Serotonin Promotes Serum Albumin Interaction with the Monomeric Amyloid β Peptide

Prevention of amyloid β peptide (Aβ) deposition via facilitation of Aβ binding to its natural depot, human serum albumin (HSA), is a promising approach to preclude Alzheimer’s disease (AD) onset and progression. Previously, we demonstrated the ability of natural HSA ligands, fatty acids, to improve the affinity of this protein to monomeric Aβ by a factor of 3 (BBRC, 510(2), 248–253). Using plasmon resonance spectroscopy, we show here that another HSA ligand related to AD pathogenesis, serotonin (SRO), increases the affinity of the Aβ monomer to HSA by a factor of 7/17 for Aβ40/Aβ42, respectively. Meanwhile, the structurally homologous SRO precursor, tryptophan (TRP), does not affect HSA’s affinity to monomeric Aβ, despite slowdown of the association and dissociation processes. Crosslinking with glutaraldehyde and dynamic light scattering experiments reveal that, compared with the TRP-induced effects, SRO binding causes more marked changes in the quaternary structure of HSA. Furthermore, molecular docking reveals distinct structural differences between SRO/TRP complexes with HSA. The disintegration of the serotonergic system during AD pathogenesis may contribute to Aβ release from HSA in the central nervous system due to impairment of the SRO-mediated Aβ trapping by HSA.


