Globular amyloid beta-peptide oligomer - a homogenous and stable neuropathological protein in Alzheimer's disease

Amyloid beta-peptide (Abeta)(1-42) oligomers have recently been discussed as intermediate toxic species in Alzheimer's disease (AD) pathology. Here we describe a new and highly stable Abeta(1-42) oligomer species which can easily be prepared in vitro and is present in the brains of patients with AD and Abeta(1-42)-overproducing transgenic mice. Physicochemical characterization reveals a pure, highly water-soluble globular 60-kDa oligomer which we named 'Abeta(1-42) globulomer'. Our data indicate that Abeta(1-42) globulomer is a persistent structural entity formed independently of the fibrillar aggregation pathway. It is a potent antigen in mice and rabbits eliciting generation of Abeta(1-42) globulomer-specific antibodies that do not cross-react with amyloid precursor protein, Abeta(1-40) and Abeta(1-42) monomers and Abeta fibrils. Abeta(1-42) globulomer binds specifically to dendritic processes of neurons but not glia in hippocampal cell cultures and completely blocks long-term potentiation in rat hippocampal slices. Our data suggest that Abeta(1-42) globulomer represents a basic pathogenic structural principle also present to a minor extent in previously described oligomer preparations and that its formation is an early pathological event in AD. Selective neutralization of the Abeta globulomer structure epitope is expected to have a high potential for treatment of AD.     

Abeta40 protects non-toxic Abeta42 monomer from aggregation

Abeta40 and Abeta42 are the predominant Abeta species in the human body. Toxic Abeta42 oligomers and fibrils are believed to play a key role in causing Alzheimer's disease (AD). However, the role of Abeta40 in AD pathogenesis is not well established. Emerging evidence indicates a protective role for Abeta40 in AD pathogenesis. Although Abeta40 is known to inhibit Abeta42 fibril formation, it is not clear whether the inhibition acts on the non-toxic monomer or acts on the toxic Abeta42 oligomers. In contrast to conventional methods that detect the appearance of fibrils, in our study Abeta42 aggregation was monitored by the decreasing NMR signals from Abeta42 monomers. In addition, differential NMR isotope labelling enabled the selective observation of Abeta42 aggregation in a mixture of Abeta42 and Abeta40. We found Abeta40 monomers inhibit the aggregation of non-toxic Abeta42 monomers, in an Abeta42/Abeta40 ratio-dependent manner. NMR titration revealed that Abeta40 monomers bind to Abeta42 aggregates with higher affinity than Abeta42 monomers. Abeta40 can also release Abeta42 monomers from Abeta42 aggregates. Thus, Abeta40 likely protects Abeta42 monomers by competing for the binding sites on pre-existing Abeta42 aggregates. Combining our data with growing evidence from transgenic mice and human genetics, we propose that Abeta40 plays a critical, protective role in Alzheimer's by inhibiting the aggregation of Abeta42 monomer. Abeta40 itself, a peptide already present in the human body, may therefore be useful for AD prevention and therapy.     

Conformation-sensitive antibodies against alzheimer amyloid-beta by immunization with a thioredoxin-constrained B-cell epitope peptide

Immunotherapy against the amyloid-beta (Abeta) peptide is a valuable potential treatment for Alzheimer disease (AD). An ideal antigen should be soluble and nontoxic, avoid the C-terminally located T-cell epitope of Abeta, and yet be capable of eliciting antibodies that recognize Abeta fibrils and neurotoxic Abeta oligomers but not the physiological monomeric species of Abeta. We have described here the construction and immunological characterization of a recombinant antigen with these features obtained by tandem multimerization of the immunodominant B-cell epitope peptide Abeta1-15 (Abeta15) within the active site loop of bacterial thioredoxin (Trx). Chimeric Trx(Abeta15)n polypeptides bearing one, four, or eight copies of Abeta15 were constructed and injected into mice in combination with alum, an adjuvant approved for human use. All three polypeptides were found to be immunogenic, yet eliciting antibodies with distinct recognition specificities. The anti-Trx(Abeta15)4 antibody, in particular, recognized Abeta42 fibrils and oligomers but not monomers and exhibited the same kind of conformational selectivity against transthyretin, an amyloidogenic protein unrelated in sequence to Abeta. We have also demonstrated that anti-Trx(Abeta15)4, which binds to human AD plaques, markedly reduces Abeta pathology in transgenic AD mice. The data indicate that a conformational epitope shared by oligomers and fibrils can be mimicked by a thioredoxin-constrained Abeta fragment repeat and identify Trx(Abeta15)4 as a promising new tool for AD immunotherapy.     

