Loss of proteins regulating synaptic plasticity in normal aging of the human brain and in Alzheimer disease.

Recent studies suggest that the cognitive impairment associated with normal aging is due to neuronal dysfunction rather than to loss of neurons or synapses. To characterize this dysfunction, molecular indices of neuronal function were quantified in autopsy samples of cerebral cortex. During normal aging, the most dramatic decline was found in levels of synaptic proteins involved in structural plasticity (remodeling) of axons and dendrites. Alzheimer disease, the most common cause of dementia in the elderly, was associated with an additional 81% decrease in levels of drebrin, a protein regulating postsynaptic plasticity. Disturbed mechanisms of plasticity may contribute to cognitive dysfunction during aging and in Alzheimer disease.

Tracing neuronal connections in postmortem human hippocampal complex with the carbocyanine dye DiI.

This report describes the ability of the carbocyanine dye DiI to trace hippocampal complex connections in a paraformaldehyde immersion-fixed human postmortem brain. Six months after the placement of DiI crystals into the hilus of the dentate gyrus, the CA1 hippocampal subfield and the lateral entorhinal cortex, 50-microns thick, vibratome cut sections were examined using an epifluorescence microscope with a rhodamine filter. In association with DiI-labeled granule, pyramidal and multipolar type neurons, we observed dendrites containing dendritic spines and axons. DiI-labeled fibers were observed coursing within classically described hippocampal pathways for at least 8 mm distal to the injection site. Photoconversion of diaminobenzidine (DAB)-treated DiI sections produced a stable record of labeled profiles. These findings indicate that DiI is a useful method for investigating intrinsic local circuit connections in normal aldehyde-fixed postmortem human brain and suggests that DiI could be a powerful tool to examine altered neural connectivity in humans with neurological disease.

Association of premorbid intellectual function with cerebral metabolism in Alzheimer's disease: implications for the cognitive reserve hypothesis.

OBJECTIVE: Clinical heterogeneity in Alzheimer's disease has been widely observed. One factor that may influence the expression of dementia in Alzheimer's disease is premorbid intellectual ability. It has been hypothesized that premorbid ability, as measured by educational experience, reflects a cognitive reserve that can affect the clinical expression of Alzheimer's disease. The authors investigated the relation between estimates of premorbid intellectual function and cerebral glucose metabolism in patients with Alzheimer's disease to test the effect of differing levels of premorbid ability on neurophysiological dysfunction. METHOD: In a resting state with eyes closed and ears occluded, 46 patients with Alzheimer's disease were evaluated with positron emission tomography and [18F]-2-fluoro-2-deoxy-D-glucose to determine cerebral metabolism. Premorbid intellectual ability was assessed by a demographics-based IQ estimate and performance on a measure of word-reading ability. RESULTS: After the authors controlled for demographic characteristics and dementia severity, both estimates of premorbid intellectual ability were inversely correlated with cerebral metabolism in the prefrontal, pre-motor, and left superior parietal association regions. In addition, the performance-based estimate (i.e., reading ability) was inversely correlated with metabolism in the anterior cingulate, paracentral, right orbitofrontal, and left thalamic regions, after demographic and clinical variables were controlled for. CONCLUSIONS: The results suggest that higher levels of premorbid ability are associated with greater pathophysiological effects of Alzheimer's disease among patients of similar dementia severity levels. These findings provide support for a cognitive reserve that can alter the clinical expression of dementia and influence the neurophysiological heterogeneity observed in Alzheimer's disease.

Three-dimensional cortical morphometry of the planum temporale in childhood-onset schizophrenia.

