Computational and Structural Biotechnology Journal 21 (2023) 1189–1204 Contents lists available at ScienceDirect Computational and Structural Biotechnology Journal journal homepage: www.elsevier.com/locate/csbj Cross interactions between Apolipoprotein E and amyloid proteins in neurodegenerative diseases ]] ]]]]]] ]] Rolf Antonie Loch a,1, Hongzhi Wang a,1, Alex Perálvarez-Marín b, Philipp Berger a, ⁎ Henrietta Nielsen d, Angeliki Chroni c, Jinghui Luo a, a Department of Biology and Chemistry, Paul Scherrer Institute, 5232 Villigen, Switzerland b Biophysics Unit, Department of Biochemistry and Molecular Biology, School of Medicine, Institut de Neurociències, Universitat Autònoma de Barcelona, 08193 Cerdanyola del Vallés, Catalonia, Spain c Institute of Biosciences and Applications, National Center for Scientific Research 'Demokritos', 15341 Athens, Greece d Department of Biochemistry and Biophysics, Stockholm University, Svante Arrhenius Väg 16B, 10691 Stockholm, Sweden a r t i cl e i nfo a bstr ac t Article history: Three common Apolipoprotein E isoforms, ApoE2, ApoE3, and ApoE4, are key regulators of lipid home­ Received 26 September 2022 ostasis, among other functions. Apolipoprotein E can interact with amyloid proteins. The isoforms differ by Received in revised form 18 January 2023 one or two residues at positions 112 and 158, and possess distinct structural conformations and functions, Accepted 18 January 2023 leading to isoform-specific roles in amyloid-based neurodegenerative diseases. Over 30 different amyloid Available online 20 January 2023 proteins have been found to share similar characteristics of structure and toxicity, suggesting a common interactome. The molecular and genetic interactions of ApoE with amyloid proteins have been extensively Keywords: ApoE studied in neurodegenerative diseases, but have not yet been well connected and clarified. Here we sum­ Amyloid proteins marize essential features of the interactions between ApoE and different amyloid proteins, identify gaps in Interaction the understanding of the interactome and propose the general interaction mechanism between ApoE iso­ Endogenous molecules forms and amyloid proteins. Perhaps more importantly, this review outlines what we can learn from the Cytotoxicity interactome of ApoE and amyloid proteins; that is the need to see both ApoE and amyloid proteins as a basis Therapeutics to understand neurodegenerative diseases. © 2023 The Authors. Published by Elsevier B.V. on behalf of Research Network of Computational and Structural Biotechnology. This is an open access article under the CC BY license (http://creative­ commons.org/licenses/by/4.0/). Contents 1. Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... ..... . ..... . ..... . ..... . ..... . ..... . ..... . . . . . 1190 2. Cross-interactions of ApoE with amyloid proteins . . . . . . . . . . . . . . . . . . . . . . . . . . ..... ..... . ..... . ..... . ..... . ..... . ..... . ..... . . . . . 1191 2.1. Interactions between ApoE and the Aβ peptide. . . . . . . . . . . . . . . . . . . . . . . ..... ..... . ..... . ..... . ..... . ..... . ..... . ..... . . . . . 1191 2.2. Interactions between ApoE and Alzheimer’s disease-associated tau . . . . . . ..... ..... . ..... . ..... . ..... . ..... . ..... . ..... . . . . . 1193 2.3. Interactions between ApoE and Parkinson’s disease-related α-Syn . . . . . . . ..... ..... . ..... . ..... . ..... . ..... . ..... . ..... . . . . . 1194 2.4. Interactions between ApoE and TSE-related prion . . . . . . . . . . . . . . . . . . . . ..... ..... . ..... . ..... . ..... . ..... . ..... . ..... . . . . . 1194 2.5. Interactions between ApoE and T2D-associated IAPP . . . . . . . . . . . . . . . . . . ..... ..... . ..... . ..... . ..... . ..... . ..... . ..... . . . . . 1195 2.6. Interactions between ApoE and ALS-related SOD1 . . . . . . . . . . . . . . . . . . . . ..... ..... . ..... . ..... . ..... . ..... . ..... . ..... . . . . . 1195 2.7. Interactions between ApoE and TDP-43 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... ..... . ..... . ..... . ..... . ..... . ..... . ..... . . . . . 1195 2.8. Interactions between ApoE and other amyloid proteins . . . . . . . . . . . . . . . . ..... ..... . ..... . ..... . ..... . ..... . ..... . ..... . . . . . 1196 3. Potential interaction mechanism between ApoE isoform and amyloid protein. . . . ..... ..... . ..... . ..... . ..... . ..... . ..... . ..... . . . . . 1196 4. Loci of cross interactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... ..... . ..... . ..... . ..... . ..... . ..... . ..... . . . . . 1197 4.1. Phase-separated droplets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... ..... . ..... . ..... . ..... . ..... . ..... . ..... . . . . . 1197 4.2. Lipid membranes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... ..... . ..... . ..... . ..... . ..... . ..... . ..... . . . . . 1197 ⁎ Corresponding author. E-mail address: Jinghui.luo@psi.ch (J. Luo). 1 These authors contributed equally. https://doi.org/10.1016/j.csbj.2023.01.022 2001-0370/© 2023 The Authors. Published by Elsevier B.V. on behalf of Research Network of Computational and Structural Biotechnology. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). R.A. Loch, H. Wang, A. Perálvarez-Marín et al. Computational and Structural Biotechnology Journal 21 (2023) 1189–1204 4.3. Cholesterol and lipidation . . . . . . . . . . ..... . ..... . ..... . ..... . ..... . ..... ..... . ..... . ..... . ..... . ..... . ..... . ..... . . . . . 1198 5. Summary and perspectives. . . . . . . . . . . . . . . ..... . ..... . ..... . ..... . ..... . ..... ..... . ..... . ..... . ..... . ..... . ..... . ..... . . . . . 1198 CRediT authorship contribution statement . . ..... . ..... . ..... . ..... . ..... . ..... ..... . ..... . ..... . ..... . ..... . ..... . ..... . . . . 1200 Declaration of Competing Interest . . . . . . . . . ..... . ..... . ..... . ..... . ..... . ..... ..... . ..... . ..... . ..... . ..... . ..... . ..... . . . . 1200 Funding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... . ..... . ..... . ..... . ..... . ..... ..... . ..... . ..... . ..... . ..... . ..... . ..... . . . . 1200 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..... . ..... . ..... . ..... . ..... . ..... ..... . ..... . ..... . ..... . ..... . ..... . ..... . . . . 1200 1. Background consist of the key fibril protein, their fragments, and several other components, for example, serum amyloid P-component (SAP) and An increasing number of people suffer from neurodegenerative heparin sulfate proteoglycan [4], which may or may not interact diseases and other amyloid-based diseases or the amyloidoses, such with the amyloid proteins, plus possible co-aggregates with other as Alzheimer’s Disease, Parkinson’s disease, amyotrophic lateral amyloid proteins [12,13]. This makes it harder to extrapolate con­ sclerosis, and prion diseases [1]. These lead to immeasurable emo­ clusions from isolated in vitro studies to physiological manifesta­ tional, social, and economic costs for the patients, their direct en­ tions. This also can explain to a large extent why none of the vironment and societies as a whole. Despite the enormous efforts developed drugs that target single amyloid proteins have been from academia and industry, no disease-modifying drugs exist and successful so far in curing amyloid-based diseases safely and effec­ treatment is limited to the management of symptoms or slowing tively, although a few may have the potential to slow down the down the progression [2]. These amyloidoses are pathologically progression when used in the early phases of the diseases [15]. manifested by the abnormal deposition of amyloid proteins in cells, It is now understood that there are many factors and multiple tissue or organs, of which the shapes and normal function are al­ molecules and pathways involved in the progression of diseases, tered [1]. There are around 30 different types of amyloid-based including pro-inflammatory and anti-inflammatory [16]. Likewise, diseases, which can be classified into localized and systemic forms, many amyloid diseases can be caused by multiple risk factors, such including various peripheral and central variations of these dis­ as multiple genetic, environmental and lifestyle factors. One of the orders, and many with different subgroups within each disease [1]. It most important genetically related compounds that can interact has been suggested that the diversity in normal functions of the with amyloid proteins is Apolipoprotein E (ApoE). ApoE is a key related native proteins requires highly dynamic structures that regulator of lipid homeostasis and the principal cholesterol carrier in comes at the cost of a higher risk to cleavage and misfolding, which the brain [17]. It is well known for its influence on Alzheimer's and is even further increased by protein mutations and post-transla­ cardiovascular diseases [18]. It has also been identified as one of the tional modifications [3]. two most robust gene loci for longevity [19], and therefore ApoE may The basic structure of all amyloids is the fibril, which is built up be a crucial factor in many other diseases. by twisted protofilaments that are stacks of protein layers in a β- In particular, ApoE polymorphism has a profound effect on sheet structure [4]. Amyloid proteins can convert stepwise from multiple molecular and cellular functions in the human body, some soluble monomers into oligomers and then into the insoluble β- of which we will discuss in this paper. There are two kinds of ApoE sheet fibrils [5]. There are several functional and pathological amy­ polymorphism, one genetically determined and one not [18]. Sia­ loids. In pathology, amyloids manifest the diseases, in which oligo­ lyated isoforms are nongenetic post-translational forms with an mers may have the most toxic properties [6], even though in nature additional negative charge and account for 10–20 % of plasma ApoE oligomeric proteins are also the most common structural form of [20]. There are three common genetic Apolipoprotein E isoforms functional proteins [7,8]. Some amyloid oligomers increase the per­ (ApoE2, ApoE3, and ApoE4) expressed by three alleles (ɛ2, ɛ3, and ɛ4) meability of cellular membranes that leads to cellular dysfunctions that allow for six phenotypes in humans: three homozygous phe­ [6]. Oligomers can also trigger other toxic pathways, like mi­ notypes (E4/4, E3/3, and E2/2) and three heterozygous phenotypes tochondrial dysfunction, synaptic dysregulation and oxidative stress (E4/3, E3/2, and E4/2) [18]. ε4, ε3, and ε2 represent the human po­ [6,9]. However, some disease-associated amyloid proteins like Aβ pulation of about 14 %, 78 %, and 8 %, respectively [21], and show a can, sometimes, display protective functions, depending on their geographical variation [21]. ApoE ε4 is considered to be the ancestral location, duration and concentration and the microscopic environ­ allele [19]. The evolution of this allele to other alleles is presumed mental conditions [10,11]. This suggests that amyloid proteins can be due to varying environmental and lifestyle conditions that lead to both protective and damaging, and amyloid homeostasis requires adaptation of diet and altered susceptibility to infections [22]. tight regulation which involves the production, post-translational Within ApoE, a low-density lipoprotein receptors (LDLR)-binding maintenance, aggregation, degradation and clearance [9]. This also site is located at the N-terminal domain, connected by the hinge to means we need to understand better how protein homeostasis, or the C-terminal domain with a lipid-binding site (see also Fig. 1 for a proteostasis, is maintained and how it is disturbed and which factors schematic representation). The three common ApoE isoforms differ can be modified. from each other at only two residues: 112 and 158 (ApoE2: 112Cys/ The study of amyloid aggregates and their roles in diseases 158Cys, ApoE3: 112Cys/158Arg, ApoE4: 112Arg/158Arg) [17,22]. Even proved extremely difficult and much of our understanding comes though multiple interactions between the N- and C-terminal do­ from in vitro studies performed under non-physiological conditions. mains are found to be the same for all apoE isoforms, the single Cys/ The formation and function of fibrils in vivo can differ profoundly Arg interchange at sites 112 and 158 does seem to slightly influence from in vitro fibrils formed from the same recombinant protein [4], the interaction between the two domains interaction in ApoE. The while studying the metastable oligomeric intermediates is even unique intrinsically disordered regions formed by hydrogen-bonds more challenging [6,11]. In addition, misfolding and aggregation can and salt-bridges in ApoE4 [17,23] have been shown also to occur in not only be seeded by the same precursor molecules but also cross- ApoE3, and these hydrogen-bonds and salt-bridges have been sug­ seeded by different protein monomers. This can alter the kinetics gested to be critical for the reversibility of the lipoprotein-binding and toxicity, possibly resulting in diseases with “mixed” pathologies activity of ApoE [23,24]. Moreover, it was also suggested that the of increased severity and faster progression, and in other cases Cys/Arg interchange affects the lipid binding and lipidation state modulate or inhibit amyloid aggregation and fibrillization or toxicity (lipid-free, partially lipidated or lipid-rich) of ApoE, which is a more [12–14]. Moreover, it is now accepted that all amyloid deposits important factor in the domain movement, binding affinity, kinetics, 1190 R.A. Loch, H. Wang, A. Perálvarez-Marín et al. Computational and Structural Biotechnology Journal 21 (2023) 1189–1204 associated with total plasma ApoE levels [33]. The 112Arg substitu­ tion in ApoE4 modulates the inter-domain interaction, which can increase the affinity of ApoE4 with lipids and ApoE receptors [23]. ApoE4 is less stable and more susceptible to proteolysis. This pro­ teolysis can lead to the degradation of ApoE4 into small neurotoxic segments, which may eventually interfere with the normal functions of ApoE4 [34]. Compared with ApoE3 and ApoE2, ApoE4 and its proteolytic, neurotoxic C-terminal truncated fragments are stronger associated with dysfunctions. A study conducted with transgenic mice expressing ApoE3 or ApoE4 in neurons or astrocytes showed that the C-terminal-truncated fragments of neuro-specific ApoE4 are related to increased tau phosphorylation [34]. Compared to ApoE2 and ApoE3, ApoE4 more efficiently induces APP transcription and amyloid-β (Aβ) production, facilitates Aβ fibrillation, interferes with cleaning up Aβ, and utilizes more lipid binding [17]. Therefore, ApoE4 is considered ‘worse’ or less protective than ApoE2 and ApoE3 for patients at the risk for neurodegenerative diseases. Both ApoE and amyloid proteins play important roles in amy­ loidoses as they form two main risk factors for these diseases Amyloid and ApoE proteins possess intrinsically disordered regions. Fig. 1. Schematic illustration of the 3D molecular structures of the ApoE 3 structure These disordered regions control structural motions and interactions (pdb ID: 2L7b) and the most studied amyloid proteins that can interact with each other: Aβ (pdb ID: 1BA4), prion protein (pdb ID: 1QLX), superoxide dismutase 1 of these proteins with lipid membranes, and thereby their func­ (SOD1) (pdb ID: 3ECU), α-syn protein (pdb ID: 1XQ8), Islet amyloid polypeptide tionality. Furthermore, amyloid and ApoE proteins interact with each precursor (Amylin) (pdb ID: 2L86), TAR DNA binding protein 43 (TDP-43) (pdb ID: other. These connections between amyloid and ApoE proteins are 5X4F). The residues 112 and 158 distinguish the ApoE isoforms and are shown in predominantly associated with amyloid deposition, with ApoE pro­ yellow spheres. ApoE2 has Cys112 