Modulating Conformation of Aβ-Peptide: An Effective Way to Prevent

3 days ago - In view of the significant inhibitory effect on the Aβ aggregation and related neurotoxicity, CAM represents a new type of leading compo...
0 downloads 0 Views 10MB Size
Download PDF
Article Cite This: Inorg. Chem. XXXX, XXX, XXX−XXX

pubs.acs.org/IC

Modulating Conformation of Aβ-Peptide: An Effective Way to Prevent Protein-Misfolding Disease Xiang Ma,†,∥ Jiai Hua,∥ Kun Wang,† Hongmei Zhang,‡ Changli Zhang,§ Yafeng He,† Zijian Guo,*,† and Xiaoyong Wang*,‡

Inorg. Chem. Downloaded from pubs.acs.org by UNIV OF SOUTH DAKOTA on 10/22/18. For personal use only.

†

State Key Laboratory of Coordination Chemistry, School of Chemistry and Chemical Engineering, Nanjing University, Nanjing 210023, P. R. China ‡ State Key Laboratory of Pharmaceutical Biotechnology, School of Life Sciences, Nanjing University, Nanjing 210023, P. R. China ∥ Chemistry and Chemical Engineering Department, Taiyuan Institute of Technology, Taiyuan 030008, P. R. China § School of Biochemical and Environmental Engineering, Nanjing Xiaozhuang University, Nanjing 210017, P. R. China S Supporting Information *

ABSTRACT: Alzheimer’s disease (AD) is a typical protein-misfolding disease. Aggregation of amyloid β-peptide (Aβ) plays a key role in the etiology of AD. The misfolding of Aβ results in the formation of β-sheet-rich aggregates and damages the function of neurons. A modified polyoxometalate (POM), [CoL(H2O)]2[CoL]2[HAsVMoV6MoVI6O40] [CAM, L = 2-(1Hpyrazol-3-yl)pyridine], was designed to disaggregate the Aβ aggregates, where L acts as an Aβ-targeting group and POM as a conformational modulator. Xray crystallography shows that CAM is composed of a ε-Keggin unit and four coordination units. CAM can disaggregate the β-sheet-rich fibrils and metalinduced or self-aggregated Aβ aggregates, and it further inhibits the production of ROS; as a result, it can protect the neurons from synaptic toxicity induced by Zn2+- or Cu2+-Aβ aggregates or Aβ self-aggregation. The mechanism of disaggregation involves a transformation of Aβ conformation from β-sheet to other conformers. The nature of the process is an interference of the β-sheet conformation by CAM via hydrogen bonding. CAM specifically interacts with Aβ aggregates but does not disturb the cerebral metal homeostasis and enzymatic systems. Molecular simulation suggests that the appropriate size of CAM and the cavity of β-sheets facilitate the interaction between CAM and Aβ aggregates; additionally, the H-bonding-favored amino acid residues in the cavity provide a precondition for the interaction. Moreover, CAM is lipophilic and capable of penetrating the blood-brain barrier, and it is metabolizable without causing an untoward effect to mice at high dosages. In view of the significant inhibitory effect on the Aβ aggregation and related neurotoxicity, CAM represents a new type of leading compounds with a distinctive mechanism of action for the treatment of Alzheimer’ disease. The conception of this study may be applied to other protein-misfolding diseases caused by conformational changes.

■

Inhibitors of β-secretase that can halt the formation of Aβ and organic chelators that can reverse the metal-induced Aβ aggregation have been designed and synthesized according to the amyloid hypothesis.12−14 However, since AD is only recognized clinically after large amounts of Aβ aggregates have formed, inhibitors of β-secretase may not be helpful at this stage; in addition, inhibiting β-secretase may incur serious and unacceptable side effects because the normal physiological role of β-secretase is not yet clearly understood.15 On the other hand, organic chelators may interrupt the homeostasis of cerebral metals,16 and inadequate disaggregation of senile plaques into oligomers may result in even more serious damage to the brain.17 Moreover, organic chelators can only inhibit the metal-induced aggregation but cannot suppress the Aβ selfaggregation, which is also a chief pathogenic factor in AD.18,19

INTRODUCTION

Protein misfolding has been associated with a variety of disorders termed as protein-misfolding diseases.1 AD is the most common protein-misfolding disease characterized by synapse loss as well as some lesions throughout the brain.2 The presence of senile plaques in the brain is the main histopathological criterion for AD, which is formed by conformational changes and aggregation of amyloid β-peptide (Aβ).3 It was found that abnormal cerebral metal ions such as Zn2+ and Cu2+ can promote the formation of Aβ aggregates,4,5 and the self-misfolded Aβ can deposit in the brain.6 The soluble neurotoxic β-sheet-rich Aβ oligomers are believed to be the main cause of synaptic dysfunction and memory loss in AD,7,8 and the reactive oxygen species (ROS) produced by the metal-induced Aβ aggregates are also causative factors of the neuronal death.4,5,9 Therefore, β-sheet-rich Aβ aggregates play a crucial role in the development of AD.10,11 © XXXX American Chemical Society

Received: July 28, 2018

A

DOI: 10.1021/acs.inorgchem.8b02115 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry Thus, therapeutic agents with Aβ per se as the primary target may avoid the adverse effects related with inhibitors of βsecretase and chelators of metal ions. We have previously designed two bifunctional platiniferous chelators to disaggregate the metal-induced Aβ aggregates through both metal chelation and peptide modification.20 Considering the potential side effect of chelators, we decide to discard the chelation tactic and adopt a noncovalent strategy to intervene with the Aβ aggregation through modulating the conformation of Aβ. Polyoxometalates (POMs) are potential drug candidates for the treatment of AD. For example, a series of POMs and transition metal-functionalized POM derivatives have been reported as inhibitors of Aβ aggregation in vitro.21 These compounds can cross the blood-brain barrier (BBB) and be metabolized in due time.22 Since POMs possess abundant naked or protonated oxygen atoms on the surface, we suppose they could interfere with Aβ through H-bonding and inhibit the formation of neurotoxic β-sheet-rich aggregates. POMs designed for weak interactions such as H-bonding with Aβ are rarely seen. Theoretically, such agents may affect the conformation of Aβ that is dominated by weak interactions and thereby inhibit the formation of β-sheet-rich aggregates. Herein we report the unique conformation-modulating property of a newly designed POM, {[CoL(H2O)]2[CoL]2[HAsVMoV6MoVI6O40]}·2.5H2O [abbreviated as CAM, L = 2-(1H-pyrazol-3-yl)pyridine], which was derived from modifying POM with an analogue of Aβ-targeting thioflavin T (ThT). In CAM, POM acts as a pharmacophore to modulate the conformation of Aβ, while L as a β-sheet targeting group to enhance the selectivity of CAM for Aβ. We chose Co, Mo, and As to synthesize POM because Co and Mo are essential elements for the body,23 and As is believed to be an essential element;23,24 in addition, AsV is far less toxic than AsIII.24 Thus, CAM may show good biocompatibility at reasonable concentrations. As expected, CAM can inhibit the formation of Aβ aggregates by modulating the β-sheet conformation through H-bonding and hence suppress the generation of ROS and reduce the toxicity of Aβ to neurons. This compound seems to be the first conformational modulator of β-sheet that takes Aβ as the direct target to treat AD.

