The Multimerization State of the Amyloid-β 42 Peptide (Aβ 42 ) Governs its Interaction Network with the Extracellular Matrix

Abstract

The goals of this work were i) to identify the interactions of amyloid-β (Aβ)₄₂ under monomeric, oligomeric, and fibrillar forms with the extracellular matrix (ECM) and receptors, ii) to determine the influence of Aβ₄₂ supramolecular organization on these interactions, and iii) to identify the molecular functions, biological processes, and pathways targeted by Aβ₄₂ in the ECM. The ECM and cell surface partners of Aβ₄₂ and its supramolecular forms were identified with protein and glycosaminoglycan (GAG) arrays (81 molecules in triplicate) probed by surface plasmon resonance imaging. The number of partners of Aβ₄₂ increased upon its multimerization, ranging from 4 for the peptide up to 53 for the fibrillar aggregates. The peptide interacted only with ECM proteins but their percentage among Aβ₄₂ partners decreased upon multimerization. Aβ₄₂ and its supramolecular forms recognized different molecular features on their partners, and the partners of Aβ₄₂ fibrillar forms were enriched in laminin IV-A, N-terminal, and EGF-like domains. Aβ₄₂ oligomerization triggered interactions with receptors, whereas Aβ₄₂ fibrillogenesis promoted binding to GAGs, proteoglycans, enzymes, and growth factors and the ability to interact with perineuronal nets. Fibril aggregation bind to further membrane proteins including tumor endothelial marker-8, syndecan-4, and discoidin-domain receptor-2. The partners of the Aβ₄₂ supramolecular forms are enriched in proteins contributing to cell growth and/or maintenance, involved in integrin cell surface interactions and expressed in kidney cancer, preadipocytes, and dentin. In conclusion, the supramolecular assembly of Aβ₄₂ governs its ability to interact in vitro with ECM proteins, remodeling and crosslinking ECM enzymes, proteoglycans, and receptors.

Keywords

Alzheimer’s disease biomolecular interactions extracellular matrix

Introduction

Alzheimer’s disease (AD), the most common neurodegenerative disease, is associated with the formation of neurofibrillary tangles in neurons and of amyloid plaques in brain paranchyma [1, 2]. Amyloid plaques are comprised of the amyloid-β (Aβ) peptide, a cleavage product of the amyloid-β protein precursor (AβPP) that forms amyloid oligomers and fibrils, and of other components including extracellular matrix (ECM) proteins and proteoglycans. The Aβ peptide also accumulates in vessel walls leading to cerebral amyloid angiopathy, which occurs in 90% of patients with AD [3]. Cerebral angiogenesis is triggered by amyloidogenesis [4] and the Aβ₄₂ peptide stimulates angiogenesis [5], which is also regulated by numerous growth factors and by ECM bioactive fragments called matricryptins [6 –8].

Laminin [9], heparan sulfate [10, 11], the heparan sulfate proteoglycan agrin [12], and collagens IV [9, 13] and XXV [14] are present in the amyloid plaques and bind to Aβ and/or to its supramolecular assemblies. Furthermore the composition of the ECM changes in the course of AD, which affects signal transmission at the synaptic level and diffusion in the extracellular space [15]. These changes are reflected in the composition of cerebrospinal fluid as we have shown for endostatin, an anti-angiogenic matricryptin of collagen XVIII, in the cerebrospinal fluid AD patients [16]. Several studies have been designed to identify at a larger scale the binding partners of the Aβ peptide(s) and its precursor (AβPP) in silico [17], in vitro [18], in brain tissue extracts [19], in serum and in cerebrospinal fluid [20] of AD patients. The vast majority of Aβ partners identified in these studies were intracellular or membrane proteins. Only eleven ECM and ECM-associated proteins out of the 2242 Aβ-binding proteins were identified in Olah’s study [18]. However, the identification of ECM proteins and ECM receptors able to bind the Aβ peptide is required to better understand the molecular mechanisms of ECM contribution to the formation of the amyloid plaques and to the neurodegenerative process occurring in AD.

Most studies have been performed with the full-length AβPP [19, 21], Aβ oligomers [18] or different Aβ peptides (1–28, 1–40, 1–42) [22, 23] proteolytically released from AβPP by β- and γ-secretases [24]. Furthermore, they have been carried with a single form, two forms or uncharacterized assemblies of the Aβ peptide. We have previously identified several ECM partners of Aβ₄₂, which was directly suspended in phosphate-buffered saline (PBS) without pretreatment with 1,1,1,3,3,3-hexafluoro-2-propanol (HFIP), and was likely under an aggregated, uncharacterized form [25]. The characterization of Aβ supramolecular assemblies is important because ECM components or receptors may bind to one form but not to the others as shown for glypican-1, which interacts with fibrillar, but not with non-fibrillar,Aβ [26].

We selected the Aβ₄₂ peptide for this study because it is more prone to aggregation than Aβ₄₀ [27], its oligomers are more neurotoxic than those of Aβ₄₀ [28], and its decreased level in cerebrospinal fluid is a clinical biomarker for AD. We prepared four forms of Aβ₄₂, namely the monomeric peptide, oligomers, fibrils, and fibrillary aggregates. The aims of this work were i) to identify the interaction repertoire of the Aβ_1–42) amyloid peptide, and its supramolecular assemblies (oligomers and fibrillar forms) with the ECM and ECM receptors using ECM arrays probed by surface plasmon resonance imaging (SPRi), ii) to determine the influence of Aβ₄₂ supramolecular organization on these interactions, and iii) to identify the molecular functions, biological processes, and pathways which are targeted by Aβ₄₂ in the ECM.

MATERIALS AND METHODS

Preparation of the different forms of the Aβ_1–42 peptide

The peptide (DAEFRHDSGYEVHHQKLVFFAEDVGSNKGAIIGLMVGGVVIA, Mw 4514Da) was synthesized by Proteogenix (Schiltigheim, France). The oligomers, fibrils, and aggregated fibrils were prepared according to the protocol of Stine et al. [29, 30]. The freeze-dried Aβ₄₂ peptide was diluted at 4.5 μg/ml in HFIP (42060, Fluka), which prevents oligomerization and fibrillogenesis of the peptide. The solution was stirred at room temperature for 30 min and aliquoted into 10 μl-fractions (0.045 mg) in Axygen^® MAXYMum recovery tubes (Corning). HFIP was evaporated overnight under a hood and the peptide film was dried in a SpeedVac for 1 h without heating. The tubes were stored at –20°C until use. A 1 mM stock solution of the peptide was prepared in dimethyl sulfoxide (DMSO, D2650, Sigma) at room temperature and was used to prepare different forms of the peptide. The stock solution was diluted in distillated water at room temperature to a final concentration of 100 μM, vortexed for 15 s to keep the peptide monomeric, form, or incubated at 4°C for 24 h to form oligomers. The 1 mM stock solution was diluted in 10 mM HCl±150 mM NaCl at room temperature to a final concentration of 100 μM, vortexed for 15 s and incubated for 24 h at 37°C to form fibrils or fibrillar aggregates respectively.

Characterization of the different forms of the Aβ_1–42 peptide by electron microscopy

Samples were adsorbed for 1 min on nickel grids coated with formvar and carbon. After a brief rinse in water, grids were floated on a 2% uranyl acetate solution for 1 min and air dried. Observations were performed with a CM120 Philips transmission electron microscope at the “Centre Technologique des Microstructures” (Université Lyon 1, Villeurbanne, France).