and Cys158 residues; ApoE3 has Cys112 and Arg158 residues; and ApoE4 has Arg112 and Arg158 residues. The N-terminal domain (or­ tein being able to directly modulate the structure and nucleation of ange) and C-terminal domain (gray) of ApoE are connected by a hinge domain (red amyloid proteins. The function and toxicity of ApoE and amyloid ribbon). proteins like Aβa tau and α-synuclein can be also influenced by their interactions in amyloidoses [35]. Though there are many therapeutic receptor interactions and consequently in the multiple functionality approaches based on targeting the amyloid protein alone [2], no of ApoE [23,24]. For example, the stability of the N-terminal domain effective therapies have been established so far for these amyloid is isoform-dependent (ApoE2 > ApoE3 > ApoE4), and ApoE4 has diseases. the greatest tendency to form a molten globule state (ApoE4 > > One reason why there is a lack of progress in treatments, comes ApoE3 > ApoE2) [17]. This results in profound differences in struc­ from the fact that the molecular and genetic cross-interactions be­ ture and function, and consequently affects the disease risks in tween ApoE and various amyloids have not been clearly interpreted. carriers. Furthermore, apolipoproteins possess amphipathic α-he­ Therefore, the main focus of this review paper is to better under­ lices, which are optimised for the functional (and protective) lipid stand the cross-interactions between ApoE and different kinds of surface binding, however, when unbound to lipids, this highly dy­ amyloid proteins, where we will discuss how ApoE interacts with the namic region also has a low thermodynamic stability and high generation, aggregation, and clearance of amyloid proteins, and how proteolysis, molten globular and amyloid-forming propensity [3]. amyloids affect the intracellular fate of ApoE (Section 2), as well as The ApoE protein is considered a major risk factor for some amyloid- where they meet (Section 3). related diseases, including Alzheimer’s disease (AD) [21], Parkinson’s disease (PD) [25], Transmissible spongiform encephalopathies (TSEs) 2. Cross-interactions of ApoE with amyloid proteins [26], type 2 diabetes (T2D) [27]. In more details, ApoE is a main component of lipoproteins in In this section, we discuss specifically the molecular and genetic plasma and the predominant apolipoprotein in the central nervous interactions of ApoE with the most studied amyloid proteins, in­ system (CNS) [28]. Peripheral and central nervous system ApoE are cluding Aβ, tau, alpha-synuclein (α-Syn), prion, amylin/Islet amyloid synthesized mainly by liver cells and macrophages, and secreted polypeptide (IAPP), superoxide dismutase 1 (SOD1), transactive re­ mainly by astrocytes and microglia [28]. It plays a crucial role in the sponse DNA-binding protein 43 (TDP-43) and a few others. A sche­ transportation and metabolism of lipids, cerebrovascular integrity matic illustration of their molecular structures is shown in Fig. 1. and cerebral blood flow, amyloid clearance, longevity and other functions [19,21,29,30]. ApoE mediates the binding of lipoproteins to 2.1. Interactions between ApoE and the Aβ peptide ApoE receptors [17]. The ApoE receptors include heparin sulfate proteoglycans (HSPGs), low-density lipoprotein receptors (LDLR), Alzheimer’s Disease (AD) is the most common form of dementia and LDL-receptor related proteins (LRP), with the latter two be­ and the Aβ42 and Aβ40 peptides are the most abundant components longing to the LDLR family [17,23]. in the amyloid plaques manifested in the brains of AD patients [36]. ApoE3 is linked to ‘normal’ plasma lipid levels because of its high In cerebral amyloid angiopathy (CAA), Aβ accumulates and deposits efficiency in promoting the clearance of triglyceride (TG)-rich lipo­ on the wall of cerebral vessels. More than 80% of autopsy-confirmed proteins [31]. ApoE2 is restricted in the clearance of TG-rich lipo­ AD cases have some degree of CAA [37,38]. It is generally accepted protein remnant particles due to its impaired ability to bind to the that the formation of Aβ oligomers and aggregates (plaques) is an LDL receptor [32]. In one study including subjects with either integrated part of AD pathophysiology [39]. These plaques mainly APOEɛ3/ɛ3 or APOEɛ4/ɛ4 genotypes, plasma triglycerides were found appear as relatively neutral ‘diffuse’ plaques or disruptive ‘neuritic’ to be associated with an ApoE isoform-dependent distribution of plaques that mostly have a fibrillary or dense-cored morphology monomers, homodimers and heterodimers, showing a difference [40,41]. However, the total amyloid-plaque load does not always between AD patients and controls [33]. In the same study, ApoE correlate well with cognitive impairment [42]. This might be due to dimers were only observed in the APOEɛ3-carriers and were the difference in initial inflammatory responses between different 1191 R.A. Loch, H. Wang, A. Perálvarez-Marín et al. Computational and Structural Biotechnology Journal 21 (2023) 1189–1204 case groups, especially the response induced by activated glial cells B103 cells transfected with human wild-type APP695 (B103-APP) [43]. Nonetheless, the misfolding of Aβ proteins is a key feature in AD [54] proposed that lipid-poor ApoE4 enhances higher Aβ production and CAA and is impacted by the type of ApoE lipoproteins, as dis­ from APP compared with lipid-poor ApoE3, which was mediated cussed next. through the low density lipoprotein receptor-related protein (LRP) The inheritance of ApoE ε4 is related to late-onset familial and pathway. In another in vitro study using human neurons induced sporadic AD [43], and ApoE ε4 increases the risk of developing CAA from embryonic stem cells and in vivo experiments in mice [55], and AD, while ApoE ε2 lowers the risk of frontotemporal dementia ApoE differentially (ApoE4 > ApoE3 > ApoE2) stimulates APP tran­ (FTD) in a very small cohort [44]. ApoE ε2 appears to act as a pro­ scription and Aβ secretion by triggering a non-canonical MAP kinase tective factor against AD and Lewy bodies (DLB) [44]. Reiman et al. cascade by binding to ApoE receptors. In addition, a recent study found that among the six genotypes of three alleles, ApoE2/2 is re­ suggested that reconstituted ApoE isoforms increase BACE1 levels lated to a low odds ratios of Alzheimer’s dementia in comparison to and promote Aβ production and oligomerization in human neuro­ other genotypes, while ApoE4/4 has the highest odds [45]. Ad­ blastoma SK-NSH cells with distinct efficiency (ApoE4 ≥ ApoE3 > ditionally, in a case report, the ApoE3 Christchurch homozygote was ApoE2) [56]. The cleavage of APP by γ-secretase and then the pro­ reported to resist autosomal major Alzheimer’s disease [46]. With duction of the APP intracellular domain (AICD) in the CNS can at­ ApoE ε4 as a main genetic risk factor for AD, the cross-interaction tenuate the expression and function of lipoprotein receptor (LRP1) between ApoE and Aβ has a major contribution to the progression of [57]. A reduced LRP1 level leads to an increased ApoE level that CAA and AD. could further increase APP transcription as mentioned above, al­ ApoE is one of many proteins that can interact with Aβ, and the though reduced LRP1 also reduces the amyloidogenic processing of interaction of ApoE with Aβ can affect various stages of Aβ home­ APP. Meanwhile, a lower LRP1 level can also decrease the cholesterol ostasis as discussed below. However, the binding strength/affinity of level within the CNS, and result in the loss of neuronal membrane ApoE/Aβ complexes varies depending on many factors such as the cholesterol that was also found to enhance the Aβ production, sug­ isoforms, lipidation state, oxidation and concentration of ApoE, as gesting a feed-forward loop [57]. The combined results can explain well as the length, concentration, morphology and source (native or to a certain degree the relationship between APP, ApoE, and AD. synthetic) of Aβ, in addition to the local environmental conditions As for the effect of ApoE on Aβ aggregation, substoichiometric like receptor concentrations, chaperone concentration, metal ions concentrations of ApoE dramatically inhibit Aβ fibril growth via and electrolyte concentration, pH, temperature, molecular crowding, binding to and stabilizing oligomers, while higher concentrations of inflammasome, and other factors that can affect proteostasis ApoE can also bind to and stabilize Aβ fibrils directly [58]. In addi­ [47–49]. For instance, recombinant ApoE2, E3 and E4 preferably bind tion, ApoE4 is the most efficient in influencing the fibrillation of to Aβ peptides containing more β-sheet structures with a binding Aβ40 while ApoE3 and ApoE2 function less efficiently [58]. Besides, affinity of 20 nM in vitro [50], and lipid-free ApoE binds stronger to ApoE3 binds preferentially to Aβ42 oligomers, leading to the in­ immobilized Aβ than lipidated ApoE does [51]. Dissimilarities in hibition of the Aβ42-induced neurotoxicity, which has been shown these stringent conditions, especially the extent of ApoE lipidation to be ApoE-isoform dependent (ApoE3 > ApoE4) [59]. This isoform- and posttranslational modifications, can, to a large extent, explain dependent effect of ApoE might arise from the differential binding the seemingly substantially contradictory results from in vitro and in affinities of the ApoE isoforms to different aggregation intermediates vivo studies. This is because different lipid-binding states can lead to or from the altered lipidation states of ApoE in the brain, which is considerable conformational differences in ApoE and the presence of regulated by ATP binding cassette transporter A1 (ABCA1) [47]. Thus, lipids can alter the binding properties of ApoE with other proteins, the destructive role of ApoE4 in AD could be partly due to the sta­ like Aβ [52]. bilization of the soluble cytotoxic oligomeric intermediates, as well We will further discuss how ApoE affects the metabolism, ag­ as the insoluble fibrils. Fragments of ApoE and other apolipoproteins, gregation, degradation and clearance of Aβ peptides. The production such as ApoA-I (the major compound in HDL) and Apo-J, may also of Aβ involves a series of sequential endoproteolytic cleavage by β- play simultaneously a protective and toxic role in Aβ aggrega­ and γ-secretases (schematically illustrated in Fig. 2) of the trans­ tion [60,61]. membrane protein, amyloid precursor protein (APP), through an In addition to the effects of ApoE, especially ApoE4, on the ki­ anti-trophic amyloidogenic pathway [53]. The influence of ApoE on netics of Aβ oligomerization and fibrillization process, Aβ in turn also Aβ production remains controversial and seems to be strongly de­ affects the oligomerization, aggregation as well as neurotoxicity of pendent on the experimental conditions [37]. Despite the fact that ApoE4. When co-incubated in vitro in human neuroblastoma SK-N- several studies [37] reported no apparent ApoE isoform-dependent SH cells, Aβ42 increases the formation of SDS-stable ApoE4 oligo­ effect on APP, one in vitro study conducted with rat neuroblastoma mers and the cytotoxicity of ApoE4 [62]. These data highlight that Fig. 2. Roles of ApoE isoforms in the processing of APP and Aβ regulations [78]: APP is initially cleaved by α-secretase in the extracellular part near the plasma membrane to generate a soluble protein and a membrane standing part. Metalloprotease of the ADAM family (e.g. ADAM10) possesses α-secretase activity. This pathway is also known as a beneficial pathway. APP can also be cleaved by β-secretase and γ-secretase leading to Aβ and soluble intra- and extracellular proteins (amyloidogenic pathway). β-secretase is an aspartic acid protease (also known as BACE1). γ-secretase is an integral membrane protein composed of PSEN1 (presenilin-1), nicastrin, APH-1 (anterior pharynx-defective 1), and PEN-2 (presenilin enhancer 2). The influence of ApoE isoforms in further processing Aβ is shown on the right side (see main text for details). 1192 R.A. Loch, H. Wang, A. Perálvarez-Marín et al. Computational and Structural Biotechnology Journal 21 (2023) 1189–1204 ApoE4 could be pathogenic due to its capacity of oligomer assem­ minimal: the uptake and subsequent degradation of Aβ in astrocytes bling that is increased by Aβ42 and leads to higher neurotoxicity. are blocked by ApoE isoforms that compete for the same cellular ApoE and its fragments can also positively or negatively affect the receptors, transporters, and/or cellular/membrane surfaces, espe­ degradation and clearance processes of Aβ, depending on the lipi­ cially the LRP1-dependent cellular uptake pathway in astrocytes. dation state of ApoE [63,64]. The promoting effect of ApoE on the Thus, the role of ApoE in the clearance of Aβ from CNS is highly cellular degradation of Aβ has been reported by in vitro studies [65], complex and remains yet to be resolved and may be profoundly and the impaired degradation may subsequently deteriorate further different in vivo from in vitro. However, this controversy has not clearance, or the other way around. There are several mechanisms prevented the use of ApoE-derived peptides and ApoE/Aβ antibodies underlying the clearance of Aβ from the brain. Of these, the receptor- as therapeutic agents for AD treatment [76,77]. Interpreting the re­ mediated clearance via microglia/astrocytes or through the blood- sults of all these studies combined, we can say little more than that brain barrier (BBB), and the clearance through the interstitial fluid for healthy functioning and homeostasis there exist an optimal (ISF) drainage play the dominant roles [66]. ApoE affects the Aβ concentration and an optimal level of lipidation of ApoE, which clearance in different ways and is dependent on several factors. appear to be isoform dependent. Different Aβ levels can be transported across the BBB in an ApoE- isoform and -lipidation state dependent manner [67]. In vitro, 2.2. Interactions between ApoE and Alzheimer’s disease-associated tau human astrocytes preferably take up oligomeric Aβ over fibrillar Aβ [68], while human microglia more efficiently take up fibrillar Aβ Tau is mainly a neuronal protein existing in axons and is mostly compared with astrocytes[69]. These uptake processes in microglia associated with the stabilization of microtubules therein. Tau can be and astrocytes are negatively affected by ApoE and ApoJ [69]. ApoE modified by phosphorylation, the state of which will in turn influ­ receptor-mediated clearance of Aβ generally occurs in microglia and ence the distribution of it. Tau tends to be phosphorylated in the astrocytes, where soluble Aβ can bind directly to LRP1 of astrocytes cytosol while dephosphorylated in the distal region of the axon [79]. in competition with ApoE isoforms [66,70]. Mouse studies suggest Furthermore, the phosphorylation state differs according to the de­ that the clearance of Aβ from ISF into the blood can be facilitated by velopment stages of tauopathic diseases and healthy states. The p-glycoprotein (Pgp) and LPR1 expressed by the vascular endothelial extent of phosphorylation is higher in fetal neurons and decreases cells [71–73]. However, ApoE4, compared to ApoE2 and ApoE3, during the development period, but increases hugely in pathological shows a relatively low affinity for LRP1 and is therefore less efficient situations (tauopathies), which may be related to the normal and in promoting Aβ excretion through the BBB [30,67,71]. Furthermore, abnormal functions of tau [80]. Under normal physiology conditions, APOE is critical for maintaining cerebrovascular integrity, as ApoE- tau binds to microtubules, as well as lipid membranes, to maintain deficient mice show more vascular defects and the BBB is broken the normal shape and function of neurons. However, under patho­ down more, even before neuronal and synaptic changes occur, al­ logical conditions, tau