas. The only significant change in multiple-system atrophy was an increase of GSH (196%) coupled with a reduction of GSSG (60%) in the globus pallidus. The only change in progressive supranuclear palsy was a reduced level of GSH in the caudate nucleus (51%). The only change in Huntington's disease was a reduction of GSSG in the caudate nucleus (50%). Despite profound nigral cell loss in the substantia nigra in Parkinson's disease, multiple-system atrophy, and progressive supranuclear palsy, the level of GSH in the substantia nigra was significantly reduced only in Parkinson's disease. This suggests that the change in GSH in Parkinson's disease is not solely due to nigral cell death, or entirely explained by drug therapy, for multiple-system atrophy patients were also treated with levodopa. The altered GSH/GSSG ratio in the substantia nigra in Parkinson's disease is consistent with the concept of oxidative stress as a major component in the pathogenesis of nigral cell death in Parkinson's disease.

The cellular localization of catechol-O-methyltransferase was analysed in the mesostriatal system of human brain post mortem by means of immunohistochemistry. In the human nigral complex, catechol-O-methyltransferase immunostaining was not detected in melanized dopaminergic neurons, except in the ventral tegmental area and substantia nigra pars lateralis, where few neurons displayed intense immunolabelling. In the striatum, cateehol-O-methyltransferase immunostaining was found in numerous cell bodies and in the neuropile. Observations at the electron microscope level revealed that catechol-O-methyltransferase immunoreactivity was present in the cell bodies of neurons and their processes, including the dendritic spines. No catechol-O-methyltransferase immunolabelling was observed in striatal nerve terminals in contact with dendritic spines, indicating that dopaminergic nerve terminals do not exhibit catechol-O-methyltransferase immunoreactivity. Catechol-O-methyltransferase-immunoreactive cell bodies and processes of glial cells were also detected in the striatum. The data suggest that catechol-O-methyltransferase is either not expressed or only slightly expressed by the dopaminergic nigrostriatal neurons, whereas it is clearly present in striatal neurons and glial cells. Thus, the catabolic degradation of striatal released dopamine by its O-methylation may involve postsynaptic neurons rather than dopaminergic presynaptic neurons. The presence of catechol-O-methyl-transferase in some dopaminergic neurons of the ventral tegmental area and substantia nigra pars lateralis suggests that methylation of dopamine may occur in these neurons, which may consequently be better protected against dopamine auto-oxidation than those of the substantia nigra pars compacta.

PSP is associated with a widespread cholinergic deficit likely corresponding to a loss in cholinergic neurons. The cholinergic damage dramatically affects the basal ganglia and specific cell groups of the mesencephalon and pons. This provides an anatomically defined basis for motor and supranuclear oculomotor syndromes characteristic of PSP. Unlike Alzheimer's disease and Parkinson's disease with dementia, the disease is not associated with a marked cholinergic deficiency in the cerebral cortex. Various peptides are present at normal concentrations in extrapyramidal and limbic subcortical areas in brains of patients with PSP. Of particular interest, is somatostatin, the levels of which are subnormal in cerebral cortex of patients with dementia of Alzheimer' or Parkinson's disease type.

Publisher Summary This chapter discusses factors associated with dopaminergic cell death in Parkinson's disease. The main biochemical characteristic of idiopathic Parkinson's disease is the degeneration of the dopaminergic nigrostriatal neurons that originate in the substantia nigra pars compacta and project to the striatum (caudate nucleus and putamen). The cause of the neuronal death remains unknown. The disease does not seem because of some viral infection. Monozygotic twin studies suggest a limited role for heredity, though there might be a genetic component because an autosomal dominant transmission has been described in some families. Abnormalities in the mitochondrial electron transport chain have been observed in idiopathic Parkinson's disease suggesting mitochondrial genome might be affected; thus, an endogenous factor may contribute to the appearance of this affection. An exogenous factor could well be responsible for the disease lead to an epidemiological search for environmental risk factors. The A number of factors may be critical in accelerating dopaminergic cell death in the mesencephalon of patients with Parkinson's disease. The pathological dopaminergic cell death in Parkinson's disease is associated to hyperoxydation phenomena, which may be one of the polyfactorial mechanisms implicated.

PSP is associated with a widespread cholinergic deficit likely corresponding to a loss in cholinergic neurons. The cholinergic damage dramatically affects the basal ganglia and specific cell groups of the mes-encephalon and pons. This provides an anatomically defined basis for motor and supranuclear oculomotor syndromes characteristic of PSP.

GAD mRNA was detected in human and monkey brain postmortem by in situ hybridization with an [35S]-labelled copy RNA corresponding to a 2.7-kb fragment of the coding region of human GAD mRNA. A characteristic and reproducible pattern of hybridization was obtained with the anti-sense, but not the sense probe in both monkey and human brain. Microscopic examination of tissue sections showed that only neuronal perikarya, not glial cells, were labelled. The data confirm the heterogeneity of GAD mRNA distribution reported in rodent brain and non-human primate brain.