Introduction
Human serum albumin (HSA; 66.5 kDa) is the most abundant blood and cerebrospinal fluid (CSF) protein and is vital to maintaining the osmotic pressure of blood and the transport of numerous substances, including fatty and amino acids, hormones, metal ions, and over 250 drugs (reviewed in [1,2]). HSA consists of three structurally similar α-helical domains comprising nine fatty-acid-binding sites, four metal-binding sites, and two highaffinity drug-binding sites [2]. One of the key participants in progression of Alzheimer's disease (AD), amyloid peptide (Aβ) [3], is mostly bound to HSA (ca 89% of Aβ in blood plasma [4]). This interaction effectively rescues Aβ from deleterious self-aggregation and cytotoxicity [5][6][7]. Thereby, HSA provides ca 60% of the total amyloid inhibitory activity of plasma proteins [5]. Furthermore, the antioxidant and detoxification activities and anti-inflammatory and blood-brain barrier-supporting properties of HSA likely prevent AD progression [8].
Based on the Aβ-buffering role of plasma HSA, removal of plasma Aβ via plasma exchange and replacement with therapeutic HSA were suggested for facilitation of Aβ   Table 1 for the fitting parameters).  [20]. Nevertheless, 1 mM TRP did not  Table 1 for the fitting parameters).
Addition of 1 mM TRP to the solution is sufficient to load HSA with TRP, considering the equilibrium association constant of 2.7 × 10 4 M −1 [20]. Nevertheless, 1 mM TRP did not affect HSA's apparent affinity to Aβ 40 /Aβ 42 (Figures 1 and 2, Table 1). Instead, it notably changed the kinetics of the HSA-Aβ 42 interaction, inducing slowdown of the association and dissociation processes (see Figure 3). Meanwhile, addition of 1 mM SRO decreased the apparent K D value for the HSA-Aβ complex by a factor of 7 and 17, for Aβ 40 and Aβ 42 , respectively ( Figure 1, Table 1), mostly due to a slowdown of the dissociation process ( Figure 3). A decrease in the SRO concentration to 10 µM was accompanied by an increase in the apparent K D up to the values observed in the absence of the ligand ( Figure 2, Table 1). Since the equilibrium dissociation constant for the HSA-SRO complex is 1.6 µM [21], 10 µM SRO is enough to saturate HSA with SRO. For this reason, the absence of noticeable effects of 10 µM SRO on HSA's affinity to Aβ indicates that SRO binding to HSA alone is insufficient for improvement of its affinity to Aβ, thereby suggesting the necessity of direct SRO interaction with Aβ for that. The latter phenomenon is indicated by the efficient inhibition of Aβ 42 fibrillation by 50-100 µM SRO [22]. Notably, TRP was markedly less effective in this sense. The direct interaction of SRO with both Aβ and HSA implies that the equilibrium and kinetic dissociation/association constants determined by SPR spectroscopy for HSA-Aβ interaction in the presence of SRO(TRP) represent the apparent values, effectively describing the complex network of the multiple chemical equilibria occurring in the system. 1). Since the equilibrium dissociation constant for the HSA-SRO complex is 1.6 µ M [21], 10 µ M SRO is enough to saturate HSA with SRO. For this reason, the absence of noticeable effects of 10 µ M SRO on HSA's affinity to Aβ indicates that SRO binding to HSA alone is insufficient for improvement of its affinity to Aβ, thereby suggesting the necessity of direct SRO interaction with Aβ for that. The latter phenomenon is indicated by the efficient inhibition of Aβ42 fibrillation by 50-100 µ M SRO [22]. Notably, TRP was markedly less effective in this sense. The direct interaction of SRO with both Aβ and HSA implies that the equilibrium and kinetic dissociation/association constants determined by SPR spectroscopy for HSA-Aβ interaction in the presence of SRO(TRP) represent the apparent values, effectively describing the complex network of the multiple chemical equilibria occurring in the system.   Table 1). The average-lg K D values for the HSA-Aβ 40 /Aβ 42 complexes in the absence of the ligands are marked by solid gray lines, and their boundary values considering standard deviations are indicated by dashed gray lines. meric/trimeric and high-molecular-weight forms (about 23% and 12%, respectively), con-sistent with the previous observations [23]. While addition of 1 mM TRP did not affect the quaternary structure of HSA, addition of 1 mM SRO was accompanied by an approximately 2-fold increase in the content of high-molecular-weight forms at the expense of the monomer. Dynamic light scattering spectroscopy confirmed that SRO binding to HSA induces more prominent accumulation of its multimeric forms, compared with the effects induced by TRP ( Figure 4B). Overall, SRO binding caused more distinct changes in the quaternary structure of HSA, relative to the TRP-induced effects.  Table  1). The average lg ka values for the HSA-Aβ40/42 complexes in the absence of the ligands are marked by solid gray lines, and their boundary values considering standard deviations are indicated by dashed gray lines. Influence of SRO/TRP on the effective kinetic dissociation constant, kd, for the HSA complex with monomeric Aβ40 (panel C) or Aβ42 (D), according to SPR data (see Table 1). The average-lg kd values for the HSA-Aβ40/42 complexes in the absence of the ligands are marked by solid gray lines, and their boundary values considering standard deviations are indicated by dashed gray lines.  Table 1). The average lg k a values for the HSA-Aβ40/42 complexes in the absence of the ligands are marked by solid gray lines, and their boundary values considering standard deviations are indicated by dashed gray lines. Influence of SRO/TRP on the effective kinetic dissociation constant, k d , for the HSA complex with monomeric Aβ40 (panel C) or Aβ42 (D), according to SPR data (see Table 1). The average-lg k d values for the HSA-Aβ40/42 complexes in the absence of the ligands are marked by solid gray lines, and their boundary values considering standard deviations are indicated by dashed gray lines.
The drastic difference between the effects exerted by SRO and TRP on the HSA-Aβ interaction may partly arise due to the structural differences between HSA complexes with SRO/TRP. The SRO/TRP-induced changes in the quaternary structure of HSA were explored by crosslinking with glutaraldehyde at 37 • C, followed by SDS-PAGE analysis ( Figure 4A). Intact HSA is mostly monomeric (66%) with minor contributions of dimeric/trimeric and high-molecular-weight forms (about 23% and 12%, respectively), consistent with the previous observations [23]. While addition of 1 mM TRP did not affect the quaternary structure of HSA, addition of 1 mM SRO was accompanied by an approximately 2-fold increase in the content of high-molecular-weight forms at the expense of the monomer. Dynamic light scattering spectroscopy confirmed that SRO binding to HSA induces more prominent accumulation of its multimeric forms, compared with the effects induced by TRP ( Figure 4B). Overall, SRO binding caused more distinct changes in the quaternary structure of HSA, relative to the TRP-induced effects. Since tertiary structures of HSA complexes with SRO/TRP have not been repor date, we predicted the location of a SRO/TRP-binding pocket in HSA using the SiteScorer algorithm [24]. According to these predictions, the binding site is located region between the IB site, 'IIA: Drug site 1′, and 'IIIA: Drug site 2′ [25] ( Figure 5). Since tertiary structures of HSA complexes with SRO/TRP have not been reported to date, we predicted the location of a SRO/TRP-binding pocket in HSA using the DoGSiteScorer algorithm [24]. According to these predictions, the binding site is located in the region between the IB site, 'IIA: Drug site 1 , and 'IIIA: Drug site 2 [25] ( Figure 5). The binding pocket is predominantly apolar, but contains charged residues and clusters of polar residues (Tyr174 of domain I; Ser216, Gln245, Ser311, Tyr365, Ser366 of domain II; Cys472, Tyr476, Ser478 of domain III): Y174, E177, S216, K219, Q220, L222, K223, F235, W238, A239, R242, L243, Q245, R246, F247, L258, L262, H266, R281, L284, A285, I288, L299, K310, S311, H312, I314, A315, E316, E318, D320, P363, Y365, S366, V367, V368, K468, P471, C472, D475, Y476, S478, V479. Molecular docking of HSA with SRO/TRP using Au-toDock Vina software [26] predicted the difference in their orientation within the binding pocket ( Figure 5; see Supplementary PDB files in Supplementary Materials). Analysis of interactions between HSA and the bound ligands using PLIP software [27] revealed specific hydrogen bonds and π-cation interactions (Table 2, Figure 5B,C). Overall, although the molecular modeling predicted the same binding HSA pocket for SRO/TRP, noticeable differences in structural peculiarities of microenvironments of the bound SRO/TRP molecules were also detected, which may have led to the differences in the HSA structural rearrangements in response to the binding of these ligands. Table 2. The interactions between SRO/TRP and HSA within the model of their complex shown in Figure 5. The distance between donor and acceptor atoms is indicated for hydrogen bonds.  The binding pocket is predominantly apolar, but contains charged residues and clusters of polar residues (Tyr174 of domain I; Ser216, Gln245, Ser311, Tyr365, Ser366 of domain II; Cys472, Tyr476, Ser478 of domain III): Y174, E177, S216, K219, Q220, L222, K223, F235, W238, A239, R242, L243, Q245, R246, F247, L258, L262, H266, R281, L284, A285, I288, L299, K310, S311, H312, I314, A315, E316, E318, D320, P363, Y365, S366, V367, V368, K468, P471, C472, D475, Y476, S478, V479. Molecular docking of HSA with SRO/TRP using AutoDock Vina software [26] predicted the difference in their orientation within the binding pocket ( Figure 5; see Supplementary PDB files in Supplementary Materials). Analysis of interactions between HSA and the bound ligands using PLIP software [27] revealed specific hydrogen bonds and π-cation interactions (Table 2, Figure 5B,C). Overall, although the molecular modeling predicted the same binding HSA pocket for SRO/TRP, noticeable differences in structural peculiarities of microenvironments of the bound SRO/TRP molecules were also detected, which may have led to the differences in the HSA structural rearrangements in response to the binding of these ligands. Table 2. The interactions between SRO/TRP and HSA within the model of their complex shown in Figure 5. The distance between donor and acceptor atoms is indicated for hydrogen bonds. To explore the conformational flexibility of the established SRO/TRP-binding pocket of HSA, we conducted a multiparametric evaluation of the protein predisposition for intrinsic disorder using a set of commonly utilized per-residue disorder predictors. In these analyses, a threshold of 0.5 is typically used to identify disordered residues and regions in a query protein. Residues with the disorder scores (DS) 0.25 ≤ DS < 0.5 are considered to be moderately disordered, whereas residues with 0.15 ≤ DS < 0.25 are flexible. The results of this analysis are shown in Figure 6, which clearly illustrate that most of the residues involved in the binding to TRP or SRO are characterized by high conformational flexibility. In fact, with a very few exceptions, all such residues are located within regions that are predicted to be flexible/disordered by at least six predictors. Importantly, the several residues are predicted to have a propensity for ordering. This observation suggests that the interplay between intrinsic (amino acid sequence-based) structural flexibility and order propensity is important to accommodating TRP/SRO in the binding site, similarly to the target recognition principle established for S100 proteins [28]. Overall, the disorder propensity predictions suggest that SRO/TRP binding to HSA is not of the "lock-and-key" type but is likely to follow the "induced fit" mechanism. To explore the conformational flexibility of the established SRO/TRP-binding pocket of HSA, we conducted a multiparametric evaluation of the protein predisposition for intrinsic disorder using a set of commonly utilized per-residue disorder predictors. In these analyses, a threshold of 0.5 is typically used to identify disordered residues and regions in a query protein. Residues with the disorder scores (DS) 0.25 ≤ DS < 0.5 are considered to be moderately disordered, whereas residues with 0.15 ≤ DS < 0.25 are flexible. The results of this analysis are shown in Figure 6, which clearly illustrate that most of the residues involved in the binding to TRP or SRO are characterized by high conformational flexibility. In fact, with a very few exceptions, all such residues are located within regions that are predicted to be flexible/disordered by at least six predictors. Importantly, the several residues are predicted to have a propensity for ordering. This observation suggests that the interplay between intrinsic (amino acid sequence-based) structural flexibility and order propensity is important to accommodating TRP/SRO in the binding site, similarly to the target recognition principle established for S100 proteins [28]. Overall, the disorder propensity predictions suggest that SRO/TRP binding to HSA is not of the "lock-and-key" type but is likely to follow the "induced fit" mechanism. Figure 6. Bioinformatics analysis of the disorder predisposition of the full-length HSA (UniProt ID: P02768), including the signal peptide (residues 1-18) and the pro-peptide (residues 19-24). Intrinsic disorder profiles were generated based on the commonly used disorder predictors PONDR ® VLXT [29], PONDR ® VL3 [30], PONDR ® VLS2B [31], PONDR ® FIT [32], IUPred2 (Short), and IUPred2 (Long) [33,34]. Mean disorder scores (DS) calculated by averaging the outputs of individual predictors are shown as well. The outputs of the disorder predictors are represented as real numbers between 1 (ideal prediction of disorder) and 0 (ideal prediction of order). A threshold of 0.5 was used to identify disordered residues and regions in HSA. Residues with 0.25 ≤ DS < 0.5 are considered as moderately disordered, whereas residues with 0.15 ≤ DS < 0.25 are flexible. The position of residues involved in TRP/SRO binding is shown by gray vertical bars.

