Protective effect of apolipoprotein E type 2 allele for late onset Alzheimer disease.

Gene dosage of the apolipoprotein E (APOE) epsilon 4 allele is a major risk factor for familial Alzheimer disease (AD) of late onset (after age 60). Here we studied a large series of 115 AD case subjects and 243 controls as well as 150 affected and 197 unaffected members of 66 AD families. Our data demonstrate a protective effect of the epsilon 2 allele, in addition to the dose effect of the epsilon 4 allele in sporadic AD. Although a substantial proportion (65%) of AD is attributable to the presence of epsilon 4 alleles, risk of AD is lowest in subjects with the epsilon 2/epsilon 3 genotype, with an additional 23% of AD attributable to the absence of an epsilon 2 allele. The opposite actions of the epsilon 2 and epsilon 4 alleles further support the direct involvement of APOE in the pathogenesis of AD.

Brain endolases as specific markers of neuronal and glial cells.

There are three distinct enolase isoenzymes in brain; neuron-specific enolase (NSE), formerly referred to as neuron-specific protein, which is specifically localized in neurons, a nonneuronal enolase (NNE), and a third hybrid form. Light microscopy with immunocytochemical techniques has permitted localization of non-neuronal enolase. The NNE is located in glial cells with no staining of endothelial cells or neurons. Thus, NSE and NNE can be used as specific metabolic markers for neurons and glial cells, respectively.

Gene dose of apolipoprotein E type 4 allele and the risk of Alzheimer's disease in late onset families.

The apolipoprotein E type 4 allele (APOE-epsilon 4) is genetically associated with the common late onset familial and sporadic forms of Alzheimer's disease (AD). Risk for AD increased from 20% to 90% and mean age at onset decreased from 84 to 68 years with increasing number of APOE-epsilon 4 alleles in 42 families with late onset AD. Thus APOE-epsilon 4 gene dose is a major risk factor for late onset AD and, in these families, homozygosity for APOE-epsilon 4 was virtually sufficient to cause AD by age 80.

Linkage and association study of late-onset Alzheimer disease families linked to 9p21.3.

A chromosomal locus for late-onset Alzheimer disease (LOAD) has previously been mapped to 9p21.3. The most significant results were reported in a sample of autopsy-confirmed families. Linkage to this locus has been independently confirmed in AD families from a consanguineous Israeli-Arab community. In the present study we analyzed an expanded clinical sample of 674 late-onset AD families, independently ascertained by three different consortia. Sample subsets were stratified by site and autopsy-confirmation. Linkage analysis of a dense array of SNPs across the chromosomal locus revealed the most significant results in the 166 autopsy-confirmed families of the NIMH sample. Peak HLOD scores of 4.95 at D9S741 and 2.81 at the nearby SNP rs2772677 were obtained in a dominant model. The linked region included the cyclin-dependent kinase inhibitor 2A gene (CDKN2A), which has been suggested as an AD candidate gene. By re-sequencing all exons in the vicinity of CDKN2A in 48 AD cases, we identified and genotyped four novel SNPs, including a non-synonymous, a synonymous, and two variations located in untranslated RNA sequences. Family-based allelic and genotypic association analysis yielded significant results in CDKN2A (rs11515: PDT p = 0.003, genotype-PDT p = 0.014). We conclude that CDKN2A is a promising new candidate gene potentially contributing to AD susceptibility on chromosome 9p.

Longitudinal follow-up of late-onset Alzheimer disease families.

Historically, data for genetic studies are collected at one time point. However, for diseases with late onset or with complex phenotypes, such as Alzheimer disease (AD), restricting diagnosis to a single ascertainment contact may not be sufficient. Affection status may change over time and some initial diagnoses may be inconclusive. Follow-up provides the opportunity to resolve these complications. However, to date, previous studies have not formally demonstrated that longitudinally re-contacting families is practical or productive. To update data initially collected for linkage analysis of late-onset Alzheimer disease (LOAD), we successfully re-contacted 63 of 81 (78%) multiplex families (two to 17 years after ascertainment). Clinical status changed for 73 of the 230 (32%) non-affected participants. Additionally, expanded family history identified 20 additional affected individuals to supplement the data set. Furthermore, fostering ongoing relationships with participating families helped recruit 101 affected participants into an autopsy and tissue donation program. Despite similar presentations, discordance between clinical diagnosis and neuropathologic diagnosis was observed in 28% of those with tissue diagnoses. Most of the families were successfully re-contacted, and significant refinement and supplementation of the data was achieved. We concluded that serial contact with longitudinal evaluation of families has significant implications for genetic analyses.

