The interactions of apolipoprotein E and inflammation in human iPSC-derived astrocytes

Abstract

Alzheimer’s disease (AD) is a devastating neurodegenerative disease characterised histologically by the presence of extracellular plaques and intracellular neurofibrillary tangles, and clinically by memory loss, cognitive decline and personality changes. Much of the research investigating AD to date has focused on the familial early-onset form of AD which accounts for <5% of cases. Most cases are sporadic with late-onset, with certain genes presenting a high risk. Apolipoprotein E (APOE) genotype is one of the major genetic risk factors for late-onset AD. There are three major APOE isoforms (APOE2, APOE3and APOE4), each differing by a single amino acid residueat the protein leveland thus inferring different AD-risk. APOE2 contains cysteineat amino acid residues 112 and 158, APOE3contains a cysteine at residue 112 and arginine at residue 158, while APOE4 contains arginine at both residue sites.APOE3 is most common in the general population and APOE4 associated with an increased AD risk. Emerging evidence supports a detrimental role of astrocytes in AD, with astrocytes developing a reactive phenotype upon exposure to inflammatory microglial factors.Furthermore, excitotoxicity has been well characterised in AD as a mechanism for neuronal deathwhich may occur as a result of alteredastrocyte function following activation.With this in mind, our study aimed to investigate the impact of APOEgenotype on astrocyte reactivity and functionality using human induced pluripotent stem cell (iPSC)-derived astrocytes with homozygous APOE3andAPOE4genotypes.Human iPSC homozygous for APOE3and APOE4were patterned towards a neural fate using dual SMAD inhibition to produce neural progenitor cells. Subsequently, astrocytes were derived using epidermal growth factor and human leukemia inhibitory factor to produce mature astrocytes after 90 days. Astrocytes were characterised by immunostaining for classic markers, then stimulated (IL1α, TNFα and C1q) to induce reactivity and a reactivity profile was generated via morphological, ELISA and qPCR analysis. The functional capacity of astrocytes was investigated via glutamate assay.iPSC-derived astrocytes are positive for astrocyte markers GFAP, S100β, Connexin-43 and EAAT1 confirming cell fate. Stimulation with microglial factors resulted in increasedIL6, RANTES and GM-CSF secretion, coupled with increased expression of genes associated with reactivity including IL6, ICAM1, LCN2and SERPINA3.APOEgenotype affected basal expression and secretion of IL6 as well as IL1B. Fibroblast growth factor 2 (FGF2) was capable of modulating some aspects of astrocyte reactivity while cannabidiol (CBD) did not. Glutamate dynamics were unaffected by stimulation,however APOE4astrocytes have altered glutamate flux relative to APOE3control.In conclusion, the ability to produce reactive astrocytes derived from genotyped human cells in vitroprovides a powerful model for researching the mechanisms underlying the detrimental role of astrocytes and the effect of APOE genotype in AD. APOE genotype appears to influence astrocyte reactivity and APOE4is strikingly detrimental to glutamate handling. While CBD did not successfully modulate astrocyte reactivity, FGF2 had some success modulating reactivity. This study presents the potential mechanisms which predispose APOE4carriers to AD and presents some therapeutic targets in order to manage AD risk.