OBJECTIVE: Anomalous planum temporale asymmetry has been linked to both schizophrenia and dyslexia. The authors examined the planum temporale of adolescents with childhood-onset schizophrenia who had a high rate of prepsychotic language disorders. METHOD: Planum temporale area and asymmetry were measured in 16 right-handed adolescent patients with schizophrenia who had experienced onset of psychosis by age 12. The same measures were made in 16 healthy adolescents matched for age, sex, and handedness. RESULTS: No differences between the healthy adolescents and those with schizophrenia in planum temporale area or asymmetry were observed. Prepsychotic language disorder predicted abnormal planum temporale asymmetry in the adolescents with schizophrenia. CONCLUSIONS: These findings do not support anomalous planum temporale asymmetry as a basis for psychopathology in childhood-onset schizophrenia.

Alz-50 immunoreactive neuropil differentiates hippocampal complex subfields in Alzheimer's disease.

The topographic distribution of Alz-50 containing profiles was determined within the hippocampal formation and anterior parahippocampal gyrus by using a monoclonal antibody directed against the A68 protein in normal and Alzheimer's diseased (AD) brains. Although there was a paucity of immunoreactive neuropil in the normal hippocampal complex, there were a few Alz-50 positive neurons that occupied the hippocampal subfield, CA2. In most AD cases, Alz-50 immunoreactive neuropil was prominent in the outer two-thirds of the molecular layer of the dentate gyrus, although a few cases exhibited staining in the inner third of the molecular layer. CA2 was characterized by an increased density of neuropil staining within stratum pyramidale. The neuropil in subfield CA1 was stained densely with Alz-50 in strata oriens, pyramidale, and at the border between strata lacunosum-moleculare and radiatum. Alz-50 immunostained neurites occupied primarily the lateral two-thirds of the subiculum proper, whereas only sparse staining was seen in the adjacent presubiculum. Alz-50 neuropil and neuronal staining displayed three distinct laminar patterns along the mediolateral extent of the entorhinal cortex, whereas the perirhinal cortex exhibited a bilaminar pattern of immunoreactivity involving heavy staining in layers 1-3 as compared to layer 5. In general, the density of Alz-50 neurite staining in the neuropil appeared inversely proportional to the distribution of Alz-50 immunoreactivity within dendritic and somal compartments. Interestingly, the patterns of Alz-50 staining observed in the hippocampal complex in AD coincides with patterns of well-characterized afferent fiber pathways to these regions, thus further supporting the suggestion that hippocampal subfield specific pathology effectively disconnects medial temporal structures from adjacent neocortex in AD.

Evidence that transmitter-containing dystrophic neurites precede paired helical filament and Alz-50 formation within senile plaques in the amygdala of nondemented elderly and patients with Alzheimer's disease.

Immunocytochemical techniques were employed to examine the temporal ordering whereby amyloid beta-protein (A beta P) and neuronal elements collectively come together to form senile plaques in Alzheimer's disease (AD). Specifically, we addressed three questions: (1) whether A beta P deposition precedes or follows neuritic changes; (2) whether paired helical filament (PHF) formation is an early or late event in the genesis of the dystrophic neurites which participate in plaque formation; and (3) whether the density of senile plaques displays any relationship with the prevalence of PHF or Alz-50 containing neurons. To address these questions we studied the amygdala from a group of patients with AD, a group of nondemented age-matched individuals exhibiting a sufficient number of senile plaques to be classified by neuropathological criteria as AD, and a group of age-matched controls without AD pathology. Amyloid-bearing plaques were demonstrated by A beta P immunolabeling and thioflavine-S staining. Neuritic changes in the form of dystrophic neurites were observed with the aid of antibodies against PHF, Alz-50, as well as antibodies against several neuropeptides (i.e., substance P, somatostatin, and neurotensin) and the acetylcholine biosynthetic enzyme, choline acetyltransferase. By using a graded range of pathologic changes both within and across the patient population to provide us with a means of evaluating plaque deposition from its earliest to most advanced stages of development, we observed in patients and/or regions of the amygdala displaying a mild degree of pathologic change A beta P deposition in the absence of any neuritic changes. With increasing density of A beta P, however, we began to observe dystrophic neurites within plaques. In regions of relatively few plaques, the dystrophic neurites were immunolabeled only with antibodies against the various neurotransmitters and they lacked evidence of cytoskeletal pathology (i.e., Alz-50 or PHF). Only as the density of A beta P increased further within a region, were dystrophic neurites observed that exhibited Alz-50 or PHF. In no instance did we observe a relationship between the density of A beta P deposition and the density of Alz-50 or PHF-immunoreactive neurons. Collectively, our data suggest that the deposition of A beta P is an early pathologic event in senile plaque formation. Thereafter, swollen neurites can be seen in the vicinity of A beta P. This early neuritic response, which can first be visualized by immunolabeling for one or another transmitter substance, is followed by alterations in the cytoskeleton as recognized initially by antibodies to Alz-50 and subsequently by the presence of PHF.