Amyloid beta protein immunotherapy neutralizes Abeta oligomers that disrupt synaptic plasticity in vivo

One of the most clinically advanced forms of experimental disease-modifying treatment for Alzheimer disease is immunization against the amyloid beta protein (Abeta), but how this may prevent cognitive impairment is unclear. We hypothesized that antibodies to Abeta could exert a beneficial action by directly neutralizing potentially synaptotoxic soluble Abeta species in the brain. Intracerebroventricular injection of naturally secreted human Abeta inhibited long-term potentiation (LTP), a correlate of learning and memory, in rat hippocampus in vivo but a monoclonal antibody to Abeta completely prevented the inhibition of LTP when injected after Abeta. Size fractionation showed that Abeta oligomers, not monomers or fibrils, were responsible for inhibiting LTP, and an Abeta antibody again prevented such inhibition. Active immunization against Abeta was partially effective, and the effects correlated positively with levels of antibodies to Abeta oligomers. The ability of exogenous and endogenous antibodies to rapidly neutralize soluble Abeta oligomers that disrupt synaptic plasticity in vivo suggests that treatment with such antibodies might show reversible cognitive deficits in early Alzheimer disease.     

High-efficiency transfection of human endothelial cells mediated by cationic lipids

Amyloid beta peptide is recognized as the main constituent of the extracellular amyloid plaques, the major neuropathological hallmark of Alzheimer's disease. Abeta is a small peptide constitutively expressed in normal cells, not toxic in the monomeric form but aggregated Abeta is assumed to be the main if not the only factor causing Alzheimer's disease. Interestingly, the new reports suggest neurotoxicity of soluble Abeta oligomers rather than amyloid fibrils. Because of the fact that fibrils were thought to be the main toxic species in AD, early structural studies focused on fibrils themselves and Abeta monomers, as their building blocks while there is practically no data on oligomer structure and mechanism of neurotoxicity. Using a model peptide spanning residues 10–30 of Abeta, obtained by overexpression in bacteria, we have employed mass spectrometry of noncovalent complexes and disulfide rearrangement assay to gain new insight into structure and dynamics of a prenucleation step of Abeta peptide oligomerisation.     

Human plasma contains cross-reactive Abeta conformer-specific IgG antibodies

Two conformers of aggregated Abeta, i.e., fibrils and oligomers, have been deemed important in the pathogenesis of Alzheimer's disease. We now report that intravenous immune globulin (IVIG) derived from pools of human plasma contains IgGs that recognize conformational epitopes present on fibrils and oligomers, but not their soluble monomeric precursor. We have used affinity chromatography to isolate these antibodies and have shown that they cross-reacted with comparable nanomolar avidity with both types of Abeta aggregates; notably, binding was not inhibited by soluble Abeta monomers. Our studies provide further support for investigating the therapeutic use of IVIG in Alzheimer's disease.     

Amyloid-beta(1-42) rapidly forms protofibrils and oligomers by distinct pathways in low concentrations of sodium dodecylsulfate

Alzheimer's disease (AD) is characterized by large numbers of senile plaques in the brain that consist of fibrillar aggregates of 40- and 42-residue amyloid-beta (Abeta) peptides. However, the degree of dementia in AD correlates better with the concentration of soluble Abeta species assayed biochemically than with histologically determined plaque counts, and several investigators now propose that soluble aggregates of Abeta are the neurotoxic agents that cause memory deficits and neuronal loss. These endogenous aggregates are minor components in brain extracts from AD patients and transgenic mice that express human Abeta, but several species have been detected by gel electrophoresis in sodium dodecylsulfate (SDS) and isolated by size exclusion chromatography (SEC). Endogenous Abeta aggregation is stimulated at cellular interfaces rich in lipid rafts, and anionic micelles that promote Abeta aggregation in vitro may be good models of these interfaces. We previously found that micelles formed in dilute SDS (2 mM) promote Abeta(1-40) fiber formation by supporting peptide interaction on the surface of a single micelle complex. In contrast, here we report that monomeric Abeta(1-42) undergoes an immediate conversion to a predominant beta-structured conformation in 2 mM SDS which does not proceed to amyloid fibrils. The conformational change is instead rapidly followed by the near quantitative conversion of the 4 kDa monomer SDS gel band to 8-14 kDa bands consistent with dimers through tetramers. Removal of SDS by dialysis gave a shift in the predominant SDS gel bands to 30-60 kDa. While these oligomers resemble the endogenous aggregates, they are less stable. In particular, they do not elute as discrete species on SEC, and they are completed disaggregated by boiling in 1% SDS. It appears that endogenous oligomeric Abeta aggregates are stabilized by undefined processes that have not yet been incorporated into in vitro Abeta aggregation procedures.     