Discussion
The literature data on HSA interaction with monomeric Aβ are contradictory [7,[35][36][37][38]. While some studies did not detect HSA interaction with monomeric Aβ [35,36], other studies confirmed it [7,37,38], but gave affinity estimates differing from those reported here and in our previous work [12]. These contradictions seem to arise due to the high propensity of Aβ for multimerization, leading to sensitivity of the experiments to many factors, including the source of Aβ samples (recombinant or synthetic), the Aβ pretreatment procedures and solvent conditions (confirmed in [12]), and the methods used for the analysis. To ensure the monomeric state of Aβ in our experiments, we used the previously validated approach based on Aβ immobilization on the surface of a SPR chip by amine coupling, followed by removal of the non-covalently bound Aβ molecules with SDS [12].  [19][20][21][22][23][24]. Intrinsic disorder profiles were generated based on the commonly used disorder predictors PONDR ® VLXT [29], PONDR ® VL3 [30], PONDR ® VLS2B [31], PONDR ® FIT [32], IUPred2 (Short), and IUPred2 (Long) [33,34]. Mean disorder scores (DS) calculated by averaging the outputs of individual predictors are shown as well. The outputs of the disorder predictors are represented as real numbers between 1 (ideal prediction of disorder) and 0 (ideal prediction of order). A threshold of 0.5 was used to identify disordered residues and regions in HSA. Residues with 0.25 ≤ DS < 0.5 are considered as moderately disordered, whereas residues with 0.15 ≤ DS < 0.25 are flexible. The position of residues involved in TRP/SRO binding is shown by gray vertical bars.

