Dendritic growth in the aged human brain and failure of growth in senile dementia.

Golgi-stained dendrites of single randomly chosen layer-II pyramidal neurons in the human parahippocampal gyrus were quantified with a computer-microscope system. In nondemented aged cases (average age, 79.6 years), dendritic trees were more extensive than in adult cases (average age, 51.2), with most of the difference resulting from increases in the number and average length of terminal segments of the dendritic tree. These results provide morphological evidence for plasticity in the mature and aged human brain. In senile dementia (average age, 76.0), dendritic trees were less extensive than in adult brains, largely because their terminal segments were fewer and shorter. Cells with shrunken dendritic trees were found in all brains. These data suggest a model of aging in the central nervous system in which one population of neurons dies and regresses and the other survives and grows. The latter appears to be the dominant population in aging without dementia.

Neurons may live for decades with neurofibrillary tangles.

Neurons containing neurofibrillary tangles (NFT) are one of the pathological hallmarks of Alzheimer disease (AD). It is known that this population of neurons express gene products and thus function to some degree, but it is unknown how long these neurons may survive with NFT. It is also thought that the formation of NFT results in the death of neurons. Using quantitative data on neuron loss and NFT formation as a function of disease duration, we have generated a computer program that models both the degeneration of CA1 hippocampal neurons and the formation of NFT in these neurons in AD. Modeling various neuron survival times with NFT and altering selected assumptions upon which the models are based, we arrive at the conclusions that 1) CA1 hippocampal neurons survive with NFT for about 20 years, and 2) NFT may not be obligatory for death of CA1 hippocampal neurons in AD.

Quantitative decrease in synaptophysin message expression and increase in cathepsin D message expression in Alzheimer disease neurons containing neurofibrillary tangles.

Combining immunocytochemistry with in situ hybridization of Alzheimer disease (AD) hippocampus demonstrated a 50% reduction in grain density for synaptophysin message over CA1 pyramidal neurons containing neurofibrillary tangles (NFT) relative to near neighbor NFT-free neurons. This decrease was not global, but was selective since message grain density for the lysosomal protein, cathepsin D, increased 33% in these neurons (relative to NFT-free neurons). Poly A+ message grain density decreased by 25% in NFT neurons. Percent of the cell body containing NFT correlated -0.35 (p < 0.0001) with grain density for synaptophysin message. These data verify the concept of altered profiles of gene expression as a function of disease state within single cells and suggest that events associated with NFT formation may lead to altered expression of synaptic messages.

Research uses of neuropathological data.

The study of relationships between neuropathological characteristics and behavioral, structural, chemical, and molecular variables offers immense promise for understanding the basic pathophysiology of Alzheimer's disease. This position paper examines the need for standardized procedures and quantitation if neuropathological data are to be optimally useful among laboratories investigating the biology of Alzheimer's disease. These requirements include standardized fixation, embedding, sectioning, staining, brain regions and sampling methods. In addition, the definition of the structures to be quantified, such as plaque type(s), needs to be rigorously specified. Unbiased stereological methods for quantification should be used. These needs for optimal research utility exceed the needs and practicality for diagnostic purposes, suggesting a two-tiered approach to the neuropathology of Alzheimer's disease: diagnostic and research.

Neuron numbers and sizes in aging brain: comparisons of human, monkey, and rodent data.

One of the several sources of interest in aging animal brains is their potential as models of the aging human brain. In this review we examine whether neuron numbers and sizes change similarly in aging human, monkey and rodent brain regions which data are available from more than one species. The number of brain regions studied in more than one species is surprisingly limited. Some regions show correspondence in age-related changes between humans and selected animal models (primary visual cortex, CA1 of hippocampus). For the majority of regions the data are conflicting, even within one species (e.g., somatosensory cortex, frontal cortex, cerebellum, cholinergic forebrain areas, locus coeruleus). Although some of the conflicting data may be attributed to procedural differences, particularly when data are expressed as density changes, much must be attributed to real species and/or strain differences in rodents. We conclude that neuron numbers and sizes may show similar age-related changes in human and animal brains only for sharply defined brain regions, animal species and/or strains, and age ranges.

Neuron numbers and dendritic extent in normal aging and Alzheimer's disease.

Factors which limit the interpretation of studies of aging brain include: secular trends, species and strain differences, effects of tissue processing, and bias which may be introduced at many levels of an experimental design. With these limitations considered, evidence is reviewed regarding neuron numbers and dendritic extent in normally aging rodent, monkey and human brain and in Alzheimer's disease. It is concluded that neuron loss and change in dendritic extent in normal aging are regionally specific, and that corresponding brain regions do not always change in similar ways in rodents and primates. It is suggested that such differences may, in part, be due to inconsistent definitions of 'aged' among species. In Alzheimer's disease there is excess neuron loss and dendritic regression in some, but not all, brain regions. Measures of the morphological substrates of brain function show appreciable overlap between AD and control groups. It is hypothesized that the static, post-mortem status of brain morphology may not adequately reflect the functional capabilities of the dynamic morphology of the living brain.