Iron is abnormally accumulated in the substantia nigra pars compacta of patients with Parkinson's disease (PD). Because neuronal and glial iron uptake seems to be mediated by the binding of ferrotransferrin to a specific high-affinity receptor on the cell surface, the number of transferrin receptors could be altered in this disease. The regional distribution of specific binding sites for human 125I-diferric transferrin has been studied in the mesencephalon, on cryostat-cut sections from autopsy brains of control subjects and parkinsonian patients by in vitro autoradiography. Densities of binding sites were highest in the central gray substance (approximately 10 fmol/mg of tissue equivalent), intermediate in the catecholaminergic cell group A8, superior colliculus, and ventral tegmental area, and almost nonexistent in the substantia nigra. The density of 125I-transferrin binding sites was not significantly different between parkinsonian and control brains in any region analyzed. These results show that in the mesencephalon the regional density of transferrin binding sites is lowest in the dopaminergic cell groups, which are the most vulnerable to PD, and suggest that iron does not accumulate through an increased density of transferrin receptors at the level of the substantia nigra.

Cholinergic neurons were studied by immunohistochemistry, with an antiserum against choline acetyltransferase (ChAT), in the basal forebrain (Ch1 to Ch4) of four patients with Alzheimer's disease (AD) and four control subjects. ChAT-positive cell bodies were mapped and counted in Ch1 (medial septal nucleus), Ch2 (vertical nucleus of the diagonal band), Ch3 (horizontal nucleus of the diagonal band) and Ch4 (nucleus basalis of Meynert). Compared to controls, the number of cholinergic neurons in AD patients was reduced by 50% on average. The interindividual variations in cholinergic cell loss were high, neuronal loss ranging from moderate (27%) to severe (63%). Despite the small number of brains studied, a significant correlation was found between the cholinergic cell loss and the degree of intellectual impairment. To determine the selectivity of cholinergic neuronal loss in the basal forebrain of AD patients, NPY-immunoreactive neurons were also investigated. The number of NPY-positive cell bodies was the same in controls and AD patients. The results (1) confirm cholinergic neuron degeneration in the basal forebrain in AD and the relative sparing of these neurons in some patients, (2) indicate that degeneration of cholinergic neurons in the basal forebrain contributes to intellectual decline, and (3) show that, in AD, such cholinergic cell loss is selective, since NPY-positive neurons are preserved in the basal forebrain.

To analyze the roles of transcriptional and post-transcriptional regulatory mechanisms of tyrosine hydroxylase (TH) gene expression during dopaminergic denervation in Parkinson's disease (PD), the cellular content of TH messenger RNA (mRNA) and TH protein in the substantia nigra were compared in control subjects and patients with PD. The average amounts of TH mRNA as well as those of TH protein per neuron were variable among controls but correlated to each other. In PD patients, both TH mRNA and TH protein content in nigral neurons were reduced relative to controls, however, the ratio between TH protein and TH mRNA levels was unaffected. The data suggest that, in PD: (1) TH protein content is decreased in the surviving nigral dopaminergic neurons, most likely as a result of a lowered TH mRNA cellular content. Thus the surviving neurons at end stages of the disease may be in a premorbid state. (2) The TH mRNA translation rate is not modified to compensate for dopamine deficiency.

Neuromelanin (NM) is an auto-oxidation by-product of catecholamine synthesis which is observed almost exclusively in primates. We have estimated the distribution and the number of NM-positive neurons of the upper brainstem and the degree of their melanization from birth to the onset of senescence in 5 monkeys (Macaca fascicularis) aged 0, 1.5, 3.5, 8 and 13 years. Series of sections taken at 640-microns intervals were examined either unstained to detect unstained NM, stained for NM with Masson silver impregnation or processed by tyrosine hydroxylase (TH) immunohistochemistry to analyze catecholaminergic neurons. The proportion of NM-containing cells among TH-positive neurons varied from one catecholaminergic region to another: low in the hypothalamus and central gray substance (cgs); moderate in the cell group A8, and high in the ventral tegmental area (VTA), locus coeruleus (LC) and substantia nigra (SN). TH-positive neurons were detected in the SN, VTA, catecholaminergic cell group A8, LC, cgs and hypothalamus. At birth, although no unstained NM-positive neurons were detected, Masson-stained cells were observed, though only in the LC. At 1.5 and 3.5 years, Masson-positive neurons were observed despite the absence of visible pigment. At 8 and 13 years, unstained NM was present in Masson-positive neurons. The number of unstained NM-positive neurons and Masson-positive neurons and the amount of NM per neuron increased with age in each subregion studied. Nevertheless, some TH-positive neurons were found to be without NM. The data indicate a differential increased NM content with age in the neurons of midbrain catecholaminergic cell groups. However, its functional significance remains to be determined.