Discussion
The literature data on HSA interaction with monomeric Aβ are contradictory [7,[35][36][37][38]. While some studies did not detect HSA interaction with monomeric Aβ [35,36], other studies confirmed it [7,37,38], but gave affinity estimates differing from those reported here and in our previous work [12]. These contradictions seem to arise due to the high propensity of Aβ for multimerization, leading to sensitivity of the experiments to many factors, including the source of Aβ samples (recombinant or synthetic), the Aβ pretreatment procedures and solvent conditions (confirmed in [12]), and the methods used for the analysis. To ensure the monomeric state of Aβ in our experiments, we used the previously validated approach based on Aβ immobilization on the surface of a SPR chip by amine coupling, followed by removal of the non-covalently bound Aβ molecules with SDS [12].
Previously, we showed the ability of linoleic and arachidonic acids to favor interaction between HSA and monomeric Aβ 40 /Aβ 42 with a 3-fold decrease in the K D value [12]. The data on SRO's effect on the HSA-Aβ equilibrium presented here illustrate the much higher potential for modulation of this equilibrium by low-molecular-weight HSA ligands with up to a 17-fold decrease in the K D value. Such a dramatic change in HSA's affinity for Aβ could be achieved via the following mechanisms or their combination: (I) an allosteric mechanism (ligand binding to HSA, altering its affinity for Aβ); or (II) ligand binding to Aβ, affecting its interaction with HSA. In the case of mechanism I alone, the SRO-induced changes in HSA's affinity for Aβ would be noticeable at SRO concentrations above 1.6 µM (the K D value for the HSA-SRO complex [21]).
The absence of a marked influence of 10 µM SRO on the HSA-Aβ interaction points out the contribution of mechanism II. This conclusion is in line with the data on the potent inhibition of Aβ 42 fibrillation by 50-100 µM SRO [22], which corresponds to the SRO level ensuring notable changes in HSA's affinity for Aβ. Despite the similar structures of SRO and TRP, the latter did not affect HSA's effective affinity for Aβ but slowed their association and dissociation. The drastic difference between SRO/TRP actions on the HSA-Aβ equilibrium can be rationalized within the same mechanisms I and II. The notable difference in quaternary structures of HSA in complex with SRO/TRP and distinct structural differences between HSA-SRO/TRP complexes predicted by molecular docking evidence that SRO binding and TRP binding differently alter HSA's structure. Meanwhile, the much less efficient inhibition of Aβ 42 fibrillation by TRP, compared with SRO [22], indicates the possibility of the differences between SRO and TRP in their interaction with monomeric Aβ.
The revealed influence of SRO on the HSA-Aβ interaction could be relevant to AD progression, considering the disintegration of the serotonergic system during AD, accompanied by a decrease in the SRO level in biofluids [15,16,39]. The SRO concentration in plasma is 0.28-1.7 µM [40], but in the synaptic cleft it may reach 6 mM and 55 nM after SRO diffuses into the extracellular compartment [41]. Hence, the peak SRO level in the central nervous system is sufficient for modulation of the HSA-Aβ interaction. Therefore, the decrease in the Aβ level in interstitial brain fluid observed in response to either administration of SSRIs or direct infusion of SRO into the hippocampus in a mouse model of AD [42] could be partly due to the facilitated trapping of Aβ by HSA. Chronic treatment with a SSRI caused a 50% reduction in plaque load in mouse brain, and the same effect has been confirmed for cognitively normal elderly participants chronically exposed to antidepressant drugs [42]. Thus, the principal finding of this work provides an alternative explanation to the available literature data on the SRO-induced reduction in the Aβ level in interstitial brain fluid and the respective decrease in brain amyloid load, thereby highlighting the importance of SRO for proper buffering of Aβ by HSA.
Since chemical structures of agents in clinical studies in AD, including SSRIs aimed at increasing the SRO level at nerve terminals [18,43], indicate their ability to interact with HSA or/and Aβ, it would be reasonable to complement their trials by tests of the ability of the drug candidates to affect the HSA-Aβ interaction.