A magnetic resonance imaging study of planum temporale asymmetry in men with developmental dyslexia.

BACKGROUND: Imaging studies have suggested anomalous anatomical asymmetries in language-related regions of the temporal and parietal lobes in individuals with developmental dyslexia. Autopsy studies have reported unusual symmetry of the planum temporale (PT) in patients with dyslexia. Methodological limitations characterize much of this literature, however. OBJECTIVE: To examine the size and asymmetry of the PT and its extension into the parietal lobe (planum parietale [PP]) in men with well-characterized, persistent dyslexia by using magnetic resonance imaging and 3-dimensional surface rendering techniques. METHODS: The brains of 16 right-handed dyslexic men aged 18 to 40 years and 14 matched control subjects were studied with magnetic resonance imaging. Most of these subjects were previously studied with positron emission tomography, which demonstrated functional abnormalities in temporal and parietal brain regions in the dyslexic group. The area of the PT was determined with the aid of 3-dimensional surface-rendering techniques. The size of the PP was estimated by measuring the length of the posterior ascending ramus on 3 parasagittal slices. RESULTS: Approximately 70% to 80% of both groups showed equivalent leftward (left > right) asymmetries of the PT; approximately 50% to 60% showed equivalent rightward (right > left) asymmetries of the PP. These asymmetries showed equivalent moderate inverse correlations with each other in both groups. CONCLUSIONS: These results challenge the notion that anomalous asymmetry of the PT is strongly associated with developmental dyslexia. Given the heterogeneity of the dyslexic population, some subgroup of dyslexic individuals (i.e., those with developmental language disorders) may show unusual symmetry or reversed asymmetry in this region. However, anomalous asymmetry of the planum did not contribute to functional abnormalities demonstrated in these patients by positron emission tomography.

A comparison of the localization of choline acetyltransferase and glutamate decarboxylase immunoreactivity in rat cerebral cortex.

The neurotransmitter-synthesizing enzymes choline acetyltransferase and glutamate decarboxylase were localized immunocytochemically at the light microscopic level. Their respective laminar distributions were compared in 17 different cytoarchitectural areas, comprising limbic and neocortical regions of rat cerebral cortex. The immunoreactive intensities within these areas were measured with an image analysis system and dark-field optics. Choline acetyltransferase and glutamate decarboxylase immunoreactivity displayed distinctive distribution patterns throughout the cerebrum. In general, limbic cortex showed greater intensity of both choline acetyltransferase and glutamate decarboxylase immunoreactivity than neocortex. For example, choline acetyltransferase immunoreactivity in pyriform and retrosplenial cortex was 54% and 29% greater, respectively, than in neocortex, and glutamate decarboxylase immunoreactivity in the same cortical areas was 5% and 17% greater, respectively. In addition to these regional differences, the marked variations of choline acetyltransferase and glutamate decarboxylase immunostaining were characterized as either coincidental or complementary when comparing their laminar distributions. The laminar pattern and relative intensities of choline acetyltransferase and glutamate decarboxylase immunostaining were coincident in some layers of all cortical regions. For example, both choline acetyltransferase and glutamate decarboxylase immunoreactive intensities were high in cellular layers II and IV of the entorhinal cortex. In contrast, examples of complementary choline acetyltransferase and glutamate decarboxylase immunoreactive patterns were observed in retrosplenial cortex and neocortex. In neocortex, layers III and part of V were intensely glutamate decarboxylase-positive, whereas these same layers were less intensely choline acetyltransferase immunoreactive than the intervening layer IV and upper part of V. Quantitatively, choline acetyltransferase immunoreactivity in layers IV and upper V was 27-37% greater than adjacent layers II and deep V. The glutamate decarboxylase immunostaining pattern was complementary in that layer IV was 19-23% less intensely stained than adjacent layers III and V. Our results demonstrate that terminals immunoreactive for choline acetyltransferase and glutamate decarboxylase, and presumably the synaptic terminals that respectively use acetylcholine or gamma-aminobutyric acid as their neurotransmitters, are distributed in distinct laminar patterns that are strategically situated for modulating either afferent information in the case of cholinergic terminals or efferent transmission for GABAergic endings.(ABSTRACT TRUNCATED AT 400 WORDS)