Identification of the partners of the different forms of the Aβ₄₂ by surface plasmon resonance imaging

The SPR imaging system (Biacore Flexchip, GE Healthcare) has been designed for screening molecular interactions using protein arrays [31]. The Biacore Flexchip platform is able to monitor simultaneously up to 400 interactions between biomolecules spotted on gold arrays (physically adsorbed on the gold surface) and a protein (the analyte) injected in buffer flow and recirculated over the array. We used this system to identify new partners of the Aβ_1–42 peptide among 81 different biomolecules spotted in triplicate onto the gold surface of a Gold Affinity chip (GE Healthcare) using a non-contact PiezoArray spotter (Scienion sciFlexarrayer S3). Several wild type and mutant protein domains and selectively desulfated heparins were also spotted on the arrays to characterize the domain, the amino acids, or the sulfate groups involved in the binding to the Aβ_1–42 peptide and its different forms. In total, 101 molecular species were spotted onto the arrays (Supplementary Table 1). Full-length proteins and protein fragments were spotted at concentrations varying from 0.060 to 1 mg/ml and glycosaminoglycans (GAGs) at 0.5 or 1 mg/ml (Supplementary Table 1) as previously described [25, 32]. Full-length proteins, protein fragments, and glycosaminoglycans spotted on the arrays were from commercial sources, except for the C-terminal domain of wild type and mutated human collagen XVIII (NC1) and human endostatin [25], von Willebrand 1 and 2-3 domains of the α1 chain of human collagen VI (amino acids 37–235 and 615–1021, respectively) and the ectodomains of human collagen XIII and XVII (amino acids 62–717 and 489–1497, respectively) and human neuropilin-1 (amino acids 23–815, generous gift of the plasmid by Prof. K. Alitalo, University of Helsinki, Finland), which were all expressed in human embryonic kidney (HEK) 293 EBNA cells in the laboratory. Full-length tenascin-X and tenascin-X deleted of the EGF-like domains were expressed as previously described [33]. Recombinant human Procollagen C-Proteinase Enhancer-1, expressed in HEK-293 EBNA cells [34], was a generous gift of Prof. E. Kessler (Sackler Faculty of Medicine, Tel Aviv University, Israel). The tags fused to the recombinant proteins spotted on the arrays were spotted as free tags (Flag, glutathion S-transferase, IgG₁) or fused to an irrelevant protein (6 His tag) to check their binding to the analyte. The role of sulfate groups in the binding to fibrils and fibrillar aggregates was investigated by spotting selectively desulfated heparins onto the arrays as previously described [35]. However, the percentages of binding inhibition calculated by comparing binding levels of desulfated heparins to Aβ₄₂ and its supramolecular forms with the binding level of high molecular weight heparin are estimates. The chips were dried at room temperature and stored under vacuum at 4°C until use. The regions of interest had four associated reference spots to correct refractive index changes as well as to evaluate non-specific binding of the analyte to the chips. The chips were blocked with a buffer containing mammalian proteins (BR-1007-08, GE-Healthcare) for 5×5 min. After the blocking step the chip was equilibrated in PBS + 0.05% Tween 20 (P3563, Sigma) at 500 μl/min for 60 min.

The different forms of the Aβ₄₂ peptide (5 μM) were diluted in PBS + 0.05% Tween 20 immediately before use and injected individually over the arrays at 25°C and at 500 μl/min for 20 min. The dissociation was monitored in PBS + 0.05% Tween 20 for 60 min. Data collected from reference spots (gold surface) for untagged proteins and from tag spots for tagged proteins were subtracted from those collected on spotted proteins to obtain sensorgrams reflecting specific binding. The four forms of the Aβ₄₂ peptide were analyzed twice. Only interactions identified in two separate experiments were considered for further analyses.

Enrichment analysis of the partners of Aβ₄₂ oligomers and fibrillar forms

Glycosaminoglycan partners were excluded from the in silico analysis because they are not annotated in the databases used for enrichment analyses. The analyses were performed for the protein partners of Aβ₄₂ oligomers and fibrillar forms with FunRich and the human proteome as a reference [36]. Only annotations associated with at least 30% of the partners and a p-value lower than 10^–2 were listed in the Results section.

RESULTS

Characterization of the supramolecular assemblies of the Aβ_1–42 peptide

Samples containing Aβ₄₂ peptide, oligomers, fibrils, and fibrillar aggregates were observed by electron microscopy after negative contrast with uranyl acetate. No fibrils were detected in the soluble Aβ₄₂ peptide preparation and in oligomers (data not shown). Fibrils were observed in samples incubated for 24 h at 37°C in 10 mM HCl without (fibrils) or with 150 mM NaCl (fibrillar aggregates) (Fig. 1). The addition of 150 mM NaCl at acidic pH induces the coalescence of soluble fibrillar Aβ_1–42 into dense, insoluble, fibril aggregates [29, 30]. Both samples exhibited a similar morphology, appearing as flexible elements with a 10 nm-diameter as expected for amyloid fibrils. Some shorter structures, 20–30 nm in length with a similar diameter, were also present in these preparations.

Identification of partners of Aβ₄₂ peptide and its supramolecular assemblies by SPR imaging

Only interactions identified in two separate experiments were considered for further analyses. Collagen III for example did not clearly bind to Aβ₄₂ in the first SPRi run and was thus not considered as a partner of the peptide although it bound to it in the second run. This was also the case for heparan sulfate from bovine kidney, which bound to fibrils in the first experiment but failed to bind in the second one. Neurexin-1β, expressed as a fusion protein with human IgG1, was not considered as a binding partner of the peptide because the SPRi signal recorded on spotted human IgG1 was either similar or higher than the signal recorded on of the IgG1-neurexin-1β fusion protein. Proteins, proteoglycans, and glycosaminoglycans that did not interact either with Aβ₄₂ or with its supramolecular assemblies are listed in Supplementary Table 2.

The partners of Aβ₄₂ and of its supramolecular forms are listed in Table 1. The number of partners identified in SPRi binding assays increased with the multimerization of the peptide, ranging from 4 partners for the monomer up to 53 partners for fibrillar aggregates. The major increases in the size of the interaction repertoires were associated with oligomerization (15 partners) and fibrillogenesis (47 partners), whereas the number of partners of fibrils and fibrillar aggregates was similar (47 and 53, respectively) (Table 1, Supplementary Figure 1). To refine the analysis of Aβ₄₂ partners, we divided them into six categories: ECM proteins, proteoglycans/GAGs, enzymes, growth factors, receptors and miscellaneous proteins (albumin, tropomyosin and calreticulin) (Fig. 2). All partners of Aβ₄₂ peptide and most partners of the oligomers (66%) were ECM proteins, whereas these proteins comprised only one third of the partners of both fibrillar forms. Fibrils and fibrillar aggregates showed similar interaction repertoires gaining the ability to interact with proteoglycans and GAGs (26% of their partners) and growth factors (4 to 6% of their partners) compared to monomers and oligomers (Fig. 2).