Apolipoprotein E associates with beta amyloid peptide of Alzheimer's disease to form novel monofibrils. Isoform apoE4 associates more efficiently than apoE3.

Late-onset and sporadic Alzheimer's disease are associated with the apolipoprotein E (apoE) type 4 allele expressing the protein isoform apoE4. Apolipoprotein E binds avidly to beta amyloid (A beta) peptide, a major component of senile plaque of Alzheimer's disease, in an isoform-specific manner. The apoE4 isoform binds to A beta peptide more rapidly than apoE3. We observed that soluble SDS-stable complexes of apoE3 or apoE4, formed by coincubation with A beta peptide, precipitated after several days of incubation at 37 degrees C with apoE4 complexes precipitating more rapidly than apoE3 complexes. A beta(1-28) and A beta(1-40) peptides were incubated in the presence or absence of apoE3, apoE4, or bovine serum albumin for 4 d at 37 degrees C (pH 7.3). Negative stain electron microscopy revealed that the A beta peptide alone self-assembled into twisted ribbons containing two or three strands but occasionally into multistranded sheets. The apoE/A beta coincubates yielded monofibrils 7 nm in diameter. ApoE4/A beta coincubates yielded a denser matrix of monofibrils than apoE3/A beta coincubates. Unlike purely monofibrillar apoE4/A beta coincubates, apoE3/A beta coincubates also contained double- and triple-stranded structures. Both apoE isoforms were shown by immunogold labeling to be uniformly distributed along the A beta peptide monofibrils. Monofibrils appeared earlier in apoE4/A beta than in apoE3/A beta in time-course experiments. Thus apoE3 and apoE4 each interact with beta amyloid peptide to form novel monofibrillar structures, apoE4 more avidly, a finding consistent with the biochemical and genetic association between apoE4 and Alzheimer's disease.

Interaction between the alpha-T catenin gene (VR22) and APOE in Alzheimer's disease.

BACKGROUND: APOE is the only gene that has been consistently replicated as a risk factor for late onset Alzheimer's disease. Several recent studies have identified linkage to chromosome 10 for both risk and age of onset, suggesting that this region harbours genes that influence the development of the disease. A recent study reported association between single nucleotide polymorphisms (SNPs) in the VR22 gene (CTNNA3) on chromosome 10 and plasma levels of Abeta42, an endophenotype related to Alzheimer's disease. OBJECTIVE: To assess whether polymorphisms in the VR22 gene are associated with Alzheimer's disease in a large sample of Alzheimer's disease families and an independent set of unrelated cases and controls. RESULTS: Several SNPs showed association in either the family based or case-control analyses (p<0.05). The most consistent findings were with SNP6, for which there was significant evidence of association in both the families and the unrelated cases and controls. Furthermore, there was evidence of significant interaction between APOE-4 and two of the VR22 SNPs, with the strongest evidence of association being concentrated in individuals carrying APOE-4. CONCLUSIONS: This study suggests that VR22 or a nearby gene influences susceptibility to Alzheimer's disease, and the effect is dependent on APOE status.

Apolipoprotein E is localized to the cytoplasm of human cortical neurons: a light and electron microscopic study.

To clarify the localization of the glial protein apolipoprotein E (apoE) in human cortical neurons, we employed specific immunoelectron microscopy using a monoclonal antibody to human apoE in surgical specimens of temporal lobe. The specimens were rapidly fixed after excision from five patients undergoing surgery for medically intractable seizures, and postmortem material was also taken from one Alzheimer's disease patient for comparison. Strong apoE immunoreactivity was observed in many astrocytes filling the perinuclear cytoplasm and distal processes completely. Some cortical neurons were also apoE-immunoreactive. ApoE immunoreactivity of neurons was less intense than glial cells and was distributed in a punctate fashion confined to the region of the cell body and proximal dendrites, but not distal processes. These findings suggest that apoE, which is presumably synthesized and stored by astrocytes, may be taken up by cortical neurons in younger adult humans. The presence of apoE in some human neurons may allow apoE to affect neuronal metabolism. Isoform-specific interactions with microtubule-associated proteins, such as tau or MAP2C, could influence the rate of pathology in neurodegenerative diseases such as Alzheimer's disease.

Intraneuronal ApoE in human visual cortical areas reflects the staging of Alzheimer disease pathology.