phosphorylation is high, leading to the re­ though these effects occur in ApoE4 mice. Thus it seems that ApoE4 duced affinity for microtubules and subsequently cytoskeleton is less protective than ApoE3 and ApoE2, and the BBB of ApoE4 is destabilization, particularly in neurons. The detached, hyperpho­ consequently more susceptible to, for example, injury [30]. This is sphorylated tau then undergoes self-aggregation, forming oligomers consistent with the observation that (in mice) ApoE2, ApoE3, Aβ/ and various higher-ordered aggregates, which are involved in AD ApoE2 and Aβ/ApoE3 complexes were cleared at the BBB via both and some other tauopathies [79]. In fact, AD patients usually have LRP1 and VLDLR (very low density lipoprotein receptors) at a faster more hyperphosphorylated brain tau [79]. Mild cognitive impair­ rate than Aβ/ApoE4 complexes that was cleared mainly via VLDLR ment (MCI) patients who are ApoE4 carriers also have high levels of alone [67]. In addition, lipidation slowed down the clearance of all CSF tau and bear the elevated risk for AD [81]. However, CSF ApoE the ApoE isoforms and ApoE/Aβ complexes at the BBB by a factor of levels do not show a consistent association with AD and MCI among 2–3 [67]. On the contrary, ApoE4-containing lipoproteins are less different studies, although CSF ApoE levels might be more closely lipidated than ApoE3 [47], which might lead to the less stable correlated with tau, phosphorylated tau, and Aβ42 in AD and MCI ApoE4/Aβ complexes, resulting in reduced ApoE4/Aβ levels, de­ patients than plasma ApoE levels [82]. In addition, it has been re­ creased clearance, and increased accumulation of Aβ, especially in ported that ApoE4 significantly aggravated tau-mediated neurode­ oligomeric form. An additional degree of complication in the inter­ generation in a tauopathy mouse model [83]. In an earlier study, a C- action of ApoE/Aβ and LRP1 is that the significantly reduced LRP1 terminal truncated ApoE form, the ApoE4 [Δ(272–299)] in neurons, levels in the vascular endothelial cells and pericytes of AD patients was found to induce the intracellular neurofibrillary tangles NFT-like are related to a noticeable decrease in the ApoE-dependent removal inclusions via interacting with neuronal phosphorylated tau as well of soluble [71–73]. This is consistent with the above-mentioned as phosphorylated neurofilaments [34], leading to mitochondrial observation that the formation of Aβ is accompanied by the APP dysfunction and a 35 % greater cell death than full-length ApoE4 processing product, the APP intracellular domain that suppresses the [84]. These studies indicate the role of ApoE in influencing tau pa­ expression of LRP1. Alternatively, Zekonyte et al. proposed that the thogenesis, in which ApoE-induced neuroinflammation is involved. efficiency of perivascular clearance of Aβ is reduced by the presence As a major genetic risk factor for AD, ApoE exerts an isoform- of ApoE4 and enhanced in the presence of ApoE3 [74]. This is be­ specific effect on the hyperphosphorylation of tau and the neuronal cause the attachment of ApoE4/Aβ complexes to the arterial base­ cytoskeleton. ApoE3 binds in vitro stably to the microtubule-binding ment membranes is weaker than that of ApoE3/Aβ complexes [74]. repeat region of tau, thus stabilizing it and inhibiting the formation In addition, the attachment of ApoE/Aβ complexes to the arterial of paired helical filaments (PHF) [85]. This interaction can be in­ basement membranes again strongly depends on the type of Aβ and hibited by the phosphorylation of tau [86], suggesting that ApoE3 the ApoE lipidation status [47], which may result from the strong may protect tau by preventing hyperphosphorylation in vivo [85]. In destabilization of the salt bridge network of ApoE4 by Aβ [75]. As a contrast to the protective role of ApoE3, ApoE4 shows reduced result of these many interactions, compared with ApoE2 and ApoE3, binding to tau. This may signify a loss of protective function when ApoE4 correlates with higher Aβ levels in cerebrospinal fluid (CSF) stimulating tau phosphorylation in a neuron-specific way, resulting and cerebral vasculature. This can be used as a biomarker in the in higher amounts of unbound tau in the CSF [87,88]. This is con­ preclinical phase of AD. sistent with the observation in AD and MCI patients as mentioned Despite the fact that the ApoE/Aβ interaction has been in­ above. vestigated in vitro, the direct molecular interaction between ApoE In vitro studies show that ApoE regulates tau phosphorylation and soluble Aβ under physiological conditions has been shown to be partly due to the modulation of the activities of tau phosphatases 1193 R.A. Loch, H. Wang, A. Perálvarez-Marín et al. Computational and Structural Biotechnology Journal 21 (2023) 1189–1204 and kinases. Dephosphorylation of (synthetic) tau was the result of ApoE4 carriers is higher in familial PD than that in sporadic PD [102]. ApoE affecting Ca2+-associated signal transduction pathways that Although not every person with ApoE4 develops PD and many PD enhance the activity of protein phosphatases 2A/2B (PP2A/2B) [89]. patients possess one of the other alleles, there appears an increased This effect is more pronounced in ApoE3 and ApoE2 than in ApoE4. risk for ApoE ε4 carriers to develop PD and they might have an Oppositely, ApoE4 enhances tau phosphorylation by promoting the earlier age of the disease onset as compared to ApoE ε3 carriers. In activation of zinc-induced extracellular signal-regulated kinase (Erk) addition, PD patients carrying ApoE ε4 might have a greater risk to in neurons [90], while there is also a reduction in PP2A activity by develop additional dementia, in comparison to non-ApoE4-carriers. ApoE4 compared with ApoE3 via epigenetic mechanisms [91]. An­ Contrary to AD, in PD patients the ApoE ε2 allele frequency is also other study suggested that ApoE controls tau phosphorylation by the higher than that in controls, and thus ApoE ε2 allele can also be assembly of the ApoE receptor/disabled-1/glycogen synthase kinase- considered as a risk factor for developing PD [103,104]. This was 3β (RAD) complex, which in turn suppresses the kinase activity [92]. attributed to a disturbance of the ApoE homeostasis and then the Furthermore, ApoE-deficient mice have hyperphosphorylated tau in subsequent dysfunction of the protein. the brain compared to wild-type controls, and show diminished As mentioned above, α-Syn acts as a lipid binding acceptor and, repair after a closed head injury that is related to transient hyper­ thus, the cross talks between α-Syn, ApoE, lipoproteins, and lipids phosphorylation [93]. Interestingly, some researchers even argued are relevant because they mutually influence both lipid (especially that only at a high concentration (2 μM), ApoE reduced the levels of cholesterol) metabolism and α-Syn properties such as the uptake tau kinases and then the levels of tau phosphorylation in primary and the tendency for aggregation [105,106]. These cross talks heavily neurons [94]. This concentration dependency of ApoE on tau illus­ depend on several other factors like the concentration and lipidation trates that there can be significant differences in ApoE-tau interac­ state of ApoE as well. Namely, it has been reported that α-Syn tions under in vivo and in vitro conditions which have to be taken strongly stimulates the outflow of cholesterol efflux [107], while a into account. It also suggests that there is an optimal concentration higher (optimal) level of plasma cholesterol is related to a lower risk of ApoE to regulate phosphorylation that is isoform dependent. of PD [108,109]. In both incidental LB disease (iLBD) and sympto­ However, there is insufficient data available on the effect of the li­ matic stages of PD, the up-regulation of ApoE and LRP1 in the pidation state of ApoE and tau phosphorylation, or any other post­ melanized neurons of the substantia nigra (SN) is associated with translational modifications of ApoE. alterations in lipoprotein homeostasis, presumably as an early event In contrast to the effect of ApoE on tau, tau also influences the in the process of PD development [110], possibly in an attempt to intracellular fate of ApoE in an ApoE isoform-specific manner. When prevent aggregation of α-Syn. An in vitro study showed that ApoE transiently expressing tau and LDLR in mammalian cells, ApoE is influences the aggregation of α-Syn in a concentration-dependent taken up from the CSF, but only ApoE3, not ApoE4, reaches the cy­ way [107]: at a low concentration of 15 nM, all three ApoE isoforms toplasm [72]. Another study showed that especially non-lipidated promote the aggregation of α-Syn where ApoE4 shows the strongest ApoE2 could form a complex with tau in-vitro [69]. This highlights efficacy. At the higher concentrations (> 15 nM), the isoforms sup­ the isoform- and lipidation-dependent cross talk between ApoE press the aggregation of α-Syn. This study suggests that the con­ and tau. centration and isoforms of ApoE significantly influence the kinetics In summary, the phosphorylation state of tau plays crucial roles of α-Syn and thus ApoE can play both protective and toxic roles. under both physiological and pathological conditions. ApoE influ­ However, how this converts to in vivo concentrations is unclear. A ences the phosphorylation state of tau in an isoform-specific way, molecular link between α-Syn, ApoE and Aβ has also been revealed with ApoE3 binding stably to tau while ApoE4 binding less effi­ in a study with transgenic mice expressing either wild type or mu­ ciently. This results in the inhibitory effect of ApoE3 and promoting tant α-Syn [111]. It was shown that increased levels of ApoE are in­ effect of ApoE4 on tau phosphorylation, influencing the cytoskeleton jurious, leading to α-Syn-induced neurodegeneration, as well as the stability and the formation of NFTs in AD. elevated accumulation of insoluble mouse Aβ. This study indicates a common pathogenic mechanism of ApoE in various neurodegen­ 2.3. Interactions between ApoE and Parkinson’s disease-related α-Syn erative diseases, including AD and PD. Although detailed mechan­ isms are still lacking, through the cross interaction between ApoE, α- α-synuclein (α-Syn) is mainly a neuronal protein in the brain with syn, and lipids (especially cholesterol), we can say that ApoE influ­ multiple functions, and it can also be generated in the intestine and ences the uptake and aggregation of α-Syn as well as the lipid me­ transported through the vagus nerve to the brain [95]. Located pri­ tabolism, and there seems to be an optimal level for ApoE since both marily in the pre-synapses, α-Syn is involved in the neurotransmitter a too low and a too high concentration is associated with increased release by interacting with lipid rafts. This interaction may be en­ amyloid aggregation. hanced by the high curvature of synaptic vesicles and the property of α-Syn to promote and sense membrane curvature. However, over­ 2.4. Interactions between ApoE and TSE-related prion expression of α-Syn inhibits exocytosis from the synaptic vesicles [95,96]. Besides the role in the maintenance of synaptic vesicles, α- The aggregation of the pathological isoform of the prion protein Syn may also function as a lipid acceptor, since it shares biophysical (PrP) in the CNS is the main hallmark of a group of neurodegen­ similarities with other lipid-carrying proteins [97]. erative disorders termed Transmissible spongiform en­ α-Syn is an intrinsically disordered protein that is prone to ag­ cephalopathies (TSEs), including Creutzfeldt–Jakob disease (CJD) gregate into intracellular inclusions [98]. Overexpression and mis­ [112]. The ApoE level is higher in the sporadic Creutzfeldt-Jakob folding of α-Syn as well as mutations within the α-Syn gene ( SNCA) disease (sCJD) cases [113]. ApoE co-occurs and co-deposits with PrP have been linked to Parkinson’s disease (PD), characterized by Lewy in TSE infected brains [114]. Furthermore, ApoE was co-purified with bodies (LB) [98,99]. PrP identified by proteomic analysis[112,113]. Biophysical and bio­ Even though it has been shown that ApoE4 exacerbates α-Syn chemical studies revealed that the recombinant PrP protein interacts pathology [25], the role of ApoE on PD pathogenesis remains to be with both recombinant ApoE and native ApoE from hamster liver further explored. The level of ApoE in CSF is higher in patients with tissue in vitro [115]. This ApoE-PrP interaction has been mapped to early PD, and α-Syn was observed to colocalize with ApoE on lipo­ the N-terminus of both proteins, namely, the residues 1–194 of ApoE protein particles [100]. In addition, the frequency of ApoE ε4 carriers and the residues 23–90 of PrP [115]. Aβ42 and ApoE were identified is higher in both PD patients with dementia (PDD) than that in in the majority of the sCJD [113] and the ε4 allele is a risk factor for healthy controls [101]. And among PD patients, the frequency of CJD, while ε2 delays the occurrence of death [26]. A meta-analysis 1194 R.A. Loch, H. Wang, A. Perálvarez-Marín et al. Computational and Structural Biotechnology Journal 21 (2023) 1189–1204 showed no significant evidence between ApoE2 and CJD, although degradation due to isoform-specific (apoE2 > apoE4 > apoE3) gly­ the ApoE 3/4 and ApoE 4/4 genotypes are risk factors for CJD, while cation in hyperglycemic diabetic patients [133,134]. ApoE 3/3 genotype may protect against CJD [116,117]. A synergistic age-dependent interaction in AD and sCJD has been observed be­ 2.6. Interactions between ApoE and ALS-related SOD1 tween ApoE and the prion protein gene (PRNP) [118]. A recent study suggested that AD/primary age-related tauopathy (AD/PART) and SOD1 is an important antioxidant enzyme that can internalize CJD pathology can occur in the same brain [119]. Still, ApoE corre­ the superoxide (O2−) into relatively nontoxic molecules [135]. There lated only to Aβ pathology and did not show to influence the age of are three known forms of SOD in humans, in which mutations in the onset and subtypes of CJD, indicating that the synergistic mechan­ gene encoding copper, zinc superoxide dismutase 1 (Cu, Zn-SOD1) isms of AD and CJD remain to be better understood. The findings of are associated with amyotrophic lateral sclerosis (ALS) [136]. ALS is a the few aforementioned studies suggest that ApoE4 does play a common motor neuron disease, and mutations in SOD1 gene are predominant role in TSEs but the exact mechanisms remain to be present in 2–6 % of all ALS patients [136]. further clarified. It has been reported that ApoE is also involved in ALS [137]. In familial ALS-linked transgenic mice, ApoE expression is strongly 2.5. Interactions between ApoE and T2D-associated IAPP upregulated in the spinal cord of end-stage disease mice with SOD1 (G93A-SOD1) mutation [137]. This upregulation is closely linked to Amylin, also called Islet amyloid polypeptide (IAPP), is cleaved the time and location of both the axonal and neuronal degeneration, from the pro-islet amyloid polypeptide (proIAPP) and co-secreted as well as the glial activation and the activation of genes in mi­ with insulin from the pancreatic β-cells. IAPP, on the one hand, plays tochondrial destruction [138,139]. This is consistent with other various functional roles physiologically (in glucose homeostasis, studies in which ApoE ε4 allele is associated with earlier onset [137] glucagon release, gastric emptying and antimicrobial activity and reduced life span in ALS patients [140]. This means that ApoE4 is [120,121]). On the other hand, it can transform from soluble, func­ more toxic or less protective in ALS. In agreement with it, ApoE ε2 tional states into highly organized amyloid deposits that contribute has been reported to delay the onset for 3 years compared to that of to the death of pancreatic β-cells and thus act as one of the