Figure 1. Polyhedral/ball-and-stick representation of CAM (A), ε[HAsVMoV6MoVI6O40]8− monomer (B), and [CoL]2+ coordination unit (C).

{AsVO4}, where AsV binds to four bridging μ4-O atoms (see Figure 2) and each μ4-O atom binds to three Mo atoms. Four

■

RESULTS AND DISCUSSION Synthesis and Characterization. CAM was synthesized by reacting (NH4)6Mo7O24·2H2O, NaAsO2, CoCl2·6H2O, and 2-(1H-pyrazol-3-yl)pyridine at 150 °C in an acidic solution. The black crystals of CAM were obtained by slow cooling the reaction solution to the room temperature. The structure of CAM was characterized by single-crystal X-ray diffraction analysis. The crystallographic data and selected bond lengths are summarized in Tables S1 and S2. Detailed information has been deposited at the Cambridge Crystallographic Data Centre with a CCDC number of 901487. As shown in Figure 1A, CAM is composed of two parts, that is, an ε-Keggin unit [HAsVMoV6MoVI6O40]8− (Figure 1B) and four coordination units of [CoL(H2O)]2+ and [CoL]2+ (Figure 1C). The Keggin-type fragment is the first defined POM and involves several isomers, with α-isomer being the most common one. So far as we know, only a few ε-isomers were reported. The ε-Keggin unit of CAM can be regarded as a derivative of the α-isomer after rotation of four Mo3O13 groups by 60° (Figure S1).25 This unit contains a tetrahedron

Figure 2. Protonation of oxygen atoms in the ε-Keggin unit of CAM. The extent of protonation for each oxygen atom is indicated by different colors.

{Mo3O13} clusters lie at the vertices of the tetrahedron and connect to each other through one edge of the cluster (Figures S2 and S3). The shared oxygen atoms can coordinate further to [CoL(H2O)]2+ or [CoL]2+, forming the structure of CAM. The Co ions adopt different coordination numbers in CAM, in that two of them are six-coordinated by three O atoms from the ε-Keggin unit, one O atom from the lattice water and two N atoms from the L; the other two are five-coordinated by three O atoms from the ε-Keggin unit and two N atoms from the L (Figure S2). CAM is roughly in a sphere shape with a diameter of 10.40 Å, which may be relevant to the interaction B

DOI: 10.1021/acs.inorgchem.8b02115 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

fluorescence intensity of Aβ40 keeps weak even after 20 h, suggesting that the conformational transformation is largely suppressed. The fluorescence increases obviously when Aβ40 is incubated with Zn2+ or Cu2+, especially with Zn2+, which suggests that these cations can promote the formation of the βsheet and the promotive effect of Zn2+ is stronger than that of Cu2+.29 By contrast, in the presence of CAM, the fluorescence of the Zn2+- or Cu2+-Aβ40 solution maintains at the same level as Aβ40 alone during the first 20 h, and remains unchanged afterward. Since CAM does not affect the fluorescence of ThT (Figure S9), the quenching of the fluorescence indicates that CAM not only suppresses the Aβ aggregation induced by the β-sheet misfolding in the presence of metal ions but also prohibits the self-β-sheet-transformation of Aβ40. It is noteworthy that the complete suppression of the self-β-sheettransformation of Aβ is rarely reported and was only observed for perphenazine at a massive dosage.30 The effect of CAM on the Zn2+-, Cu2+-, or self-induced Aβ40 aggregates was further studied by turbidimetry, which reflects the level of all types of aggregates.31 The UV−vis absorbance at 405 nm (A405) was recorded to represent the turbidity of the Aβ40 solution in the absence and presence of metal ions and CAM. In the absence of CAM, the results are in agreement with the reported observations;32 in the presence of CAM, however, the turbidity of Zn2+- or Cu2+-Aβ40 solutions decreased markedly, though it does not come to the level of Aβ40 (Figure S10). The results indicate that CAM only partially inhibits the Aβ40 aggregation induced by Zn2+ or Cu2+. As for the pure Aβ40 solution, addition of CAM to it increased the turbidity. To further clarify the effect of CAM on the Aβ aggregates, Aβ40 was incubated with Zn2+ to obtain a suspension with abundant aggregates. Different concentrations of CAM were then added to the suspension to test its effect on the aggregates. As shown in Figure 4, in the ThT assay, the

with Aβ aggregates. In comparison with other ε-Keggin-type POMs reported previously,26 CAM possesses some unique characteristics. For example, it is lipophilic (Kow = 19.0) and electroneutral, which may make it penetrate the cellular membrane and approach the negatively charged Aβ more readily at pH 7.4 (pI 5.1). More importantly, 2-(1H-pyrazol-3yl)pyridine is an analogue of ThT, which has been shown to specifically bind to the β-sheet fibrils. Therefore, CAM may exhibit some targeting ability toward Aβ. The surface oxygen and the delocalized hydrogen atoms lingering on the oxygen atoms of the ε-Keggin unit could play key roles in the interaction of CAM with Aβ. As shown in Figure 2, the 42 oxygen atoms, excluding O in 2.5 crystallized H2O, in CAM can be classified into terminal Ot, bridging μ2-O, μ3-O and μ4-O. Aside from the inside μ4-O, the surface oxygen atoms are expected to interfere with Aβ via H-bonding. We thus calculated the protonation level (ΣH) of oxygen atoms in CAM using the crystallographic data (Table S2 and Table S3), X-ray photoelectron spectra (XPS) (Figure S4), and bond valence sums (Σs) (Figure S5, Table S3) according to the reported method.27 The results are shown in Figure 2. The O atoms with ΣH of 0.5−2.0 could act as H-donors owing to the delocalized protons on them, whereas the O atoms with ΣH of 0−0.1 possess dense electron cloud and could act as Hacceptors. As for the remaining O atoms, the electron cloud varies greatly (ΣH = 0.2−0.5), suggesting that they could act either as H-donors or H-acceptors. Anyway, the O atoms on the surface of CAM could potentially form H-bonds with Aβ. The stability of CAM was studied by using UV−vis spectroscopy according to the literature method.22 CAM remains intact in the solution at pH 7.4 for more a week and is stable at pH 5.0−9.0 (Figures S6 and S7). Furthermore, the stability of CAM in the presence of Cu2+ or Zn2+ was also studied in water solution. No covalent interaction was observed for them (Figure S8). Inhibition Effect on Aβ Aggregation. The inhibitory effect of CAM on the β-sheet formation of Aβ was first studied by the ThT assay, which has been widely used to detect the βsheet content in Aβ aggregates.28 As shown in Figure 3, the fluorescence intensity of Aβ40 alone is very weak during the first 20 h, indicating that Aβ40 exists primarily in its initial form. After 20 h, the fluorescence intensity begins to ascend, suggesting that increasing amounts of Aβ40 are transformed into the β-sheet. In the presence of CAM, however, the