Neurons bearing neurofibrillary tangles are responsible for selected synaptic deficits in Alzheimer's disease.

The observation that neurons containing neurofibrillary tangles are usually adjacent to neurons free of any morphological indication of disease, suggests the hypothesis that it is NFT-bearing neurons that are primarily responsible for the loss of function in AD. Quantitative Golgi postmortem studies from our laboratories have indicated that there is in many regions of the brains of nondemented humans an age-related increase in dendritic extent of single neurons. In Alzheimer's disease, this normal, age-related increase in dendritic extent was not found, leading to the hypothesis that one of the neurobiological defects in AD is a failure of neuronal plasticity. Message levels of the growth-associated protein, GAP-43, in frontal association cortex (area 9/46) indicated that AD brains with the highest density of neurofibrillary tangle-bearing neurons, showed GAP-43 message levels decreased of the order of 6-fold relative to AD brains with the lowest density of NFT. Combined immunocytochemistry to differentiate tangle-bearing from tangle-free neurons with in situ hybridization to define relative GAP-43 message levels in single neurons revealed that grain density over tangle-bearing neurons containing nuclei was reduced 3-fold compared to that over adjacent tangle-free neurons. This reduction in expression of GAP-43 message in tangle-bearing neurons was selective, because using probes for other messages showed that grain density over tangle-bearing neurons was, on average, increased or similar to that over adjacent non-tangle-bearing neurons. Message levels for the synaptic vesicle-associated protein, synaptophysin, have also been found to be reduced in tangle-bearing neurons relative to adjacent tangle-free neurons.(ABSTRACT TRUNCATED AT 250 WORDS)

Dendritic regression dissociated from neuronal death but associated with partial deafferentation in aging rat supraoptic nucleus.

As neurons are lost in normal aging, the dendrites of surviving neighbor neurons may proliferate, regress, or remain unchanged. In the case of age-related dendritic regression, it has been difficult to distinguish whether the regression precedes neuronal death or whether it is a consequence of loss of afferent supply. The rat supraoptic nucleus (SON) represents a model system in which there is no age-related loss of neurons, but in which there is an age-related loss of afferents. The magnocellular neurosecretory neurons of the SON, that produce vasopressin and oxytocin for release in the posterior pituitary, were studied in male Fischer 344 rats at 3, 12, 20, 27, 30, and 32 months of age. Counts in Nissl-stained sections showed no neuronal loss with age, and confirmed similar findings in other strains of rat and in mouse and human. Nucleolar size increased between 3 and 12 months of age, due, in part, to nucleolar fusion, and was unchanged between 12 and 32 months of age, indicating maintenance of general cellular function in old age. Dendritic extent quantified in Golgi-stained tissue increased between 3 and 12 months of age, was stable between 12 and 20 months, and decreased between 20 and 27 months. We interpret the increase between 3 and 12 months as a late maturational change. Dendritic regression between 20 and 27 months was probably the result of deafferentation due to the preceding age-related loss of the noradrenergic input to the SON from the ventral medulla.

Neuron numbers in locus coeruleus do not change with age in Fisher 344 rat.

Counts of neurons in locus coeruleus of Fisher 344 rats of ages from 12 to 32 months showed no loss with advancing age. Species comparisons of age changes in locus coeruleus neuron counts and measures of catecholamine systems suggested that the human and nonhuman primate locus coeruleus shows age changes and seen in rodent locus coeruleus. It is suggested that although rodent models of human aging have value under some circumstances, rodent and primate brain should not be considered equivalent with regard to phenomena related to noradrenergic mechanisms.

Divergence of biological and chronological aging: evidence from rodent studies.

Literature on aging populations of rodents supports the intuitive view that significant functional variation exists among like-aged, elderly individuals: chronological age as a solitary measure is a poor indicator of biological age. In this report, we review a variety of studies which classify aged rodents based on genetic and/or behavioral similarities, in addition to chronological age, and have provided valuable neurobiological and physiological information on age-related changes which accompany functional impairments, or the lack of them. Beyond their descriptive value for gerontological research, these findings suggest ways in which biological aging can be manipulated to promote good function in aged individuals.

Net dendritic stability of layer II pyramidal neurons in F344 rat entorhinal cortex from 12 to 37 months.

