Search

Three decades of research, both in vitro and in vivo, have demonstrated the conformational heterogeneity that is displayed by the amyloid β peptide (Aβ) in Alzheimer's disease (AD). Understanding the distinct properties between Aβ conformations and how conformation may impact cellular activity remain open questions, yet still continue to provide new insights into protein misfolding and aggregation. In particular, there is interest in the group of soluble oligomeric prefibrillar Aβ species comprising lower molecular weight oligomers up to larger protofibrils. In the current study, a number of strategies were utilized to separate Aβ protofibrils and oligomers and show that the smaller Aβ oligomers have a much different conformation than Aβ protofibrils. The differences were consistent for both Aβ40 and Aβ42. Protofibrils bound thioflavin T to a greater extent than oligomers, and were highly enriched in β-sheet secondary structure. Aβ oligomers possessed a more open structure with significant solvent exposure of hydrophobic domains as determined by tryptophan fluorescence and bis-ANS binding, respectively. The protofibril-selective antibody AbSL readily discerned conformational differences between protofibrils and oligomers. The more developed structure for Aβ protofibrils ultimately proved critical for provoking the release of tumor necrosis factor α from microglial cells. The findings demonstrated a dependency on β-sheet structure for soluble Aβ aggregates to cause a microglial inflammatory response. The Aβ aggregation process yields many conformationally-varied species with different levels of β-structure and exposed hydrophobicity. The conformation elements likely determine biological activity and pathogenicity.

Prior work suggests that amyloid precursor protein (APP) can function as a proinflammatory receptor on immune cells, such as monocytes and microglia. Therefore, we hypothesized that APP serves this function in microglia during Alzheimer9s disease. Although fibrillar amyloid β (Aβ)-stimulated cytokine secretion from both wild-type and APP knock-out (mAPP −/− ) microglial cultures, oligomeric Aβ was unable to stimulate increased secretion from mAPP −/− cells. This was consistent with an ability of oligomeric Aβ to bind APP. Similarly, intracerebroventricular infusions of oligomeric Aβ produced less microgliosis in mAPP −/− mice compared with wild-type mice. The mAPP −/− mice crossed to an APP/PS1 transgenic mouse line demonstrated reduced microgliosis and cytokine levels and improved memory compared with wild-type mice despite robust fibrillar Aβ plaque deposition. These data define a novel function for microglial APP in regulating their ability to acquire a proinflammatory phenotype during disease. SIGNIFICANCE STATEMENT A hallmark of Alzheimer9s disease (AD) brains is the accumulation of amyloid β (Aβ) peptide within plaques robustly invested with reactive microglia. This supports the notion that Aβ stimulation of microglial activation is one source of brain inflammatory changes during disease. Aβ is a cleavage product of the ubiquitously expressed amyloid precursor protein (APP) and is able to self-associate into a wide variety of differently sized and structurally distinct multimers. In this study, we demonstrate both in vitro and in vivo that nonfibrillar, oligomeric forms of Aβ are able to interact with the parent APP protein to stimulate microglial activation. This provides a mechanism by which metabolism of APP results in possible autocrine or paracrine Aβ production to drive the microgliosis associated with AD brains.

Accumulation of aggregated amyloid-β protein (Aβ) is an important feature of Alzheimer’s disease. There is significant interest in understanding the initial steps of Aβ aggregation due to the recent focus on soluble Aβ oligomers. In vitro studies of Aβ aggregation have been aided by the use of conformation-specific antibodies which recognize shape rather than sequence. One of these, OC antiserum, recognizes certain elements of fibrillar Aβ across a broad range of sizes. We have observed the presence of these fibrillar elements at very early stages of Aβ incubation. Using a dot blot assay, OC-reactivity was found in size exclusion chromatography (SEC)-purified Aβ(1-42) monomer fractions immediately after isolation (early-stage). The OC-reactivity was not initially observed in the same fractions for Aβ(1-40) or the aggregation-restricted Aβ(1-42) L34P but was detected within 1–2 weeks of incubation. Stability studies demonstrated that early-stage OC-positive Aβ(1-42) aggregates were resistant to 4M urea or guanidine hydrochloride but sensitive to 1% sodium dodecyl sulfate (SDS). Interestingly, the sensitivity to SDS diminished over time upon incubation of the SEC-purified Aβ(1-42) solution at 4° C. Within 6–8 days the OC-positive Aβ42 aggregates were resistance to SDS denaturation. The progression to, and development of, SDS resistance for Aβ(1-42) occurred prior to thioflavin T fluorescence. In contrast, Aβ(1-40) aggregates formed after 6 days of incubation were sensitive to both urea and SDS. These findings reveal information on some of the earliest events in Aβ aggregation and suggest that it may be possible to target early-stage aggregates before they develop significant stability.

Neuroinflammation triggered by accumulation of amyloid-β protein (Aβ) is a significant component of the Alzheimer's disease (AD) brain. Senile plaques composed of Aβ attract and activate microglia cells resulting in cytokine secretion and a proinflammatory environment. The mechanism by which Aβ activates microglia is complex and involves numerous cellular components. One receptor potentially involved in Aβ recognition and the ensuing microglia proinflammatory response is CD47. Since there is significant interest in soluble aggregated Aβ species, we sought to determine if CD47 plays a key role in microglia cytokine release stimulated by soluble Aβ(1-42) protofibrils. Pretreatment of primary murine microglia with the CD47 antagonist peptide 4N1K significantly and potently inhibited both tumor necrosis factor-α (TNFα) and interleukin-1β (IL-1β) secretion stimulated by Aβ(1-42) protofibrils. 4N1K displayed toxicity to the microglia but only at concentrations much higher than the observed inhibition. Surprisingly, 4N1K also potently inhibited TNFα secretion triggered by lipopolysaccharide which is not known to signal through CD47. Treatment of the microglia with a neutralizing anti-CD47 antibody failed to block the Aβ protofibril response even though comparable samples were completely inhibited by 4N1K. Finally, Aβ(1-42) protofibrils stimulated similar levels of secreted TNFα production in both wild-type and CD47(-/-) microglia and 4N1K still potently inhibited the Aβ protofibril response even in the CD47(-/-) microglia. The overall findings demonstrated that the microglial proinflammatory response to Aβ(1-42) protofibril is not dependent on CD47 and that 4N1K exhibits CD47-independent inhibitory activity.