Expression and Purification of Recombinant Human Aβ 40
Human Aβ 40 was expressed in E. coli and purified as described earlier [12] with some modifications. The 2-Mercaptoethanol (2-ME) concentration in the buffers was 5 mM. The 6×His-ubiquitin-Aβ 40 fusion protein was dialyzed twice against 50 mM Tris-HCl and 5 mM 2-ME (pH 8.6) buffers containing 6 M and 4 M urea, respectively. Aβ 40 was excised from the fusion protein via incubation with Usp2-cc (50-fold molar excess of the fusion over the enzyme) at room temperature for 3 h. Urea was added to the solution up to 8 M. The hydrolysate was passed through a Profinity IMAC column equilibrated with 50 mM Tris-HCl, 5 mM 2-ME, and 8 M urea (pH 8.0-8.5) to remove His-tagged ubiquitin and Usp2-cc. Aβ 40 was purified by chromatography on an HPLC Jupiter C18 column (buffer A: 20 mM NH 4 + , 6 mM acetic acid, pH 9.6; buffer B: 70% acetonitrile; gradient steps: loading of the sample in 15% buffer B, flushing by 25% buffer B for 10 min, linear gradient from 28% to 45% buffer B for 80 min). Precise chain cleavage by Usp2-cc was confirmed by electrospray ionization mass spectrometry (LCMS-2010EV, Shimadzu, Kyoto, Japan). The purified Aβ 40 samples were freeze-dried and stored at −70 • C.