Figure 4. Solubilizing effect of CAM on the Aβ40 (20 μM) aggregates induced by Zn2+ (40 μM) determined by ThT assay (λex = 415 nm, λem = 480 nm) and turbidimetry.

fluorescence decreases dramatically with the increase of CAM, indicating that the aggregates derived from the misfolded βsheet of Aβ40 were dissociated. In contrast, the turbidity of the corresponding solution increases slowly with the increase of CAM. Actually, even for the Aβ40 aggregates without Zn2+, similar phenomena were also observed (Figure S11). The results imply that CAM cannot completely solubilize the Aβ aggregates in all forms, on the contrary, it aggravates some kind of aggregation probably due to the nucleating effect.32 It seems that the aggregates are not only composed of the misfolded β-

Figure 3. ThT fluorescence intensity (λex = 415 nm) of Aβ40 (20 μM) solutions in the absence and presence of Zn2+ or Cu2+ after incubation with or without CAM at 37 °C and pH 7.4 for 10−3750 min ([Aβ40]: [metal ion]: [CAM] = 1:2:0.6). C

DOI: 10.1021/acs.inorgchem.8b02115 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

Figure 5. TEM images of Aβ40 in the absence or presence of Zn2+, Cu2+, and CAM after incubation at 37 °C and pH 7.4 for 24 h. (A) Aβ40; (B) Aβ40 + Zn2+; (C) Aβ40 + Cu2+; (D) Aβ40 + CAM; (E) Aβ40 + Zn2+ + CAM; (F) Aβ40 + Cu2+ + CAM ([Aβ40]: [metal ion]: [CAM] = 1:2:0.6).

complex.20 As shown in Figure 6, a strong fluorescence of DCF is observed at 525 nm for the Cu2+-Aβ40 system as compared

sheet of Aβ but also contain some amorphous Aβ aggregates. It is known that the β-sheet-rich oligomers are more toxic than the amyloid plaques,33 and the dissociation of the β-sheetrelated aggregates may suggest that CAM can eliminate the major toxic species in AD and leave the less toxic ones intact in the aggregates. Currently, the senile plaques are widely accepted as the primary pathologic indicator of AD; however, their content cannot correlate well with the cognitive impairment because many humans who showed abundant senile plaques at death did not suffer dementia.7,34 Here we demonstrate that some Aβ deposits are actually not composed of the toxic β-sheet aggregates. Considering the seemingly opposite effect on the β-sheet and amorphous aggregates, we choose 12 μM CAM to perform the following experiments. Morphological Analysis. The morphology of Aβ40 and Zn2+- or Cu2+-induced Aβ40 aggregates in the absence or presence of CAM was analyzed by transmission electron microscopy (TEM). Some fibrils are seen in the solution of Aβ40 (Figure 5A); however, large amounts of aggregates are observed after Aβ40 was incubated with Zn2+ (Figure 5B), which is consistent with the previous literature.35 A great number of fibrils are formed when Aβ40 was incubated with Cu2+ (Figure 5C), indicating that abundant soluble β-sheetrich Aβ oligomers are in the solution.36 In the presence of CAM, Aβ40 and Zn2+- or Cu2+-induced Aβ40 aggregates are all changed into granule-like species (Figure 5D−F). The morphological changes imply that CAM can inhibit the formation of β-sheet-rich fibrils or Aβ aggregates and fold them into dispersive small granules possibly due to the conformational changes of Aβ, no matter in the presence or absence of metal ions. Although a large number of amorphous aggregates still exist in the solution after the incubation with CAM, according to the above ThT assay, they are non-β-sheet aggregates. Inhibition of ROS Generation. ROS are the major species accounting for the neurotoxicity in AD.4 The effect of CAM on the Cu2+-Aβ mediated H2O2 generation was thus investigated by a DCF assay. DCF is a fluorescent marker derived from the reaction of nonfluorescent 2′,7′-dichlorofluorecin (DCFH) with H2O2 in the presence of horseradish peroxidase, which can indicate the generation of H2O2 from the Cu2+-Aβ

Figure 6. Fluorescence of DCF (λex = 485 nm) induced by Aβ40 or Cu2+-Aβ40 complex in the absence and presence of CAM.

to the Aβ40 solution, indicating that the Cu2+-Aβ complex is an effective ROS producer in solution. However, in the presence of CAM, the fluorescence intensity decreased by 50%, showing that CAM can inhibit the generation of ROS induced by the Cu2+-Aβ complex. The results suggest that CAM may impede the formation of the Cu2+-Aβ complex and thereby reduce the production of ROS. However, the ROS induced by metal-Aβ complexes is not the only species that cause the death of neuron cells; Aβ aggregates may also induce the cell death via other pathways.37,38 Inhibition of Neurotoxicity. The neurotoxicity of Aβ40 alone and Zn2+- or Cu2+-induced Aβ40 aggregates in the absence and presence of CAM toward neuronal pheochromocytoma cells (PC12) was first investigated by the MTT assay. As shown in Figure 7A, the cell viability in the presence of Aβ40 or Aβ40 plus Zn2+ or Cu2+ is quite low (40%). These results demonstrate that CAM can effectively inhibit the neurotoxicity induced by Aβ40 or metal-Aβ40 aggregates. D

DOI: 10.1021/acs.inorgchem.8b02115 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

the uptake of CAM in the brain. As shown in Figure 9, the content of Mo in the brain homogenates increases from ca.

Figure 7. Cell viability of PC12 cells after incubation with Aβ40 (20 μM) without or with metal ions (40 μM) for 72 h (A) and that of PC12 cells after incubation with Aβ40 without or with metal ions for 36 h and then with CAM (12 μM) for 36 h (B). Untreated cells are used as the control, and the results are mean ± SD of at least three experiments.

Figure 9. Time-dependent uptake of Mo in the brain and blood of B6 mice after intravenous injection of CAM (80 mg kg−1). Inset shows the dose-dependent uptake of Mo in the brain at 15 min after intravenous injection of CAM.