Parvalbumin-immunoreactive neurons in the hippocampal formation of Alzheimer's diseased brain.

The number and topographic distribution of immunocytochemically stained parvalbumin interneurons was determined in the hippocampal formation of control and Alzheimer's diseased brain. In control hippocampus, parvalbumin interneurons were aspiny and pleomorphic, with extensive dendritic arbors. In dentate gyrus, parvalbumin cells, as well as a dense plexus of fibers and puncta, were associated with the granule cell layer. A few cells also occupied the molecular layer. In strata oriens and pyramidale of CA1-CA3 subfields, parvalbumin neurons gave rise to dendrites that extended into adjacent strata. Densely stained puncta and beaded fibers occupied stratum pyramidale, with less dense staining in adjacent strata oriens and radiatum. Virtually no parvalbumin profiles were observed in stratum lacunosum-moleculare or the alveus. Numerous polymorphic parvalbumin neurons and a dense plexus of fibers and puncta characterized the deep layer of the subiculum and the lamina principalis externa of the presubiculum. In Alzheimer's diseased hippocampus, there was an approximate 60% decrease in the number of parvalbumin interneurons in the dentate gyrus/CA4 subfield (P<0.01) and subfields CA1-CA2 (P<0.01). In contrast, parvalbumin neurons did not statistically decline in subfields CA3, subiculum or presubiculum in Alzheimer's diseased brains relative to controls. Concurrent staining with Thioflavin-S histochemistry did not reveal degenerative changes within parvalbumin-stained profiles. These findings reveal that parvalbumin interneurons within specific hippocampal subfields are selectively vulnerable in Alzheimer's disease. This vulnerability may be related to their differential connectivity, e.g., those regions connectionally related to the cerebral cortex (dentate gyrus and CA1) are more vulnerable than those regions connectionally related to subcortical loci (subiculum and presubiculum).

Decreased expression of nuclear and mitochondrial DNA-encoded genes of oxidative phosphorylation in association neocortex in Alzheimer disease.

We recently reported 50% decreases in mRNA levels of mitochondrial DNA (mtDNA)-encoded cytochrome oxidase (COX) subunits I and III in Alzheimer disease (AD) brains. The decreases were observed in an association neocortical region (midtemporal cortex) affected in AD, but not in the primary motor cortex unaffected in AD. To investigate whether the decreases are specific to mtDNA-encoded mRNA, we extended this analysis to nuclear DNA (nDNA)-encoded subunits of mitochondrial enzymes of oxidative phosphorylation (OXPHOS). Brains from five AD patients showed 50-60% decreases in mRNA levels of nDNA-encoded subunit IV of COX and the beta-subunit of the F0F1-ATP synthase in midtemporal cortex compared with mRNA levels from midtemporal cortex of control brains. In contrast, these mRNAs were not reduced in primary motor cortices of the AD brains. The amount of nDNA-encoded beta-actin mRNA and the amount of 28S rRNA were not altered in either region of the AD brain. The results suggest that coordinated decreases in expression of mitochondrial and nuclear genes occur in association cortex of AD brains and are a consequence of reduced neuronal activity and downregulation of OXPHOS machinery.