most non-ApoE ε2 carriers [141]. Because of the limited number of important hallmarks of type 2 diabetes (T2D), together with insulin available studies, the role of ApoE in ALS and the ApoE interaction resistance and deficiency [120–125]. It is worth mentioning that with SOD1 remains to be further investigated. lifestyle and environmental factors are accepted as the main causes for developing T2D [123,126], while T2D doubles the risk for de­ 2.7. Interactions between ApoE and TDP-43 veloping dementia and AD, and a subgroup among AD associated with insulin resistance has been dubbed type 3 diabetes [124]. In TDP-43 is a 414–amino acid RNA-binding protein involved in addition to the aforementioned cross interactions between Aβ, tau, several cellular processes [142,143]. In addition, during cellular α-Syn, or prion and ApoE, an association between ApoE and IAPP also stress, TDP-43 enters into the cytoplasm as a component of insoluble exists. In particular, the colocalization of ApoE and IAPP suggests aggregates with a ubiquitinated, hyper-phosphorylated and cleaved that ApoE plays an important role in the development of multiple form of TDP-43 with a higher molecular mass at ∼ 45 kDa [144]. The amyloid deposits [127]. However, it remains to be clarified whether latter form is the major disease protein in some sporadic and familial the role is protective, toxic or a loss of protection. ALS, most cases of frontotemporal dementia (FTLD-TDP) [144], and ApoE was found not to be crucial for the formation of IAPP in AD and some PD patients [145–147]. Moreover, it was shown to be amyloid fibrils in a transgenic mouse model of T2D [128], unlike the associated with memory loss and progressive hippocampal atrophy case in AD. However, a clinical study with T2D patients from China [148]. Recently, a strong association was reported between ApoE4 indicated that ApoE ε4 enhances islet amyloidosis [129]. Further­ and TDP-43 [149]. In particular, the ApoE4 isoform and TDP-43 ap­ more, it has been shown in human samples and AD mouse models pear to synergistically affect the cognitive impairment of AD pa­ that ApoE4 attenuates the IAPP-assisted clearance of Aβ40 from the tients, independent of Aβ. In a study with AD subjects with and AD brain into blood [130], suggesting the risk factor of IAPP and without TDP-43 having similar degrees of AD pathology, especially ApoE4 interaction in AD. The effect of ApoE on T2D may be related to during the early phases of neurodegeneration, while normal cogni­ the modulation of IAPP aggregation in a concentration-dependent tion was strongly associated with the absence of TDP-43 [150]. TDP- manner. ApoE and the incompletely processed proIAPP promote the 43 was also found to be co-localized with tau, and both amyloids are fibrillation of IAPP and coexist with the amyloid [120]. The fibrilla­ present in a number of diseases [151]. tion of IAPP can be inhibited at a low concentration of ApoE4 but is Furthermore, abnormal levels of TDP-43 were observed in the promoted at a high concentration of ApoE4 [131]. This implies that majority of patients with AD, ALS, and FTLD, and there is a sub­ under healthy physiological conditions ApoE4 may prevent IAPP stantial increase (200 %) in TDP-43 levels in cortical autopsies of late aggregation by efficiently binding and sequestering IAPP, however, stage AD patients [150,152]. It was suggested that overexpression of the enhanced binding affinity between ApoE4 and IAPP in T2D re­ TDP-43 might be harmful and induce neurodegeneration, possibly sults in the critical accumulation of IAPP and subsequent islet with or without formation of insoluble protein aggregates. The amyloid formation [132]. Like the interaction between ApoE and Aβ mutant TDP-43 (A315T, G348C and A382T) may cause disease via [50], the interaction between ApoE and IAPP partially depends on both loss-of-positive-function and gain-of-toxicity [153]. The loss of the state of ApoE (concentration and lipidation). These studies functional TDP-43 may disrupt key nuclear functions, resulting in suggest again that the cross interactions of ApoE and amyloid pro­ the loss of cellular homeostasis, transcriptional deregulation, disin­ teins presumably rely on specific (largely unknown) microscopic tegration of nuclear bodies, or aberrant messenger RNA splicing conditions that differ between healthy and sick people. In addition, [142,144]. Expression of Aβ42 (independent of tau) was associated cholesterol homeostasis is strongly related to ApoE, and recent with TDP-43 phosphorylation, up-regulation, cleavage and accu­ studies suggest that gangliosides and cholesterol affect hIAPP- mulation in the cytosol, while α-Syn does not seem to affect TDP-43 membrane interaction and the uptake and clearance of hIAPP much [152]. TDP-43 oligomers can also cross-seed amyloid oligo­ [133,134]. This implies an indirect role of ApoE in IAPP clearance, mers with Aβ [154]. which requires further investigation. Further research is also re­ Furthermore, there exists an indirect link between ApoE4, TDP- quired to understand investigation remains to be understood for the 43 and Aβ. The depletion, aggregation, as well as the disease asso­ indirect effect of impaired ApoE functionality, stability and ciated mutation of TDP‐43 can disturb intracellular sorting and 1195 R.A. Loch, H. Wang, A. Perálvarez-Marín et al. Computational and Structural Biotechnology Journal 21 (2023) 1189–1204 the functioning of Aβ. TTR levels in the CSF of AD patients are de­ creased and even more among ApoE ε4 allele carriers. Reduction in CSF TTR levels leads to decreased levels of CSF Aβ42 and increased levels of senile plaque (SP) abundance in the AD brain [160]. Due to the facts that ApoE induces the accumulation of Aβ and that TTR facilitates the transport of Aβ from the brain and prevents fibril formation [161], ApoE and TTR may act competitively in the ag­ gregation of Aβ and its deposition in the SP of AD brain. ApoE was also found to be co-located with β2m in amyloid de­ positions [162]. Furthermore, ApoE can enhance β2m fibrillogenesis processes by stabilizing preformed seeds [163,164], and then stabi­ lize β2m fibrils by inhibiting its depolymerization [165]. The ex­ istence of ApoE induces AA amyloid deposition in a mouse model [166]. Another study employing ApoE-deficient mice with or without injecting adenoviruses expressing the human ApoE3 gene showed that ApoE (ApoE3) promotes the deposition of AA [167]. Fig. 3. Diagram illustrating the interactions between ApoE and tau, α-Syn, prion, IAPP, SOD1, and TDP-43. For details, please refer to Sections 2.2–2.7. The question marker 3. Potential interaction mechanism between ApoE isoform and means that even though most studies agree, a consensus has not been reached in the amyloid protein whole field. Specific disease-related amyloid proteins are labeled with the same color as the labels with their disease abbreviation. To discriminate the different interaction mechanisms of ApoE isoforms with amyloid proteins, we propose the energy landscapes subsequent activity-dependent secretion of the neurotrophin brain- of ApoE isoform protein folding and amyloid aggregation in Fig. 5. derived neurotrophic factor (BDNF) [155]. ApoE4 subjects were The energy landscapes describe the relative stabilities of different found to have significantly lower serum BDNF levels [156]. The in­ aggregated and folded states. Different energy states are linked to terruption of BDNF signaling in hippocampal neurons was found to structural, dynamical and functional differences. Intrinsically dis­ rapidly activate the amyloidogenic pathway for Aβ production, ordered amyloid proteins have a high degree of entropy and free causing neuronal apoptotic death [53]. ApoE4 subjects were found to energy and will transform to lower free energy states by aggregating, have significantly lower serum BDNF levels [132], and a BDNF re­ usually via transient intermediates. These transient states co-ag­ duction contributes to lower cognitive performance in both AD gregate to each other and form oligomers and then fibrils. This is apathetic patients and female ApoE4 carriers [157] and non-de­ schematically illustrated in Fig. 5A. In the process of protein folding, mented subjects [158]. However, how these processes are linked to the number of conformational states reduces and, hence, the entropy ApoE is currently unknown at the molecular level. A schematic and the free energy as well in Fig. 5C. Intramolecular interactions summary of the above-mentioned amyloid proteins and their in­ lead to protein folding toward the natively soluble state. This com­ teractions with ApoE is provided in Fig. 3. petes with intermolecular interactions in Fig. 5B that can lead to different amyloid states, which can be disordered (amorphous ag­ 2.8. Interactions between ApoE and other amyloid proteins gregates) or highly ordered (polymorphic amyloid fibrils) and are less soluble due to exposure of hydrophobic amino acid residues. In addition to the interactions with the aforementioned amyloid The three ApoE isoforms, on the other hand, also have multiple proteins, ApoE also interacts with other amyloids such as trans­ intermediate states at different temperatures with varying thermal thyretin (TTR), β2-microglobulin (β2m), Prolactin, and Amyloid A stability, according to molecular dynamics simulations [168]. ApoE4 (AA), as schematically illustrated in Fig. 4. However, there is much has a lower thermal stability as compared to ApoE2 and ApoE3. It less data available on these interactions, and further research is re­ suggests that, despite minimal differences in the primary structures, quired to draw any conclusions. the differences in physiological behaviour among the isoforms can The co-localization of ApoE and TTR has been reported in the be explained by the difference in folding mechanisms related to the systemic organs except for the liver and spleen of an aged male minimal mutations in the amino acid sequences. In particular, all of vervet monkey especially in the extracellular stroma among muscle the isoforms could be characterized by different intermediate states fibers and external tunica of arterioles [159]. Both ApoE and TTR are for structural rearrangement upon binding events. They identified a Aβ sequestering proteins forming a stable compound and affecting unique, more compact and stable ApoE4 misfolded intermediate state with increased inter-domain interactions. This more stable intermediate state was speculated to change the interaction with lipids and lipoprotein and thereby modify the lipid transport effi­ ciency and affect the kinetics of aggregation and clearance me­ chanisms of amyloid proteins. Other studies also suggest that ApoE4 has a higher population of a more chemically and thermally stable intermediate in the form of a flexible molten globule state that exposes the hydrophobic core due to a partial opening of the four-helix bundle [169,170]. It was also suggested that the open conformation of the intermediate is similar to the conformation change of ApoE once it binds to lipids. These stable intermediate molten globules have nearly native structural features and are responsible for the stronger interdomain FRET sig­ nals of ApoE4 in mildly denaturing conditions, in particular at a low pH, and even under native conditions as opposed to ApoE3 [169]. Molten globule states have several physiological roles. They have Fig. 4. Interactions between ApoE and other amyloid proteins. increased affinity for membranes and bind to liposomes and 1196 R.A. Loch, H. Wang, A. Perálvarez-Marín et al. Computational and Structural Biotechnology Journal 21 (2023) 1189–1204 Fig. 5. Proposed mechanism of the interaction between amyloid proteins and ApoE isoforms. A, C). In amyloid protein aggregation, monomeric proteins search low-energy states and usually deposit as amyloid fibrils through transit and high-entropy oligomers. Compared to ApoE2 and ApoE3, ApoE4 is likely to fold as a molten globule state where the N- and C-terminal domains unfold with higher entropy. B). Alternatively, both ApoE4 and amyloid protein, with high entropy and co-occurrence in the brain, are prone to cross reaction for seeking a lower entropy and free energy. Due to the higher entropy state, the interaction may be more easily modulated by other factors, like lipidation discussed in this review. phospholipids, although they are also known to be prone to mis­ cellular reservoirs and are associated with ApoE and neurodegen­ folding, aggregation, post-translational modifications and proteo­ erative diseases [175]. Farmer et al. found several differences in li­ lytic cleavage [170]. We also agree that ApoE4 molten globules quid droplets formation and fatty acid metabolism with astrocytes within lysosomes and endosomes may act as a possible route for the expressing ApoE4 or ApoE3. Phase-separated droplets or organelles commonly observed pathological features in neurodegenerative apparently offer ideal mesoscopic locations for cross-interactions of diseases. ApoE with amyloid proteins, although how the interactions in the To summarize, besides the higher susceptibility of ApoE4 to form droplets take place remains to be investigated. toxic fragments, the formation of a higher amount of more stable intermediate molten globules with ApoE4 could be an important factor to explain the differences in the interactions between ApoE 4.2. Lipid membranes isoforms and amyloid proteins as described in Section 2. Another important factor to take into account, is the loci of cross-interac­ Playing a key role in lipid metabolism, ApoE binds and redis­ tions, which will briefly be discussed in the next section. tributes lipids, especially from astrocytes in the CNS to other types of cells, including neurons [176]. Likewise, most amyloid proteins ex­ perience conformational changes upon binding lipid membranes 4. Loci of cross interactions and self-propagate through a lipid membrane-enriched environ­ ment such as membrane boundaries, cellular compartments and 4.1. Phase-separated droplets tissues [12,14,177] Lipid induced conformational changes of amyloid proteins, on the other hand, modulate amyloid protein aggregation Through phase separation, an emerging paradigm underlies in­ and toxicity, which varies with lipid composition and metal ions tracellular processes, such as amyloid protein aggregation in neu­ [13]. The co-occurrence, accumulation and interaction of amyloid rodegenerative diseases [171]. Phase separation enhances the proteins and ApoE in/on lipid membranes or the membrane concentration of amyloid proteins and promotes their self-assembly, boundaries seem highly plausible at the microscopic level. The af­ typically in membrane-less organelles. For example, full-length tau finity of ApoE and its signal peptide toward membranes depends on proteins (tau441-GFP) were observed to form droplets in cultured the lipid composition of the membrane, and it was shown that the neurons, with more rapid droplet formation in the presence of he­ ApoE signal peptide binds more tightly to membranes with low parin and RNA, or under higher temperature or phosphorylation cholesterol concentrations, such as found in the endoplasmic re­ [172]. The C-terminal domain of TDP-43 undergoes phase separation ticulum (ER) [178]. Further research is needed in this direction. The into droplets, with slower dynamics or faster aggregation in all fact that molten globule intermediates have a higher affinity for lipid disease mutants [173]. A series of other amyloid proteins have been membranes and that such a state is more readily formed with found to form droplets summarized in the review paper by Elbaum- ApoE4, as discussed above, emphasizes the need for further research Garfinkle [174]. Similar phase-separation droplets, lipid droplets in this direction. (LDs), surrounded by a layer of lipids and proteins, act