Dendritic extent of Golgi-Cox stained layer II entorhinal cortex pyramidal neurons was quantified in five groups of male F344 rats aged 12, 20, 27, 30 and 37 months. Over the age range studied, neither the apical nor the basal dendritic trees showed any statistically significant change in total dendritic length, numbers of segments or average segment length. This finding of average stability of the dendritic tree does not imply absence of remodelling of connections, but does require that if remodeling does occur, retraction and proliferation of dendrites must, on average, be equal. We hypothesized that in groups of animals with similar genetic and environmental histories neighbor neuron death provides the major stimulus for dendritic proliferation. Since we found dendritic stability in the cells reported here, we would predict that there should be no age-related loss of layer II pyramidal neurons in the entorhinal cortex of the normally aging F344 male rat between 12 and 37 months. This hypothesis may be tested by counting neurons within this region.

Prostaglandin G/H synthase-2 (cyclooxygenase-2) mRNA expression is decreased in Alzheimer's disease.

Recent evidence suggests that the use of nonsteroidal anti-inflammatory drugs (NSAIDS) is beneficial for therapy or prevention of Alzheimer's disease (AD). The major anti-inflammatory action of NSAIDS is to inhibit prostaglandin G/H synthase-2 (PGHS-2), the first committed enzymatic step for prostaglandin biosynthesis. We have previously shown that PGHS-2 message is induced by Interleukin-1 beta and other inflammatory mediators in primary cultures of rodent astrocytes. To determine whether similar elevations of PGHS-2 occur as part of the gliosis in AD, we quantified PGHS-2 mRNA levels in control and AD brain by Northern hybridization analysis. To our surprise we found that PGHS-2 mRNA levels were reduced threefold in AD neocortex relative to control brain tissue. In contrast, levels were not reduced in putamen, an area that is relatively spared in AD. To localize PGHS-2 mRNA production in control and AD brain, sections of neocortex and hippocampus were hybridized with a 35S-labeled riboprobe for human PGHS-2 followed by immunocytochemistry with antibodies against neuron specific enolase (NSE) or glial fibrillary acidic protein (GFAP). Our findings indicate that PGHS-2 message is primarily localized to cells that stain for NSE rather than GFAP. Furthermore, in the three cases we examined, PGHS-2 hybridization per neuron appeared to be reduced in AD. Thus, the decrease we observe in overall PGSH-2 mRNA levels is likely to reflect both the known decline in numbers of neurons in AD as well as a lowered capacity for neuronal synthesis of PGHS-2, perhaps due to dysfunction or a loss of synaptic input.

Characterization of brain samples in studies of aging, Alzheimer's, and other neurodegenerative diseases.

We review current understanding of the clinical and pathologic information needed for the determination of optimal brain tissue samples for the conduct of studies of Alzheimer's disease (AD). Characteristics that may distinguish AD from other dementing disorders are discussed. Selected considerations in the conduct of basic neurobiological studies are also outlined. Although the 28 NIA-funded Alzheimer's Centers can provide excellent clinical and neuropathological data, studies conducted outside these centers should also strive to gather the information suggested here. Clinical and neuropathological data should be used not only to classify subjects as control or AD, but also as variables that may significantly contribute to the analysis of neurobiological data obtained in the laboratory.

Decreased sympathetic innervation of spleen in aged Fischer 344 rats.

Splenic noradrenergic innervation in young adult and aged Fischer 344 rats was examined using fluorescence histochemistry for catecholamines and high performance liquid chromatography with electrochemical detection (LCEC) for the quantitation of norepinephrine (NE). In young adult rats, abundant noradrenergic plexuses followed the vasculature and trabeculae into splenic white pulp. In aged rats, noradrenergic innervation was reduced in density and in overall intensity of fluorescence, and splenic NE levels were significantly lower. The relationship between diminished noradrenergic innervation and diminished immune responsiveness in aging mammals, while not clear on a causal level, is presented as a hypothesis for further testing.

A Golgi study of hypothalamic transplants in young and old host rats.

The supraoptic nucleus of the F344 rat shows an age-related dendritic regression. In order to determine whether this previously observed dendritic regression may have been related to extrinsic (to the cell) hormonal, neurotoxic, or other circulating factors unique to the hypothalamus of older brains, we conducted a quantitative Golgi study of F344 embryonic anterior hypothalamic transplants into the third ventricle of young adult (5 months) and older (25 months) male F344 rats. Three months following transplantation there were no qualitative effects of host age on neuronal morphology, nor were there quantitative effects on transplant size, dendritic length or branching frequency within the transplanted tissue. These results suggest that either (a) there were no age-related changes in factors in the host brain which were sufficient to significantly affect dendritic extent or, (b) intrinsic connections or other properties within the transplant may be important in moderating the effect of the milieu of the aged brain on the transplanted tissue.

Stability of numbers but not size of mouse forebrain cholinergic neurons to 53 months.