No association between Alzheimer plaques and decreased levels of cytochrome oxidase subunit mRNA, a marker of neuronal energy metabolism.

It has been proposed that neuritic plaques or toxic substances diffusing from them contribute to neurodegeneration in Alzheimer disease. We examined this hypothesis by looking for evidence of decreased neuronal energy metabolism in the proximity of neuritic plaques. Levels of mitochondrial DNA-encoded mRNA for subunit III of cytochrome oxidase, a marker of neuronal energy metabolism, were determined in post mortem brain samples. Consistent with earlier results, overall cytochrome oxidase subunit III mRNA levels were decreased in Alzheimer midtemporal cortex compared with controls. However, this reduction did not correlate with plaque density. In Alzheimer brains, cytochrome oxidase subunit III mRNA levels in neurons bearing neurofibrillary tangles were lower than in tangle-free neurons. However, neuronal cell bodies in close proximity of neuritic plaques showed no decrease in cytochrome oxidase subunit III mRNA or total polyadenylated mRNA compared with more distant neurons. Cytochrome oxidase enzyme activity in neuronal processes also showed no local reduction around neuritic plaques. These results suggest that neuritic plaques do not contribute to reduced neuronal energy metabolism in Alzheimer disease.

Impairment in mitochondrial cytochrome oxidase gene expression in Alzheimer disease.

Brains from 5 patients with Alzheimer's disease (AD) showed a 50%-65% decrease in mRNA levels of the mitochondrial-encoded cytochrome oxidase (COX, a marker of oxidative metabolism) subunits I and III in the middle temporal association neocortex, but not in the primary motor cortex, as compared to 5 control brains. The amount of mitochondrial-encoded 12S rRNA was not altered, nor was the amount of nuclear-encoded lactate dehydrogenase B mRNA (a marker of glycolytic metabolism). These data suggest that the decrease in COX I and III subunits mRNA in affected brain regions may contribute to reduced brain oxidative metabolism in AD.

Development and application of a modified monoclonal hybridoma technique for isolating monoclonal antibodies to human brain regions.

We developed a modified monoclonal hybridoma technique that combines two conventional methods: a conventional immunosuppression method with cyclophosphamide treatment and an in vitro immunization method. This technique is advantageous over conventional methodologies because it requires a shorter period for immunization of mice and a smaller quantity of antigen, and gives rise to antibody-secreting hybridomas with higher efficiency. One monoclonal hybridoma line, designated as BG5, was established by this technique after activation of lymphocytes with muramyl dipeptide and with the immunogen obtained from human entorhinal cortex. Western blot analysis showed a relatively high expression of BG5 antigen in human entorhinal cortex. Our results suggest that this modified hybridoma technique may rapidly facilitate the acquisition of brain region-specific antibodies. We call this technique 'suppression immunization followed by in vitro stimulation procedure' (SOFISTIC).

Development and application of a modified monoclonal hybridoma technique for isolating monoclonal antibodies to human brain regions.

We developed a modified monoclonal hybridoma technique that combines two conventional methods: a conventional immunosuppression method with cyclophosphamide treatment and an in vitro immunization method. This technique is advantageous over conventional methodologies because it requires a shorter period for immunization of mice and a smaller quantity of antigen, and gives rise to antibody-secreting hybridomas with higher efficiency. One monoclonal hybridoma line, designated as BG5, was established by this technique after activation of lymphocytes with muramyl dipeptide and with the immunogen obtained from human entorhinal cortex. Western blot analysis showed a relatively high expression of BG5 antigen in human entorhinal cortex. Our results suggest that this hybridoma technique may rapidly facilitate the acquisition of brain region-specific antibodies. We call this technique 'suppression immunization followed by in vitro stimulation procedure' (SOFISTIC).