as inert 1197 R.A. Loch, H. Wang, A. Perálvarez-Marín et al. Computational and Structural Biotechnology Journal 21 (2023) 1189–1204 4.3. Cholesterol and lipidation Aβ and whether cholesterol is a causal or non-causal factor for AD or both remains uncertain [198], since cholesterol may be both dele­ The disturbance of lipid-related molecules and biological pro­ terious (when its homeostasis is perturbed) and favorable (when cesses are deeply involved in neurodegenerative diseases, especially this homeostasis is maintained). A study in astrocyte-conditioned AD [179,180]. There exist an optimum level for cholesterol as both media suggested that ER–mitochondrial communication and MAM too low and too high cholesterol are recognised risk factors for AD, (mitochondria-associated ER membranes) function, which is re­ resulting in pro-inflammatory responses and increased Aβ and tau sponsible for lipid regulation and cholesterol trafficking, is upregu­ amyloids, as reviewed by Gupta and Weaver [181]. With molecular lated with lipoproteins containing ApoE4 as compared to ApoE3, dynamics simulations of model neuronal membranes, they showed while no difference was observed with lipid-free ApoE [199]. An­ that too high cholesterol concentrations make the membrane less other study has reported that it is the transfer of cholesterol from the permeable by compacting and stiffening it, and that too low cho­ cytofacial leaflet to the exofacial leaflet (from inside to outside the lesterol results in compromised membrane integrity via higher cell), not the change in total cholesterol concentration, that pro­ molecular disorder and tilting, and increased fluidity and inter­ motes APP and Aβ expressions [200]. Besides, this transfer could digitation of the leaflet lipids [181]. The cholesterol concentration disrupt membrane structure including lipid rafts and different kinds then also affects the lipid rafts and the proteins associated with it of cell functions including disruption in cholesterol homeostasis [182], which are also the most common modulators for both amyloid [200]. This is consistent with the fact that both increasing age and and ApoE proteins. Cholesterol is an important constituent of the the inheritance of the ApoE4 allele can lead to the doubling of plasma membrane. As a molecular spacer, it can stabilize lipid rafts cholesterol levels in the exofacial leaflet [200]. On the other hand, by occupying the space between raft proteins and other raft lipids cholesterol impacts the lateral organization of lipid membranes with [183]. There is an interplay between raft proteins and raft lipids. On condensation effect and therefore induces the interaction of amyloid the one hand, the regulation of cholesterol homeostasis can lead to protein and ApoE protein in lipid draft or lipidated complex. These the dissociation, dysregulation, and/or inactivation of raft proteins. suggest that lipid-related modulators, such as cholesterol, are in­ On the other hand, raft proteins including amyloids can influence volved in the interactions between ApoE and amyloid proteins, cholesterol homeostasis. The cholesterol/amyloids interplay is en­ perhaps occurring in lipid raft and lipidated complex. hanced in the cholesterol-rich brain [184,185], and includes the in­ To explain the interactions of ApoE and amyloid proteins in the teraction of cholesterol with tau [186,187], prion [188], Aβ and presence of lipids, a theoretical model can be used. Assuming a [184,189] and also plays a role in other amyloids such as IAPP and coupling, amyloid aggregates are stabilized through intramolecular depends on the type and consistency of lipid membranes [123]. and intermolecular interactions [201]. Intramolecular interaction Measuring lipids alterations in lipid rafts is even suggested as a occurs among amyloid or ApoE protein, due to their disordered and potential biomarker for neurodegenerative diseases. Interestingly, flexible conformations. These conformational states are affected by the amyloidogenic processing of APP takes place in lipid rafts while ordered alpha-helix/β-sheet and disordered turn/coil structures the non-amyloidogenic processing predominantly happens in the [202]. The conformational interchanges may minimize the total free non-raft regions [184]. energy, by coupling to intermolecular assembly. Once amyloid pro­ ApoE, together with other apolipoproteins, lipoprotein receptors, tein interacts with ApoE, the conformational interchanges and in­ and lipid transporters, is also important for the maintenance of termolecular assembly spontaneously emerge to reach a lower total cholesterol homeostasis in the brain [190]. Generally, ApoE is lipi­ free energy than ApoE or amyloid protein alone. Furthermore, a dated, and lipid association can enhance the affinity of ApoE for the lipid-enriched environment converts the intramolecular conforma­ LDL receptors via the lipid binding-induced conformational changes tions of ApoE and amyloid proteins with more ordered alpha-helix/ [191]. Lipidated ApoE is mostly produced by microglia and astro­ β-sheet structures and lower energy [23,203]. This may explain why cytes, illustrated in Fig. 5. It can bind to soluble Aβ in an isoform- amyloid protein interaction with ApoE happens and is prone to the dependent manner (ApoE2 > ApoE3 > ApoE4), facilitating Aβ de­ lipid-related loci mentioned above. gradation via receptor-mediated endocytosis in various brain cells or clearance via the BBB [184]. In contrast to this differential binding affinity of lipidated ApoE for soluble Aβ, ApoE4 is less lipidated than 5. Summary and perspectives E3 and E2. Furthermore, ApoE4 cannot efficiently protect neuronal cells by assisting liquid droplet formation in glia in response to ApoE interacts with amyloid proteins in multiple ways, as de­ elevated reactive oxygen species (ROS) [192,193]. In addition, lipid- scribed in this paper, and these interactions play an important role in poor ApoE is more prone to cleavage and misfolding [3]. Together the pathogenesis of several neurodegenerative diseases. Although these factors indicate the intensive involvement of ApoE4 and in­ not every person with ApoE4 develops neurodegenerative diseases creased risk in AD and another amyloidosis. However, the effect of and many patients with neurodegenerative diseases possess one of cholesterol on the Aβ processing pathway may be both harmful and the other ApoE alleles, there clearly appears an increased risk for favorable, probably due to the promotion effect of high cholesterol ApoE ε4 carriers. Therefore, the modulation of ApoE structure, ex­ levels within lipid rafts on Aβ generation (harmful) as well as de­ pression, lipidation and function could be utilized for developing gradation or clearance (favorable) [184], suggesting the importance novel therapeutic approaches for degenerative diseases. More spe­ of maintaining lipid homeostasis (Fig. 6). cifically, ApoE-based therapies are currently being investigated, As mentioned before, α-Syn potently stimulates cholesterol efflux mainly in the treatment for AD, with major focus on: a) the use of in a dose- and time-dependent manner via ABCA1 transporter [194], small molecule structural correctors that are able to convert the which regulates the lipidation state of ApoE [195]. Tau pathology ApoE4 structure and function into those of ApoE3 [204,205]; b) the promotes the level of cellular unesterified cholesterol in neurons overexpression of protective ApoE2, by adeno-associated virus de­ that is independent of apolipoprotein E deficiency. Synergically, a livery in brain [206] and the downregulation of ApoE4, by ApoE4 loss of cholesterol homeostasis may simulate tau pathology [186]. antisense oligonucleotides [207]; c) the inhibition of the binding Obviously, interactions between cholesterol, ApoE and amyloid between ApoE and Aβ, by using antibodies against Aβ targeting the proteins like Aβ [196], tau [197] and α-Syn occur in various neuro­ ApoE binding site or antibodies against ApoE targeting the Aβ degenerative disorders, and lipid rafts may provide an important site binding site [76]; d) increase of ApoE lipidation via ABCA1 [208]. It for this kind of interaction. However, whether lipid raft is a specific should be noted that the overexpression of ApoE2 may not be a mechanism on how manipulation of cholesterol levels modifies APP/ reasonable approach as ApoE2 has been associated with the disorder 1198 R.A. Loch, H. Wang, A. Perálvarez-Marín et al. Computational and Structural Biotechnology Journal 21 (2023) 1189–1204 Fig. 6. Secretion of lipidated ApoE by microglia and astrocytes and complex formation with extracellular and intracellular amyloid proteins. type III hyperlipoproteinemia and with both increased and de­ Why there occurs a loss of homeostasis in amyloid diseases in the creased risk for atherosclerosis and PD. first place remains to be further investigated. There is an increasing The study of ApoE and amyloid proteins and their roles in dis­ awareness that neurodegenerative disorders including AD and PD eases proved extremely difficult mainly because the structures of are multifactorial diseases [210,211]. The majority of complex and these proteins are very heterogeneous and their aggregation and chronic diseases are the result of the interplay between multiple function are prone to profound changes under physiological and (risk) factors, which can be categorized in genetic, environmental pathological conditions. Additionally, there are multiple in vivo and lifestyle factors. Together they influence the microscopic and components involved that are often left out in many studies of ApoE dynamic environment of cells and their (dys)function, including the and amyloid proteins. Furthermore, many studies ignore or fail to (mal)functioning of the identified target molecules (amyloid pro­ report specific parameters, such as the lipidation, sialylation, oxi­ teins) related to specific diseases. This awareness has resulted in an dation or glycation state of ApoE, which makes it more difficult to increasing interest in systems biology, personalized medicine and compare results and draw consistent conclusions. Despite com­ multimodal therapy [212], because any therapy for multifactorial bining all the studies, we are still not able to conclude what is the diseases that focuses only on a single factor has a low chance of isoform-dependent optimal range of ApoE concentration and lipi­ success. We therefore suggest that some of the aforementioned dation state. However, we elaborated that the isoform-dependent ApoE-based therapies might be more successful when combined lipidation state of ApoE and its intermediates affect the binding to with other therapies or by addressing other known risk factors (e.g., cholesterol, which in turn differently affects the production, stabi­ endogenous factors) that are most likely being ignored in mono­ lization, degradation, clearance, binding and transport of different therapies. amyloids in various aggregation states (monomers, oligomers or fi­ ApoE-based therapies for amyloidosis show great therapeutic brils). As one of the strongest genetic risk factors for neurodegen­ potential in this field, especially after the failure of therapies by erative diseases, ApoE4 affects normal brain function in mice and targeting only amyloids. However, developing an ApoE-based humans long before obvious pathological changes are noticeable therapy targeting only the pathology-related function is challenging. [209]. ApoE4 is less stable and (its fragments) more toxic/less pro­ On the one hand, ApoE has various functions, on the other hand, tective in comparison with ApoE2 and ApoE3, although more pro­ ApoE is closely interwoven with amyloid proteins involved in various tective/less toxic compared to the ApoE knocked-out mice. Because amyloidosis. Even though confronted with challenges, there are not every person with ApoE4 develops diseases, it is most likely that important points that can be taken into account when developing all ApoE isoforms are highly functional within homeostasis. The ApoE-based therapies. Firstly, try to understand as much as possible detrimental effects of ApoE can be subscribed to either the loss of the exact normal functions of ApoE and their roles in amyloidosis, protective function or the gain of toxic effect under disturbed and and to restore the balance of ApoE. Secondly, learn more about the physiologically stressed conditions. And this change in functions of cross-interactions between ApoE and amyloids and the stress con­ ApoE will occur when changes in the concentration, morphology ditions that ApoE is involved in, which is the main purpose of this and/or modification levels of ApoE are too dramatic. Besides, it is review. Thirdly, recognize the discrepancies in the conditions of the probable that compared to ApoE3 and ApoE2, ApoE4 seems more in vitro and in vivo studies, as well as the differences between an­ susceptible and shows a stronger negative response to an induced imal models and clinical trials, when interpreting the investigation loss of homeostasis. ApoE4 is prone to formation of high entropic results. Otherwise, the intended therapeutic effects may not be and partially folded structures [17]. This enhanced susceptibility of achieved and unintended negative consequences might occur. ApoE4 can be induced mainly by other components, like amyloid Moreover, the connection between ApoE/amyloid proteins and proteins and can be considered as an important factor for the un­ chronic inflammation, oxidative stress and metal homeostasis derstanding of neurodegeneration as well as the development of [16,78,124] as well as the microbiome–gut–brain axis (which is also potential therapeutics. ApoE-dependent) are strongly associated with neurodegenerative 1199 R.A. Loch, H. Wang, A. Perálvarez-Marín et al. Computational and Structural Biotechnology Journal 21 (2023) 1189–1204 diseases [213], which are other complex factors in most diseases that [16] Furman D, Campisi J, Verdin E, Carrera-Bastos P, Targ S, Franceschi C, et al. go beyond the scope of this paper. Chronic inflammation in the etiology of disease across the life span. Nat Med 2019;25:1822–32. https://doi.org/10.1038/s41591-019-0675-0 [17] Zhong N, Weisgraber KH. Understanding the association of apolipoprotein E4 CRediT authorship contribution statement with Alzheimer disease: clues from its structure. J Biol Chem 2009;284:6027–31. https://doi.org/10.1074/jbc.R800009200 [18] Mahley RW, Rall SC. Apolipoprotein E: far more than a lipid transport protein. Rolf Antonie Loch: Writing – original draft. Hongzhi Wang: Annu Rev Genom Hum Genet 2000;1:507–37. https://doi.org/10.1146/annurev. Writing – original draft. Alex Perálvarez Marín: Writing – review & genom.1.1.507 editing. Philipp Berger: Writing – review & editing. Henrietta [19] Sebastiani P, Gurinovich A, Nygaard M, Sasaki T, Sweigart B, Bae H, et al. APOE alleles and extreme human longevity. J Gerontol - Ser A Biol Sci Med Sci Nielsen: Writing – review & editing. Angeliki Chroni: Writing – 2019;74:44–51. https://doi.org/10.1093/gerona/gly174 review & editing. Jinghui Luo: Conceptualization, Investigation, [20] Moon HJ, Haroutunian V, Zhao L. Human apolipoprotein E isoforms are dif­ Supervision, Writing – review & editing. ferentially sialylated and the sialic acid moiety in ApoE2 attenuates ApoE2-Aβ interaction and Aβ fibrillation. Neurobiol Dis 2022;164:105631https://doi.org/ 10.1016/j.nbd.2022.105631 Declaration of Competing Interest [21] Emrani S, Arain HA, DeMarshall C, Nuriel T. APOE4 is associated with cognitive and pathological heterogeneity in patients with Alzheimer’s disease: a sys­ The authors declare that they have no known competing fi­ tematic review. Alzheimer’s Res Ther 2020;12:1–19. https://doi.org/10.1186/ s13195-020-00712-4 nancial interests or personal relationships that could have appeared [22] Yassine HN, Finch CE. APOE alleles and diet in brain aging and Alzheimer’s to influence the work reported in this paper. disease. Front Aging Neurosci 2020;12:1–18. https://doi.org/10.3389/fnagi. 