Abeta42 is more rigid than Abeta40 at the C terminus: implications for Abeta aggregation and toxicity

Abeta40 and Abeta42 are the major forms of amyloid beta peptides (Abeta) in the brain. Although Abeta42 differs from Abeta40 by only two residues, Abeta42 is much more prone to aggregation and more toxic to neurons than Abeta40. To probe whether dynamics contribute to such dramatic difference in function, backbone ps-ns dynamics of native Abeta monomers were characterized by 15N spin relaxation at 273.3 K and 800 MHz. Abeta42 aggregates much faster than Abeta40 in the NMR tube. The effect of Abeta aggregation was removed from the relaxation measurement by interleaved data collection. R1, R2 and nuclear Overhauser enhancement (NOE) values are similar in Abeta40 and Abeta42, except at the C terminus, indicating Abeta42 and Abeta40 monomers have identical global motions. Comparisons of the spectral density function J(0.87omegaH) and order parameters (S2) indicate that the Abeta42 C terminus is more rigid than the Abeta40 C terminus. At 280.4 K and 287.6 K, the Abeta42 C terminus remains more rigid than the Abeta40 C terminus, suggesting such a dynamical difference is likely present at the physiological temperature. The Abeta42 monomer likely has less configurational entropy due to restricted motion in the C terminus and may pay a smaller entropic price to form fibrils than the Abeta40 monomer. We hypothesize that the entropic difference between Abeta40 and Abeta42 monomers might partly account for the fact that Abeta42 is the major Abeta species in parenchymal senile plaques in most Alzheimer's diseased brains in spite of the predominance of Abeta40 in plasma. The increased rigidity of the Abeta42 C terminus is likely due to its pre-ordering for beta-conformation present in soluble oligomers and fibrils. The Abeta42 C terminus may therefore serve as an internal seed for aggregation.     

Induction of NADPH cytochrome P450 reductase by the Alzheimer β-protein. Amyloid as a 'foreign body'

A large body of data suggests that the Alzheimer's amyloid peptide (Abeta) causes degeneration and death of neurons by mechanisms that involve reactive oxygen species. The pathways involved in Abeta-mediated oxidative injury are only partially understood. We theorized that abnormal microaggregates and/or pathological conformations of Abeta peptides may behave as xenobiotics and trigger the induction of NADPH cytochrome P450 reductase (CP450r), an enzyme which, if induced by non-physiological substrates (such as xenobiotics like drugs or other 'foreign molecules'), is known to cause oxidative stress. In order to test this hypothesis, i.e. that Abeta can increase the expression of CP450r, SK-N-SH human neuroblastoma cells were exposed to Abeta25-35 and Abeta1-42 and then examined for induction of this enzyme in immunoblots, using specific antibodies. Following exposure to Abeta peptides, neuroblastoma cells showed a clear-cut induction of CP450r. To determine whether this mechanism is operational in vivo, we investigated the expression of CP450r in a transgenic mouse model of Alzheimer's disease (AD) and in brains from patients afflicted with AD, using an immunocytochemical approach. Tissue sections from brains of transgenic mice exhibited strong immunoreactivity for CP450r, surrounding amyloid deposits. The pattern of expression of CP450r was similar to that exhibited by neuritic and oxidative stress markers. Sections from non-transgenic mice showed no detectable immunoreactivity. Immunostaining of sections from four brains with neuropathologically confirmed AD showed a pattern of abnormality different from transgenic mice that was characterized by abnormal immunoreactivity for CP450r within the cytoplasm of cortical neurons. No labeling was seen in sections from aged-matched control brains. The data showed that CP450r is induced by Alzheimer amyloid peptide and that such a response must be considered as one possible mechanism whereby Abeta causes oxidative stress.     