Localization of cytochrome oxidase (COX) activity and COX mRNA in the perirhinal and superior temporal sulci of the monkey brain.

Cytochrome oxidase (COX) activity and COX II mRNA expression were localized in the perirhinal and superior temporal sulci of the rhesus monkey brain. In both regions, a laminar distribution of COX activity and COX II mRNA was observed. COX activity was intense in layers I and IV and were localized to the neuropil. In contrast, COX II mRNA was localized to neuronal cell bodies. In the prorhinal region, highest levels of COX II mRNA was detected in cell bodies of layers II and IV, and in the perirhinal region, in cell bodies of layers III and V-VI. In the superior temporal sulcus, COX II mRNA was detected in cell bodies of layers III and V-VI. Thus, COX II mRNA and COX activity are uniquely localized in the cortical layers and to those neurons that support cortico-cortical connections.

Localization of cytochrome oxidase (COX) activity and COX mRNA in the hippocampus and entorhinal cortex of the monkey brain: correlation with specific neuronal pathways.

Cytochrome oxidase (COX) activity and COX II mRNA expression were localized in the hippocampal formation and entorhinal cortex of the rhesus monkey brain by means of enzyme histochemistry and in situ hybridization, respectively. Within the hippocampal formation, the terminal field of the perforant pathway showed the highest levels of COX activity, whereas COX II mRNA was localized mainly in neuronal cell bodies. In the entorhinal cortex. COX II mRNA was detected in neuronal cell bodies of layers II and IV. These results indicate that the pattern of localization of COX and its mRNA in entorhinal cortex correlates with the input and output pathways of the hippocampus.

Reduced nicotinamide adenine dinucleotide phosphate-diaphorase (NADPH-d) profiles in the amygdala of human and New World monkey (Saimiri sciureus).

The topographic distribution of nicotinamide adenine dinucleotide-diaphorase (NADPH-d) stained profiles in the amygdala of the human and new world monkey (Saimiri sciureus) were studied histochemically. Fiber and terminal staining were heterogeneously distributed within the amygdala. The most intense staining occurred in the basolateral subdivision, consisting of the lateral, basolateral and accessory basal nuclei. Moderate staining intensity was observed throughout the cortical and media nuclei and cortical transition area, constituents of the corticomedial subdivision. The central amygdaloid area was characterized by minimal NADPH-d histochemical reactivity. NADPH-d positive neurons were pleomorphic and divisible into two classes based on their staining characteristics: intensely or lightly stained neurons. Their distribution was generally complementary, with the majority of intensely stained neurons occupying the basolateral subdivision. There were no appreciable species differences in the patterns of neuronal, fiber and terminal staining between monkey or human amygdala. These results may be relevant to our understanding of the selective vulnerability of neural systems within the human amygdala in neurodegenerative diseases.

Reduced nicotinamide adenine dinucleotide phosphate-diaphorase (NADPH-d) histochemistry in the hippocampal formation of the New World monkey (Saimiri sciureus).