To further investigate the protective effect of CAM on neurons, we analyzed the morphological changes of PC12 cells before and after the treatment with CAM. As shown in Figure 8, the untreated PC12 cells take polygonal shapes with neurites (Figure 8A). After treatment with Aβ40 or metal-Aβ40 complexes for 48 h, the cell body and the neurites shrank, and the dendritic networks were disrupted (Figure 8B−D). However, in the presence of CAM, PC12 cells after coincubation with Aβ40 or Aβ40 plus Zn2+ or Cu2+ almost take the same shape as the untreated cells (Figure 8E−G). Recent studies showed that Aβ can induce synaptic toxicity, leading to the dysfunction of neuron cells.39,40 The above results indicate that CAM can protect the neuron cells from synaptic toxicity induced by Zn2+- or Cu2+-Aβ40 aggregates or that induced by the self-aggregation of Aβ40. BBB Penetration Ability. BBB is the main obstruction in developing anti-AD drugs because most drug candidates for AD cannot cross it.41 To test whether CAM can pass through the BBB or not, we injected CAM intravenously into the B6type mice and determined the content of Mo as a substitute for

2.34 μg kg−1 (control value) to ca. 0.25 mg kg−1 after 2 min of the injection; however, that in the plasma homogenates increases to ca. 110 mg kg−1. The content of Mo in the brain homogenates reaches its maximum at about 15 min and keeps deceasing until 48 h, and it does not change synchronously with that in the blood. The distribution behavior of CAM is similar to the previous descriptions for POMs,22 which may imply that the metabolic mechanism of CAM is similar to that of other POMs in crossing the BBB. We also studied the uptake of Mo in the brain at other dosages. As shown in the inset of Figure 9, the content of Mo in the brain homogenates increases with the dosage of CAM, which suggests that the uptake of CAM in the brain is positively related to its initial concentrations. As compared with the mice treated with physiological saline, no untoward effect was observed on the mice after the injection of CAM at a dosage of 80 mg kg−1 every 3 days for 2 weeks. The mice only showed a mild lethargy when treated at a dose of 160 mg kg−1. These results

Figure 8. Photomicrographs of PC12 cells after incubation with Aβ40 (20 μM), or Zn2+- or Cu2+-induced (40 μM) Aβ40 aggregates with or without CAM (12 μM) for 48 h. Images a−g are close-up views of the typical areas in A−G. E

DOI: 10.1021/acs.inorgchem.8b02115 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry indicate that CAM is not only capable of crossing the BBB through intravenous injection but is also highly biocompatible. Analysis of Brain Homogenates. The brain homogenates of 7-month-old APPswe transgenic mice were analyzed by Western blotting after treatment with CAM to further verify the effect of CAM on the Aβ aggregation in real biosamples. Glyceraldehyde-phosphate dehydrogenase (GAPDH) at the same concentration as in the mouse homogenate was used as an internal reference to ensure the equal protein loading. As shown in Figure 10A, two dark bands are observed at ca. 130

Figure 11. CD spectra of Aβ40 (20 μM) and CAM (12 μM), respectively, and those of Aβ40 in the presence of Zn2+ or Cu2+ (40 μM) with or without CAM (12 μM) after incubation at 37 °C.

incubated alone for 2 days, which indicates that Aβ is in the βsheet conformation.42 The CD spectrum became more negative after Aβ40 was coincubated with Zn2+, indicating that the conformational transformation is greatly aggravated by the metal ion, which is consistent with the above ThT fluorescence assay. However, in the presence of CAM, the spectra of Aβ40 show a dramatic inversion from the negative bands at ca. 215 nm to the positive bands at 220−230 nm, indicating that new conformations are formed. These results indicate that CAM could act as a modulator to transform the β-sheet conformation of Aβ to other conformers. In the case of chelators reported in the literature,43 the CD spectrum of Aβ usually shows a gradual recession in the negative band, reflecting the disaggregation of the β-sheet-rich aggregates to monomers of Aβ. The different effects of CAM and organic chelators on Aβ aggregates suggest that CAM does not act as a chelating agent but as a conformational transformation agent (CTA) when interacts with the Aβ aggregates. As a CTA, CAM is expected to eradicate the toxic β-sheet-rich aggregates as well as to avoid the transformation of Aβ plaques into the baneful oligomers.17 In addition, indiscriminative chelators are prone to remove the essential metals from the metalloproteins and to disturb the normal physiological functions,44 whereas CAM acts directly on Aβ aggregates without disturbing the cerebral metal homeostasis, which may be beneficial to prevent the unexpected side effects induced by the chelation therapy.16,17,44 The effect of CAM on the conformational transformation of Aβ was also confirmed by the IR spectra. A strong peak is observed at 1620 cm−1 after Aβ40 was incubated for 2 days (Figure S12), which indicates that the β-sheet conformation is formed.45 However, in the presence of CAM, the intensity of this peak declined obviously, suggesting that the β-sheet conformers have been disintegrated. To test whether CAM can disturb other conformations of proteins such as the α-helix, the CD spectra of myoglobin (Mb) and bovine serum albumin (BSA) were determined in the presence of CAM. As shown in Figure 12, two negative peaks are observed at about 210 and 222 nm when Mb is incubated alone, which are assigned to the native α-helix motifs of Mb. In the presence of CAM, the negative bands remain almost unchanged, which indicate that CAM barely interfere the conformation of α-helix unless at very high concentrations (∼200 μM, Figure S13). Similarly, the interfering effect of CAM on the conformation of BSA is also negligible. The weak interaction between CAM and Mb or BSA may arise from their unmatchable size or unfavorable solubility. For instance, α-

Figure 10. Immunoblots of brain homogenates from APPswe transgenic mice in SDS-PAGE after treatment with CAM analyzed by Western blotting with the anti-Aβ antibody 6E10. (A) Brain homogenates of four 7-month-old APPswe transgenic mice (Lanes 1− 4) and those with CAM (0.6 μM, Lanes 5−8); (B) Low MW brain homogenates after treating with different concentrations of CAM (Lanes 0−5, 0, 0.2, 0.4, 0.6, 0.8, 1.0 μM, respectively).

and 55 kDa in the control group (Lanes 1−4), and two other pale bands are observed at ca. 40 kDa; in addition, some fine ambiguous bands are observed in the high molecular weight (MW) region (55−130 kDa). These results indicate that large amount of Aβ aggregates in different maturity stages coexist in the brain homogenates of APP mice, with the high MW species being the dominant components. After treatment with CAM (Lanes 5−8), however, the dark bands in the high MW region (130−55 kDa) become blurry and weak, while the bands at ca. 40 kDa darkened, and a new band is observed at ca. 35 kDa, indicating that CAM can intervene with the high MW Aβ species and reduce the aggregates. Figure 10B shows that as the concentration of CAM increases, the Aβ species at ca. 15 and 35 kDa also arise. These results indicate that CAM can disaggregate the Aβ aggregates with high MW and convert them into low MW species, which is in accord with the morphological analysis in vitro. By contrast, the intensity of GAPDH bands is not interfered by CAM even at high concentrations, which suggests that CAM possesses a high specificity for Aβ aggregates. Conformational Transformation of Aβ. The conversion of an α-helix or unordered conformation into the β-sheet is an acknowledged conception in the formation of toxic Aβ aggregates, which is a crucial process in the development of AD.7,8,33 Since CAM can inhibit the β-sheet-related Aβ aggregation, we hence studied the effect of CAM on the selfor metal-induced conformational transition of Aβ40 using circular dichroism (CD) spectrometry. As shown in Figure 11, a negative band is observed at 215 nm after Aβ40 was F

DOI: 10.1021/acs.inorgchem.8b02115 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

Figure 14. 1H NMR spectra of Aβ40 (0.2 mM) in the absence (A) and presence (B) of CAM (0.1 mM) after incubation at 37 °C and pH 7.4 for 24 h. Figure 12. CD spectra of Mb and BSA (20 μM) alone and with CAM (12 μM) after incubation at 37 °C.