2020.00150 [23] Frieden C, Wang H, Ho CMW. A mechanism for lipid binding to apoE and the Funding role of intrinsically disordered regions coupled to domain-domain interactions. Proc Natl Acad Sci USA 2017;114:6292–7. https://doi.org/10.1073/pnas. We acknowledge financial support from the Paul Scherrer 1705080114 [24] Chen J, Li Q, Wang J. Topology of human apolipoprotein E3 uniquely regulates Institute, the Swiss National Scientific Foundation (310030_197626 its diverse biological functions. Proc Natl Acad Sci USA 2011;108:14813–8. to J.L.), the BrightFocus Foundation (A20201759S to J.L.), and the https://doi.org/10.1073/pnas.1106420108 Spanish Ministry of Science and Innovation and Agencia Estatal de [25] Davis AA, Inman CE, Wargel ZM, Dube U, Freeberg BM, Galluppi A, et al. APOE genotype regulates pathology and disease progression in synucleinopathy. Sci Investigación (MCIN/AEI/ 10.13039/501100011033 project PID2020- Transl Med 2020;12:1–15. https://doi.org/10.1126/scitranslmed.aay3069 120222GB-I00 to A.P.-M.). [26] Amouyel P, Vidal O, Launay JMLJ. The apolipoprotein E alleles as major sus­ ceptibility factors for Creutzfeldt-Jakob disease. Lancet 1994;344:1315–8. [27] Wu L, Zhang Y, Zhao H, Rong G, Huang P, Wang F, et al. Dissecting the asso­ References ciation of apolipoprotein E gene polymorphisms with type 2 diabetes mellitus and coronary artery disease. Front Endocrinol 2022;13:1–8. https://doi.org/10. [1] Chiti F, Dobson CM. Protein misfolding, amyloid formation, and human disease: 3389/fendo.2022.838547 a summary of progress over the last decade. Annu Rev Biochem 2017;86:27–68. [28] Fernandez CG, Hamby ME, McReynolds ML, Ray WJ. The role of apoE4 in dis­ https://doi.org/10.1146/annurev-biochem-061516-045115 rupting the homeostatic functions of astrocytes and microglia in aging and [2] Ankarcrona M, Winblad B, Monteiro C, Fearns C, Powers ET, Johansson J, et al. Alzheimer’s disease. Front Aging Neurosci 2019;10:1–18. https://doi.org/10. Current and future treatment of amyloid diseases. J Intern Med 3389/fnagi.2019.00014 2016;280:177–202. https://doi.org/10.1111/joim.12506 [29] Kockx M, Traini M, Kritharides L. Cell-specific production, secretion, and [3] Lewkowicz E, Gursky O. Dynamic protein structures in normal function and function of apolipoprotein E. J Mol Med 2018;96:361–71. https://doi.org/10. pathologic misfolding in systemic amyloidosis. Biophys Chem 1007/s00109-018-1632-y 2022;280:106699https://doi.org/10.1016/j.bpc.2021.106699 [30] Bell RD, Winkler EA, Singh I, Sagare AP, Deane R, Wu Z, et al. Apolipoprotein e [4] Benson MD, Buxbaum JN, Eisenberg DS, Merlini G, Saraiva MJM, Sekijima Y, controls cerebrovascular integrity via cyclophilin A. Nature 2012;485:512–6. et al. Amyloid nomenclature 2020: update and recommendations by the https://doi.org/10.1038/nature11087 International Society of Amyloidosis (ISA) nomenclature committee. Amyloid [31] Mahley RW, Weisgraber KH, Huang Y. Apolipoprotein E: structure determines 2020;27:217–22. https://doi.org/10.1080/13506129.2020.1835263 function, from atherosclerosis to Alzheimer’s disease to AIDS. J Lipid Res [5] Wu J, Cao C, Loch RA, Tiiman A, Luo J. Single-molecule studies of amyloid 2009;50:183–8. https://doi.org/10.1194/jlr.R800069-JLR200 proteins: from biophysical properties to diagnostic perspectives. Q Rev Biophys [32] Weisgraber KH, Innerarity TL, Mahley RW. Abnormal lipoprotein receptor- 2020;53:e12https://doi.org/10.1017/S0033583520000086 binding activity of the human E apoprotein due to cysteine-arginine inter­ [6] Wu J, Blum TB, Farrell DP, DiMaio F, Abrahams JP, Luo J. Cryo-electron micro­ change at a single site. J Biol Chem 1982;257:2518–21. https://doi.org/10.1016/ scopy imaging of Alzheimer’s amyloid-beta 42 oligomer displayed on a func­ s0021-9258(18)34954-8 tionally and structurally relevant scaffold. Angew Chem - Int Ed [33] Patra Kalicharan, Giannisis Andreas, Edlund Anna K, Sando Sigrid Botne, 2021;60:18680–7. https://doi.org/10.1002/anie.202104497 Lauridsen Camilla, Berge Guro, et al. Plasma apolipoprotein E monomer and [7] David Goodsell AO. Structural symmetry and protein function. Annu Rev dimer profile and relevance to Alzheimer’s disease. J Alzheimer’s Dis Biophys Biomol Struct 2000;29:105–53. 2019;71:1217–31. https://doi.org/10.1073/pnas.1220484110 [8] Gwyther REA, Dafydd Jones D, Worthy HL. Better together: building protein [34] Huang Y, Liu XQ, Wyss-Coray T, Brecht WJ, Sanan DA, Mahley RW. oligomers naturally and by design. Biochem Soc Trans 2019;47:1773–80. Apolipoprotein E fragments present in Alzheimer’s disease brains induce https://doi.org/10.1042/BST20190283 neurofibrillary tangle-like intracellular inclusions in neurons. Proc Natl Acad Sci [9] Chen GF, Xu TH, Yan Y, Zhou YR, Jiang Y, Melcher K, et al. Amyloid beta: USA 2001;98:8838–43. https://doi.org/10.1073/pnas.151254698 structure, biology and structure-based therapeutic development. Acta Pharm [35] Wisniewski T, Drummond E. APOE-amyloid interaction: therapeutic targets. Sin 2017;38:1205–35. https://doi.org/10.1038/aps.2017.28 Neurobiol Dis 2020:138. https://doi.org/10.1016/j.nbd.2020.104784 [10] Pearson HA, Peers C. Physiological roles for amyloid β peptides. J Physiol [36] Selkoe DJ, Hardy J. The amyloid hypothesis of Alzheimer’s disease at 25 years. 2006;575:5–10. https://doi.org/10.1113/jphysiol.2006.111203 EMBO Mol Med 2016;8:1–14. https://doi.org/10.15252/emmm.201606210 [11] Cawood EE, Karamanos TK, Wilson AJ, Radford SE. Visualizing and trapping [37] Huynh TPV, Davis AA, Ulrich JD, Holtzman DM. Apolipoprotein E and transient oligomers in amyloid assembly pathways. Biophys Chem Alzheimer’s disease: the influence of apolipoprotein E on amyloid-β and other 2021;268:106505https://doi.org/10.1016/j.bpc.2020.106505 amyloidogenic proteins. J Lipid Res 2017;58:824–36. https://doi.org/10.1194/jlr. [12] Ivanova MI, Lin Y, Lee YH, Zheng J, Ramamoorthy A. Biophysical processes un­ R075481 derlying cross-seeding in amyloid aggregation and implications in amyloid pa­ [38] Jellinger KA. Alzheimer disease and cerebrovascular pathology: an update. J thology. Biophys Chem 2021;269:106507https://doi.org/10.1016/j.bpc.2020.106507 Neural Transm 2002;109:813–36. https://doi.org/10.1007/s007020200068 [13] Nguyen PH, Ramamoorthy A, Sahoo BR, Zheng J, Faller P, Straub JE, et al. [39] Haass C, Selkoe DJ. Soluble protein oligomers in neurodegeneration: lessons Amyloid oligomers: a joint experimental/computational perspective on from the Alzheimer’s amyloid beta-peptide. Nat Rev Mol Cell Biol Alzheimer’s disease, Parkinson’s disease, type II diabetes, and amyotrophic 2007;8:101–12. https://doi.org/10.1038/nrm2101 lateral sclerosis. Chem Rev 2021;121:2545–647. https://doi.org/10.1021/acs. [40] Liu J, Costantino I, Venugopalan N, Fischetti RF, Hyman BT, Frosch MP, et al. chemrev.0c01122 Amyloid structure exhibits polymorphism on multiple length scales in human [14] Luo J, Wärmländer SKTS, Gräslund A, Abrahams JP, Cross. interactions between brain tissue. Sci Rep 2016;6:1–11. https://doi.org/10.1038/srep33079 the Alzheimer′s disease amyloid-β peptide and other amyloid proteins: a fur­ [41] Dickson TC, Vickers JC. The morphological phenotype of β-amyloid plaques and ther aspect of the amyloid cascade hypothesis. J Biol Chem 2016;291:16485–93. associated neuritic changes in Alzheimer’s disease. Neuroscience https://doi.org/10.1074/jbc.R116.714576 2001;105:99–107. https://doi.org/10.1016/S0306-4522(01)00169-5 [15] van Dyck Christopher H, Swanson Chad J, Aisen Paul, Bateman Randall J, Chen [42] Price JL, Morris JC. Tangles and plaques in nondemented aging and “preclinical” Christopher, Gee Michelle, et al. Lecanemab in early Alzheimer’s disease. N Engl Alzheimer’s disease. Ann Neurol 1999;45:358–68. https://doi.org/10.1002/ J Med 2023:9–21. 1531-8249(199903)45:3<358::AID-ANA12>3.0.CO;2-X 1200 R.A. Loch, H. Wang, A. Perálvarez-Marín et al. Computational and Structural Biotechnology Journal 21 (2023) 1189–1204 [43] Darocha-Souto B, Scotton TC, Coma M, Serrano-Pozo A, Hashimoto T, Serenó L, [65] Jiang Q, Lee CYD, Mandrekar S, Wilkinson B, Cramer P, Zelcer N, et al. ApoE et al. Brain oligomeric β-amyloid but not total amyloid plaque burden correlates promotes the proteolytic degradation of Aβ. Neuron 2008;58:681–93. https:// with neuronal loss and astrocyte inflammatory response in amyloid precursor doi.org/10.1016/j.neuron.2008.04.010 protein/tau transgenic mice. J Neuropathol Exp Neurol 2011;70:360–76. https:// [66] Wolfe CM, Fitz NF, Nam KN, Lefterov I, Koldamova R. The role of APOE and doi.org/10.1097/NEN.0b013e318217a118 TREM2 in Alzheimer′s disease—current understanding and perspectives. Int J [44] Bernardi L, Maletta RG, Tomaino C, Smirne N, Di Natale M, Perri M, et al. The Mol Sci 2019;20:65–70. https://doi.org/10.3390/ijms20010081 effects of APOE and tau gene variability on risk of frontotemporal dementia. [67] Deane R, Sagare A, Hamm K, Parisi M, Lane S, Finn MB, et al. apoE isoform- Neurobiol Aging 2006;27:702–9. https://doi.org/10.1016/j.neurobiolaging.2005. specific disruption of amyloid β peptide clearance from mouse brain. J Clin 03.008 Invest 2008;118:4002–13. https://doi.org/10.1172/JCI36663 [45] Reiman EM, Arboleda-Velasquez JF, Quiroz YT, Huentelman MJ, Beach TG, [68] Nielsen HM, Mulder SD, Beliën JAM, Musters RJP, Eikelenboom P, Veerhuis R. Caselli RJ, et al. Exceptionally low likelihood of Alzheimer’s dementia in APOE2 Astrocytic Aβ1-42 uptake is determined by Aβ-aggregation state and the pre­ homozygotes from a 5000-person neuropathological study. Nat Commun sence of amyloid-associated proteins. Glia 2010;58:1235–46. https://doi.org/10. 2020:11. https://doi.org/10.1038/s41467-019-14279-8 1002/glia.21004 [46] Arboleda-velasquez JF, Lopera F, Hare MO, Marino C, Chmielewska N, Saez- [69] Mulder SD, Nielsen HM, Blankenstein MA, Eikelenboom P, Veerhuis R. torres KL, et al. Resistance to autosomal dominant Alzheimer’s in an APOE3- Apolipoproteins E and J interfere with amyloid-beta uptake by primary human Christchurch homozygote: a case report. Nat Med 2019;25:1680–3. https://doi. astrocytes and microglia in vitro. Glia 2014;62:493–503. https://doi.org/10. org/10.1038/s41591-019-0611-3.Resistance 1002/glia.22619 [47] Tai LM, Mehra S, Shete V, Estus S, Rebeck GW, Bu G, et al. Soluble apoE/Aβ [70] Liu CC, Hu J, Zhao N, Wang J, Wang N, Cirrito JR, et al. Astrocytic LRP1 mediates complex: mechanism and therapeutic target for APOE4-induced AD risk. Mol brain Aβ clearance and impacts amyloid deposition. J Neurosci Neurodegener 2014:9. https://doi.org/10.1186/1750-1326-9-2 2017;37:4023–31. https://doi.org/10.1523/JNEUROSCI.3442-16.2017 [48] Kanekiyo T, Xu H, Bu G. ApoE and Aβ in Alzheimer’s disease: accidental en­ [71] Louveau A, Da Mesquita S, Kipnis J. Lymphatics in neurological disorders: a counters or partners? Neuron 2014;81:740–54. https://doi.org/10.1016/j. neuro-lympho-vascular component of multiple sclerosis and Alzheimer’s dis­ neuron.2014.01.045 ease? Neuron 2016;91:957–73. https://doi.org/10.1016/j.neuron.2016.08.027 [49] Luo J, Wang H, Wu J, Romankov V, Daffé N, Dreiser J. Amyloid-beta–copper [72] Erickson MA, Hansen K, Banks WA. Inflammation-induced dysfunction of the interaction studied by simultaneous nitrogen K and copper L2,3-edge soft X-ray low-density lipoprotein receptor-related protein-1 at the blood-brain barrier: absorption spectroscopy. IScience 2021;24:103465https://doi.org/10.1016/j.isci. protection by the antioxidant N-acetylcysteine. Brain Behav Immun 2021.103465 2012;26:1085–94. https://doi.org/10.1016/j.bbi.2012.07.003 [50] Golabek AA, Soto C, Vogel T, Wisniewski T. The Interaction between apolipo­ [73] Hartz AMS, Zhong Y, Wolf A, LeVine H, Miller D, Bauer B. Aβ40 reduces p-gly­ protein E and Alzheimer’s amyloid β-peptide is dependent on β-peptide con­ coprotein at the blood–brain barrier through the ubiquitin–proteasome formation. J Biol Chem 1996;271:10602–6. https://doi.org/10.1074/jbc.271.18. pathway. J Neurosci 2016;36:1930–41. https://doi.org/10.1523/JNEUROSCI. 10602 0350-15.2016 [51] Verghese PB, Castellano JM, Garai K, Wang Y, Jiang H, Shah A, et al. ApoE in­ [74] Zekonyte J, Sakai K, Nicoll JAR, Weller RO, Carare RO. Quantification of mole­ fluences amyloid-β (Aβ) clearance despite minimal apoE/Aβ association in cular interactions between ApoE, amyloid-beta (Aβ) and laminin: relevance to physiological conditions. Proc Natl Acad Sci USA 2013:110. https://doi.org/10. accumulation of Aβ in Alzheimer’s disease. Biochim Biophys Acta - Mol Basis Dis 1073/pnas.1220484110 2016;1862:1047–53. https://doi.org/10.1016/j.bbadis.2015.08.025 [52] Kim J, Basak JM, Holtzman DM. The role of apolipoprotein E in Alzheimer’s [75] Luo J, Maréchal J-D, Wärmländer S, Gräslund A, Perálvarez-Marín A. In silico disease. Neuron 2009;63:287–303. https://doi.org/10.1016/j.neuron.2009.06. analysis of the apolipoprotein E and the amyloid beta peptide interaction: 026 misfolding induced by frustration of the salt bridge network. PLoS Comput Biol [53] Matrone C, Ciotti MT, Mercanti D, Marolda R, Calissano P. NGF and BDNF sig­ 2010;6:e1000663https://doi.org/10.1371/journal.pcbi.1000663 naling control amyloidogenic route and Aβ production in hippocampal neurons. [76] Kim J, Eltorai AEM, Jiang H, Liao F, Verghese PB, Kim J, et al. Anti-apoE im­ Proc Natl Acad Sci USA 2008;105:13139–44. https://doi.org/10.1073/pnas. munotherapy inhibits amyloid accumulation in a transgenic mouse model of Aβ 0806133105 amyloidosis. J Exp Med 2012;209:2149–56. https://doi.org/10.1084/jem. [54] Ye S, Huang Y, Müllendorff K, Dong L, Giedt G, Meng EC, et al. Apolipoprotein 20121274 (apo) E4 enhances amyloid β peptide production in cultured neuronal cells: [77] Sadowski MJ, Pankiewicz J, Scholtzova H, Mehta PD, Prelli F, Quartermain D, ApoE structure as a potential therapeutic target. Proc Natl Acad Sci USA et al. Blocking the apolipoprotein E/amyloid-β interaction as a potential ther­ 2005;102:18700–5. https://doi.org/10.1073/pnas.0508693102 apeutic approach for Alzheimer’s disease. Proc Natl Acad Sci USA [55] Huang YWA, Zhou B, Wernig M, Südhof TC. ApoE2, ApoE3, and ApoE4 differ­ 2006;103:18787–92. https://doi.org/10.1073/pnas.0604011103 entially stimulate APP transcription and Aβ secretion. Cell 2017;168:427–41. [78] Dorey E, Chang N, Liu QY, Yang Z, Zhang W. Apolipoprotein E, amyloid-beta, and https://doi.org/10.1016/j.cell.2016.12.044. [.e21]. neuroinflammation in Alzheimer’s disease. Neurosci Bull 2014;30:317–30. [56] Dafnis I, Raftopoulou C, Mountaki C, Megalou E, Zannis VI, Chroni A. ApoE https://doi.org/10.1007/s12264-013-1422-z isoforms and carboxyl-terminal-truncated apoE4 forms affect neuronal BACE1 [79] Avila J, Lucas JJ, Pérez M, Hernández F. Role of Tau protein in both physiological levels and Aβ production independently of their cholesterol efflux capacity. and pathological conditions. Physiol Rev 2004;84:361–84. https://doi.org/10. Biochem J 2018;475:1839–59. https://doi.org/10.1042/BCJ20180068 1152/physrev.00024.2003 [57] Liu Q, Zerbinatti CV, Zhang J, Hoe HS, Wang B, Cole SL, et al. Amyloid precursor [80] Arendt T, Stieler JT, Holzer M. Tau and tauopathies. Brain Res Bull protein regulates brain apolipoprotein E and cholesterol metabolism through 2016;126:238–92. https://doi.org/10.1016/j.brainresbull.2016.08.018 lipoprotein receptor LRP1. Neuron 2007;56:66–78. https://doi.org/10.1016/j. [81] Herukka SK, Helisalmi S, Hallikainen M, Tervo S, Soininen H, Pirttilä T. CSF Aβ42, neuron.2007.08.008 Tau and phosphorylated Tau, APOE ε4 allele and MCI type in progressive MCI. [58] Garai K, Verghese PB, Baban B, Holtzman DM, Frieden C. The binding of apo­ Neurobiol Aging 2007;28:507–14. https://doi.org/10.1016/j.neurobiolaging. lipoprotein E to oligomers