Alzheimer disease (AD) is marked by progressive loss of cortical neurons with associated cognitive decline. Multiple genetic and environmental factors likely contribute to this progressive loss. Such genetic factors include the polymorphic locus (APOE) that encodes apolipoprotein E (apoE). In order to investigate a possible correspondence between cellular localization of apoE and the neuropathology of AD, we examined the distribution of apoE-immunoreactive neurons in visual cortical areas with different apparent susceptibility to AD neuropathology (areas 17-primary sensory, 18-secondary sensory, and inferior temporal-association cortex) at different stages of AD pathology as described by Braak and Braak. We found that intraneuronal apoE was present at all these stages, however, only in visual cortical regions known to be vulnerable to AD. In the late stages, the laminar distribution of apoE-immunoreactivity matched the distribution of other markers of AD pathology, especially modified tau. These data support previous findings that intraneuronal apoE in neocortex is common in aged, nondemented controls and demonstrate that it may be more common in regions at risk for AD pathology. Thus, intraneuronal accumulation of apoE may be an attribute of cortical neurons that are more vulnerable to age-related injury with the presence of apoE antedating the classical indices of late-onset AD pathology.

Neuropathological features of frontotemporal dementia and parkinsonism linked to chromosome 17q21-22 (FTDP-17): Duke Family 1684.

Frontotemporal dementia with parkinsonism (FTDP-17) is an autosomal dominant disorder that presents clinically with dementia, extrapyramidal signs, and behavioral disturbances in mid-life and progresses to death within 5 to 10 years. Pathologically, the disorder is characterized by variable neuronal loss and gliosis in the frontal and temporal lobes, limbic structures, and the midbrain. Autopsied individuals from some kindreds display abundant neurofibrillary change while others, including a single affected individual from Duke Family 1684, lack distinctive histological features and exhibit only mild neuronal loss and gliosis in limbic structures and subcortical nuclei when examined by routine silver stain. Recently, mutations in the microtubule associated protein tau have been shown to segregate with the disease in this family and in many other affected kindreds. In order to examine the distribution of tau deposits, we performed tau immunohistochemistry, immunoblotting, and immunoelectron microscopy of tau-containing filaments. Immunohistochemistry revealed numerous tau deposits within glial cells and within neurons. Twisted ribbon-like filaments observed by immunoelectron microscopy were immunodecorated with tau AT8 antibody. Sarkosyl-insoluble tau extracted from the hippocampus and cortex migrated as 2 major bands at 64 and 68 kilodaltons and a minor band at 72 kilodaltons, which after alkaline phosphatase treatment appeared to contain mainly tau isoforms with 4 repeats. Furthermore, the ratio of soluble tau with 4 to 3 microtubule-binding repeats was increased. The role of tau mutations in this disorder is discussed in this paper.

The existence and neurobiological significance of neuronal and glial forms of the glycolytic enzyme enolase.

The isoenzymes of the glycolytic enzyme enolase have been separated and purified. The structural and functional properties of two brain enolases are described. Immunocytochemical techniques have established that one brain enolase is restricted to neuronal cells (neuron-specific enolase, NSE) while the other is localized in glial cells (nonneuronal enolase, NNE). The brain enolases, therefore, represent the first example of functional markers for neuronal and glial cell types in brain. The two enzymes are structurally distinct with the evidence establishing that they are products of separate genes. Functionally, the neuronal enolase has been demonstrated to be uniquely stable to concentrations of chloride salts that rapidly inactivate the glial enzyme. NSE may therefore represent an adaptation of this enzyme that is specifically suited to the neuronal milieu. A specific radioimmunoassay is described for NNE and NSE with the studies reported indicating that neuronal enzyme levels vary considerably when different brain areas are compared, suggesting a relationship between functional activity and levels of NSE. In addition to being a marker for neuronal cells, NSE has also been found to be present in various glands. The cells of the APUD series (amine precursor uptake and decarboxylation cells) in the pituitary, adrenal medulla, pineal, thyroid, and pancreas have been shown to contain NSE. NSE is, therefore, also a marker for these neuronlike endocrine cells since they are the only cells other than neurons that contain this protein.

Apolipoprotein E isoform-specific differences in outcome from focal ischemia in transgenic mice.