Molecular Simulation. To further understand the interaction between CAM and Aβ40, we adopted the crystallographic data of CAM and Aβ40 to perform the blind docking calculations using the Autodock software.49 The calculated docking energies are summarized in Table 1. For the

helix is hydrophilic while β-sheet is lipophilic, and thus, the lipophilic CAM can more readily approach the β-sheet structure. Nature of Interactions. The interactions between CAM and Aβ40 were first investigated by ESI-MS. As shown in Figure 13, two peaks corresponding to different species of Aβ

Table 1. Lowest Docking Energy (kJ mol−1) for the β-Sheet and α-Helix Conformation of Aβ in the Reaction with CAM Calculated by the Autodock Program conformer

EHVda

Ee

Eimb

Etc

ΔGd

Ka

β-sheet α-helix

−43.51 −34.39

−0.08 −0.79

−43.59 −35.18

+7.94 +4.98

−36.74 −27.70

3.15 × 106 7.09 × 104

EHVd = EH + EV + Ed; bEim = EHVd + Ee; ctorsional energy; dΔG = Eim + Et a

β-sheet, the H-bonding energy (EH), van der Waals energy (EV), and desolvo energy (Ed) are the main contributors for the intermolecular energy (Eim), whereas the electrostatic energy (Ee) hardly contributes to the Eim. Since H-bonding is the strongest force among the intra- or intermolecular weak interactions, it is the primary factor responsible for the interaction between CAM and the β-sheet of Aβ aggregates. The negative free energy (ΔG) indicates that the interaction between CAM and Aβ aggregates proceed spontaneously. The equilibrium constant (Ka) of the interaction between CAM and the β-sheet of Aβ is calculated to be 3.15 × 106, indicating that the interaction is fairly strong. By contrast, the Ka of the interaction between CAM and the α-helix of Aβ is 2 orders of magnitude less than that with the β-sheet. These results provide further explanation for the above observation that CAM can effectively disturb the β-sheet of Aβ rather than the α-helix of proteins. The binding mode of CAM to Aβ40 was simulated by the PyMOL software. As shown in Figures 15A and S14, a cavity with a diameter of 12−16 Å was formed during the β-sheet aggregation, which could suitably accommodate CAM with a diameter of 10.40 Å (vide ante). Furthermore, most of the amino acid residues in the cavity, such as Gln15, His14, and Leu17 (Figure 15B), could potentially interact with CAM through hydrogen bonds. According to the simulated binding mode, the fit sizes of CAM and Aβ cavity as well as the Hbonding-favored amino acid residues play pivotal roles in the interaction between CAM and Aβ40 aggregates.

Figure 13. ESI-MS spectra of Aβ40 in the absence and presence of CAM after incubation at 37 °C and pH 7.4 for 24 h.

are observed. After Aβ40 was coincubated with CAM for 24 h, no peak for new peptide or peptide-CAM adduct was observed. The results indicate that the interaction between CAM and Aβ is neither a coordination interaction nor a cleavage action, but some kind of weak interactions.20,46 The weak interaction between CAM and Aβ40 was further studied by 1H NMR. As shown in Figure 14, the proton signals assignable to the imidazole of His residues only slightly shift to the lower magnetic field in the presence of CAM,47 which suggest that CAM may affect the chemical environment of protons in Aβ40 through H-bonding interactions.48 So far as we know, modulators of Aβ aggregates working on the weak interactions such as H-bonding are rarely seen. In comparison with metal chelators or other agents that work on covalent interactions, the reversible association of CAM with Aβ40 suggest that its impact on the structure and function of biomolecules is mild, which explains why CAM shows an excellent biocompatibility in vivo.

■

CONCLUSIONS The accumulation of misfolded and misassembled proteins plays a key role in the mechanism of Alzheimer’s disease (AD). G

DOI: 10.1021/acs.inorgchem.8b02115 Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry

■

Article

EXPERIMENTAL SECTION

Materials. Reagents used in this study were all of analytical grade, purchased from commercial suppliers and used as received unless otherwise stated. Human Aβ40 and Aβ42 was purchased from GL Biochem Ltd. (China) and verified by HPLC and electrospray ionization mass spectrometry (ESI-MS). Thioflavine T (ThT), 2′,7′dichlorofluorescin diacetate (DCFH-DA), 3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2-H-tetrazolium bromide (MTT), nerve growth factor 7S (NGF-7S), and tris(hydromethyl)aminomethane (Tris) were purchased from Sigma-Aldrich. Zn(OAC)2, CuCl2, CoCl2, (NH4)6Mo7O24·2H2O, NaAsO2, and 2-(1H-pyrazol-3-yl)pyridine (C8H7N3) were purchased from J&K. Stock solutions of Aβ40, Zn2+ and Cu2+ were prepared according to the reported procedures;31 that of CAM was prepared by dissolving the compound in dimethyl sulfoxide (DMSO) to give a final concentration of 5 mM and filtered through a filter (0.22 μm, organic system). All the solutions were prepared with Milli-Q water and filtered through a 0.22 μm filter (Millipore). PC12 cells were purchased from American Type Culture Collection (ATCC). C57BL/B6 mice were purchased from the Model Animal Research Center of Nanjing University (MARC). All the animal tests were performed according to the regulations approved by MARC. X-ray Single Crystal Diffraction. The single crystal data of CAM were collected on a Bruker CCD, Apex-II diffractometer with graphite monochromated Mo Kα (λ = 0.71073 Å) radiation at room temperature. Routine Lorentz and polarization corrections were applied and an absorption correction was performed using the SADABS program. The structure was solved by direct methods and refined using full-matrix least-squares on F2. All calculations were performed using the SHELXL-97 program package. Synthesis of CAM. (NH4)6Mo7O24·2H2O (0.145 g), NaAsO2 (0.182 g), CoCl2·6H2O (0.225 g), and 2-(1H-pyrazol-3-yl)pyridine (C8H7N3, 0.145 g) were dissolved in a C2H5OH/water (2.0/10 mL) solution and was adjusted to pH 3.65 by a HCl solution (4 mol·L−1). After the solution was stirred for 1 h, it was sealed in a Teflon-lined autoclave (15 mL) and kept at 150 °C for 5 days. The black block crystals of CAM suitable for the X-ray crystallographic analysis were obtained by slow cooling the reactive solution to room temperature (yield: ca. 21% based on (NH4)6Mo7O24·2H2O). Lipophilicity of CAM. The octyl alcohol/water distribution coefficient (Kow) of CAM was determined by the shake-flask method.53 Calculation of Bond Valence Sum (Σs). The bond valence sum and protonation level are calculated according to the reported method.27 Briefly, the oxidation states of the oxygen atoms in CAM were calculated on the following formula, which could be used to estimate the protonation of oxygen atoms for locating the H-acceptors or H-donors:54

Figure 15. Overview of the CAM-Aβ complex represented in a spacefilling mode (A) and the detailed surroundings of the CAM-Aβ complex represented in a ball-and-stick mode (B).