Interactions of Aβ₄₂ and its assemblies with ECM proteins

All forms of Aβ₄₂ interacted with fibrillar collagen I and collagen II, which is expressed in brain parenchyma [37]. Oligomers acquired the ability to bind to collagens III, IV, and XIII, expressed by neurons. Fibrillogenesis promoted interactions with collagens V, VI, and XVII, which is expressed by neurons, and with the C-terminal domain of collagen XVIII, which contains the endostatin sequence (Table 1). Endostatin has been previously shown to bind to the Aβ_1–42 peptide by SPR imaging when injected in buffer flow over spotted Aβ₄₂ peptide. The binding in reverse orientation was also observed when Aβ₄₂ peptide was injected over spotted endostatin [25]. However, in this study no form of the Aβ₄₂ peptide interacted with spotted endostatin. Several differences between both series of SPRi experiments might explain this discrepancy. The lyophilized Aβ_1–42 peptide from Sigma used in our previous work was solubilized in PBS and was not characterized thereafter. It might have been heterogeneous with some amount of preformed peptide assemblies and β-sheet structures. In contrast, the peptide used in this study was 95.21% pure and any preexisting structures was removed by dissolution of the peptide in HFIP to minimize formation of β-sheet structures (<1%), yield predominately α-helix and random coil [29, 30], and monomerize the peptide.

All forms of the Aβ peptide also bound to tropoelastin and tenascin-X, which is expressed in the connective tissue of choroid plexus in the brain [38].Tenascin-X has elastic properties and associates with both collagen fibrils and elastic fibers [39]. All multimeric assemblies of the Aβ peptide bound to procollagen C-proteinase enhancer-1, which enhances C-terminal processing of procollagens by bone morphogenetic protein-1 [40] and is expressed in the choroid plexus and leptomeninges [41]. In contrast, only fibrils and fibrillar aggregates interacted with fibronectin, laminin-111, the matricellular protein SPARC and dermatopontin, which regulate cell adhesion (Table 1). ECM-1, a pro-angiogenic ECM protein, interacted only with Aβ fibrillar aggregates (Table 1).

Interactions of Aβ oligomers and fibrils with membrane proteins

SPRi binding assays were carried out with the ectodomains of membrane proteins. The amyloid peptide did not bind to any membrane proteins spotted on the arrays, whereas receptors comprised between 15 and 20% of the partners of the Aβ₄₂ multimers (Fig. 2). Oligomers bound to α5β1 and αvβ5 integrins and to the macrophage receptor MARCO, which contains a triple-helical domain and belongs to the class A of scavenger receptors (Table 1). Fibrils and fibrillary aggregates interacted with two further integrins (α4β1 and αvβ3), neuropilin-1, and the receptor 2 of the vascular endothelial growth factor (VEGFR-2) (Table 1). Fibrillar aggregates bound with three other membrane proteins, the discoidin domain receptor-2, syndecan-4, and TEM-8 (tumor endothelial marker-8, also known as anthrax toxin receptor-1) (Table 1). TEM-8 is an essential regulator of connective tissue homeostasis [42], which binds to the C-terminus of collagen VI and might thus interfere with the neuroprotective effect of this collagen reported in AD [43]. Only receptors lacking tyrosine kinase activity (integrins, MARCO, neuropilin-1) bind to Aβ oligomers, whereas both tyrosine kinase receptors (VEGFR2, and DDR2) and receptors lacking intrinsic tyrosine kinase activity (integrins, MARCO, neuropilin-1, syndecan-4 and TEM-8) interact with Aβ fibrillar forms.

Interactions of Aβ oligomers and fibrils with enzymes

Aβ₄₂ oligomerization promoted interaction with plasminogen, the precursor of the serine protease plasmin, whereas fibrils and fibrillar aggregates bound to seven enzymes including one ECM remodeling enzyme (MMP-2), two ECM cross-linking enzymes, lysyl oxidase like-3 (LOXL-3), expressed in the central nervous system and neurons [44], and transglutaminase-2, and the serine protease reelin. Fibrillar forms interacted with both active and inactive, mutant, transglutaminase-2, which suggests that the catalytic activity of transglutaminase-2 was not required for the binding. Another serine protease, the coagulation factor X, interacted only with fibrillar aggregates (Table 1) in agreement with its reported adsorption to amyloid fibrils [45].

Aβ fibrillogenesis is associated with binding to glycosaminoglycans, proteoglycans, and growth factors

In our experimental conditions, no proteoglycan or glycosaminoglycan bound either to the Aβ₄₂ peptide or to its oligomers (Fig. 2). Heparin bound to both Aβ fibrillar forms, whereas heparan sulfate bound only to fibrillar aggregates (Table 1) in agreement with the fact that heparin-binding properties of the Aβ peptide depends on its aggregation state [46]. Both heparin and heparan sulfate favor amyloid fibril formation [47]. Low molecular weight heparin (3 kDa, corresponding approximately to a decasaccharide) bound in higher amount to fibrils and fibrillar aggregates than high molecular weight heparin. Proteoglycans interacting with fibrils and fibrillar aggregates were hyalectans (brevican and neurocan), small leucine-rich proteoglycans (decorin, fibromodulin, and lumican), a basement membrane heparan sulfate proteoglycan (agrin), and membrane (neuroglycan) or membrane-associated (glypican-1, –3, –5, and –6) proteoglycans (Table 1). The interactions of fibrillar forms with neuroglycan, a chondroitin sulfate proteoglycan, was likely mediated by its core protein because chondroitin sulfate did not interact with fibrils and aggregated fibrils (Supplementary Table 2). Only fibrillar aggregates interacted with the membrane proteoglycan syndecan-4 (Table 1).

Aβ₄₂ fibrillogenesis also induced binding to the growth factors FGF-2 and EGF (Fig. 2). The soluble form of epigen, a member of the mammalian family of EGFR ligands, which is mitogenic for fibroblasts and epithelial cells, interacted with Aβ fibrils but not with fibrillar aggregates (Table 1).

Aβ₄₂ and its supramolecular forms recognize different molecular sites on their partners

The deletion of the EGF domains from tenascin-X prevented its interaction with the Aβ₄₂ peptide but not with its multimers suggesting that the peptide and its multimers recognized different molecular features on tenascin-X. Fibrils bound to the fibronectin fragment III1-C, corresponding to anastellin, but not to full-length fibronectin, whereas fibrillar aggregates interacted with both anastellin and full-length fibronectin. The von Willebrand A domain 1 of the α1 chain of collagen VI participated in the interaction with Aβ₄₂ fibrillar assemblies but not with Aβ₄₂ oligomers. Oligomers bound to α5β1 and αvβ5 integrins but not to α4β1 and αvβ3 integrins, which suggests that the binding site is not comprised of a single integrin subunit, such as αv and β1, but involved both α and β subunits. In addition, only fibrillar aggregates were able to bind to the uncharacterized form of the Aβ₄₂ peptide resuspended in PBS and spotted on the array without further treatment. In the same way, the contribution of the sulfate groups of heparin to the binding of Aβ fibrillar forms was modulated upon fibril aggregation. The N-desulfation of heparin inhibited its binding to fibrils and fibrillar aggregates by 80% (data not shown). The removal of 2-O- and 6-O-sulfate groups led to the inhibition of heparin binding to fibrils by ~46% and 69% respectively. In contrast 2-O-desulfated heparin bound to fibrillar aggregates to the same extent than heparin, and 6-O-desulfation of heparin decreased its interaction with fibrillar aggregates by ~26% (data not shown). N-sulfate groups of heparin play thus a crucial role in the binding of heparin to both fibrillar forms of the Aβ₄₂ peptide, whereas 2-O-sulfate groups contributed only to its interaction with the fibrils and 6-O-sulfation mostly to its interactions with fibrils. All the above results show that the molecular features recognized by Aβ₄₂ on its partners depend at least in part on its multimerization state.

