Prolonged seizures produce death of hippocampal neurons, which is thought to initiate epileptogenesis and cause a disruption of hippocampally mediated behaviors. This study aimed to evaluate behavioral and neuroanatomical changes induced by brief seizures and to compare them with changes induced by prolonged seizures. Adult rats were administered 6 brief seizures, elicited by electroshock (ECS). Prolonged seizures (status epilepticus, SE) were induced by pilocarpine. Two months later, the rats’ behavior was tested using the Morris water maze, passive avoidance and active avoidance tests. The number of neurons in the hippocampal formation was estimated using stereological methods. ECS seizures produced loss of neurons, ranging between 14% and 26%, in the dentate hilus, subiculum, presubiculum, parasubiculum, and entorhinal layers III and V/VI. However, the neuron loss caused by SE in the same structures, as well as in the hippocampal CA3 and CA1 fields, ranged between 34% and 50%. SE additionally killed many neurons in the dentate granular layer, postsubiculum and entorhinal layer II. ECS treatment caused mild impairments in spatial learning and passive avoidance, but it was not associated with spontaneous motor seizures. In contrast, SE produced a severe disruption of spatial learning, passive and active avoidance, and led to the development of spontaneous seizures. These data show that both prolonged seizure activity and brief seizures result in structural and functional alterations in the temporal lobe circuits, but those caused by prolonged seizures are considerably more severe. Hippocampal damage elicited by brief seizures does not necessarily lead to spontaneous motor seizures.

A novel presenilin-1 (PS1) mutation (P117S) in an American pedigree is described. We compare clinical, neuropathological and cell culture phenotypes produced by this mutation with another codon 117 mutation that was earlier discovered by our group in a Polish kindred. Both mutations are associated with an unusually severe Alzheimer disease (AD) phenotype, with the onset starting before the third decade of life, rapid disease progression and acute presentation of clinical symptoms. The severity of clinical phenotype was closely correlated with the abundance of pathology: massive deposition of Abeta42 in plaques, severe neurofibrillary degeneration and neuronal loss. When overexpressed in mouse neuroblastoma N2a cells, both mutations caused loss of an ability to promote neurite outgrowth and produced an increase in the ratio of secreted Abeta42/40 amyloid peptides. In stably transfected N2a cell lines only mutant proteins were endoproteolytically cleaved indicating some dependability of this process on the presence of mutation. Taken together, our results show that clinical and cell culture phenotypes produced by these 2 codon 117 mutations are closely related suggesting that the pathogenic action of PS1 may involve effect on neurite outgrowth and endoproteolytic cleavage of the full-length protein. Given the high potency in vivo and in vitro of both codon 117 mutations, this site of PS1 must be particularly important for its normal/pathogenic function.

The minibrain kinase (Mnb/Dyrk1A) gene is localized in the Down syndrome (DS) critical region of chromosome 21. This gene encodes a proline-directed serine/threonine protein kinase (minibrain kinase-Mnb/Dyrk1A), which is required for the proliferation of distinct neuronal cell types during postembryonic neurogenesis. To study the distribution of Mnb/Dyrk1A during human brain development and aging, we raised Mnb/Dyrk1A-specific antibody (mAb 7F3) and examined 22 brains of normal subjects from 8 months to 90 years of age. We found that neurons were the only cells showing the presence of 7F3-positive product in both cell nucleus and cytoplasm. Nuclear localization supports the concept that Mnb/Dyrk1A may be involved in control of gene expression. Synaptic localization of Mnb/Dyrk1A also supports our previous studies suggesting that Mnb/Dyrk1A is a regulator of assembly of endocytic apparatus and appears to be involved in synaptic vesicle recycling and synaptic signal transmission. Accumulation of numerous 7F3-positive corpora amylacea in the memory and motor system subdivisions in subjects older than 33 years of age indicates that Mnb/Dyrk1A is colocalized with markers of astrocyte and neuron degeneration. Differences in the topography and the amount of Mnb/Dyrk1A in neurons, astrocytes, and ependymal and endothelial cells appear to reflect cell type- and brain structure-specific patterns in trafficking and utilization of Mnb/Dyrk1A.

