Abstract Body

The apolipoprotein E (APOE) gene is the strongest genetic risk factor for Alzheimer’s disease (AD) with APOE4 strongly increases the risk and APOE2 being protective. Physiologically, apoE protein is a lipid carrier transporting cholesterol and other lipids in periphery and brain by binding to cell surface apoE receptors. Using human APOE allele-specific targeted replacement mice, we found that apoE2 reduces the cholesterol levels in brain parenchyma but increases those in the cerebral spinal fluid, both correlate with better memory performance and longevity when compared to apoE3 and apoE4. The increased lipid-association and metabolism is also seen with an APOE3-Jacksonville variant, which is protective against AD risk. The apoE3-Jac variant reduces self-aggregation as the underlying mechanism for enhanced lipidation and the ability to reduce brain amyloids. Using human induced pluripotent stem cell (iPSC)-derived cerebral organoid models, we found that deficiency of apoE leads to intracellular accumulation of lipids and α-synuclein, supporting a critical role of apoE in promoting lipid efflux to maintain cellular homeostasis. The differential effects of apoE isoforms on lipid metabolism and distribution are also seen in brains from humans and mice through lipidomic analysis. Our studies collective support a critical role of apoE in brain lipid metabolism essential for brain health and reveal apoE isoform-specific effects impacting AD pathogenesis.

Abstract Body

The APOE4 allele increases the risk of Alzheimer's disease (AD). Similarly, low plasma apolipoprotein E (apoE) levels increase the risk of AD and other types of dementia. Plasma apoE levels, mainly derived from the liver, are known to vary in an APOE genotype-dependent manner however our recent results suggest that this effect may vary between ethnicities. Plasma apoE cannot cross the blood-brain-barrier, nevertheless our results expose an unfavorable association between low plasma apoE levels, cognition and cerebrospinal fluid (CSF) AD biomarkers. Recent results further propose that plasma apoE levels are linked to plasma glucose but not insulin levels which may help explain why a higher relative ratio of the apoE4 isoform over apoE3 in cognitively healthy APOE3/4 carriers was associated with glucose hypometabolism and gray matter volume reductions in AD relevant brain areas. Furthermore, APOE4-carriers exhibit higher plasma triglyceride levels and plasma apoE4 was positively correlated with plasma levels of total cholesterol and low-density lipoprotein. With the liver as an essential player in the peripheral glucose and lipid metabolism we are investigating a potential liver-phenotype which could explain the APOE4-promoted risk of AD. Our preliminary results suggest altered hepatic protein levels mainly of the amyloid-beta degrading enzyme cathepsin-D in an APOE4-dependent manner, and we speculate in line with other recent studies that altered hepatic amyloid-beta metabolism may be linked to AD brain pathology. Our ongoing lipidomic profiling of primary human hepatocytes from APOE4 versus non-APOE4 carriers will further aid our efforts of finding an APOE4 liver phenotype that drives the increased risk of AD.

Aims

Low plasma levels of apolipoprotein E (apoE) were previously linked to a higher risk of Alzheimer’s disease (AD).Plasma apoE is mainly derived from the liver and APOEε4-carriers frequently exhibit lower levels than non-carriers.We speculate that a liver APOEε4-phenotype contributes to the increased risk of AD.The liver is essential to various physiological processes including lipid-metabolism and peripheral amyloid-β (Aβ) clearance.Previous studies have shown that Aβ-degrading enzymes in the liver are altered in AD however;a potential contribution of APOEε4 genotype to these findings has not been investigated.

Methods

Pellets of primary human hepatocytes (total of:n=69) isolated from liver tissues acquired through liver transplantation were APOE genotyped using real-time polymerase-chain-reaction.Protein levels of Aβ-degrading enzymes cathepsin-D (CatD), insulin-degrading enzyme (IDE) and neprilysin were assessed using western blot in RIPA buffered cell lysates.