Domains of protein partners of Aβ assemblies

We looked for the domain composition of Aβ₄₂ partners, listed in UniProtKB, to determine if it is related to their ability to interact with a specific Aβ form. Collagen domains (triple-helical domain, NC1 fibrillar collagen, and vWC) and tenascin-X domains (fibronectin III, fibrinogen C-terminal, and EGF-like) were found in partners common to all the Aβ forms (Fig. 3). Peptidase S1, fibronectin II, CUB (present in PCPE-1, and neuropilin-1), scavenger receptor cysteine-rich (SRCR, found in LOXL3 and MARCO), and vWA domains (the most abundant) were present in partners of Aβ oligomers and fibrillar forms (Fig. 3). A variety of protein domains was found in the partners of Aβ₄₂ fibrillar forms only, the most abundant being EGF-like, and laminin domains (G, A, and NT) (Fig. 3), which suggests that other proteins containing these domains are potential partners of Aβ₄₂ fibrillar forms. ECM-1 and albumin, which bind only to fibrillar aggregates, share a common cysteine pattern characteristic of the albumin protein family, CC-(X₎-C, X being any amino acid, which generates double-loop domains involved in ligand binding [48].

Enrichment analysis of the functions, pathways, sites of expression and clinical phenotypes associated with Aβ partners

Enrichment analyses were performed using FunRich as described in the experimental section. The genes included in the enrichment analysis are listed in Supplementary Table 3. Annotations of the partners of Aβ₄₂ oligomers, Aβ₄₂ fibrils, and Aβ₄₂ fibrillar aggregates meeting the criteria defined in the Material and Methods sections are listed in Tables 2–4. No transcription factor matched these criteria either for Aβ₄₂ oligomers or for fibrillar forms. Enrichment analysis of the Gene Ontology (GO) terms “Cellular component” and Molecular Function” will not be discussed here because our study was focused on extracellular matrix proteins and their receptors. Indeed, and as expected, there was a strong bias in the enrichment analysis towards “Extracellular matrix”, “Extracellular”, “Extracellular space” and “Extracellular region”, and “Plasma membrane” for the GO term “Cellular component”, and towards “Extracellular matrix structural constituent” and “Receptor activity” for the GO term “Molecular function”. The partners of Aβ oligomers and fibrillar forms were enriched in proteins involved in “Cell growth and/or maintenance”. The pathways enriched in the partners of the three Aβ₄₂ supramolecular forms were related to integrins (“Integrin cell surface interactions”, “Integrin family cell surface interactions” and “β1 integrin cell surface interactions”). The partners of Aβ₄₂ oligomers were specifically enriched in pathways associated with endothelium (“Cell surface interactions at the vascular wall” and “VEGFR3 signaling in lymphatic endothelium”). The partners of Aβ₄₂ fibrils were also enriched in “VEGF and VEGFR signaling network” but only one of the corrected p-values reached statistical significance. The partners of Aβ₄₂ fibrillar forms were associated with pathways related to proteoglycans (“Proteoglycan syndecan-mediated signaling events”, “Glypican pathway” and “Glypican 1 network” but here again only one of the corrected p-values reached statistical significance. The enrichment analysis of the sites of expression showed a significant enrichment in kidney cancer, pre-adipocytes, dentin, and in amniotic and cerebrospinal fluids for the partners of the three Aβ₄₂ supramolecular forms. The expression in aorta was specifically enriched in the partners of Aβ oligomers, and the expression in the PC3 prostate cell line was enriched in the partners of both Aβ₄₂ fibrillar forms. The expression in neutrophil was enriched in the partners of Aβ oligomers and fibrils, whereas the enrichment in plasma was significant for Aβ₄₂ oligomers and fibrillar aggregates. Regarding the enrichment analysis of clinical phenotypes only the partners of Aβ₄₂ oligomers met the criteria we defined (Table 2). The phenotypes enriched in their partners were associated with cardiovascular diseases (“Mitral valve prolapse” and “Cardiovascular”), skin defects (“Hyperextensible skin”, “Soft skin”, and “Easy bruisability”) and also with premature osteoarthritis and inguinal hernia.

DISCUSSION

We have identified binding partners of the Aβ₄₂ peptide, oligomers, fibrils, and fibrillar aggregates and checked by electron microscopy that peptide and oligomer samples did not contain fibrils and/or other aggregated material. This allowed us to characterize and compare for the first time the interactions established by the different forms of the peptide, oligomers, fibrils, and fibrillar aggregates, with the extracellular matrix, receptors, and growth factors. Several proteins identified in this study have been previously reported to bind Aβ in vitro or to be associated with it in senile plaques. They include catalase [49], integrins [50], agrin [12], glypican-1 [26], laminin [9], collagen VI [43], heparan sulfate [10, 11], transglutaminase-2 [51], MARCO [52], and reelin [53]. However, these studies did not report the binding of these proteins to the three supramolecular assemblies formed by the Aβ peptide. The number of partners of the Aβ_1–42 amyloid peptide increases upon its multimerization, ranging from 4 for the monomeric peptide up to 53 for the fibrillar aggregates. This is in agreement with a study designed to identify the protein partners of protofibrils (oligomers) in serum and cerebrospinal fluid, which showed that aggregation of Aβ as protofibrils enhances protein binding in human body fluids compared to the monomer [20].

Aβ oligomerization and fibrillogenesis are associated with a 4-fold and 10-fold increase in the number of partners respectively compared to the peptide. However, Aβ oligomers and fibrils do not simply have an extended interaction repertoire but they target different ECM and membrane protein families. Oligomerization promotes interactions with membrane proteins, whereas fibrillogenesis triggers interactions with glycosaminoglycans, proteoglycans, enzymes and growth factors. Fibrillar aggregates gain the ability to bind to further ECM (ECM-1), secreted (factor X, albumin), and membrane (Discoidin Domain Receptor-2 or DDR-2, syndecan-4, TEM-8) proteins. Furthermore, Aβ oligomers and fibrillar forms recognize different molecular features on some of their common partners as shown for the sulfate groups of heparin, the EGF domains of tenascin-X, the anastellin domain of fibronectin and the vWA domain of collagen VI, which are not recognized by all Aβ forms but only by some of them. It is thus unlikely that the increase in Aβ partners observed upon its multimerization results from a mere avidity effect due the presence of increased number of potential multiple identical binding sites on oligomers and fibrillar assemblies.