Loss of midbrain dopaminergic neurons in Parkinson’s disease is associated to hyperoxidation phenomena. In the human substantia nigra, free radicals may be produced in large quantities i.e, during degradation of dopamine, synthesis and accumulation of neuromelanin, through iron, present in high concentrations. Under normal conditions, production of free radicals is compensated by powerful protective enzymes: superoxide dismutase is detected in neurons, and expressed at high levels in those of the substantia nigra; gluthatione peroxidase is exclusively detected in glial cells. A low density of glial cells surround the substantia nigra neurons relatively to other midbrain areas. Thus nigral dopaminergic neurons may be less protected against deleterious action of free radicals. This may explain their preferential susceptibility to oxidative stress. In Parkinson’s disease, an overproduction of free radicals, suggested by the increased level of lipid peroxidation in the substantia nigra, might accelerate the rate of dopaminergic cells death. Impairment of the oxygen toxicity protective mechanisms may be responsible. Besides, the above normal levels of iron in the substantia nigra may contribute to free radical production and have a role in the toxic process, though one cannot exclude the increased iron content may be a non-specific product of cellular degeneration, a consequence of the gliosis.

The purpose of a brain bank is to collect specimens in order to investigate human brain and pathologies of the central nervous system. The organization of the brain bank in Paris, devoted to the study of Parkinson's disease, is described, with it's evolutions with years, to respond to the exigencies of the new neurochemical methodologies.

The mechanism by which cholinergic neurons degenerate in Alzheimer's disease is not known. Some of these neurons depend, however, on trophic support from NGF via a membrane receptor. We have analyzed the state of these receptors by autoradiography, with 125I-NGF as the ligand, in the caudate nucleus, putamen, ventral striatum, nucleus basalis of Meynert, and nucleus tegmenti pedunculopontinus of six patients with Alzheimer's disease and five controls, matched for age and postmortem delay. The binding characteristics were similar in the striatum (including caudate nucleus, putamen, and ventral striatum) and basal forebrain of control subjects and patients with Alzheimer's disease (Kd = 2.5-4 x 10(-11) M). In control brains, high levels of 125I-NGF binding were observed in the basal forebrain and striatum (0.32–0.49 fmol/mg tissue equivalent), but no specific binding was detected in the nucleus tegmenti pedunculopontinus. NGF binding sites were distributed heterogeneously in the striatum with patches of low density, corresponding to AChE-poor striosomes, surrounded by a matrix in which receptor density was significantly greater. In Alzheimer's disease, the density of NGF receptors was markedly decreased in the caudate nucleus, putamen, ventral striatum, and nucleus basalis of Meynert. In contrast, AChE staining decreased less in the nucleus basalis of Meynert in all Alzheimer's disease patients, and in the ventral striatum of those most severely affected. These results indicate that if NGF receptors are located on cholinergic neurons, receptor loss and the consequent decrease in trophic support may precede cell degeneration in Alzheimer's disease. The relationship between the loss of these receptors and the pathogenesis of the disease remains to be determined.

The contribution of neuromelanin (NM) to the pathogenesis of Parkinson's disease (PD) has long been suspected. In particular, a correlation has been reported between the estimated cell loss in the mesencephalic dopaminergic cell groups and the percentage of NM-pigmented neurons in these cell groups. To test whether the amount of pigment per cell is a critical factor or whether the presence of NM within a neuron is sufficient to account for the degeneration of dopaminergic neurons, the NM content was measured in each neuron from representative sections throughout the ventral mesencephalon of four controls subjects and four patients with PD. Intraneuronal NM was quantified by a densitometric method, using known amounts of synthetic melanin as standards. In control brains, the distribution of melanized neurons in the nigral complex showed a high proportion of lightly melanized neurons in the ventral tegmental area and the pars alpha and gamma of the substantia nigra (SN), whereas heavily melanized neurons were mostly located in the pars beta and lateralis of the SN. An inverse relationship was observed between the percentage of surviving neurons in PD compared with controls and the amount of NM they contain, suggesting that the vulnerability of the dopaminergic neurons is related to their NM content. Factors other than NM may be involved in the differential vulnerability of catecholaminergic neurons in PD. In particular, the constant topography of the cell loss suggests that cell position within the nigral complex is a key factor.

Muscarinic cholinergic receptors and choline acetyltransferase (ChAT) activity were studied in postmortem brain tissue from patients with Huntington's disease and matched control subjects. In comparison with controls, reductions in ChAT activity were found in the hippocampus, but not in the temporal cortex in Huntington's disease. Patients with Huntington's disease showed reduced densities of the total number of muscarinic receptors and of M-2 receptors in the hippocampus while the density of M-1 receptors was unaltered. Muscarinic receptor binding was unchanged in the temporal cortex. These results indicate a degeneration in Huntington's disease of the septo-hippocampul cholinergic pathway, but no impairment of the innominato-cortical cholinergic system.