Apolipoprotein E (apoE), a 34-KD glycosylated lipid-binding protein, is expressed as three common isoforms in humans (E2, E3, or E4). Clinical evidence suggests that the apoE genotype (APOE) may be a risk factor for poor outcome after acute central nervous system injury. This was examined further in transgenic mice constructed with the human APOE3 or APOE4 gene under the control of human promoter and tissue expression elements. Presence of human apoE3 and apoE4 proteins in brains of human APOE homozygous transgenic mice was confirmed by Western blotting. APOE3 (n = 12) and APOE4 (n = 10) mice underwent 60 minutes of middle cerebral artery occlusion. After 24-hour recovery, infarct size was measured. Infarct volumes (mean +/- standard deviation) were smaller in the APOE3 group (cortex: APOE3 = 18 +/- 4 mm3; APOE4 = 30 +/- 11 mm3, P = 0.04; subcortex: APOE3 = 12 +/- 4 mm3; APOE4 = 18 +/- 4 mm3, P = 0.003). Hemiparesis was less severe in APOE3 mice (P = 0.02). These data indicate that human isoform-specific effects of apoE are relevant to acute pathomechanisms of focal ischemic brain damage when examined in the mouse. APOE transgenic mice may provide an appropriate model to examine the mechanistic basis for the differential effects of human apoE isoforms in acute central nervous system injury.

Cis-acting human ApoE tissue expression element is associated with human pattern of intraneuronal ApoE in transgenic mice.

Apolipoprotein E polymorphic variants (ApoE-epsilon2, epsilon3, and epsilon4) are associated with the age of onset distribution and risk of Alzheimer disease. The question of whether ApoE is expressed at a comparatively low level in human neurons compared to astrocytes, or whether ApoE enters neuronal cytoplasm via altered endosomal metabolism is important to understanding potential pathogenic roles for ApoE as a susceptibility gene in Alzheimer disease. ApoE deficient ("knock-out") mice received large human genomic DNA fragment transgenes for each of the three common apoE alleles. All transgenic mice demonstrated glial/astrocytic (normal rodent pattern), as well as cortical intraneuronal ApoE immunoreactivity with all three human isoforms and at multiple ApoE human allele doses (Xu et al. (32)). To test whether ApoE intraneuronal immunoreactivity was due to ApoE gene sequences between mouse and human, we examined another set of mice constructed using targeted replacement so that the human ApoE gene was placed under mouse gene promoters. Current analyses show that targeted replacement recombinant mice show normal rodent glial expression pattern, but no ApoE neuronal immunoreactivity through six months of age compared to large human genomic DNA fragment transgenic mice, which show neuronal content of ApoE throughout adult life. We conclude that cis-acting DNA sequences, rather than the specific sequence of the ApoE gene, may be responsible for low levels of transcription activity in cortical neurons.

Distribution of GABAergic neurons and axon terminals in the macaque striate cortex.

Antisera to glutamic acid decarboxylase (GAD) and gamma-aminobutyric acid (GABA) have been used to characterize the morphology and distribution of presumed GABAergic neurons and axon terminals within the macaque striate cortex. Despite some differences in the relative sensitivity of these antisera for detecting cell bodies and terminals, the overall patterns of labeling appear quite similar. GABAergic axon terminals are particularly prominent in zones known to receive the bulk of the projections from the lateral geniculate nucleus; laminae 4C, 4A, and the cytochrome-rich patches of lamina 3. In lamina 4A, GABAergic terminals are distributed in a honeycomb pattern which appears to match closely the spatial pattern of geniculate terminations in this region. Quantitative analysis of axon terminals that contain flat vesicles and form symmetric synaptic contacts (FS terminals) in lamina 4C beta and in lamina 5 suggest that the prominence of GAD and GABA axon terminal labeling in the geniculate recipient zones is due, at least in part, to the presence of larger GABAergic axon terminals in these regions. GABAergic cell bodies and their initial dendritic segments display morphological features characteristic of nonpyramidal neurons and are found in all layers of striate cortex. The density of GAD and GABA immunoreactive neurons is greatest in laminae 2-3A, 4A, and 4C beta. The distribution of GABAergic neurons within lamina 3 does not appear to be correlated with the patchy distribution of cytochrome oxidase in this region; i.e., there is no significant difference in the density of GAD and GABA immunoreactive neurons in cytochrome-rich and cytochrome-poor regions of lamina 3. Counts of labeled and unlabeled neurons indicate that GABA immunoreactive neurons make up at least 15% of the neurons in striate cortex. Layer 1 is distinct from the other cortical layers by virtue of its high percentage (77-81%) of GABAergic neurons. Among the other layers, the proportion of GABAergic neurons varies from roughly 20% in laminae 2-3A to 12% in laminae 5 and 6. Finally, there are conspicuous laminar differences in the size and dendritic arrangement of GAD and GABA immunoreactive neurons. Lamina 4C alpha and lamina 6 are distinguished from the other layers by the presence of populations of large GABAergic neurons, some of which have horizontally spreading dendritic processes. GABAergic neurons within the superficial layers are significantly smaller and the majority appear to have vertically oriented dendritic processes.(ABSTRACT TRUNCATED AT 400 WORDS)