Pretreatment of Aβ Samples for SPR Experiments
Recombinant human Aβ 40 /Aβ 42 samples were pretreated for SPR measurements mainly as described in [46]. The freeze-dried Aβ sample was dissolved in neat TFA at a concentration of 0.5-1 mg/mL. The solution was sonicated for 30 s. TFA was evaporated using Eppendorf Concentrator plus. The dried Aβ sample was dissolved in DMSO at a concentration of 2 mg/mL, and stored at −20 • C.

Surface Plasmon Resonance Studies
SPR measurements were performed at 25 • C using a Bio-Rad ProteOn™ XPR36 instrument mainly as previously described [12]. Ligand (57 µg/mL recombinant human Aβ 40 /Aβ 42 in 10 mM sodium acetate, pH 4.5 buffer) was immobilized on a ProteOn GLH sensor chip surface (up to 7000-9000 RUs) by amine coupling. The remaining activated amine groups on the chip surface were blocked by 1 M ethanolamine solution. The noncovalently bound Aβ 40 /Aβ 42 molecules were washed from the chip surface with 2% SDS water solution until stabilization of the signal [12]. The monomeric state of Aβ was previously verified by use of a polyclonal antibody selective to Aβ oligomers (A11) or an anti-Aβ 20 monoclonal antibody (7N22) as the analyte [12]; the immobilized Aβ sample did not reveal changes in the SPR signal upon application of A11 but interacted with 7N22. Analyte (5-60 µM HSA) in the running buffer (20 mM Tris-HCl, 140 mM NaCl, 4.9 mM KCl, 1 mM MgCl 2 , 2.5 mM CaCl 2 , pH 7.4) was passed over the chip at a rate of 30 µL/min for 300 s, followed by flushing the chip with the running buffer for 2400 s. The sensor chip surface was regenerated by passage of 10 mM EDTA pH 7.9 solution containing 2% SDS. The influence of HSA ligands on its interaction with Aβ 40 /Aβ 42 was studied in the presence of TRP/SRO (10 µM, 100 µM, or 1 mM). The double-referenced SPR sensograms were globally fitted according to the one-site binding model using Bio-Rad ProteOn Manager™ v.3.1 software. The effective equilibrium (K D ) and kinetic (k d and k a ) dissociation/association constants were evaluated for each HSA concentration, followed by averaging of the resulting values (standard deviations are indicated).

Evaluation of the Per-Residue Intrinsic Disorder Predisposition of HSA
The intrinsic disorder predisposition of HSA was analyzed by commonly used perresidue predictors, including PONDR ® VLXT [29], PONDR ® VSL2 [31], and PONDR ® VL3 [30], which are available on the PONDR site (http://www.pondr.com, accessed on 28 April 2021), PONDR ® FIT, which is a meta-predictor that incorporates predictions from several different sources [32], and the IUPred computational platform that allows for identification of either short or long regions of intrinsic disorder, IUPred Long and IUPred_Short [33]. We utilized the DiSpi web crawler, which was designed for the rapid prediction and comparison of protein disorder profiles. It aggregates the results from a number of well-known disorder predictors: PONDR ® VLXT [29], PONDR ® VL3 [30], PONDR ® VLS2B [31], PONDR ® FIT [32], IUPred2_Short, and IUPred2_Long [33,34]. The outputs of the disorder predictors are represented as real numbers between 1 (ideal prediction of disorder) and 0 (ideal prediction of order). A threshold of ≥0.5 was used to identify disordered residues and regions in HSA. Residues with the disorder scores (DS) 0.25 ≤ DS < 0.5 are considered to be moderately disordered, whereas residues with 0.15 ≤ DS < 0.25 are flexible.

Conclusions
The presented example of ligand-induced facilitation of HSA interaction with monomeric Aβ demonstrates the high potential for pharmacological modulation of this interaction aimed at decreasing the free Aβ concentration, leading to the inhibition of malicious Aβ deposits. The revealed effect seems to involve ligand binding to both participants of the reaction. The data deepen our understanding of the SRO-mediated effects reported in animal and clinical studies.

Acknowledgments:
We are indebted to Victoria A. Rastrygina (Pushchino, Russia) for help in the preparation of Aβ samples, and to Eugene A. Permyakov (Pushchino, Russia) for language editing of the manuscript.

Conflicts of Interest:
The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.