Aβ fibrillar forms bound to basement membrane components (agrin and collagens IV and XVIII) and to proteins expressed by neurons (e.g., membrane collagens XIII and XVII, LOXL3, [44 , 55]) and astrocytes (proteoglycans of the hyalectan family [56], and PCPE-1 [57]). Aβ oligomers and fibrillar forms interact with membrane proteins, ECM proteins (collagens IV and XIII), proteoglycans (agrin, brevican, glypican-1, –3, and –6), and integrins located at synapses [50 , 58–60]. These interactions might contribute to the dysfunction of synapses and impaired synaptic plasticity, which are part of the neurodegenerative process in AD, the synapse loss strongly correlating with dementia [61]. This is in agreement with previous studies implicating Aβ oligomers in synapse dysfunction and loss in AD models [61]. Several partners of Aβ fibrillar forms (brevican and neurocan) are located in perineuronal nets, which are the brain-specific ECM structures surrounding neurons, controlling plasticity and exerting neuroprotective functions [62 –65]. The functional consequences of these interactions are not clear since there is no significant change in the number and distribution of perineuronal nets in AD patients [60, 66]. Last, several partners of Aβ fibrillar forms (decorin, agrin, glypican-1, reelin and catalase) are deposited in senile plaques [15 , 67–69]. Partners of Aβ peptide and supramolecular assemblies are thus located in several brain structures, in amyloid plaques and at the cell surface, whereas fibrillogenesis provides Aβ with the ability to interact with perineural nets.

Triple-helical peptides delay nucleation and amyloid fibril growth [70] suggesting that collagens I, II, III, XIII, XVII, and MARCO may delay or inhibit amyloid fibril formation. In contrast, agrin increases fibrillogenesis rate [15] and heparin and heparan sulfate favor amyloid fibril formation and stabilization [47]. Tropoelastin, which binds to all Aβ forms, contains amyloidogenic sequence(s) [71 –73], which might be deposited in plaques as individual amyloid fibrils or as mixed fibrils with Aβ after elastin degradation. This might also be the case for other protein partners of Aβ forms identified here because most of them contain aggregation-prone sequences predicted with the web-based software AGGRESCAN [74–75], except for the mature form of collagens I-III and the extracellular domain of syndecan-4 (data not shown).

Aβ supramolecular forms interact with several receptors regulating angiogenesis (VEGFR2, neuropilin-1, [76, 77], α5β1 and αvβ3 integrins, TEM-8 [78], and DDR-2 [79, 80]) and pro-angiogenic proteins (FGF-2, EGF, and angiopoietin-like-4). This is in agreement with the fact that the modulation of angiogenesis may repair damage in the AD brain [81] and that vascular activation may be a novel therapeutic target in AD [82]. In addition several partners of Aβ fibrils (EGF, decorin and epigen [83, 84]), and fibrillar aggregates (syndecan-4) bind to EGFR, which is a target for treating Aβ-induced memory loss [85]. They may compete with Aβ oligomers or fibrils for EGFR binding and inhibit Aβ_1–42-induced EGFR activation.

The protein partners of the three Aβ supramolecular forms are enriched in proteins expressed in kidney cancer. There is an inverse association between cancer and AD [86 –88]. Anti-cancer drugs might be useful to treat AD patients [89] and cancer immunotherapy drug reduces symptoms of AD in mice [90]. However, the molecular mechanisms linking ECM proteins expressed in kidney cancer and AD remain to be deciphered. The partners of the three Aβ supramolecular forms are also enriched in proteins expressed in pre-adipocytes, in keeping with the fact that obesity and diabetes increase the risk of developing late-life dementia [91].

In conclusion the supramolecular assembly of the Aβ_1–42 peptide controls its interaction with the ECM and its receptors in agreement with its control of neurotoxicity and binding to cellular proteins [28]. The interactions identified in this study may affect Aβ oligomerization and fibrillogenesis, the stability and clearance of Aβ fibrils and the signaling pathways triggered by FGF-2 and EGF. They open new perspectives to develop therapeutic strategies by targeting Aβ interactions specific of an Aβ oligomerization state and/or location. The roles of decorin and collagen VI in AD brain, which both bind to Aβ supramolecular forms, warrant further investigation Indeed both proteins activate autophagy [92], and defects in autophagy are likely to contribute to the neurodegenerative processes in AD. Autophagy enhancers may thus be used in the treatment ofAD [93].

Footnotes

ACKNOWLEDGMENTS

Protein and glycosaminoglycan arrays were analyzed in the Biacore Flexchip system of the UMS 3444 Facility (Lyon, France). We thank Bertrand Duclos (ICBMS, UMR 5246 CNRS-University Lyon 1, Villeurbanne, France) and Efrat Kessler for their critical reading of the manuscript, Sylvain D. Vallet (ICBMS, UMR 5246 CNRS-University Lyon 1, Villeurbanne, France) for his help in formatting the manuscript and for building the interaction networks, Christophe Marquette and Loïc Blum (ICBMS, UMR 5246 CNRS-University Lyon 1, Villeurbanne, France) and AXO Science (Villeurbanne, France) for providing access to the spotter. We thank Professor Kari Alitalo (University of Helsinki, Finland) for providing the cDNA encoding human neuropilin-1, Professor Leena Bruckner-Tudermann (Münster, Germany) for providing HEK 293 cells expressing the ectodomain of collagen XVII, Professor Efrat Kessler (Sackler Faculty of Medicine, Tel Aviv University, Israel) for the generous gift of human procollagen C-proteinase enhancer-1, and Professor Björn Olsen for providing HEK 293 cells expressing humain endostatin and the NC1 domain of collagen XVIII. The financial support of the Fondation pour la Recherche Médicale to SRB is gratefully acknowledged (grant n° DBI20141231336).

Authors’ disclosures available online ().

Appendix

References

Mandelkow

E-M

, Mandelkow

(2012) Biochemistry and cell biology of tau protein in neurofibrillary degeneration. Cold Spring Harb Perspect Med 2, a006247.

Masters

, Selkoe

(2012) Biochemistry of amyloid β-protein and amyloid deposits in Alzheimer disease., a. Cold Spring Harb Perspect Med 2, 006262–.

Love

, Miners

(2016) Cerebrovascular disease in ageing and Alzheimer’s disease. Acta Neuropathol (Berl) 131, 645–658.

Biron

, Dickstein

, Gopaul

, Jefferies

(2011) Amyloid triggers extensive cerebral angiogenesis causing blood brain barrier permeability and hypervascularity in Alzheimer’s disease. PloS One 6, e23789.

Boscolo

, Folin

, Nico

, Grandi

, Mangieri

, Longo

, Scienza

, Zampieri

, Conconi

, Parnigotto

, Ribatti

(2007) Beta amyloid angiogenic activity in vitro and in vivo. Int J Mol Med 19, 581–587.

Ricard-Blum

, Salza

(2014) Matricryptins and matrikines: Biologically active fragments of the extracellular matrix. Exp Dermatol 23, 457–463.

Ricard-Blum

, Vallet

(2016) Proteases decode the extracellular matrix cryptome. Biochimie 122, 300–313.

Ricard-Blum

, Vallet

(2016) Matricryptins network with matricellular receptors at the surface of endothelial and tumor cells. Front Pharmacol 7, 11.

Perlmutter

, Barrón

, Saperia

, Chui

(1991) Association between vascular basement membrane components and the lesions of Alzheimer’s disease. J Neurosci Res 30, 673–681.

10.

, Cummings

, Cotman

(1992) Localization of heparan sulfate glycosaminoglycan and proteoglycan core protein in aged brain and Alzheimer’s disease. Neuroscience 51, 801–813.