and fibrils of amyloid-β alters the kinetics of amyloid 2006.02.001 aggregation. Biochemistry 2014;53:6323–31. https://doi.org/10.1021/ [82] Martínez-Morillo E, Hansson O, Atagi Y, Bu G, Minthon L, Diamandis EP, et al. bi5008172 Total apolipoprotein E levels and specific isoform composition in cerebrospinal [59] Pankiewicz JE, Guridi M, Kim J, Asuni AA, Sanchez S, Sullivan PM, et al. Blocking fluid and plasma from Alzheimer’s disease patients and controls. Acta the apoE/Aβ interaction ameliorates Aβ-related pathology in APOE ε2 and ε4 Neuropathol 2014;127:633–43. https://doi.org/10.1007/s00401-014-1266-2 targeted replacement Alzheimer model mice. Acta Neuropathol Commun [83] Shi Y, Yamada K, Liddelow SA, Smith ST, Zhao L, Luo W, et al. ApoE4 markedly 2014;2:1–11. https://doi.org/10.1186/s40478-014-0075-0 exacerbates tau-mediated neurodegeneration in a mouse model of tauopathy. [60] Camacho J, Moliné T, Bonaterra-Pastra A, Cajal SRY, Martínez-Sáez E, Nature 2017;549:523–7. https://doi.org/10.1038/nature24016 Hernández-Guillamon M. Brain ApoA-I, ApoJ and ApoE immunodetection in [84] Chang S, Ma TR, Miranda RD, Balestra ME, Mahley RW, Huang Y. Lipid- and cerebral amyloid angiopathy. Front Neurol 2019;10:1–12. https://doi.org/10. receptor-binding regions of apolipoprotein E4 fragments act in concert to cause 3389/fneur.2019.00187 mitochondrial dysfunction and neurotoxicity. Proc Natl Acad Sci USA [61] Sahoo BR, Bekier ME, Liu Z, Kocman V, Stoddard AK, Anantharamaiah GM, et al. 2005;102:18694–9. https://doi.org/10.1073/pnas.0508254102 Structural interaction of apolipoprotein A-I mimetic peptide with amyloid-β [85] Strittmatter WJ, Weisgraber KH, Goedert M, Saunders AM, Huang D, Corder EH, generates toxic hetero-oligomers. J Mol Biol 2020;432:1020–34. https://doi.org/ et al. Hypothesis: microtubule instability and paired helical filament formation 10.1016/j.jmb.2019.12.005 in the alzheimer disease brain are related to apolipoprotein E genotype. Exp [62] Dafnis I, Argyri L, Chroni A. Amyloid-peptide β 42 enhances the oligomerization Neurol 1994;125:163–71. https://doi.org/10.1006/exnr.1994.1019 and neurotoxicity of apoE4: the C-terminal residues Leu279, Lys282 and [86] Huang DY, Weisgraber KH, Goedert M, Saunders AM, Roses AD, Strittmatter WJ. Gln284 modulate the structural and functional properties of apoE4. ApoE3 binding to tau tandem repeat I is abolished by tau serine262 phos­ Neuroscience 2018;394:144–55. https://doi.org/10.1016/j.neuroscience.2018.10. phorylation. Neurosci Lett 1995;192:209–12. https://doi.org/10.1016/0304- 026 3940(95)11649-H [63] Zhang X, Fu Z, Meng L, He M, Zhang Z. The early events that initiate β-amyloid [87] Brecht WJ, Harris FM, Chang S, Tesseur I, Yu GQ, Xu Q, et al. Neuron-specific aggregation in Alzheimer’s disease. Front Aging Neurosci 2018;10:1–13. https:// apolipoprotein E4 proteolysis is associated with increased Tau phosphorylation doi.org/10.3389/fnagi.2018.00359 in brains of transgenic mice. J Neurosci 2004;24:2527–34. https://doi.org/10. [64] Mouchard A, Boutonnet MC, Mazzocco C, Biendon N, Macrez N. ApoE-fragment/ 1523/JNEUROSCI.4315-03.2004 Aβ heteromers in the brain of patients with Alzheimer’s disease. Sci Rep [88] Harris FM, Brecht WJ, Xu Q, Tesseur I, Kekonius L, Wyss-Coray T, et al. Carboxyl- 2019;9:1–13. https://doi.org/10.1038/s41598-019-40438-4 terminal-truncated apolipoprotein E4 causes Alzheimer’s disease-like neuro­ degeneration and behavioral deficits in transgenic mice. Proc Natl Acad Sci USA 2003;100:10966–71. https://doi.org/10.1073/pnas.1434398100 1201 R.A. Loch, H. Wang, A. Perálvarez-Marín et al. Computational and Structural Biotechnology Journal 21 (2023) 1189–1204 [89] Wang XS, Luebbe P, Gruenstein E, Zemlan F. Apolipoprotein E (ApoE) peptide [114] Nakamura S, Takahashi K, Yoshikawa Y, Ono F, Hamano M, Odagiri K, et al. regulates tau phosphorylation via two different signaling pathways. J Neurosci Immunohistochemical detection of apolipoprotein E within prion-associated Res 1998;51:658–65. https://doi.org/10.1002/(SICI)1097-4547(19980301) lesions in squirrel monkey brains. Acta Neuropathol 2000;100:365–70. https:// 51:5<658::AID-JNR13>3.0.CO;2-Z doi.org/10.1007/s004010000200 [90] Harris FM, Brecht WJ, Xu Q, Mahley RW, Huang Y. Increased tau phosphor­ [115] Gao C, Lei YJ, Han J, Shi Q, Chen L, Guo Y, et al. Recombinant neural protein PrP ylation in apolipoprotein E4 transgenic mice is associated with activation of can bind with both recombinant and native apolipoprotein E in vitro. Acta extracellular signal-regulated kinase. Modulation by zinc. J Biol Chem Biochim Biophys Sin 2006;38:593–601. https://doi.org/10.1111/j.1745-7270. 2004;279:44795–801. https://doi.org/10.1074/jbc.M408127200 2006.00209.x [91] Theendakara V, Bredesen DE, Rao RV. Downregulation of protein phosphatase [116] Van Everbroeck B, Croes EA, Pals P, Dermaut B, Jansen G, Van Duijn CM, et al. 2A by apolipoprotein E: implications for Alzheimer’s disease. Mol Cell Neurosci Influence of the prion protein and the apolipoprotein E genotype on the 2017;83:83–91. https://doi.org/10.1016/j.mcn.2017.07.002 Creutzfeldt-Jakob disease phenotype. Neurosci Lett 2001;313:69–72. https:// [92] Ohkubo N, Lee YD, Morishima A, Terashima T, Kikkawa S, Tohyama M, et al. doi.org/10.1016/S0304-3940(01)02264-9 Apolipoprotein E and Reelin ligands modulate tau phosphorylation through an [117] Wei Y, Tang Y, He W, Qu Z, Zeng J, Qin C. APOE gene polymorphisms and sus­ apolipoprotein E receptor/disabled-1/glycogen synthase kinase-3beta cascade. ceptibility to Creutzfeldt-Jakob disease. J Clin Neurosci 2014;21:390–4. https:// FASEB J 2003;17:295–7. https://doi.org/10.1096/fj.02-0434fje doi.org/10.1016/j.jocn.2013.07.019 [93] Genis L, Chen Y, Shohami E, Michaelson DM. Tau hyperphosphorylation in [118] Calero O, Bullido MJ, Clarimón J, Frank-García A, Martínez-Martín P, Lleó A, et al. apolipoprotein E-deficient and control mice after closed head injury. J Neurosci Genetic cross-interaction between apoe and prnp in sporadic alzheimer’s and Res 2000;60:559–64. https://doi.org/10.1002/(SICI)1097-4547(20000515) Creutzfeldt-Jakob diseases. PLoS One 2011;6:1–9. https://doi.org/10.1371/ 60:4<559::AID-JNR15>3.0.CO;2-K journal.pone.0022090 [94] Hoe HS, Freeman J, Rebeck GW. Apolipoprotein e decreases tau kinases and [119] Rossi M, Kai H, Baiardi S, Bartoletti-Stella A, Carlà B, Zenesini C, et al. The phospho-Tau levels in primary neurons. Mol Neurodegener 2006;1:1–8. characterization of AD/PART co-pathology in CJD suggests independent pa­ https://doi.org/10.1186/1750-1326-1-18 thogenic mechanisms and no cross-seeding between misfolded Aβ and prion [95] Sulzer D, Edwards RH. The physiological role of α-synuclein and its relationship proteins. Acta Neuropathol Commun 2019;7:53. https://doi.org/10.1186/ to Parkinson’s disease. J Neurochem 2019;150:475–86. https://doi.org/10.1111/ s40478-019-0706-6 jnc.14810 [120] Akter R, Cao P, Noor H, Ridgway Z, Tu LH, Wang H, et al. Islet amyloid poly­ [96] Larsen KE, Schmitz Y, Troyer MD, Mosharov E, Dietrich P, Quazi AZ, et al. peptide: structure, function, and pathophysiology. J Diabetes Res 2016:2016. Synuclein overexpression in PC12 and chromaffin cells impairs catecholamine https://doi.org/10.1155/2016/2798269 release by interfering with a late step in exocytosis. J Neurosci [121] Cao P, Abedini A, Raleigh DP. Aggregation of islet amyloid polypeptide: from 2006;26:11915–22. https://doi.org/10.1523/JNEUROSCI.3821-06.2006 physical chemistry to cell biology. Curr Opin Struct Biol 2013;23:82–9. https:// [97] Fantini J, Carlus D, Yahi N. The fusogenic tilted peptide (67–78) of α-synuclein is doi.org/10.1016/j.sbi.2012.11.003 a cholesterol binding domain. Biochim Biophys Acta - Biomembr [122] Last NB, Miranker AD. Common mechanism unites membrane poration by 2011;1808:2343–51. https://doi.org/10.1016/j.bbamem.2011.06.017 amyloid and antimicrobial peptides. Proc Natl Acad Sci USA 2013;110:6382–7. [98] Goedert M. Alpha-synuclein and neurodegenerative diseases. Nat Rev Neurosci https://doi.org/10.1073/pnas.1219059110 2001;2:492–501. https://doi.org/10.1038/35081564 [123] Sciacca MFM, Lolicato F, Di Mauro G, Milardi D, D’Urso L, Satriano C, et al. The [99] Pals P, Lincoln S, Manning J, Heckman M, Skipper L, Hulihan M, et al. α- role of cholesterol in driving IAPP-membrane interactions. Biophys J Synuclein promoter confers susceptibility to Parkinson’s disease. Ann Neurol 2016;111:140–51. https://doi.org/10.1016/j.bpj.2016.05.050 2004;56:591–5. https://doi.org/10.1002/ana.20268 [124] Milardi D, Gazit E, Radford SE, Xu Y, Gallardo RU, Caflisch A, et al. Proteostasis of [100] Paslawski W, Zareba-Paslawska J, Zhang X, Hölzl K, Wadensten H, Shariatgorji islet amyloid polypeptide: a molecular perspective of risk factors and protec­ M, et al. α-synuclein–lipoprotein interactions and elevated ApoE level in cer­ tive strategies for type II diabetes. Chem Rev 2021;121:1845–93. https://doi. ebrospinal fluid from Parkinson’s disease patients. Proc Natl Acad Sci USA org/10.1021/acs.chemrev.0c00981 2019;116:15226–35. https://doi.org/10.1073/pnas.1821409116 [125] Luo J, Wärmländer SKTS, Gräslund A, Abrahams JP. Reciprocal molecular in­ [101] Parsian A, Racette B, Goldsmith LJ, Perlmutter JS. Parkinson’s disease and teractions between the Aβ peptide linked to Alzheimer’s disease and insulin apolipoprotein E: possible association with dementia but not age at onset. linked to diabetes mellitus type II. ACS Chem Neurosci 2016;7:269–74. https:// Genomics 2002;79:458–61. https://doi.org/10.1006/geno.2002.6707 doi.org/10.1021/acschemneuro.5b00325 [102] Blázquez L, Otaegui D, Sáenz A, Paisán-Ruiz C, Emparanza JI, Ruiz-Martinez J, [126] Kolb H, Martin S. Environmental/lifestyle factors in the pathogenesis and pre­ et al. Apolipoprotein E ε4 allele in familial and sporadic Parkinson’s disease. vention of type 2 diabetes. BMC Med 2017;15:1–11. https://doi.org/10.1186/ Neurosci Lett 2006;406:235–9. https://doi.org/10.1016/j.neulet.2006.07.037 s12916-017-0901-x [103] Pulkes T, Papsing C, Mahasirimongkol S, Busabaratana M, Kulkantrakorn K, [127] Chargé SBP, Esiri MM, Bethune CA, Hansen BC, Clark A. Apolipoprotein E is Tiamkao S. Association between apolipoprotein e genotypes and Parkinson’s associated with islet amyloid and other amyloidoses: implications for disease. J Clin Neurosci 2011;18:1333–5. https://doi.org/10.1016/j.jocn.2011.01. Alzheimer’s disease. J Pathol 1996;179:443–7. https://doi.org/10.1002/(SICI) 028 1096-9896(199608)179:4<443::AID-PATH612>3.0.CO;2-7 [104] Williams-Gray CH, Goris A, Saiki M, Foltynie T, Compston DAS, Sawcer SJ, et al. [128] Vidal J, Verchere CB, Andrikopoulos S, Wang F, Hull RL, Cnop M, et al. The effect Apolipoprotein e genotype as a risk factor for susceptibility to and dementia in of apolipoprotein E deficiency on islet amyloid deposition in human islet Parkinson’s disease. J Neurol 2009;256:493–8. https://doi.org/10.1007/s00415- amyloid polypeptide transgenic mice. Diabetologia 2003;46:71–9. https://doi. 009-0119-8 org/10.1007/s00125-002-0984-5 [105] Emamzadeh FN, Allsop D. α-Synuclein interacts with lipoproteins in plasma. J [129] Guan J, Zhao HL, Sui Y, He L, Lee HM, Lai FMM, et al. Histopathological corre­ Mol Neurosci 2017;63:165–72. https://doi.org/10.1007/s12031-017-0967-0 lations of islet amyloidosis with apolipoprotein e polymorphisms in type 2 [106] Ruipérez V, Darios F, Davletov B. Alpha-synuclein, lipids and Parkinson’s dis­ diabetic Chinese patients. Pancreas 2013;42:1129–37. https://doi.org/10.1097/ ease. Prog Lipid Res 2010;49:420–8. https://doi.org/10.1016/j.plipres.2010.05. MPA.0b013e3182965e6e 004 [130] Qiu WQ, Wallack M, Dean M, Liebson E, Mwamburi M, Zhu H. Association [107] Emamzadeh FN, Aojula H, McHugh PC, Allsop D. Effects of different isoforms of between amylin and amyloid-β peptides in plasma in the context of apolipo­ apoE on aggregation of the α-synuclein protein implicated in Parkinson’s dis­ protein E4 allele. PLoS One 2014;9:1–7. https://doi.org/10.1371/journal.pone. ease. Neurosci Lett 2016;618:146–51. https://doi.org/10.1016/j.neulet.2016.02. 0088063 042 [131] Gharibyan AL, Islam T, Pettersson N, Golchin SA, Lundgren J, Johansson G, et al. [108] De Lau LML, Koudstaal PJ, Hofman A, Breteler MMB. Serum cholesterol levels Apolipoprotein E interferes with IAPP aggregation and protects pericytes from and the risk of Parkinson’s disease. Am J Epidemiol 2006;164:998–1002. IAPP-induced toxicity. Biomolecules 2020;10:1–14. https://doi.org/10.3390/ https://doi.org/10.1093/aje/kwj283 biom10010134 [109] Huang X, Abbott RD, Petrovitch H, Mailman RB, Ross GW. Low LDL cholesterol [132] Lei P, Wu W hui, Li R wang, Ma J wen, Yu Y ping, Cui W, et al. Prevention and and increased risk of Parkinson’s disease: prospective results from Honolulu- promotion effects of apolipoprotein E4 on amylin aggregation. Biochem Asia aging study. Mov Disord 2008;23:1013–8. https://doi.org/10.1002/mds. Biophys Res Commun 2008;368:414–8. https://doi.org/10.1016/j.bbrc.2008.01. 22013 103 [110] Wilhelmus MMM, Bol JGJM, Van Haastert ES, Rozemuller AJM, Bu G, Drukarch [133] Wakabayashi M, Matsuzaki K. Ganglioside-induced amyloid formation by B, et al. Apolipoprotein E and LRP1 increase early in Parkinson’s disease pa­ human islet amyloid polypeptide in lipid rafts. FEBS Lett 2009;583:2854–8. thogenesis. Am J Pathol 2011;179:2152–6. https://doi.org/10.1016/j.ajpath.2011. https://doi.org/10.1016/j.febslet.2009.07.044 07.021 [134] Cao P, Marek P, Noor H, Patsalo V, Tu L-H, Wang H, et al. Islet amyloid: from [111] Gallardo G, Schlüter OM, Südhof TC. A molecular pathway of neurodegeneration fundamental biophysics to mechanisms of cytotoxicity. FEBS Lett linking alpha-synuclein to ApoE and Abeta peptides. Nat Neurosci 2013;587:1106–18. https://doi.org/10.1016/j.febslet.2013.01.046 2008;11:301–8. https://doi.org/10.1038/nn2058 [135] Hayyan M, Hashim MA, Alnashef IM. Superoxide ion: generation and chemical [112] Giorgi A, Di Francesco L, Principe S, Mignogna G, Sennels L, Mancone C, et al. implications. Chem Rev 2016;116:3029–85. https://doi.org/10.1021/acs. Proteomic profiling of PrP27-30-enriched preparations extracted from the brain chemrev.5b00407 of hamsters with experimental scrapie. Proteomics 2009;9:3802–14. https:// [136] PM A. Amyotrophic lateral sclerosis associated with mutations in the CuZn doi.org/10.1002/pmic.200900085 superoxide dismutase gene. Curr Neurol Neurosci Rep 2006;6:37–46. [113] Moore RA, Choi YP, Head MW, Ironside JW, Faris R, Ritchie DL, et al. Relative [137] Zetterberg H, Jacobsson J, Rosengren L, Blennow K, Andersen PM. Association of abundance of apoE and Aβ1-42 associated with abnormal prion protein differs APOE with age at onset of sporadic amyotrophic lateral sclerosis. J Neurol Sci between Creutzfeldt-Jakob disease subtypes. J Proteome Res 2016;15:4518–31. 