Results

Regardless of APOEε4 genotype, higher levels of IDE were linked to higher levels of both CatD (Spearman’s-ρ=0.404,p=0.001) and neprilysin (Spearman’s-ρ=0.501,p<0.001).These associations remained even after separating the subjects based on their APOEε4 allele status.Interestingly, hepatocytes isolated from APOEε4-carriers (n=52) exhibited 73% higher levels of CatD compared to hepatocytes from non-carriers (n=17) (p=0.004) with the highest levels found in APOEε4 homozygous carriers.Presence of APOEε4 had no effect on the levels of neprilysin and IDE.

Conclusions

The significant associations between neprilysin, IDE and CatD might indicate that APOEε4 genotype affects the Aβ-degrading enzymes neprilysin and IDE through CatD.Previous studies have also proposed that CatD can degrade apoE and the observed APOEε4-dependent increase in this enzyme may help explain the lower plasma apoE levels documented in APOEε4 subjects.

Aims

The locus coeruleus (LC) is a small pontine nucleus and primary source of norepinephrine in the human brain. In Alzheimer’s disease (AD), the LC is one of the first structures to accumulate hyperphosphorylated tau, resulting in marked reductions in size and volume. As the LC projects to and provides norepinephrine throughout corticolimbic structures, we aimed to examine the relationship of LC neuronal loss with corticolimbic patterns among AD subtypes and to investigate clinicopathologic predictors of LC neuronal loss.

Methods

Corticolimbic tangle patterns were used to classify AD subtypes as hippocampal sparing (HpSp), typical, and limbic predominant. An H&E-stained pons was digitally scanned and neuronal density measured in 783 AD cases.

Results

Among AD subtypes, limbic predominant had higher number of LC neurons compared to typical AD with HpSp AD having the fewest LC neurons (p=0.013). Multivariable regression analysis in typical AD accounted for 24% (R²=0.24) of the variability in LC neuronal loss (p<0.001): 10 years younger at death equated to 4 fewer LC neurons, 5 years longer in disease duration equated to 1 fewer neuron, and Braak tangle stage increase equated to 4 fewer neurons. No association was found between LC neuronal loss and sex, brain weight, or Thal phase. Interestingly, APOE ε4 status was associated with 2 more LC neurons.

Conclusions

Younger age onset, longer disease duration, and lack of an APOE ε4 allele may influence LC vulnerability. Our findings highlight the potential biomarker relevance of LC atrophy for young onset AD, which may inform corticolimbic tau patterns.

Aims

The association of Apolipoprotein E (APOE) ɛ4 with Alzheimer's disease risk is well-established however, the impact of APOE ɛ4 on human brain cell function remains unclear. We hypothesized that human APOE ε4/ε4 genotype contributes to disease risk through cell autonomous and non-cell autonomous mechanisms in a human specific manner.

Methods

Global transcriptomic analyses were performed on astrocytes, microglia, mixed cortical cultures (neurons and astrocytes) and brain microvascular endothelial cells derived from human induced pluripotent stem cells (iPSCs), post-mortem brains and microglia and astrocytes from APOE targeted replacement (TR) mice. Lipid pathway dysregulation identified through GSEA, WGCNA and cell type deconvolution was validated in vitro using Gas Chromatography-Mass Spectrometry (GC-MS) to measure intracellular cholesterol and filipin staining with immunocytochemistry and western blotting to determine the subcellular compartment that accumulates cholesterol in iPSC-based models, and further confirmed by immunohistochemical analysis in human APOE4 post-mortem brain.

Results

Global transcriptomic analyses reveal that APOE 44 drives lipid metabolic dysregulation in astrocytes and microglia from human iPSCs and cell type-deconvolved post-mortem AD brain but not in primary murine APOE 44 astrocytes and microglia. APOE ε4 leads to elevated de novo cholesterol synthesis despite the intracellular accumulation of free cholesterol coupled with decreased efflux due to lysosomal cholesterol sequestration in iPSC-derived astrocytes and in hippocampal astrocytes from APOE ε4 human brain.

Conclusions

Human APOE ε4 causes human glia-specific, region-specific, cell autonomous lipid dysregulation that may increase AD risk.

Aims

APOE4 is the strongest genetic risk factor for late-onset Alzheimer’s disease (AD). However, the molecular mechanism associated with its deleterious effect in AD and its protective role in glaucoma, is unknown.