11.

Ariga

, Miyatake

, Yu

(2010) Role of proteoglycans and glycosaminoglycans in the pathogenesis of Alzheimer’s disease and related disorders: Amyloidogenesis and therapeutic strategies–a review. J Neurosci Res 88, 2303–2315.

12.

Donahue

, Berzin

, Rafii

, Glass

, Yancopoulos

, Fallon

, Stopa

(1999) Agrin in Alzheimer’s disease: Altered solubility and abnormal distribution within microvasculature and brain parenchyma. Proc Natl Acad Sci U S A 96, 6468–6472.

13.

Kiuchi

, Isobe

, Fukushima

(2002) Type IV collagen prevents amyloid beta-protein fibril formation. Life Sci 70, 1555–1564.

14.

Hashimoto

, Wakabayashi

, Watanabe

, Kowa

, Hosoda

, Nakamura

, Kanazawa

, Arai

, Takio

, Mann

DMA

, Iwatsubo

(2002) CLAC: A novel Alzheimer amyloid plaque component derived from a transmembrane precursor, CLAC-P/collagen type XXV. EMBO J 21, 1524–1534.

15.

Morawski

, Filippov

, Tzinia

, Tsilibary

, Vargova

(2014) ECM in brain aging and dementia. Prog Brain Res 214, 207–227.

16.

Salza

, Oudart

J-B

, Ramont

, Maquart

F-X

, Bakchine

, Thoannès

, Ricard-Blum

(2015) Endostatin level in cerebrospinal fluid of patients with Alzheimer’s disease. J Alzheimers Dis 44, 1253–1261.

17.

Perreau

, Orchard

, Adlard

, Bellingham

, Cappai

, Ciccotosto

, Cowie

, Crouch

, Duce

, Evin

, Faux

, Hill

, Hung

, James

, Li

Q-X

, Mok

, Tew

, White

, Bush

, Hermjakob

, Masters

(2010) A domain level interaction network of amyloid precursor protein and Abeta of Alzheimer’s disease. Proteomics 10, 2377–2395.

18.

Oláh

, Vincze

, Virók

, Simon

, Bozsó

, Tõkési

, Horváth

, Hlavanda

, Kovács

, Magyar

, Szũcs

, Orosz

, Penke

, Ovádi

(2011) Interactions of pathological hallmark proteins: Tubulin polymerization promoting protein/p25, beta-amyloid, and alpha-synuclein. J Biol Chem 286, 34088–34100.

19.

Cottrell

, Galvan

, Banwait

, Gorostiza

, Lombardo

, Williams

, Schilling

, Peel

, Gibson

, Koo

, Link

, Bredesen

(2005) A pilot proteomic study of amyloid precursor interactors in Alzheimer’s disease. Ann Neurol 58, 277–289.

20.

Rahman

, Zetterberg

, Lendel

, Härd

(2015) Binding of human proteins to amyloid-β protofibrils. ACS Chem Biol 10, 766–774.

21.

Hoe

H-S

, Lee

, Carney

RSE

, Lee

, Markova

, Lee

J-Y

, Howell

, Hyman

, Pak

DTS

, Bu

, Rebeck

(2009) Interaction of reelin with amyloid precursor protein promotes neurite outgrowth. J Neurosci 29, 7459–7473.

22.

Snow

, Kinsella

, Parks

, Sekiguchi

, Miller

, Kimata

, Wight

(1995) Differential binding of vascular cell-derived proteoglycans (perlecan, biglycan, decorin, and versican) to the beta-amyloid protein of Alzheimer’s disease. Arch Biochem Biophys 320, 84–95.

23.

Calero

, Rostagno

, Ghiso

(2012) Search for amyloid-binding proteins by affinity chromatography. Methods Mol Biol 849, 213–223.

24.

Zolezzi

, Bastías-Candia

, Santos

, Inestrosa

(2014) Alzheimer’s disease: Relevant molecular and physiopathological events affecting amyloid-β brain balance and the putative role of PPARs. Front Aging Neurosci 6, 176.

25.

Faye

, Chautard

, Olsen

, Ricard-Blum

(2009) The first draft of the endostatin interaction network. J Biol Chem 284, 22041–22047.

26.

Watanabe

, Araki

, Chui

D-H

, Makifuchi

, Ihara

, Tabira

(2004) Glypican-1 as an Abeta binding HSPG in the human brain: Its localization in DIG domains andossible roles in the pathogenesis of Alzheimer’s disease. FASEB J 18, 1013–1015.

27.

Chow

, Mattson

, Wong

, Gleichmann

(2010) An overview of APP processing enzymes and products. Neuromolecular Med 12, 1–12.

28.

Manzoni

, Colombo

, Bigini

, Diana

, Cagnotto

, Messa

, Lupi

, Bonetto

, Pignataro

, Airoldi

, Sironi

, Williams

, Salmona

(2011) The molecular assembly of amyloid aβ controls its neurotoxicity and binding to cellular proteins. PloS One 6, e24909.

29.

Stine

, Dahlgren

, Krafft

, LaDu

(2003) In vitro characterization of conditions for amyloid-beta peptide oligomerization and fibrillogenesis. J Biol Chem 278, 11612–11622.

30.

Stine

, Jungbauer

, Yu

, LaDu

(2011) Preparing synthetic Aβ in different aggregation states. Methods Mol Biol 670, 13–32.

31.

Rich

, Myszka

(2007) Higher-throughput, label-free, real-time molecular interaction analysis. Anal Biochem 361, 1–6.

32.

Salza

, Peysselon

, Chautard

, Faye

, Moschcovich

, Weiss

, Perrin-Cocon

, Lotteau

, Kessler

, Ricard-Blum

(2014) Extended interaction network of procollagen C-proteinase enhancer-1 in the extracellular matrix. Biochem J 457, 137–149.

33.

Lethias

, Carisey

, Comte

, Cluzel

, Exposito

J-Y

(2006) A model of tenascin-X integration within the collagenous network. FEBS Lett 580, 6281–6285.

34.

Weiss

, Ricard-Blum

, Moschcovich

, Wineman

, Mesilaty

, Kessler

(2010) Binding of procollagen C-proteinase enhancer-1 (PCPE-1) to heparin/heparan sulfate: Properties and role in PCPE-1 interaction with cells. J Biol Chem 285, 33867–33874.

35.

Fatoux-Ardore

, Peysselon

, Weiss

, Bastien

, Pratlong

, Ricard-Blum

(2014) Large-scale investigation of Leishmania interaction networks with host extracellular matrix by surface plasmon resonance imaging. Infect Immun 82, 594–606.

36.

Pathan

, Keerthikumar

, Ang

C-S

, Gangoda

, Quek

CYJ

, Williamson

, Mouradov

, Sieber

, Simpson

, Salim

, Bacic

, Hill

, Stroud

, Ryan

, Agbinya

, Mariadason

, Burgess

, Mathivanan

(2015) FunRich: An open access standalone functional enrichment and interaction network analysis tool. Proteomics 15, 2597–2601.

37.

Leung

, Ng

, Ho

, Tam

, Cheah

(1998) Different cis-regulatory DNA elements mediate developmental stage- and tissue-specific expression of the human COL2A1 gene in transgenic mice. J Cell Biol 141, 1291–1300.