In normal mammalian aging there is a reduction of cholinergic markers in a variety of regions. To determine whether this reduction is related to reduced numbers of basal forebrain cholinergic neurons, we counted the number and measured the sizes of the magnocellular acetylcholinesterase-positive neurons in this region of 7, 15, and 53-month-old C57Bl/6NNIA mice. Data were collected from coded slides containing the medial septum, nucleus of the diagonal band, magnocellular preoptic nucleus, and nucleus basalis magnocellularis. There was no decline in numbers of basal forebrain acetylcholinesterase-positive neurons in any of the regions studied. However, cell sizes showed a progressive age-related decline which was greatest in the nucleus basalis magnocellularis.

Stability of dendrites in cortical barrels of C57BL/6N mice between 4 and 45 months.

Qualitative and quantitative studies of dendritic parameters were conducted on Golgi-impregnated layer IV spiny stellate neurons in the posteromedial barrel subfield (PMBSF) of somatosensory cortex of the C57B1/6N mouse. Three mice each, at 4, 12, 22, 26, 30, 36 and 45 months of age were studied. No qualitative changes were observed among animals of different ages. The quantitative data indicated that dendritic length and numbers of segments remained unchanged over all ages studied.

Clathrin assembly protein AP-2 is detected in both neurons and glia, and its reduction is prominent in layer II of frontal cortex in Alzheimer's disease.

There are several adaptor proteins associated with clathrin coated vesicles. Among them are AP180 and AP-2. We and others have previously described synaptic localization of AP180. AP180 immunoreactivity is altered in both the superior frontal gyrus and hippocampus in Alzheimer's disease (AD). We here investigate the location and alteration of another adaptor protein, AP-2. In contrast to AP180, we have found that AP-2 is expressed by both neurons and glia. Furthermore, the only noticeable change of AP-2 in AD is a loss of its immunoreactivity in layer II of the superior frontal gyrus.

Reduced GAP-43 message levels are associated with increased neurofibrillary tangle density in the frontal association cortex (area 9) in Alzheimer's disease.

We previously suggested the hypothesis that defective neuronal plasticity is a major neurobiological deficit causing the dementia of Alzheimer's disease (AD). We used message levels of the growth-associated protein, GAP-43, as a marker of axonal plasticity to examine the hypothesis of defective neuronal plasticity in AD. When all AD cases are combined, the average level of GAP-43 message in area 9 of the AD frontal association cortex was not significantly different from the level in the comparably aged control cortex. Differentiation of AD cases on the basis of neurofibrillary tangle (NFT) density revealed that in AD cases with high tangle density average GAP-43 message level was reduced fivefold relative to levels in AD cases with low NFT density. AD cases with low neurofibrillary tangle density had levels of GAP-43 message that were not significantly different from the levels of normal controls. Differentiation of AD cases on the basis of neuritic plaque density did not indicate as strong a relationship to GAP-43 message level. The association between neurofibrillary tangle density and GAP-43 message level suggests the hypothesis that neurofibrillary tangles may reduce GAP-43 expression. Data of others show a relationship between high NFT density and reduced levels of synaptophysin-like immunoreactivity and reduced cerebral glucose metabolism. These data combine to suggest a set of AD cases with high NFT density, reduced axonal plasticity, reduced synaptic density, and reduced cerebral glucose metabolism--all variables that may be directly related to the functioning of the brain.(ABSTRACT TRUNCATED AT 250 WORDS)

Preliminary evidence: decreased GAP-43 message in tangle-bearing neurons relative to adjacent tangle-free neurons in Alzheimer's disease parahippocampal gyrus.

Loss of synapses has been shown to correlate with the severity of dementia in Alzheimer's disease (AD). Intracellular neurofibrillary tangles (NFTs) have also been shown to correlate to the severity of AD dementia. We have been investigating the influence of NFTs on mRNAs related to neuronal plasticity and synaptic function. We recently reported a decrease in message for the plasticity marker, GAP-43, in AD cases with high tangle densities. The study did not permit us to determine if: a) the decrease in GAP-43 message was specific to the NFT-bearing neurons, b) a general decrease in GAP-43 message was occurring in all surviving neurons, or c) the decrease in GAP-43 message was due to a loss of neurons. It is unlikely a loss of neurons could explain the sixfold GAP-43 message loss we reported, because only a 19% excess decrease in density of hippocampal neurons occurs in AD cases with high tangle densities. Consequently, the study reported here was undertaken to determine if a general decrease in GAP-43 message was occurring in all surviving AD neurons or if the decrease in GAP-43 message was specific to NFT-bearing neurons. We combined immunocytochemistry for neurofibrillary tangles with in situ hybridization for GAP-43 message. We report here preliminary evidence indicating a decrease in GAP-43 message in NFT-bearing neurons compared to adjacent nontangle bearing neurons in parahippocampal cortex of AD patients.