Amyloid β-peptide (Aβ) in β-sheet conformation is the core structure of the toxic aggregates. In this paper, we described a modified polyoxometalate (CAM) as a conformational modulator to disaggregate the β-sheet-rich Aβ aggregates that are crucial to the etiology of AD. CAM can prevent the self-misfolded or metal-induced Aβ aggregates and inhibit the ROS-associated neurotoxicity induced by Aβ aggregates. Importantly, CAM is biocompatible and capable of crossing the BBB. The uniqueness of CAM is that it specifically influences the β-sheet conformation of Aβ aggregates through H-bonding without disturbing the cerebral metal homeostasis and the conformation of other proteins, which may be beneficial to circumvent the unexpected side effects induced by chelation therapy or covalent agents. Aside from AD, a lot of other diseases such as Parkinson’s disease, Creutzfeldt−Jacob disease, new-variant Creutzfeldt− Jakob disease, Huntington’s disease, and type II diabetes are also remarkable examples of protein-misfolding diseases.50,51 In these diseases, large quantities of misfolded proteins destroy brain cells or other tissues, and the pathogenic processes are similar; that is, they convert the conformation of proteins to βsheet. Since hydrogen bond is the major force maintaining the secondary structure of misfolded proteins,52 the mechanism of action of CAM might be applicative to other proteinmisfolding diseases. As a proof of concept, the result of this study suggests that modulating the conformation of misfolded proteins could be a practical and powerful therapeutic strategy for the prevention of protein-misfolding diseases.

Vi =

i r0′ − rij zy zz zz k B {

∑ sij = ∑ expjjjjj j

j

where r0′ represents the theoretical value of bond distance between two atoms, and rij represents the observed values of bond distance that are listed in Table S2; B was set to 0.37.54 As shown in Figure S4, the results of XPS indicate that the oxidation states of Mo, As, Co in CAM are +6/+5, +5, and +2, repectively. Accordingly, the theoretical value of Mo−O, As−O, and Co−O can be obtained from the literature,27,54 that is, r0′(Mo6+−O) = 1.907 (2) Å, r0′(Mo5+−O) = 1.879 (7) Å, r0′(As5+−O) = 1.7673 (5) Å, and r0′(Co2+−O) = 1.625 (2) Å. The results are summarized in Table S3. Stability of CAM. The stability of CAM was studied by using UV−vis spectroscopy according to the reported method.22 Inhibition Effect on Aβ Aggregation. ThT assay. Aβ40 (20 μM) in buffer solution (20 mM Tris-HCl/150 mM NaCl, 993.6 μL) was incubated with or without Zn(OAc)2 or CuCl2 (4 μL, 10 mM) at 37 °C for 5 min. CAM (2.4 μL, 5 mM) or DMSO (2.4 μL, final concentration 0.24%) was added into the system with stirring. Each sample (300 μL) was transferred to a well of a flat-bottomed 96-well H

DOI: 10.1021/acs.inorgchem.8b02115 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry black plate (Corning Costar Corp). ThT (2 μL, 5 mM) was added to each solution simultaneously in the dark and incubated at 37 °C for 0.5 h. The fluorescence intensity (λex = 415 nm, λem = 485 nm) was measured by a Varioskan Flash microplate reader (Thermo Scientific) every 15 min from 2 to 24 h. Turbidity. Sample solutions were prepared as described above except the incubation time was extended to 24 h. Each sample was transferred to a well of a flat-bottomed 96-well plate. Turbidity of the solution was measured using the absorbance at 405 nm. Data were expressed as mean ± standard deviations of at least three independent experiments. Solubilizing effect. Aβ40 (20 μM) in buffer solution (20 mM TrisHCl/150 mM NaCl, 992 μL) was incubated with Zn(OAc)2 (4 μL, 10 mM) at 37 °C for 24 h. CAM solutions with the final concentration of 2−18 μM were added to each sample, respectively, and incubated at 37 °C for another 24 h. All the control groups were treated by DMSO with a final concentration of 0.36%. The solutions were divided into two parts: one for the ThT assay and the other for the turbidity test. Morphological Analysis. Sample solutions were prepared in the same way as described in the turbidity test.35 An aliquot of solution (10 μL) was spotted on the 300-mesh carbon-coated copper grids for 2 min at room temperature and the excess sample was removed. Each grid was stained with uranyl acetate (10 μL, 1%, w/v) for 1 min and washed with Milli-Q water (10 μL) and then was examined on a JEOL JEM-2100 LaB6 (HR) transmission electron microscope. Inhibition of ROS Generation. DCFH solution (1 mM) was prepared with a buffer (20 mM Tris-HCl/150 mM NaCl, pH 7.4) according to the reported procedures.20 Horseradish peroxidase (HRP) stock solution (4 μM) was prepared with the same buffer. Sample solutions containing Aβ40 (20 μM) and CuCl2 (40 μM) were incubated with or without CAM (12 μM) at 37 °C for 23 h. Ascorbate (10 μM) was added to each sample and incubated at 37 °C for 1 h. All the controls were treated by DMSO with the final concentration of 0.24%. The sample (200 μL) was transferred to the wells of a flat-bottomed 96-well black plate. HRP (0.04 μM) and DCFH (100 μM) were added to each solution and incubated in the dark at 37 °C for 1 h. Fluorescence spectra (λex = 485 nm) in the range of 505−650 nm were measured by a Varioskan Flash microplate reader (Thermo Scientific). Inhibition of Neurotoxicity. The inhibition effect of CAM on the neurotoxicity was evaluated using PC12 cells by the MTT assay. PC12 cells were pretreated with Aβ40 alone or with Zn2+- or Cu2+induced Aβ40 complexes for 36 h before the assay and then treated by CAM (12 μM) for another 36 h. The PC12 cells used for the morphological analysis were prepared as described previously.55 Aβ40 (20 μM), Zn2+ or Cu2+ (40 μM) were incubated with the cell culture medium for 15 min, and CAM (12 μM) was added to the medium. After incubation for 48 h, the morphological pictures of the cells were captured by a microscope. Control cells were treated with DMSO at the final concentration of 0.24%. BBB Penetration Ability. CAM in saline solution (DMSO 5%) was administered to the B6-mice through the tail vein injection at a dose of 80 mg·kg−1. The blood and brain tissues were collected at different time points and homogenized. The homogenates were nitrated and the content of Mo was determined by ICP-MS. The dose-dependent uptake of Mo in the brain was determined in the same way. The blank group was treated with saline containing 5% DMSO. Analysis of Brain Homogenates. Brain extracts of 7-month-old APPswe transgenic mice and samples for Western blot analysis were prepared according to the reported procedure.20 Briefly, the extracted brain tissue pieces were homogenized in NP40 buffer (Tris-HCl, 50 mM; NaCl, 150 mM; NP-40, 1%; Na-deoxycholate, 0.25%; 1 mM of EDTA, Na3VO4, NaF; 1 μg mL−1 of aprotinin, leupeptin, pepstatin; and 1 mM of fresh PMSF was added along with the buffer solution). The homogenate solutions were centrifuged at 12 000 rpm and 4 °C for 10 min, and the resulting supernatants were collected. CAM (1 μL, 60 μM) or DMSO (1 μL, final concentration 1%) were added,