38.

Kawakami

, Matsumoto

(2011) Behavioral alterations in mice lacking the gene for tenascin-x. Biol Pharm Bull 34, 590–593.

39.

Valcourt

, Alcaraz

, Exposito

J-Y

, Lethias

, Bartholin

(2015) Tenascin-X: Beyond the architectural function. Cell Adhes Migr 9, 154–165.

40.

Kessler

, Takahara

, Biniaminov

, Brusel

, Greenspan

(1996) Bone morphogenetic protein-1: The type I procollagen C-proteinase. Science 271, 360–362.

41.

Lecain

, Zelenika

, Laine

, Rhyner

, Pessac

(1991) Isolation of a novel cDNA corresponding to a transcript expressed in the choroid plexus and leptomeninges. J Neurochem 56, 2133–2138.

42.

Besschetnova

, Ichimura

, Katebi

, St Croix

, Bonventre

, Olsen

(2015) Regulatory mechanisms of anthrax toxin receptor 1-dependent vascular and connective tissue homeostasis. Matrix Biol 42, 56–73.

43.

Cheng

, Dubal

, Kim

, Legleiter

, Cheng

, Yu

G-Q

, Tesseur

, Wyss-Coray

, Bonaldo

, Mucke

(2009) Collagen VI protects neurons against Abeta toxicity. Nat Neurosci 12, 119–121.

44.

Jourdan-Le Saux

, Tomsche

, Ujfalusi

, Jia

, Csiszar

(2001) Central nervous system, uterus, heart, and leukocyte expression of the LOXL3 gene, encoding a novel lysyl oxidase-like protein. Genomics 74, 211–218.

45.

Furie

, Voo

, McAdam

, Furie

(1981) Mechanism of factor X deficiency in systemic amyloidosis. N Engl J Med 304, 827–830.

46.

Watson

, Lander

, Selkoe

(1997) Heparin-binding properties of the amyloidogenic peptides Abeta and amylin. Dependence on aggregation state and inhibition by Congo red. J Biol Chem 272, 31617–31624.

47.

Iannuzzi

, Irace

, Sirangelo

(2015) The effect of glycosaminoglycans (GAGs) on amyloid aggregation and toxicity. Molecules 20, 2510–2528.

48.

Smits

, Ni

, Feng

, Wauters

, Van Hul

, Boutaibi

, Dillon

, Merregaert

(1997) The human extracellular matrix gene 1 (ECM1): Genomic structure, cDNA cloning, expression pattern, and chromosomal localization. Genomics 45, 487–495.

49.

Milton

(1999) Amyloid-beta binds catalase with high affinity and inhibits hydrogen peroxide breakdown. Biochem J 344(Pt 2), 293–296.

50.

Denda

, Reichardt

(2007) Studies on integrins in the nervous system. Methods Enzymol 426, 203–221.

51.

Wilhelmus

MMM

, de Jager

, Bakker

ENTP

, Drukarch

(2014) Tissue transglutaminase in Alzheimer’s disease: Involvement in pathogenesis and its potential as a therapeutic target. J Alzheimers Dis 42(Suppl 3), S289–S303.

52.

, Ye

(2015) Microglial Aβ receptors in Alzheimer’s disease. Cell Mol Neurobiol 35, 71–83.

53.

Pujadas

, Rossi

, Andrés

, Teixeira

, Serra-Vidal

, Parcerisas

, Maldonado

, Giralt

, Carulla

, Soriano

(2014) Reelin delays amyloid-beta fibril formation and rescues cognitive deficits in a model of Alzheimer’s disease. Nat Commun 5, 3443.

54.

Sund

, Väisänen

, Kaukinen

, Ilves

, Tu

, Autio-Harmainen

, Rauvala

, Pihlajaniemi

(2001) Distinct expression of type XIII collagen in neuronal structures and other tissues during mouse development. Matrix Biol 20, 215–231.

55.

Seppänen

, Autio-Harmainen

, Alafuzoff

, Särkioja

, Veijola

, Hurskainen

, Bruckner-Tuderman

, Tasanen

, Majamaa

(2006) Collagen XVII is expressed in human CNS neurons. Matrix Biol 25, 185–188.

56.

Wiese

, Karus

, Faissner

(2012) Astrocytes as a source for extracellular matrix molecules and cytokines. Front Pharmacol 3, 120.

57.

Takahara

, Kessler

, Biniaminov

, Brusel

, Eddy

, Jani-Sait

, Shows

, Greenspan

(1994) Type I procollagen COOH-terminal proteinase enhancer protein: Identification,rimary structure, and chromosomal localization of the cognate human gene (PCOLCE). J Biol Chem 269, 26280–26285.

58.

Singhal

, Martin

(2011) Role of extracellular matrix proteins and their receptors in the development of the vertebrate neuromuscular junction. Dev Neurobiol 71, 982–1005.

59.

Frischknecht

, Seidenbecher

(2012) Brevican: A key proteoglycan in the perisynaptic extracellular matrix of the brain. Int J Biochem Cell Biol 44, 1051–1054.

60.

Soleman

, Filippov

, Dityatev

, Fawcett

(2013) Targeting the neural extracellular matrix in neurological disorders. Neuroscience 253, 194–213.

61.

Spires-Jones

, Hyman

(2014) The intersection of amyloid beta and tau at synapses in Alzheimer’s disease. Neuron 82, 756–771.

62.

Zimmermann

, Dours-Zimmermann

(2008) Extracellular matrix of the central nervous system: From neglect to challenge. Histochem Cell Biol 130, 635–653.

63.

Dansie

, Ethell

(2011) Casting a net on dendritic spines: The extracellular matrix and its receptors. Dev Neurobiol 71, 956–981.

64.

Kwok

JCF

, Dick

, Wang

, Fawcett

(2011) Extracellular matrix and perineuronal nets in CNS repair. Dev Neurobiol 71, 1073–1089.

65.

Wlodarczyk

, Mukhina

, Kaczmarek

, Dityatev

(2011) Extracellular matrix molecules, their receptors, and secreted proteases in synaptic plasticity. Dev Neurobiol 71, 1040–1053.

66.

Morawski

, Brückner

, Jäger

, Seeger

, Matthews

, Arendt

(2012) Involvement of perineuronal and perisynaptic extracellular matrix in Alzheimer’s disease neuropathology. Brain Pathol 22, 547–561.

67.

Snow

, Mar

, Nochlin

, Kresse

, Wight

(1992) Peripheral distribution of dermatan sulfate proteoglycans (decorin) in amyloid-containing plaques and their presence in neurofibrillary tangles of Alzheimer’s disease. J Histochem Cytochem 40, 105–113.

68.

Verbeek

, Otte-Höller

, van den Born

, van den Heuvel

, David

, Wesseling

, de Waal

(1999) Agrin is a major heparan sulfate proteoglycan accumulating in Alzheimer’s disease brain. Am J Pathol 155, 2115–2125.

69.

Doehner

, Madhusudan

, Konietzko

, Fritschy

J-M

, Knuesel

(2010) Co-localization of Reelin and proteolytic AbetaPP fragments in hippocampal plaques in aged wild-type mice. J Alzheimers Dis 19, 1339–1357.

70.

Parmar

, Nunes

, Baum

, Brodsky

(2012) A peptide study of the relationship between the collagen triple-helix and amyloid. Biopolymers 97, 795–806.

