**6. ABCA1 in Neurological Disease**

Cholesterol homeostasis is essential for the central nervous system (CNS). Approximately 23% of total body cholesterol is found in the CNS. Brain cholesterol is mainly synthesized in situ, as essentially no cholesterol enters the brain from the peripheral circulation [144]. Moreover, CNS growth and differentiation requires cholesterol produced by de novo synthesis [144,145]. The capability of neurons to biosynthesize cholesterol decreases in adulthood and depends mainly on glial cells [146]. ABCA1 is expressed in neurons and astrocytes, where it promotes the efflux of phospholipids and unesterified cholesterol to glia-derived apolipoprotein E (apoE) [147]. ApoE is the main apolipoprotein found and synthesized in the brain and is found in the interstitial and cerebrospinal fluid in the form of lipid-rich ApoE particles. The density and size of these particles are similar to those of plasma HDL [148]. ABCA1 contributes to cholesterol homeostasis and participates in the pathophysiology of neurological diseases involving the accumulation of proteins in brain cells, such as traumatic brain injury, stroke sequelae, Parkinson's disease, and Alzheimer's disease (AD) [149–157].

AD is a neurodegenerative disorder clinically characterized by progressive memory loss, disorientation and cognitive decline [158]. At the histopathological level, characteristic amyloid plaques and neurofibrillary tangles are found in the brain tissue [159–161]. Amyloid plaques develop from the accumulation of amyloid β peptide (Aβ) [161]. ApoE plays a crucial role in the proteolytic degradation of soluble forms of Aβ, and this effect is dependent of apoE lipidation by ABCA1-mediated cholesterol and phospholipid transfer [162]. The ABCA1 protein participates in this process by regulating apoE levels and function in the CNS [163–167].

In murine models, ABCA1 deficiency (*Abca1*−*/*−) was found to reduce apoE protein levels in the brain, to decrease lipidation of astrocyte-secreted apoE and to favor rapid apoE degradation [167,168]. *Abca1* deficiency may also increase amyloid burden in certain AD mouse models. Specifically, in a transgenic AD mouse model (APP23), targeted *Abca1* disruption (APP23/*Abca1*−*/*−) increased amyloid deposition, increased the level of cerebral amyloid angiopathy, exacerbated cerebral amyloid angiopathy-related microhemorrhage, and caused a sharp decrease of soluble, but not of insoluble brain apoE levels [167]. Conversely, selective ABCA1 overexpression in AD mouse models led to increased CNS apoE lipidation and sharply decreased amyloid deposition [168], while ABCA1 upregulation by miRNA-33 inhibition was found to increase apoE lipidation and to decrease Aβ levels in the brain [169]. Notably, Fitz et al. reported that while *Abca1* deletion in transgenic APP mice caused cognitive deficit at a stage of early amyloid pathology, these characteristics were not observed in *Abca1*−/−/wildtype mice. However, intra-hippocampal infusion of scrambled A oligomers affected cognitive performance of *Abca1* KO mice, which also showed altered neurite architecture in the hippocampus, suggesting that mice lacking ABCA1 have basal cognitive deficits that prevent them from coping with additional stressors [170].

Neuroinflammation and glucose metabolism are also important pathophysiological features in AD. Aβ deposits induce infiltration of immune cells such as T-helper 17 to the brain parenchyma and the secretion of proinflammatory cytokines such as interleukin 17A (IL-17A), which contribute to AD progression [171,172]. Interestingly, Yang et al. demonstrated that intracranial IL-17A overexpression increased ABCA1 protein levels in the hippocampus protein but not in cortex, decreased soluble Aβ levels in the hippocampus and cerebrospinal fluid, and improved glucose metabolism, suggesting that IL-17A may play a protective role in the pathogenesis of AD [173]. Moreover, hyperglycemic states are associated with greater severity of AD [174,175]. In this context, Lee et al. reported that in Zucker diabetic fatty rats (*fa*−/*fa*−) and in human neuroblastoma cells, exposure to

high glucose levels increased Aβ deposition in the brain and decreased ABCA1 expression through JNK-reduced LXRα expression and binding to the *abca1* gene promoter [176].

Genetic studies support a role of ABCA1 in AD. Firstly, loss-of-function *ABCA1* mutations (N1800H) have been associated with low plasma apoE and increased AD risk in humans [177,178]. Moreover, although GWAS have consistently shown the crucial relevance of the *APOE4* variant in increasing AD risk across populations, *ABCA1* gene variation (rs3905000, rs27772082, rs2740488) has also been found to contribute to AD susceptibility in some GWAS [179–185]. Candidate gene studies analyzing the R219K polymorphism (rs2230806) and AD risk have reported conflicting results. This variant was associated with an increased risk of AD in Caucasian [186–189] and Chinese [51] populations, found to be a protective variant to AD in Chinese-Han and Hungarian individuals [190,191], and found not to be associated with AD risk in the German population [192]. However, two meta-analyses failed to find significant associations between *ABCA1* polymorphisms and AD [193,194].

It has been suggested that upregulation of ABCA1 expression or function may be a therapeutic target for AD and other diseases where Aβ plays a pathophysiological role. Interestingly, ABCA1 mediates the effect of some drugs proposed for AD treatment, such as bexarotene [195] and the liver X receptor agonist GW-3965 [164]. In addition, cyclodextrin [196], ondansetron [197], prostaglandin A1 [198], the purinergic receptor antagonist P2X7 [199], and the CS-6253 peptide [200] increase *ABCA1* gene expression in brain cells, although not all of these drugs improved cognitive function in vivo. Furthermore, Sarlak Z et al. reported that aerobic exercise significantly increases *Abca1* mRNA expression and decreases soluble Aβ1-42 in the hippocampus of rats with and without AD diagnosis. Aerobic training also improved cognitive function (learning and memory) [201]. ABCA1 and ApoE are currently the matter of intensive research for AD treatment [202].