Variability in the terminations of GABAergic chandelier cell axons on initial segments of pyramidal cell axons in the monkey sensory-motor cortex.

Chandelier cell axons were studied in the sensory-motor cortex of adult monkeys. The axonal fields of Golgi-impregnated chandelier cells in layer II in motor cortex are flattened sagittally. The vertical terminal portions of the axons varied both in length and in the numbers converging to form terminations of greater or lesser complexity. Golgi-impregnated plexuses were embedded in plastic and resectioned serially at 2.5-3.0 micrograms. A single axonal field could have as many as 400 terminal rows. All lie 3-13 micrograms beneath pyramidal cell somata. These terminations are not randomly distributed but instead, form clusters. Further resectioning the plastic sections for electron microscopy revealed that all the terminations are on the initial axon segments of pyramidal cells and all form symmetric synaptic contacts. In immunocytochemical material stained for glutamic acid decarboxylase (GAD), the enzyme involved in the synthesis of GABA, GAD-positive boutons were found to form symmetric synaptic contacts with a variety of postsynaptic elements including the axon hillocks and axon initial segments of pyramidal cells. Serial reconstructions from electron micrographs revealed GAD-positive terminals synapsing with the axon initial segment of pyramidal cells joined by cytoplasmic bridges and forming vertically oriented rows identical to those of chandelier cell terminals identified positively in the resectioned Golgi material. The GAD-positive terminals forming initial segment synapses were never continuous with GAD-positive terminals forming axo hillock synapses. The latter probably arise from basket cell axons. Initial segments of pyramidal cell axons in layers II and III were contacted by more GAD-positive terminals than the initial segments of pyramidal cell axons in layer V. The largest pyramidal cells in layer III received the most synapses. Many larger pyramidal cells, identified as callosally projecting cells by the retrograde transport of horseradish peroxidase (HRP), were shown in serial electron micrographs to possess large numbers of initial segment synapses, comparable to those seen in the immunocytochemical material. Serial reconstructions of pyramidal cell axons from axon hillock to the first myelin internode in resectioned Golgi, immunocytochemical and HRP material showed that the number of synapses varied from 2 to 52 for layers II and III and from 2 to 26 for layer V. The number of synapses on the axon hillocks varied from zero to 12. The variability in these terminations may be an important factor in the shaping of the functional properties of the pyramidal cells.

The distribution of glutamic acid decarboxylase immunoreactivity in the diencephalon of the opossum and rabbit.

We have examined the distribution of neurons and terminals immunoreactive for glutamic acid decarboxylase (GAD) in the thalamus and adjacent structures of the opossum (Didelphis virginiana) and the rabbit and have compared this distribution with the distributions we described previously for the cat and bushbaby (Galago senegalensis). The significance of these experiments depends, first, on the fact that GAD is the synthetic enzyme for GABA, and therefore that GAD immunoreactivity is a marker for GABAergic inhibitory neurons, and second, on previous findings that suggest that GABAergic neurons in the dorsal thalamus are local circuit neurons. In both cat and Galago, GAD-immunoreactive neurons are distributed essentially throughout the entire thalamus. In the opossum, GAD neurons are chiefly confined to the dorsal lateral geniculate nucleus and the lateral extremity of the lateral posterior nucleus. The distribution of GAD neurons in the rabbit is intermediate between that found in the opossum on the one hand and cat and Galago on the other. Like opossum, about 25% of the neurons in the lateral geniculate nucleus of rabbit are GAD immunoreactive. Unlike opossum, however, as many as 18% of the cells in the ventral posterior nucleus of the rabbit are GAD immunoreactive, and scattered cells are also labeled in other thalamic areas, such as the medial geniculate and the lateral group. Aside from the findings in the dorsal thalamus, the chief observation is that GAD-immunoreactive neurons and/or terminals densely fill all principal targets of the optic tract, including the ventral lateral geniculate nucleus; the superficial gray layer of the superior colliculus; the anterior, posterior, and olivary pretectal nuclei; the nucleus of the optic tract; and the medial and lateral terminal nuclei of the accessory optic tract. These results support the idea first put forward by Cajal that local circuit neurons increase in number during the course of the evolution of complex mammalian brains. If we can assume that the conservative opossum retains characteristics reflecting an early stage of mammalian evolution, the results suggest that thalamic local circuit neurons arose first in the visual system and only later in evolution spread throughout the thalamus.