71.

Tamburro

, Pepe

, Bochicchio

, Quaglino

, Ronchetti

(2005) Supramolecular amyloid-like assembly of the polypeptide sequence coded by exon 30 of human tropoelastin. J Biol Chem 280, 2682–2690.

72.

del Mercato

, Maruccio

, Pompa

, Bochicchio

, Tamburro

, Cingolani

, Rinaldi

(2008) Amyloid-like fibrils in elastin-related polypeptides: Structural characterization and elastic properties. Biomacromolecules 9, 796–803.

73.

Ostuni

, Bochicchio

, Armentano

, Bisaccia

, Tamburro

(2007) Molecular and supramolecular structural studies on human tropoelastin sequences. Biophys J 93, 3640–3651.

74.

Conchillo-Solé

, de Groot

, Avilés

, Vendrell

, Daura

, Ventura

(2007) AGGRESCAN: A server for the prediction and evaluation of “hot spots” of aggregation in polypeptides. BMC Bioinformatics 8, 65.

75.

Pawlicki

, Le Béchec

, Delamarche

(2008) AMYPdb: A database dedicated to amyloid precursor proteins. BMC Bioinformatics 9, 273.

76.

Carmeliet

, Ruiz de Almodovar

, Carmen R de

(2013) VEGF ligands and receptors: Implications in neurodevelopment and neurodegeneration. Cell Mol Life Sci 70, 1763–1778.

77.

Gelfand

, Hagan

, Tata

, Oh

W-J

, Lacoste

, Kang

K-T

, Kopycinska

, Bischoff

, Wang

J-H

, Gu

(2014) Neuropilin-1 functions as a VEGFR2 co-receptor to guide developmental angiogenesis independent of ligand binding. eLife 3, e03720.

78.

Jinnin

, Medici

, Park

, Limaye

, Liu

, Boscolo

, Bischoff

, Vikkula

, Boye

, Olsen

(2008) Suppressed NFAT-dependent VEGFR1 expression and constitutive VEGFR2 signaling in infantile hemangioma. Nat Med 14, 1236–1246.

79.

Zhang

, Bu

, Zhao

, Yu

, Wang

, Li

, Zhu

, Ren

, Liu

, Yao

, Su

(2014) A host deficiency of discoidin domain receptor 2 (DDR2) inhibits both tumour angiogenesis and metastasis. J Pathol 232, 436–448.

80.

Zhu

, Zhu

, Bu

, Zhao

, Zhang

, Chang

, Li

, Yao

, Wang

, Su

(2015) The anti-angiogenic role of discoidin domain receptor 2 (DDR2) in laser-induced choroidal neovascularization. J Mol Med 93, 187–198.

81.

Biron

, Dickstein

, Gopaul

, Fenninger

, Jefferies

(2013) Cessation of neoangiogenesis in Alzheimer’s disease follows amyloid-beta immunization. Sci Rep 3, 1354.

82.

Grammas

, Sanchez

, Tripathy

, Luo

, Martinez

(2011) Vascular signaling abnormalities in Alzheimer disease. Cleve Clin J Med 78(Suppl 1), S50–S53.

83.

Neill

, Schaefer

, Iozzo

(2012) Decorin: A guardian from the matrix. Am J Pathol 181, 380–387.

84.

Schneider

, Yarden

(2014) Structure and function of epigen, the last EGFR ligand. Semin Cell Dev Biol 28, 57–61.

85.

Wang

, Chiang

H-C

, Wu

, Liang

, Xie

, Yao

, Ma

, Du

, Zhong

(2012) Epidermal growth factor receptor is a preferred target for treating amyloid-β-induced memory loss. Proc Natl Acad Sci U S A 109, 16743–16748.

86.

Driver

, Beiser

, Au

, Kreger

, Splansky

, Kurth

, Kiel

, Lu

, Seshadri

, Wolf

(2012) Inverse association between cancer and Alzheimer’s disease: Results from the Framingham Heart Study. BMJ 344, e1442.

87.

Musicco

, Adorni

, Di Santo

, Prinelli

, Pettenati

, Caltagirone

, Palmer

, Russo

(2013) Inverse occurrence of cancer and Alzheimer disease: A population-based incidence study. Neurology 81, 322–328.

88.

Shi

, Tang

, Liu

Y-W

, Wang

X-F

, Chen

G-J

(2015) Alzheimer disease and cancer risk: A meta-analysis. J Cancer Res Clin Oncol 141, 485–494.

89.

Araki

(2013) Potential repurposing of oncology drugs for the treatment of Alzheimer’s disease. BMC Med 11, 82.

90.

Malkki

(2016) Alzheimer disease: Cancer immunotherapy drug reduces symptoms of Alzheimer disease in mice. Nat Rev Neurol 12, 126.

91.

Ishii

, Iadecola

(2016) Adipocyte-derived factors in age-related dementia and their contribution to vascular and Alzheimer pathology. Biochim Biophys Acta 1862, 966–974.

92.

Neill

, Schaefer

, Iozzo

(2014) Instructive roles of extracellular matrix on autophagy. Am J Pathol 184, 2146–2153.

93.

Silva

, Reis

, Teixeira

, Borges

(2014) Alzheimer’s disease, enzyme targets and drug discovery struggles: From natural products to drug prototypes. Ageing Res Rev 15, 116–145.

The Multimerization State of the Amyloid-β 42 Peptide (Aβ 42 ) Governs its Interaction Network with the Extracellular Matrix

Abstract

Keywords

Introduction

MATERIALS AND METHODS

Preparation of the different forms of the Aβ1–42 peptide

Characterization of the different forms of the Aβ1–42 peptide by electron microscopy

Identification of the partners of the different forms of the Aβ42 by surface plasmon resonance imaging

Enrichment analysis of the partners of Aβ42 oligomers and fibrillar forms

RESULTS

Characterization of the supramolecular assemblies of the Aβ1–42 peptide

Identification of partners of Aβ42 peptide and its supramolecular assemblies by SPR imaging

Interactions of Aβ42 and its assemblies with ECM proteins

Interactions of Aβ oligomers and fibrils with membrane proteins

Interactions of Aβ oligomers and fibrils with enzymes

Aβ fibrillogenesis is associated with binding to glycosaminoglycans, proteoglycans, and growth factors

Aβ42 and its supramolecular forms recognize different molecular sites on their partners

Domains of protein partners of Aβ assemblies

Enrichment analysis of the functions, pathways, sites of expression and clinical phenotypes associated with Aβ partners

DISCUSSION

Footnotes

ACKNOWLEDGMENTS

Appendix

References

Preparation of the different forms of the Aβ_1–42 peptide

Characterization of the different forms of the Aβ_1–42 peptide by electron microscopy

Identification of the partners of the different forms of the Aβ₄₂ by surface plasmon resonance imaging

Enrichment analysis of the partners of Aβ₄₂ oligomers and fibrillar forms

Characterization of the supramolecular assemblies of the Aβ_1–42 peptide

Identification of partners of Aβ₄₂ peptide and its supramolecular assemblies by SPR imaging

Interactions of Aβ₄₂ and its assemblies with ECM proteins

Aβ₄₂ and its supramolecular forms recognize different molecular sites on their partners