Methods

Here we aimed to dissect the impact of microglial APOE4 on AD pathology and glaucoma models, using CX3CR1-CRE^ERT2 mice crossed to APOE-KI(E3 and E4)^fl/fl:APP/PS1.

Results

We identified reduced numbers of Clec7a⁺ MGnD-microglia per plaque in APP/PS1:APOE4-KI mice, despite their increased plaque load, compared with APP/PS1:APOE3-KI mice. APOE4 microglia accumulate lipids and represent a dysfunctional metabolic state. Conditional genetic deletion of APOE4 in microglia resulted in increased numbers of MGnD-microglia per plaque in APP/PS1 mice. scRNAseq analysis showed increased proportion of MGnD-microglia in APP/PS1:APOE4 conditional KO mice, associated with reduced plaque pathology. Furthermore, we show impaired induction of MGnD signature in AD brains of APOE4 carriers. We observed a similar microglial phenotype in both mouse model and human glaucoma. However, this microglial signature showed impaired MGnD response in glaucoma mice expressing APOE4, which was associated with protection of retinal ganglion cells (RGC), despite elevated intraocular pressure. Importantly, genetic and pharmacologic targeting of LGASL3 ameliorated RGC degeneration, and LGALS3 expression was attenuated in human APOE4 glaucoma samples.

Conclusions

We show that APOE4 is a negative regulator of MGnD in AD, and that its genetic deletion restores an MGnD signature and reduce plaque pathology. Taken together, these findings identify a beneficial role of MGnD in AD and their detrimental role in glaucoma, which may provide new molecular targets to modulate and restore functional microglia in AD.

Abstract Body

ApoE4 is the most prevalent and clinically most important risk factor for late-onset Alzheimer’s disease (AD). Three major ApoE isoforms exist in humans which differ by the presence of a positively charged arginine or a neutral cysteine at residues 112 and 158, respectively. The main biophysical difference between these different ApoE isoforms therefore is their net charge, and disease risk correlates with their isoelectric point (IEP). The higher the IEP, the greater the disease risk. The IEP of ApoE4 closely matches the pH of the early endosomes, which undergo swelling and impaired trafficking upon endocytosis of ApoE4. We showed previously that this trafficking defect can be completely abolished by pharmacologically inhibiting or genetically reducing NHE6, the main proton leak channel specifically in early endosomes. We now show that genetic disruption of NHE6 prevents amyloid accumulation in a humanized APP knocking mouse model, abolishes all functional differences between ApoE isoforms and restores synaptic dysfunction in humanized ApoE4 knockin mice. These findings reveal NHE6 as a novel rational therapeutic target for the prevention of late-onset AD.

Abstract Body

Objectives: The overall objective of this study was to develop a low molecular weight and brain permeable druggable ABCA1 activator that can reverse the hypo-lipidation of apoE4 and the associated brain pathological effects.

Methods: High-ThroughputScreening (HTS) was first performed utilizing a cellular cholesterol efflux assay which was based on measurements of hit derived and ABCA1 driven increased lipidation of apoE4 as well as on assessment of the size of the lipidated particle utilizing fluorescence polarization measurements .Potential hits were then subjected to an orthogonal assay (i.e. digestion of senile plaques) in order to assess the extent to which the hits can reverse an efflux-independent apoE4 driven phenotype.

Results: Analysis of the HTS ,plaque digestion, and ABCA1 target recognition results led to the discovery of a lead compound which reacted optimally in theses assays, Further mechanistic studies revelled that the decreased levels of apoE and growth factor receptors which are induced by apoE4 in primary neuronal cultures are reversed by the lead compound. Preliminary in vivo studies revealed that injection of the lead compound into the brains of the apoE4 TR mice with Alzet mini osmotic pumps reversed the hypo-lipidation of apoE4 and increased the brain levels of apoE4

Conclusions: The results obtained show that the pathological effects of apoE4 can be reduced by reversal of the hypolipidation of apoE4 and suggest a novel approach for the treatment of the pathological effects of apoE4 in AD