Nicotinamide adenine dinucleotide phosphate-diaphorase (NADPH-d) containing fibers and neurons within the hippocampal formation and entorhinal cortex of the new world monkey were determined using a direct histochemical procedure. Occasional intensely stained bipolar NADPH-d positive neurons were seen in the polymorphic zone within the hilus of the dentate gyrus and molecular layer of the hippocampus. Although virtually no intensely stained cells were seen in the CA subfields, a few small oval lightly stained NADPH-d perikarya were found subjacent to CA2. An occasional intensely stained multipolar NADPH-d containing neuron was observed in the subiculum, presubiculum and parasubiculum. In the entorhinal cortex, NADPH-d cells were scattered in all layers with the greatest preponderance in layers 5-6 and underlying white matter. Dense bands of NADPH-d fibers occurred in the outer layer of the molecular layer of the dentate gyrus and the hippocampo-subicular border. NADPH-d fibers also were seen in pre- and parasubicular regions. NADPH-d fiber staining in entorhinal cortex varied mediolaterally with an increasing laminar distribution more caudally. The heaviest bands of NADPH-d fibers occurred in layers 1 and 4 and the white matter-layer 6 border. The distribution patterns of this select neuronal population may be relevant to the study of hippocampal and entorhinal areas in neurodegenerative diseases.

Evidence that transmitter-containing dystrophic neurites precede those containing paired helical filaments within senile plaques in the entorhinal cortex of nondemented elderly and Alzheimer's disease patients.

Within the amygdala of elderly subjects and patients with Alzheimer's disease (AD), we recently found evidence suggesting amyloid beta-protein (A beta P) deposition occurs before the appearance of dystrophic neurites. Moreover, these data suggested dystrophic neurites initially lack evidence of cytoskeletal pathology although with time and further maturation, the dystrophic neurites display an altered cytoskeleton as evidenced by their immunoreactivity to Alz-50 and paired-helical filaments (PHF). These findings are of particular relevance to our understanding of the sequence of pathologic events in AD and thus it has become important to determine whether these events are unique to the amygdala or are representative of a more general pattern which can be found throughout the brain. Using a battery of antibodies to markers that are characteristic of AD pathology (i.e., A beta P, PHF, and Alz-50), three peptidergic neurotransmitters (neurotensin, somatostatin, and substance P), and one neurotransmitter biosynthetic enzyme (choline acetyltransferase), we examined the entorhinal cortex (EC) of three groups of subjects (AD, normal elderly, and a group of nondemented elderly with numerous senile plaques). The EC was studied, in part, because it is well recognized as a brain region displaying severe and, most importantly, early pathologic changes. Like the amygdala, we found evidence that amyloid beta-protein immunoreactive (A beta P-IR) and thioflavine-S-positive senile plaques occur within the EC prior to the appearance of transmitter-, Alz-50-, or PHF-immunoreactive dystrophic neurites. We also observed transmitter-immunoreactive dystrophic neurites in the absence of Alz-50 or PHF-immunolabeled dystrophic neurites and transmitter- and Alz-50-IR dystrophic neurites in the absence of those containing PHF. Collectively, these findings were similar to those seen within the amygdala and thus reinforced the concept that A beta P deposition is the primary event in plaque pathology, and this deposition is subsequently followed by the appearance of dystrophic neurites which retain their transmitter phenotype yet lack an altered cytoskeleton. With time, these dystrophic neurites develop cytoskeletal alterations and become immunoreactive to Alz-50 and PHF.

Downregulation of oxidative phosphorylation in Alzheimer disease: loss of cytochrome oxidase subunit mRNA in the hippocampus and entorhinal cortex.

Messenger RNA (mRNA) for cytochrome oxidase subunit II (COX II) was localized by in situ hybridization in the entorhinal cortex and hippocampal formation of postmortem brain tissue from normal human subjects and from patients with Alzheimer disease (AD). In the control entorhinal cortex, COX II mRNA was detected mainly in neuronal cell bodies of layers II and IV. In control hippocampal formation, highest levels were localized in neuronal cell bodies of the dentate gyrus and the CA3 and CA1 regions, neurons that are involved in the major input and output pathways of the hippocampal formation. In AD brain, COX II mRNA was markedly reduced in the entorhinal cortex and the hippocampal formation compared with control brain. In the AD hippocampal formation, reductions were in regions severely affected by AD pathology as well as in regions that were relatively spared. These results are consistent with the hypothesis that reduced mitochondrial energy metabolism reflects loss of neuronal connections in AD.