I32E and V36K double mutation in beta2-sheet abrogates amyloid beta peptide toxicity

Alzheimer's disease (AD) is the most common cause of dementia in the elderly, wherein, the accumulation of amyloid beta (Abeta) peptide as cytotoxic oligomers leads to neuropathologic changes. Transgenic mice with brain Abeta plaques immunized with aggregated Abeta have reduced amyloid burden and improved cognitive functions. However, such active immunization in humans led to a small but significant occurrence of meningoencephalitis in 6% AD volunteers due to Abeta induced toxicity. In an attempt to develop safer alternative vaccines, the design of a highly soluble peptide homologous to Abeta (Abeta-EK), that has a reduced amyloidogenic potential while maintaining the major immunogenic epitopes of Abeta is reported. More importantly, this homologue has been shown to be non-toxic, as this peptide failed to exert any observable effect on erythrocytes. The results of the present study suggests that immunization with non-toxic Abeta derivative may offer a safer therapeutic approach to AD, instead of using toxic Abeta fibrils.     

Compensatory function of transthyretin in Alzheimer's disease

Alzheimer's disease (AD) is the most common form of age-related primary neurodegenerative diseases characterized by progressive memory loss, aphasia, and intellectual and mental breakdown. Pathogenesis of AD is based on the early synaptic dysfunction following neurodegeneration and neuronal death. According to modern concepts, the development of neuropathological processes is due to progressively deposited intermediates of amyloid fibrils that represent oligomers consisting of short peptide named amyloid beta protein (Abeta). In this context, it is reasonable to propose that one of the compensatory mechanisms of AD might be inhibition of Abeta oligomerization by sequestration or clearance of Abeta. Experiments with transgenic animals and epidemiological studies demonstrate that major protein of cerebrospinal fluid, transthyretin, is a natural neuroprotector that inhibits Abeta amyloid formation and restore cognitive functions. The study of Abeta-transthyretin complexes allowed to create peptides that are mimetics of transthyretin. These mimetics inhibit amyloid formation in vitro and, therefore, could be used in therapeutic treatment of AD.     

Immunotherapeutic strategies for prevention and treatment of Alzheimer's disease.

Pathologic examination in Alzheimer's disease (AD) shows a significant correlation between beta-amyloid peptide (AbetaP) deposition and the clinical severity of dementia. Formation of beta-amyloid (Abeta) is a complex kinetic and thermodynamic process, dependent on peptide-peptide interactions that may be modulated by other proteins. We found that site-directed antibodies toward peptide EFRH sequences 3-6 of the N-terminal region of AbetaP suppress in vitro formation of Abeta and dissolve already-formed fibrillar amyloid. These so-called chaperone-like properties of monoclonal antibodies led to the development of a new immunologic approach to AD treatment. The immunization procedure, based on phages displaying the EFRH epitope as antigen, induced anti-AbetaP antibodies that recognized the whole AbetaP and exhibited antiaggregating properties similar to those of antibodies obtained by injection of Abeta fibrils. Production and performance of anti-beta-amyloid antibodies in the transgenic mouse model of AD showed that these antibodies may be delivered from the periphery to the central nervous system, preventing the formation of Abeta and dissolving already-present aggregates. Moreover, immunization with Abeta protected transgenic mice from the learning and age-related memory deficits that occur in AD. These data support the hypotheses that Abeta plays a central role in AD and that site-directed antibodies that modulate Abeta conformation may provide immunotherapy of the disease.     

Mitochondrial Ca2+ overload underlies Abeta oligomers neurotoxicity providing an unexpected mechanism of neuroprotection by NSAIDs

Dysregulation of intracellular Ca(2+) homeostasis may underlie amyloid beta peptide (Abeta) toxicity in Alzheimer's Disease (AD) but the mechanism is unknown. In search for this mechanism we found that Abeta(1-42) oligomers, the assembly state correlating best with cognitive decline in AD, but not Abeta fibrils, induce a massive entry of Ca(2+) in neurons and promote mitochondrial Ca(2+) overload as shown by bioluminescence imaging of targeted aequorin in individual neurons. Abeta oligomers induce also mitochondrial permeability transition, cytochrome c release, apoptosis and cell death. Mitochondrial depolarization prevents mitochondrial Ca(2+) overload, cytochrome c release and cell death. In addition, we found that a series of non-steroidal anti-inflammatory drugs (NSAIDs) including salicylate, sulindac sulfide, indomethacin, ibuprofen and R-flurbiprofen depolarize mitochondria and inhibit mitochondrial Ca(2+) overload, cytochrome c release and cell death induced by Abeta oligomers. Our results indicate that i) mitochondrial Ca(2+) overload underlies the neurotoxicity induced by Abeta oligomers and ii) inhibition of mitochondrial Ca(2+) overload provides a novel mechanism of neuroprotection by NSAIDs against Abeta oligomers and AD.     