2008;273:67–9. https://doi.org/10.1016/j.jns.2008.06.025 https://doi.org/10.1021/acs.jproteome.6b00633 1202 R.A. Loch, H. Wang, A. Perálvarez-Marín et al. Computational and Structural Biotechnology Journal 21 (2023) 1189–1204 [138] Haasdijk ED, Vlug A, Mulder MT, Jaarsma D. Increased apolipoprotein E ex­ [162] Jadoul M, Garbar C, Strihou CVY De. Pathological aspects of beta-2-micro­ pression correlates with the onset of neuronal degeneration in the spinal cord globulin amyloidosis. Semin Dial 2001;14:86–9. of G93A-SOD1 mice. Neurosci Lett 2002;335:29–33. https://doi.org/10.1016/ [163] Gejyo F, Narita I. Current clinical and pathogenetic understanding of β2-m S0304-3940(02)01159-X amyloidosis in long-term haemodialysis patients. Nephrology 2003;8:45–9. [139] Olsen MK, Roberds SL, Ellerbrock BR, Fleck TJ, McKinley DK, Gurney ME. Disease https://doi.org/10.1046/j.1440-1797.8.s.10.x mechanisms revealed by transcription profiling in SOD1-G93A transgenic [164] Myers SL, Jones S, Jahn TR, Morten IJ, Tennent GA, Hewitt EW, et al. A sys­ mouse spinal cord. Ann Neurol 2001;50:730–40. https://doi.org/10.1002/ana. tematic study of the effect of physiological factors on β2-microglobulin amyloid 1252 formation at neutral pH. Biochemistry 2006;45:2311–21. https://doi.org/10. [140] Drory VE, Birnbaum M, Korczyn AD, Chapman J. Association of APOE ε4 allele 1021/bi052434i with survival in amyotrophic lateral sclerosis. J Neurol Sci 2001;190:17–20. [165] Yamaguchi I, Hasegawa K, Takahashi N, Gejyo F, Naiki H. Apolipoprotein E in­ https://doi.org/10.1016/S0022-510X(01)00569-X hibits the depolymerization of β2-microglobulin-related amyloid fibrils at a [141] Li YJ, Pericak-Vance MA, Haines JL, Siddique N, McKenna-Yasek D, Hung WY, neutral pH. Biochemistry 2001;40:8499–507. https://doi.org/10.1021/ et al. Apolipoprotein E is associated with age at onset of amyotrophic lateral bi0027128 sclerosis. Neurogenetics 2004;5:209–13. https://doi.org/10.1007/s10048-004- [166] Hoshii Y, Kawano H, Cui D, Takeda T, Gondo T, Takahashi M, et al. Amyloid A 0193-0 protein amyloidosis induced in apolipoprotein-E-deficient mice. Am J Pathol [142] Heyburn L, Moussa CEH. TDP-43 in the spectrum of MND-FTLD pathologies. 1997;151:911–7. Mol Cell Neurosci 2017;83:46–54. https://doi.org/10.1016/j.mcn.2017.07.001 [167] Kindy MS, Rader DJ. Reduction in amyloid A amyloid formation in apolipo­ [143] Youmans KL, Wolozin B. TDP-43: a new player on the AD field. Exp Neurol protein-E-deficient mice. Am J Pathol 1998;152:1387–95. 2012;237:90–5. https://doi.org/10.1016/j.expneurol.2012.05.018 [168] Williams B, Convertino M, Das J, Dokholyan NV. ApoE4-specific misfolded in­ [144] Neumann M, Kwong LK, Sampathu DM, Trojanowski JQ, Lee VM-Y. TDP-43 termediate identified by molecular dynamics simulations. PLoS Comput Biol proteinopathy in frontotemporal lobar degeneration and amyotrophic lateral 2015;11:1–20. https://doi.org/10.1371/journal.pcbi.1004359 sclerosis. Arch Neurol 2007;64:1388. https://doi.org/10.1001/archneur.64.10. [169] Dolai S, Cherakara S, Garai K. Apolipoprotein E4 exhibits intermediates with 1388 domain interaction. Biochim Biophys Acta - Proteins Proteom [145] Lagier-Tourenne C, Polymenidou M, Cleveland DW. TDP-43 and FUS/TLS: 2020;1868:140535https://doi.org/10.1016/j.bbapap.2020.140535 emerging roles in RNA processing and neurodegeneration. Hum Mol Genet [170] Morrow JA, Hatters DM, Lu B, Höchtl P, Oberg KA, Rupp B, et al. Apolipoprotein 2010;19:46–64. https://doi.org/10.1093/hmg/ddq137 E4 forms a molten globule: a potential basis for its association with disease. J [146] Uryu K, Nakashima-Yasuda H, Forman MS, Kwong LK, Clark CM, Grossman M, Biol Chem 2002;277:50380–5. https://doi.org/10.1074/jbc.M204898200 et al. Concomitant TAR-DNA-binding protein 43 pathology is present in [171] Shorter J. Phase separation of RNA-binding proteins in physiology and disease: Alzheimer disease and corticobasal degeneration but not in other tauopathies. J an introduction to the JBC reviews thematic series. J Biol Chem Neuropathol Exp Neurol 2008;67:555–64. https://doi.org/10.1097/NEN. 2019;294:7113–4. https://doi.org/10.1074/jbc.REV119.007944 0b013e31817713b5 [172] Wegmann S, Eftekharzadeh B, Tepper K, Zoltowska KM, Bennett RE, Dujardin S, [147] Kadokura A, Yamazaki T, Lemere CA, Takatama M, Okamoto K. Regional dis­ et al. Tau protein liquid–liquid phase separation can initiate tau aggregation. tribution of TDP-43 inclusions in Alzheimer disease (AD) brains: their relation EMBO J 2018;37:1–21. https://doi.org/10.15252/embj.201798049 to AD common pathology. Orig Artic Neuropathol 2009;29:566–73. https://doi. [173] Conicella AE, Zerze GH, Mittal J, Fawzi NL. ALS mutations disrupt phase se­ org/10.1111/j.1440-1789.2009.01017.x paration mediated by α-helical structure in the TDP-43 low-complexity C- [148] Josephs KA, Dickson DW, Tosakulwong N, Weigand SD, Murray ME, Petrucelli L, terminal domain. Structure 2016;24:1537–49. https://doi.org/10.1016/j.str.2016. et al. Rates of hippocampal atrophy and presence of post-mortem TDP-43 in 07.007 patients with Alzheimer’s disease: a longitudinal retrospective study. Lancet [174] Elbaum-Garfinkle S. Matter over mind: liquid phase separation and neurode­ Neurol 2017;16:917–24. https://doi.org/10.1016/S1474-4422(17)30284-3 generation. J Biol Chem 2019;294:7160–8. https://doi.org/10.1074/jbc.REV118. [149] Yang HS, Yu L, White CC, Chibnik LB, Chhatwal JP, Sperling RA, et al. Evaluation 001188 of TDP-43 proteinopathy and hippocampal sclerosis in relation to APOE ε4 [175] Farmer B, Kluemper J, Johnson L. Apolipoprotein E4 alters astrocyte fatty acid haplotype status: a community-based cohort study. Lancet Neurol metabolism and lipid droplet formation. Cells 2019;8:182. https://doi.org/10. 2018;17:773–81. https://doi.org/10.1016/S1474-4422(18)30251-5 3390/cells8020182 [150] Josephs KA, Whitwell JL, Weigand SD, Murray ME, Tosakulwong N, Liesinger [176] Yamazaki Y, Zhao N, Caulfield TR, Liu CC, Bu G. Apolipoprotein E and Alzheimer AM, et al. TDP-43 is a key player in the clinical features associated with disease: pathobiology and targeting strategies. Nat Rev Neurol 2019;15:501–18. Alzheimer’s disease. Acta Neuropathol 2014;127:811–24. https://doi.org/10. https://doi.org/10.1038/s41582-019-0228-7 1007/s00401-014-1269-z [177] Jucker M, Walker LC. Self-propagation of pathogenic protein aggregates in [151] Chornenkyy Y, Fardo DW, Nelson PT. Tau and TDP-43 proteinopathies: kindred neurodegenerative diseases. Nature 2013;501:45–51. https://doi.org/10.1038/ pathologic cascades and genetic pleiotropy. Lab Invest 2019;99:993–1007. nature12481 https://doi.org/10.1038/s41374-019-0196-y [178] Mirdha L, Sengupta T, Chakraborty H. Lipid composition dependent binding of [152] Herman AM, Khandelwal PJ, Stanczyk BB, Rebeck GW, Moussa CEH. β-Amyloid apolipoprotein E signal peptide: importance of membrane cholesterol in pro­ triggers ALS-associated TDP-43 pathology in AD models. Brain Res tein trafficking. Biophys Chem 2022;291:106907https://doi.org/10.1016/j.bpc. 2011;1386:191–9. https://doi.org/10.1016/j.brainres.2011.02.052 2022.106907 [153] Lee EB, Lee VMY, Trojanowski JQ. Gains or losses: molecular mechanisms of [179] Schneider A, Schulz-Schaeffer W, Hartmann T, Schulz JB, Simons M. Cholesterol TDP43-mediated neurodegeneration. Nat Rev Neurosci 2012;13:38–50. https:// depletion reduces aggregation of amyloid-beta peptide in hippocampal neu­ doi.org/10.1038/nrn3121 rons. Neurobiol Dis 2006;23:573–7. https://doi.org/10.1016/j.nbd.2006.04.015 [154] Fang YS, Tsai KJ, Chang YJ, Kao P, Woods R, Kuo PH, et al. Full-length TDP-43 [180] Corraliza-Gomez M, Sanchez D, Ganfornina MD. Lipid-binding proteins in brain forms toxic amyloid oligomers that are present in frontotemporal lobar de­ health and disease. Front Neurol 2019:10. https://doi.org/10.3389/fneur.2019. mentia-TDP patients. Nat Commun 2014:5. https://doi.org/10.1038/ 01152 ncomms5824 [181] Gupta M, Weaver DF. Axonal plasma membrane-mediated toxicity of choles­ [155] Tann JY, Wong L, Sajikumar S, Ibáñez CF. Abnormal TDP‐43 function impairs terol in Alzheimer’s disease: a microsecond molecular dynamics study. Biophys activity‐dependent BDNF secretion, synaptic plasticity, and cognitive behavior Chem 2022;281:106718https://doi.org/10.1016/j.bpc.2021.106718 through altered Sortilin splicing. EMBO J 2019;38:1–18. https://doi.org/10. [182] Simons K, Vaz WLC. Model systems, lipid rafts, and cell membranes. Annu Rev 15252/embj.2018100989 Biophys Biomol Struct 2004;33:269–95. https://doi.org/10.1146/annurev. [156] Liu YH, Jiao SS, Wang YR, Bu XLe, Yao XQ, Xiang Y, et al. Associations between biophys.32.110601.141803 ApoEε4 carrier status and serum BDNF levels—new insights into the molecular [183] Browning JL. Motions and interactions of phospholipid head groups at the mechanism of ApoEε4 actions in Alzheimer’s disease. Mol Neurobiol membrane surface. 3. Dynamic properties of amine-containing head groups. 2015;51:1271–7. https://doi.org/10.1007/s12035-014-8804-8 Biochemistry 1981;20:7144–51. https://doi.org/10.1021/bi00528a014 [157] Álvarez A, Aleixandre M, Linares C, Masliah E, Moessler H. Apathy and APOE4 [184] Wong BX, Hung YH, Bush AI, Duce JA. Metals and cholesterol: two sides of the are associated with reduced BDNF levels in Alzheimer’s disease. J Alzheimer’s same coin in Alzheimer’s disease pathology. Front Aging Neurosci 2014;6:1–17. Dis 2014;42:1347–55. https://doi.org/10.3233/JAD-140849 https://doi.org/10.3389/fnagi.2014.00091 [158] Forlenza OV, Miranda AS, Barbosa IG, Talib LL, Diniz BS, Gattaz WF, et al. [185] Mesa-Herrera F, Taoro-González L, Valdés-Baizabal C, Diaz M, Marín R. Lipid Decreased neurotrophic support is associated with cognitive decline in non- and lipid raft alteration in aging and neurodegenerative diseases: a window for demented subjects. J Alzheimer’s Dis 2015;46:423–9. https://doi.org/10.3233/ the development of new biomarkers. Int J Mol Sci 2019:20. https://doi.org/10. JAD-150172 3390/ijms20153810 [159] Nakamura S, Okabayashi S, Ageyama N, Koie H, Sankai T, Ono F, et al. [186] Glöckner F, Ohm TG. Tau pathology induces intraneuronal cholesterol accu­ Transthyretin amyloidosis and two other aging-related amyloidoses in an aged mulation. J Neuropathol Exp Neurol 2014;73:846–54. https://doi.org/10.1097/ vervet monkey. Vet Pathol 2008:67–72. NEN.0000000000000103 [160] Merched A, Serot JM, Visvikis S, Aguillon D, Faure G, Siest G. Apolipoprotein E, [187] Rahman A, Akterin S, Flores-Morales A, Crisby M, Kivipelto M, Schultzberg M, transthyretin and actin in the CSF of Alzheimer’s patients: relation with the et al. High cholesterol diet induces tau hyperphosphorylation in apolipoprotein senile plaques and cytoskeleton biochemistry. FEBS Lett 1998;425:225–8. e deficient mice. FEBS Lett 2005;579:6411–6. https://doi.org/10.1016/j.febslet. https://doi.org/10.1016/S0014-5793(98)00234-8 2005.10.024 [161] Schwarzman AL, Gregori L, Vitek MP, Lyubski S, Strittmatter WJ, Enghilde JJ, [188] Cui HL, Guo B, Scicluna B, Coleman BM, Lawson VA, Ellett L, et al. Prion infection et al. Transthyretin sequesters amyloid β protein and prevents amyloid for­ impairs cholesterol metabolism in neuronal cells. J Biol Chem mation. Proc Natl Acad Sci USA 1994;91:8368–72. https://doi.org/10.1073/pnas. 2014;289:789–802. https://doi.org/10.1074/jbc.M113.535807 91.18.8368 1203 R.A. Loch, H. Wang, A. Perálvarez-Marín et al. Computational and Structural Biotechnology Journal 21 (2023) 1189–1204 [189] Beel AJ, Sakakura M, Barrett PJ, Sanders CR. Direct binding of cholesterol to the [201] La Rosa C, Condorelli M, Compagnini G, Lolicato F, Milardi D, Do TN, et al. amyloid precursor protein: an important interaction in lipid-Alzheimer’s dis­ Symmetry-breaking transitions in the early steps of protein self-assembly. Eur ease relationships? Biochim Biophys Acta - Mol Cell Biol Lipids Biophys J 2020;49:175–91. https://doi.org/10.1007/s00249-020-01424-1 2010;1801:975–82. https://doi.org/10.1016/j.bbalip.2010.03.008 [202] Scollo F, Rosa CLa. Amyloidogenic intrinsically disordered proteins: new in­ [190] Hung YH, Bush AI, Fontaine SLa. Links between copper and cholesterol in sights into their self-assembly and their interaction with membranes. Life Alzheimer’s disease. Front Physiol 2013;4 MAY:1–18. https://doi.org/10.3389/ 2020;10:1–20. https://doi.org/10.3390/life10080144 fphys.2013.00111 [203] Walter J, Van Echten-Deckert G. Cross-talk of membrane lipids and Alzheimer- [191] Hauser PS, Narayanaswami V, Ryan RO. Apolipoprotein E: from lipid transport related proteins. Mol Neurodegener 2013;8:1–12. https://doi.org/10.1186/1750- to neurobiology. Prog Lipid Res 2011;50:62–74. https://doi.org/10.1016/j. 1326-8-34 plipres.2010.09.001 [204] Mahley RW, Huang Y. Small-molecule structure correctors target abnormal [192] Liu L, Zhang K, Sandoval H, Yamamoto S, Jaiswal M, Sanz E, et al. Glial lipid protein structure and function: structure corrector rescue of apolipoprotein E4- droplets and ROS induced by mitochondrial defects promote neurodegenera­ associated neuropathology. J Med Chem 2012;55:8997–9008. https://doi.org/ tion. Cell 2015;160:177–90. https://doi.org/10.1016/j.cell.2014.12.019 10.1021/jm3008618 [193] Liu L, MacKenzie KR, Putluri N, Maletić-Savatić M, Bellen HJ. The glia-neuron [205] Wang C, Najm R, Xu Q, Jeong DE, Walker D, Balestra ME, et al. Gain of toxic lactate shuttle and elevated ROS promote lipid synthesis in neurons and lipid apolipoprotein E4 effects in human iPSC-derived neurons is ameliorated by a droplet accumulation in glia via APOE/D. Cell Metab 2017;26:719–37. https:// small-molecule structure corrector article. Nat Med 2018;24:647–57. https:// doi.org/10.1016/j.cmet.2017.08.024. [.e6]. doi.org/10.1038/s41591-018-0004-z [194] Hsiao JHT, Halliday GM, Kim WS. α-Synuclein regulates neuronal cholesterol [206] Rosenberg JB, Kaplitt MG, De BP, Chen A, Flagiello T, Salami C, et al. AAVrh.10- efflux. Molecules 2017;22:1–8. https://doi.org/10.3390/molecules22101769 mediated APOE2 central nervous system gene therapy for APOE4-associated [195] Serrano-Pozo A, Mielke ML, Gómez-Isla T, Betensky RA, Growdon JH, Frosch Alzheimer’s disease. Hum Gene Ther Clin Dev 2018;29:24–47. https://doi.org/ MP, et al. Reactive glia not only associates with plaques but also parallels 10.1089/humc.2017.231 tangles in Alzheimer’s disease. Am J Pathol 2011;179:1373–84. https://doi.org/ [207] Huynh TPV, Liao F, Francis CM, Robinson GO, Serrano JR, Jiang H, et al. Age- 10.1016/j.ajpath.2011.05.047 dependent effects of apoE reduction using antisense oligonucleotides in a [196] Lahdo R, De La Fournière-Bessueille L. Insertion of the amyloid precursor model of β-amyloidosis. Neuron 2017;96:1013–23. https://doi.org/10.1016/j. protein into lipid monolayers: effects of cholesterol and apolipoprotein E. neuron.2017.11.014. [.e4]. Biochem J 2004;382:987–94. https://doi.org/10.1042/BJ20040777 [208] Suidan GL, Ramaswamy G. Targeting apolipoprotein E for Alzheimer’s disease: an [197] Kawarabayashi T, Shoji M, Younkin LH, Wen-Lang L, Dickson DW, Murakami T, industry perspective. Int J Mol Sci 2019:20. https://doi.org/10.3390/ijms20092161 et al. Dimeric amyloid β protein rapidly accumulates in lipid rafts followed by [209] Di Battista A M, Heinsinger M, William Rebeck G N. Alzheimer’s disease genetic apolipoprotein E and phosphorylated Tau accumulation in the Tg2576 mouse risk factor APOE-ε4 also affects normal brain function. Curr Alzheimer Res model of Alzheimer’s disease. J Neurosci 2004;24:3801–9. https://doi.org/10. 2016;13:1200–7. https://doi.org/10.2174/1567205013666160401115127 1523/JNEUROSCI.5543-03.2004 [210] Peterson SL, Benowitz LI. Mammalian dendritic regrowth: a new perspective on [198] Wood WG, Li L, Müller WE, Eckert GP. Cholesterol as a causative factor in neural repair. Brain 2018;141:1891–4. https://doi.org/10.1093/brain/awy165 Alzheimer’s disease: a debatable hypothesis. J Neurochem 2014;129:559–72. [211] Blennow K, de Leon MJ, Zetterberg H. Alzheimer’s disease. Lancet https://doi.org/10.1111/jnc.12637 2006;368:387–403. https://doi.org/10.1016/S0140-6736(06)69113-7 [199] Tambini MD, Pera M, Kanter E, Yang H, Guardia‐Laguarta C, Holtzman D, et al. [212] National Academies of Sciences, Engineering and MH and MDB on HSPF on N ApoE4 upregulates the activity of mitochondria‐associated ER membranes. and NSD. Developing multimodal therapies for brain disorders; 2017. 〈https:// EMBO Rep 2016;17:27–36. https://doi.org/10.15252/embr.201540614 doi.org/10.17226/23657〉. [200] Wood WG, Eckert GP, Igbavboa U, Müller WE. Amyloid beta-protein interac­ [213] Tran TTT, Corsini S, Kellingray L, Hegarty C, Le Gall G, Narbad A, et al. APOE tions with membranes and cholesterol: causes or casualties of Alzheimer’s genotype influences the gut microbiome structure and function in humans and disease. Biochim Biophys Acta - Biomembr 2003;1610:281–90. https://doi.org/ mice: relevance for Alzheimer’s disease pathophysiology. FASEB J 10.1016/S0005-2736(03)00025-7 2019;33:8221–31. https://doi.org/10.1096/fj.201900071R 1204