respectively, to the supernatant (100 μL). After incubation at 37 °C for 24 h, the samples were dissolved in loading buffer containing βmercaptoethanol (5%) and boiled at 95 °C for 10 min. Each sample was separated by SDS polyacrylamide gel electrophoresis (PAGE) and transferred onto polyvinylidene difluoride (PVDF) membranes. The membranes were blocked for 0.5 h at room temperature with fat-free milk (5%) and then incubated at 4 °C for 12 h with monoclonal antiAβ antibody 6E10 (1:1000, Covance Inc.). The membranes were then incubated with the HRP-conjugated goat antimouse antibody (1:1000) for 1 h at room temperature. Bands were visualized using ChemiScope 3400 mini (CLiNX Science Instruments Co., Ltd.). The concentration gradient analysis was performed according to the above methods, except that CAM was used at 0, 0.2, 0.4, 0.6, 0.8, and 1.0 μM, respectively. The membranes were then developed using an enhanced chemiluminescence detection kit, stripped and reprobed with GAPDH antibody (1:5000; Meridian Life Sciences, Brockville, ON, Canada) to ensure equal protein loading. Conformational Transformation. Aβ40, Mb or BSA (20 μM) was dissolved in the buffer solution (20 mM Tris-HCl/150 mM NaCl) respectively and incubated with or without Zn(OAc)2 or CuCl2 (40 μM) at 37 °C for 2 days. CAM (12 μM) was added to each solution and incubated at 37 °C for 1 day. The CD spectra of the solutions were recorded on a JASCO J-810 automatic recording spectropolarimeter (Tokyo, Japan) in the range of 200−260 nm. The data acquired in the absence of protein were subtracted from the spectrum. The samples for IR spectra were prepared in the same way as described above. The IR spectra were recorded on a NICOLET iS10 spectrometer in the range of 400−4000 cm−1. Controls were treated with DMSO at the final concentration of 0.24%. In all the above control tests, DMSO gave negative results. Nature of Interactions. Aβ40 (20 μM) was dissolved in the buffer solution (20 mM Tris-HCl/150 mM NaCl, pH 7.4, 996 μL) and incubated with CAM (4 μL, 5 mM) at 37 °C for 24 h. The solution was loaded on the chromatographic column and the saline ions were washed off by Milli-Q water. The products were collected using an acid solution composed of 75% acetonitrile, 20% Milli-Q water and 5% acetic acid. The peptide solution was examined on an LCQ Fleet electrospray mass spectrometer. The samples of 1H NMR spectra were prepared by dissolving CAM (0.1 mM) in a mixture containing Aβ40 (0.2 mM), 10% D2O, 85% H2O and 5% DMSO-d6 and incubating at 37 °C for 24 h and then were centrifuged to get the soluble samples. The 1H NMR spectra were recorded on a Bruker DRX-600 spectrometer. Molecular Simulation. The molecular simulations were performed by the blind docking calculations using the Autodock software.49 The crystallographic data of Aβ40 fibrils (PBD ID 2lmo) were obtained from the Protein Data Bank (http://www.rcsb.org/ pdb/explore/explore.do?structureId=2lmo). Gaussian 09 was used to optimize the three-dimensional structure of CAM (CCDC 901487) at DFT B3LYP/LanL2DZ base level. Autodock 4.2.3 program was used to perform the blind docking calculations of Aβ fibrils with CAM. A grid box of 80 Å × 80 Å × 80 Å with spacing of 0.357 Å was used to enclose compound and Aβ40 fibrils. The Lamarckian Genetic Algorithm method was used as the searching algorithm. PyMOL software was used to analyze the predicated binding mode [Delano, W.L. The PyMol Molecular Graphic System, DeLano Scientific, San Carlos, CA, USA, 2004, http://pymol.sourceforge.net.].

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.inorgchem.8b02115. Crystallographic data and structural refinements for CAM; Selected bond length for CAM; Polyhedral representation of the α-Keggin and ε-Keggin unit of CAM; XPS spectra of As, Co and Mo; The location and numbering of Ot, Oμ2 and Oμ3; Bond valence sum (Σs) I

DOI: 10.1021/acs.inorgchem.8b02115 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

(11) Selkoe, D. J. Cell biology of protein misfolding: the examples of Alzheimer’s and Parkinson’s diseases. Nat. Cell Biol. 2004, 6, 1054− 1061. (12) Hardy, J.; Selkoe, D. J. The amyloid hypothesis of Alzheimer’s disease: progress and problems on the road to therapeutics. Science 2002, 297, 353−356. (13) Ghosh, A. K.; Osswald, H. L. BACE1 (β-secretase) inhibitors for the treatment of Alzheimer’s disease. Chem. Soc. Rev. 2014, 43, 6765−6813. (14) Savelieff, M. G.; De Toma, A. S.; Derrick, J. S.; Lim, M. H. The ongoing search for small molecules to study metal-associated amyloidβ species in Alzheimer’s disease. Acc. Chem. Res. 2014, 47, 2475− 2482. (15) Potter, H.; Dressler, D. The potential of BACE inhibitors for Alzheimer’s therapy. Nat. Biotechnol. 2000, 18, 125−126. (16) Barnham, K. J.; Bush, A. I. Biological metals and metal-targeting compounds in major neurodegenerative diseases. Chem. Soc. Rev. 2014, 43, 6727−6749. (17) Sharma, A. K.; Pavlova, S. T.; Kim, J.; Finkelstein, D.; Hawco, N. J.; Rath, N. P.; Kim, J.; Mirica, L. M. Bifunctional compounds for controlling metal-mediated aggregation of the Aβ42 peptide. J. Am. Chem. Soc. 2012, 134, 6625−6636. (18) Park, L.; Wang, G.; Moore, J.; Girouard, H.; Zhou, P.; Anrather, J.; Iadecola, C. The key role of transient receptor potential melastatin-2 channels in amyloid-β-induced neurovascular dysfunction. Nat. Commun. 2014, 5, 5318−5350. (19) Ratnayaka, J. A.; Serpell, L. C.; Lotery, A. J. Dementia of the eye: the role of amyloid-β in retinal degeneration. Eye 2015, 29, 1013−1026. (20) Wang, X. H.; Wang, X. Y.; Zhang, C. L.; Jiao, Y.; Guo, Z. J. Inhibitory action of macrocyclic platiniferous chelators on metalinduced Aβ aggregation. Chem. Sci. 2012, 3, 1304−1312. (21) Geng, J.; Li, M.; Ren, J. S.; Wang, E. B.; Qu, X. G. Polyoxometalates as inhibitors of the aggregation of amyloid-β peptides associated with Alzheimer’s disease. Angew. Chem., Int. Ed. 2011, 50, 4184−4188. (22) Gao, N.; Sun, H. J.; Dong, K.; Ren, J. S.; Duan, T. C.; Xu, C.; Qu, X. G. Transition-metal-substituted polyoxometalate derivatives as functional anti-amyloid agents for Alzheimer’s disease. Nat. Commun. 2014, 5, 3422−3431. (23) Malone, R. Bioinorganic Chemistry, 2nd ed.; John Wiley & Sons, Inc: Hoboken, NJ, 2007; pp 1−3. (24) Sun, H. Z. Biological Chemistry of Arsenic, Antimony and Bismuth; John Wiley & Sons, Ltd.: Chichester, U.K., 2011; pp 113− 128. (25) Baker, L. C.; Glick, D. C. Present general status of understanding of heteropoly electrolytes and a tracing of some major highlights in the history of their elucidation. Chem. Rev. 1998, 98, 3−50. (26) Baskar, V.; Shanmugam, M.; Helliwell, M.; Teat, S. J.; Winpenny, R. E. Reverse-Keggin ions: polycondensation of antimonate ligands give inorganic cryptand. J. Am. Chem. Soc. 2007, 129, 3042−3043. (27) Zheng, S. T.; Zhang, J.; Clemente-Juan, J. M.; Yuan, D. Q.; Yang, G. Y. Poly(polyoxotungstate)s with 20 nickel centers: from nanoclusters to one-dimensional chains. Angew. Chem., Int. Ed. 2009, 48, 7176−7179. (28) Noël, S.; Cadet, S.; Gras, E.; Hureau, C. The benzazole scaffold: a SWAT to combat Alzheimer’s disease. Chem. Soc. Rev. 2013, 42, 7747−7762. (29) Miller, Y.; Ma, B. Y.; Nussinov, R. Zinc ions promote Alzheimer Aβ aggregation via population shift of polymorphic states. Proc. Natl. Acad. Sci. U. S. A. 2010, 107, 9490−9495. (30) Zhu, L.; Song, Y.; Cheng, P. N.; Moore, J. S. Molecular design for dual modulation effect of amyloid protein aggregation. J. Am. Chem. Soc. 2015, 137, 8062−8068. (31) Chen, T. T.; Wang, X. Y.; He, Y. F.; Zhang, C. L.; Wu, Z. Y.; Liao, K.; Wang, J. J.; Guo, Z. J. Effects of cyclen and cyclam on

