It is suspected that excess of brain cholesterol plays a role in Alzheimer’s disease (AD). Membrane-associated cholesterol was shown to be increased in the brain of individuals with sporadic AD and to correlate with the severity of the disease. We hypothesized that an increase of membrane cholesterol could trigger sporadic AD early phenotypes. We thus acutely loaded the plasma membrane of cultured neurons with cholesterol to reach the 30% increase observed in AD brains. We found changes in gene expression profiles that are reminiscent of early AD stages. We also observed early AD cellular phenotypes. Indeed we found enlarged and aggregated early endosomes using confocal and electron microscopy after immunocytochemistry. In addition amyloid precursor protein vesicular transport was inhibited in neuronal processes, as seen by live-imaging. Finally transient membrane cholesterol loading lead to significantly increased amyloid-β42 secretion. Membrane cholesterol increase in cultured neurons reproduces most early AD changes and could thus be a relevant model for deciphering AD mechanisms and identifying new therapeutic targets.

The proinflammatory cytokine interleukin-1β (IL-1β) is overexpressed in Alzheimer disease (AD) as a key regulator of neuroinflammation. Amyloid-β (Aβ) peptide triggers activation of inflammasomes, protein complexes responsible for IL-1β maturation in microglial cells. Downregulation of NALP3 (NACHT, LRR, and PYD domains-containing protein 3) inflammasome has been shown to decrease amyloid load and rescue cognitive deficits in a mouse model of AD. Whereas activation of inflammasome in microglial cells has been described in AD, no data are currently available concerning activation of inflammasome in astrocytes, although they are involved in inflammatory response and phagocytosis. Here, by targeting the inflammasome adaptor protein ASC (apoptosis-associated speck-like protein containing a CARD domain), we investigated the influence of activation of the inflammasome on the phagocytic activity of astrocytes. We used an ASC knockout mouse model, as ASC is a central protein in the inflammasome, acting as an adaptor and stabilizer of the complex and thus critical for its activation. Lipopolysaccharide (LPS)-primed primary cultures of astrocytes from newborn mice were utilized to evaluate Aβ-induced inflammasome activation by measuring IL-1β release by ECLIA (electro-chemiluminescence immunoassay). Phagocytosis efficiency was measured by incorporation of bioparticles, and the release of the chemokine CCL3 (C-C motif ligand 3) was measured by ECLIA. ASC mice were crossbred with 5xFAD (familial Alzheimer disease) mice and tested for spatial reference memory using the Morris water maze (MWM) at 7–8 months of age. Amyloid load and CCL3 were quantified by thioflavine S staining and quantitative real-time polymerase chain reaction (qRT-PCR), respectively. Cultured astrocytes primed with LPS and treated with Aβ showed an ASC-dependent production of IL-1β resulting from inflammasome activation mediated by Aβ phagocytosis and cathepsin B enzymatic activity. ASC+/− astrocytes displayed a higher phagocytic activity as compared to ASC+/+ and ASC −/− cells, resulting from a higher release of the chemokine CCL3. A significant decrease in amyloid load was measured in the brain of 7–8-month-old 5xFAD mice carrying the ASC +/− genotype, correlated with an increase in CCL3 gene expression. In addition, the ASC +/− genotype rescued spatial reference memory deficits observed in 5xFAD mice. Our results demonstrate that Aβ is able to activate astrocytic inflammasome. Downregulation of inflammasome activity increases phagocytosis in astrocytes due to the release of CCL3. This could explain why downregulation of inflammasome activity decreases amyloid load and rescues memory deficits in a mouse model of AD.

Alzheimer’s disease (AD) is the leading cause of dementia in the elderly. Mutations in genes encoding proteins involved in amyloid-β peptide (Aβ) production are responsible for inherited AD cases. The amyloid cascade hypothesis was proposed to explain the pathogeny. Despite the fact that Aβ is considered as the main culprit of the pathology, most clinical trials focusing on Aβ failed and suggested that earlier interventions are needed to influence the course of AD. Therefore, identifying risk factors that predispose to AD is crucial. Among them, the epsilon 4 allele of the apolipoprotein E gene that encodes the major brain lipid carrier and metabolic disorders such as obesity and type 2 diabetes were identified as AD risk factors, suggesting that abnormal lipid metabolism could influence the progression of the disease. Among lipids, fatty acids (FAs) play a fundamental role in proper brain function, including memory. Peroxisome proliferator-activated receptor α (PPARα) is a master metabolic regulator that regulates the catabolism of FA. Several studies report an essential role of PPARα in neuronal function governing synaptic plasticity and cognition. In this review, we explore the implication of lipid metabolism in AD, with a special focus on PPARα and its potential role in AD therapy.

Mechanisms driving cognitive improvements following nuclear receptor activation are poorly understood. The peroxisome proliferator–activated nuclear receptor alpha (PPARα) forms heterodimers with the nuclear retinoid X receptor (RXR). We report that PPARα mediates the improvement of hippocampal synaptic plasticity upon RXR activation in a transgenic mouse model with cognitive deficits. This improvement results from an increase in GluA1 subunit expression of the alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptor, eliciting an AMPA response at the excitatory synapses. Associated with a two times higher PPARα expression in males than in females, we show that male, but not female, PPARα null mutants display impaired hippocampal long-term potentiation. Moreover, PPARα knockdown in the hippocampus of cognition-impaired mice compromises the beneficial effects of RXR activation on synaptic plasticity only in males. Furthermore, selective PPARα activation with pemafibrate improves synaptic plasticity in male cognition-impaired mice, but not in females. We conclude that striking sex differences in hippocampal synaptic plasticity are observed in mice, related to differences in PPARα expression levels.

Among genetic susceptibility loci associated with late-onset Alzheimer disease (LOAD), genetic polymorphisms identified in genes encoding lipid carriers led to the hypothesis that a disruption of lipid metabolism could promote disease progression. We previously reported that amyloid precursor protein (APP) involved in Alzheimer disease (AD) physiopathology impairs lipid synthesis needed for cortical networks’ activity and that activation of peroxisome proliferator–activated receptor α (PPARα), a metabolic regulator involved in lipid metabolism, improves synaptic plasticity in an AD mouse model. These observations led us to investigate a possible correlation between PPARα function and full-length APP expression. Here, we report that PPARα expression and activation were inversely related to APP expression both in LOAD brains and in early-onset AD cases with a duplication of the APP gene, but not in control human brains. Moreover, human APP expression decreased PPARA expression and its related target genes in transgenic mice and in cultured cortical cells, while opposite results were observed in APP-silenced cortical networks. In cultured neurons, APP-mediated decrease or increase in synaptic activity was corrected by a PPARα-specific agonist and antagonist, respectively. APP-mediated control of synaptic activity was abolished following PPARα deficiency, indicating a key function of PPARα in this process.