Study of the hippocampal formation of 82 subjects, including 25 control subjects from 33 to 83 years of age, 34 subjects with Alzheimer disease (AD) from 65 to 89 years of age, and 23 subjects with Down syndrome (DS) from 33 to 72 years of age, revealed hippocampal vasculopathy with fibrosis and calcification (VFC) in 40% of control, 59% of AD, and 4% of DS subjects. VFC starts in the precapillaries/capillaries in the molecular layer of the dentate gyrus (DG) and expands to the granule cell and polymorphic cell layer of the DG, and to the stratum lacunosum/moleculare in the CA1 sector. Vasculopathy spreads from the tail to the body and, in a few cases, to the head of the hippocampal formation. Light and electron microscopy reveal thickening of the vascular wall with fibrosis, calcification, and enforcement of the astrocyte interface with vessels with anchorage densities associated with hemidesmosome-like structures. In moderately and severely affected cases, fragmentation and removal of calcified and occluded vessels result in local reduction of vascular network. In two AD subjects, severe vascular calcification extending from the tail to the head of the hippocampal formation was associated with loss of almost all neurons in the CA1 sector and in the subiculum proper, corresponding to hippocampal sclerosis. The topography of affected vessels and the patterns of neuronal loss reflect the middle hippocampal artery distribution with its precapillary/capillary network. The similar prevalence of vasculopathy in the AD group and in the age-matched control group, and the presence of hippocampal VFC in only one subject in the DS cohort, 96% of which is affected by Alzheimer-type pathology, oppose the link between AD and this form of vasculopathy. However, severe VFC affects the pattern of AD pathology locally by deletion of neurofibrillary degeneration and β-amyloidosis in the CA1 sector, subiculum proper, and the molecular layer of the dentate gyrus. Hippocampal VFC appears to be a form of vascular pathology with a unique predilection for the middle hippocampal artery and corresponding capillary network, which results in patchy neuronal loss in moderately affected subjects and in almost total neuronal loss in the area of impaired blood supply in severely affected subjects. These observations suggest an etiologic link between hippocampal VFC and hippocampal sclerosis.

The decrease in the number of neurons free of neurofibrillary changes, neurons with neurofibrillary degeneration, and the total volume of beta-amyloid (A beta) deposits in the amygdala of people with Down syndrome and in late stages of Alzheimer disease were estimated by using morphometry and regression analysis. This model predicts that the duration of neurofibrillary changes from the pretangle stage to ghost tangles is approximately 4.7 years. The correlation between the decrease in the number of neurons and the decrease in the amount of A beta indicates that amyloid deposition is associated with neurons and that loss of neurons causes decrease in A beta deposition. The presence of neurons only with neurofibrillary tangles, and the absence of the amyloid deposits predicted by regression analysis suggest that neurons with tangles are not engaged in amyloid deposition. The disappearance of amyloid by approximately 2.2 years after loss of neurons free of neurofibrillary changes indicates that A beta deposits are degradable and removable and that even in severely atrophic amygdala, there are mechanisms of amyloid resolution. This study shows that in normal aging in the amygdala, extracellular A beta appears later than neurofibrillary changes.

Although the molecular basis of programmed cell death, a pivotal process for the proper development, organization and future functioning of the brain, has been studied extensively in different experimental models and in various species, it is not entirely understood. The principles that govern this process in the developing human brain are even less elucidated. To provide more insight into the mechanisms underlying programmed cell death in the human brain, we have studied the temporal and spatial pattern of immunolocalization in the developing normal and Down syndrome brain of two proteins that are supposed to act either as a neuronal survival promoter-Bcl-2 or a neuronal survival suppressor- Bax. Our data show a dramatic decrease in immunoreactivity to Bcl-2 and, at the same time, a gradual increase in immunoreactivity to Bax in the human brain around 30-32 week’s gestation; according to recent data, this gestational period is accompanied by the most intense reduction of cortical neurons. However, our obser vations also suggest that at least at certain stages of human brain development, both proteins, and especially Bax, may also be involved in the final differentiation and maturation of central nervous system neurons. In addition, altered patterns of Bcl-2 and, to a much lesser degree, also of Bax immunoreactivity in developing Down syndrome brain allow us to propose that mechanisms underlying programmed cell death may contribute to the pathology in Down syndrome brain.