Surface structure of amyloid-beta fibrils contributes to cytotoxicity

Amyloid beta (Abeta) toxicity has been hypothesized to initiate the pathogenesis of Alzheimer's disease (AD). The characteristic fibrillar morphology of Abeta-aggregates, that constitute the main components of senile plaque, has long been considered to account for the neurotoxicity. But recent reports argue against a primary role for mature fibrils in AD pathogenesis because of the lack of a robust correlation between the severity of neurological impairment and the extent of amyloid deposition. Toxicity from the soluble prefibrillar intermediate entity of aggregates often called oligomer has recently proposed a plausible explanation for this inconsistency. An alternative explanation is based on the observation that certain amyloid fibril morphologies are more toxic than others, indicating that not all amyloid fibrils are equally toxic. Here, we report that it is not only the beta-sheeted fibrillar structure but also the surface physicochemical composition that affects the toxicity of Abeta fibrils. For the first time, colloidal gold was used to visualize by electron microscopy positive-charge clusters on Abeta fibrils. Chemical modifications as well as point-mutated Abeta synthesis techniques were applied to change the surface structures of Abeta and to show how local structure affects surface properties that are responsible for electrostatic and hydrophobic interactions with cells. We also report that covering the surface of Abeta fibers with myelin basic protein, which has surface properties contrary to those of Abeta, suppresses Abeta toxicity. On the basis of these results, we propose that the surface structure of Abeta fibrils plays an important role in Abeta toxicity.     

Abeta42 is essential for parenchymal and vascular amyloid deposition in mice

Considerable circumstantial evidence suggests that Abeta42 is the initiating molecule in Alzheimer's disease (AD) pathogenesis. However, the absolute requirement for Abeta42 for amyloid deposition has never been demonstrated in vivo. We have addressed this by developing transgenic models that express Abeta1-40 or Abeta1-42 in the absence of human amyloid beta protein precursor (APP) overexpression. Mice expressing high levels of Abeta1-40 do not develop overt amyloid pathology. In contrast, mice expressing lower levels of Abeta1-42 accumulate insoluble Abeta1-42 and develop compact amyloid plaques, congophilic amyloid angiopathy (CAA), and diffuse Abeta deposits. When mice expressing Abeta1-42 are crossed with mutant APP (Tg2576) mice, there is also a massive increase in amyloid deposition. These data establish that Abeta1-42 is essential for amyloid deposition in the parenchyma and also in vessels.     

Monoclonal antibodies that target pathological assemblies of Abeta

Amyloid beta (Abeta) immunotherapy for Alzheimer's disease has shown initial success in mouse models of Alzheimer's disease and in human patients. However, because of meningoencephalitis in clinical trials of active vaccination, approaches using therapeutic antibodies may be preferred. As a novel antigen to generate monoclonal antibodies, the current study has used Abeta oligomers (amyloid beta-derived diffusible ligands, ADDLs), pathological assemblies known to accumulate in Alzheimer's disease brain. Clones were selected for the ability to discriminate Alzheimer's disease from control brains in extracts and tissue sections. These antibodies recognized Abeta oligomers and fibrils but not the physiologically prevalent Abeta monomer. Discrimination derived from an epitope found in assemblies of Abeta1-28 and ADDLs but not in other sequences, including Abeta1-40. Immunoneutralization experiments showed that toxicity and attachment of ADDLs to synapses in culture could be prevented. ADDL-induced reactive oxygen species (ROS) generation was also inhibited, establishing this response to be oligomer-dependent. Inhibition occurred whether ADDLs were prepared in vitro or obtained from Alzheimer's disease brain. As conformationally sensitive monoclonal antibodies that selectively immunoneutralize binding and function of pathological Abeta assemblies, these antibodies provide tools by which pathological Abeta assemblies from Alzheimer's disease brain might be isolated and evaluated, as well as offering a valuable prototype for new antibodies useful for Alzheimer's disease therapeutics.     

Oligomerization and toxicity of beta-amyloid-42 implicated in Alzheimer's disease

beta-Amyloid protein (Abeta) is the major component of senile plaques found in the brains of Alzheimer's patients. A novel ELISA has been developed which probes the early stages of oligomerization of Abeta. Incubation of Abeta solutions at 37 degrees C and pH 7.4 produces soluble oligomers in a concentration-dependent manner. Fresh Abeta42 solutions rapidly form soluble oligomers, whereas Abeta40 solutions require prolonged incubation to produce oligomers. Fresh Abeta42 solutions are more toxic to human neuroblastoma SH-SY5Y cells than Abeta40 solutions, possibly mediated by soluble oligomers. The differences between Abeta42 and Abeta40 could explain the association of the longer form with familial early-onset Alzheimer's disease. We also report a new strategy for solid-phase synthesis of Abeta peptides which gives high yield and purity of the initial crude preparation.     