Glutamic acid decarboxylase and somatostatin immunoreactivities in rat visual cortex.

Antibodies to glutamic acid decarboxylase (GAD) and somatostatin (SS) were used to determine the laminar distribution and morphology of GAD- and SS-immunoreactive neurons and terminals in rat visual cortex. The present study demonstrates that GAD-immunoreactive neurons constitute several morphologically distinct subclasses of neurons in rat visual cortex. These subclasses of neurons can be distinguished by differences in soma size, soma shape, dendritic branching patterns, axonal arborizations, and location in the neuropil. GAD-immunoreactive neurons are found throughout all layers of visual cortex. They have nonpyramidal morphology and constitute roughly 15% of the total neuronal population. The laminar pattern of GAD-immunoreactive puncta is uneven, with a prominent band of terminals in layer IV. Numerous large GAD-positive puncta surround the somata and proximal dendrites of pyramidal cells in layers II, III, and V. SS-immunoreactive neurons constitute a less numerous and more restricted population of nonpyramidal neurons. Their somata are located mainly in layers II, III, V, and VI. Very few, if any, SS-immunoreactive neurons are found in layers I and IV. SS-immunoreactive terminals are arranged along vertical and diagonal collateral branches that have a beaded appearance. Finally, many neurons in the supra- and infragranular layers and in the white matter are immunoreactive to both glutamic acid decarboxylase and somatostatin. This coexistence of immunoreactivity to both GAD and SS may characterize a broad subclass of cortical nonpyramidal neurons.

Apolipoprotein E, survival in Alzheimer's disease patients, and the competing risks of death and Alzheimer's disease.

The apolipoprotein E (APOE) epsilon 4 allele carries an increased risk of a patient developing Alzheimer's disease (AD) while the epsilon 2 allele carries a decreased risk. We compared survival from the onset of AD in subjects with different numbers of epsilon 4 alleles and evaluated changes in genotypic frequencies with age. Two subject groups were investigated: unrelated AD case and control subjects, and affected and unaffected members from 74 multiplex AD families. In both subject groups, survival from onset decreased with increasing onset age, was longer in women, and was unrelated to epsilon 4 gene dose. The epsilon 2/epsilon 3 genotype became more common with age (p = 0.004). The epsilon 4 allele decreased in frequency with age in all patient groups but, unexpectedly, remained unchanged in control subjects. We conclude that the progression of AD is not strongly related to epsilon 4 gene dose, that the higher prevalence of AD in women may involve the longer survival of affected women, and that AD and death are competing risks involving APOE that change over time.

Association of apolipoprotein E allele epsilon 4 with late-onset familial and sporadic Alzheimer's disease.

Apolipoprotein E, type epsilon 4 allele (APOE epsilon 4), is associated with late-onset familial Alzheimer's disease (AD). There is high avidity and specific binding of amyloid beta-peptide with the protein ApoE. To test the hypothesis that late-onset familial AD may represent the clustering of sporadic AD in families large enough to be studied, we extended the analyses of APOE alleles to several series of sporadic AD patients. APOE epsilon 4 is significantly associated with a series of probable sporadic AD patients (0.36 +/- 0.042, AD, versus 0.16 +/- 0.027, controls [allele frequency estimate +/- standard error], p = 0.00031). Spouse controls did not differ from CEPH grandparent controls from the Centre d'Etude du Polymorphisme Humain (CEPH) or from literature controls. A large combined series of autopsy-documented sporadic AD patients also demonstrated highly significant association with the APOE epsilon 4 allele (0.40 +/- 0.026, p < or = 0.00001). These data support the involvement of ApoE epsilon 4 in the pathogenesis of late-onset familial and sporadic AD. ApoE isoforms may play an important role in the metabolism of beta-peptide, and APOE epsilon 4 may operate as a susceptibility gene (risk factor) for the clinical expression of AD.