Characterization of amyloid beta peptides from brain extracts of transgenic mice overexpressing the London mutant of human amyloid precursor protein

Alzheimer's disease (AD) is marked by the presence of neurofibrillary tangles and amyloid plaques in the brain of patients. To study plaque formation, we report on further quantitative and qualitative analysis of human and mouse amyloid beta peptides (Abeta) from brain extracts of transgenic mice overexpressing the London mutant of human amyloid precursor protein (APP). Using enzyme-linked immunosorbant assays (ELISAs) specific for either human or rodent Abeta, we found that the peptides from both species aggregated to form plaques. The ratios of deposited Abeta1-42/1-40 were in the order of 2-3 for human and 8-9 for mouse peptides, indicating preferential deposition of Abeta42. We also determined the identity and relative levels of other Abeta variants present in protein extracts from soluble and insoluble brain fractions. This was done by combined immunoprecipitation and mass spectrometry (IP/MS). The most prominent peptides truncated either at the carboxyl- or the amino-terminus were Abeta1-38 and Abeta11-42, respectively, and the latter was strongly enriched in the extracts of deposited peptides. Taken together, our data indicate that plaques of APP-London transgenic mice consist of aggregates of multiple human and mouse Abeta variants, and the human variants that we identified were previously detected in brain extracts of AD patients.     

Lipids revert inert Abeta amyloid fibrils to neurotoxic protofibrils that affect learning in mice

Although soluble oligomeric and protofibrillar assemblies of Abeta-amyloid peptide cause synaptotoxicity and potentially contribute to Alzheimer's disease (AD), the role of mature Abeta-fibrils in the amyloid plaques remains controversial. A widely held view in the field suggests that the fibrillization reaction proceeds 'forward' in a near-irreversible manner from the monomeric Abeta peptide through toxic protofibrillar intermediates, which subsequently mature into biologically inert amyloid fibrils that are found in plaques. Here, we show that natural lipids destabilize and rapidly resolubilize mature Abeta amyloid fibers. Interestingly, the equilibrium is not reversed toward monomeric Abeta but rather toward soluble amyloid protofibrils. We characterized these 'backward' Abeta protofibrils generated from mature Abeta fibers and compared them with previously identified 'forward' Abeta protofibrils obtained from the aggregation of fresh Abeta monomers. We find that backward protofibrils are biochemically and biophysically very similar to forward protofibrils: they consist of a wide range of molecular masses, are toxic to primary neurons and cause memory impairment and tau phosphorylation in mouse. In addition, they diffuse rapidly through the brain into areas relevant to AD. Our findings imply that amyloid plaques are potentially major sources of soluble toxic Abeta-aggregates that could readily be activated by exposure to biological lipids.     

Self-assembly of Abeta(1-42) into globular neurotoxins

Amyloid beta 1-42 (Abeta(1-42)) is a self-associating peptide that becomes neurotoxic upon aggregation. Toxicity originally was attributed to the presence of large, readily formed Abeta fibrils, but a variety of other toxic species are now known. The current study shows that Abeta(1-42) can self-assemble into small, stable globular assemblies free of fibrils and protofibrils. Absence of large molecules was verified by atomic force microscopy (AFM) and nondenaturing gel electrophoresis. Denaturing electrophoresis revealed that the globular assemblies comprised oligomers ranging from trimers to 24mers. Oligomers prepared at 4 degrees C stayed fibril-free for days and remained so when shifted to 37 degrees C, although the spectrum of sizes shifted toward larger oligomers at the higher temperature. The soluble, globular Abeta(1-42) oligomers were toxic to PC12 cells, impairing reduction of MTT and interfering with ERK and Rac signal transduction. Occasionally, oligomers were neither toxic nor recognized by toxicity-neutralizing antibodies, suggesting that oligomers could assume alternative conformations. Tests for oligomerization-blocking activity were carried out by dot-blot immunoassays and showed that neuroprotective extracts of Ginkgo biloba could inhibit oligomer formation at very low doses. The observed neurotoxicity, structure, and stability of synthetic Abeta(1-42) globular assemblies support the hypothesis that Abeta(1-42) oligomers play a role in triggering nerve cell dysfunction and death in Alzheimer's disease.