and protonation level (ΣH) of Ot, Oμ2 and Oμ3; UV−vis spectra of CAM after incubation and at different pH values; IR spectra for Aβ40 and Aβ40 incubated with CAM (PDF) Accession Codes

CCDC 901487 contains the supplementary crystallographic data for this paper. These data can be obtained free of charge via www.ccdc.cam.ac.uk/data_request/cif, or by emailing [email protected], or by contacting The Cambridge Crystallographic Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK; fax: +44 1223 336033.

■

AUTHOR INFORMATION

Corresponding Authors

*Tel: +862589684549. Fax: +862583314502. E-mail: [email protected] (X.W.). *Tel: +862589684549. Fax: +862583314502. E-mail: zguo@ nju.edu.cn (Z.G.). ORCID

Zijian Guo: 0000-0003-4986-9308 Xiaoyong Wang: 0000-0002-8338-9773 Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We appreciate the financial support from the National Natural Science Foundation of China (21877059, 31570809, 21731004), the National Basic Research Program of China (2015CB856300), the Natural Science Foundation of Jiangsu Province (BK20150054), and the Research Foundation for Advanced Talents of Taiyuan Institute of Technology (1800006003).

■

REFERENCES

(1) Silva, J. L.; Vieira, T. C. R. G.; Gomes, M. P. B.; Bom, A. P. A.; Lima, L. M. T. R.; Freitas, M. S.; Ishimaru, D.; Cordeiro, Y.; Foguel, D. Ligand binding and hydration in protein misfolding: insights from studies of prion and p53 tumor suppressor proteins. Acc. Chem. Res. 2010, 43, 271−279. (2) Vinters, H. V. Emerging concepts in Alzheimer’s disease. Annu. Rev. Pathol.: Mech. Dis. 2015, 10, 291−319. (3) Selkoe, D. J. Folding proteins in fatal ways. Nature 2003, 426, 900−904. (4) Kepp, K. P. Bioinorganic chemistry of Alzheimer’s disease. Chem. Rev. 2012, 112, 5193−5239. (5) Faller, P.; Hureau, C.; La Penna, G. Metal ions and intrinsically disordered proteins and peptides: from Cu/Zn amyloid-β to general principles. Acc. Chem. Res. 2014, 47, 2252−2259. (6) Aguzzi, A. Alzheimer’s disease under strain. Nature 2014, 512, 32−34. (7) Selkoe, D. J. Resolving controversies on the path to Alzheimer’s therapeutics. Nat. Med. 2011, 17, 1060−1065. (8) Guo, J. L.; Lee, V. M. Y. Cell-to-cell transmission of pathogenic proteins in neurodegenerative diseases. Nat. Med. 2014, 20, 130−138. (9) Gaggelli, E.; Kozlowski, H.; Valensin, D.; Valensin, G. Copper homeostasis and neurodegenerative disorders (Alzheimer’s, prion, and Parkinson’s diseases and amyotrophic lateral sclerosis). Chem. Rev. 2006, 106, 1995−2044. (10) Jang, H.; Teran Arce, F.; Ramachandran, S.; Kagan, B. L.; Lal, R.; Nussinov, R. Disordered amyloidogenic peptides may insert into the membrane and assemble into common cyclic structural motifs. Chem. Soc. Rev. 2014, 43, 6750−6764. J

DOI: 10.1021/acs.inorgchem.8b02115 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry zinc(II)- and copper(II)-induced amyloid β-peptide aggregation and neurotoxicity. Inorg. Chem. 2009, 48, 5801−5809. (32) Harper, J. D., Jr; Lansbury, P. T. Models of amyloid seeding in Alzheimer’s disease and scrapie: mechanistic truths and physiological consequences of the time-dependent solubility of amyloid proteins. Annu. Rev. Biochem. 1997, 66, 385−407. (33) Ono, K.; Condron, M. M.; Teplow, D. B. Structureneurotoxicity relationships of amyloid β-protein oligomers. Proc. Natl. Acad. Sci. U. S. A. 2009, 106, 14745−14750. (34) Selkoe, D. J.; Hardy, J. The amyloid hypothesis of Alzheimer’s disease at 25 years. EMBO Mol. Med. 2016, 8, 595−608. (35) Kochi, A.; Eckroat, T. J.; Green, K. D.; Mayhoub, A. S.; Lim, M. H.; Garneau-Tsodikova, S. A novel hybrid of 6-chlorotacrine and metal-amyloid-β modulator for inhibition of acetylcholinesterase and metal-induced amyloid-β aggregation. Chem. Sci. 2013, 4, 4137−4145. (36) Musiek, E. S.; Holtzman, D. M. Three dimensions of the amyloid hypothesis: time, space and ‘wingmen’. Nat. Neurosci. 2015, 18, 800−806. (37) Woo, J. A.; et al. Slingshot-cofilin activation mediates mitochondrial and synaptic dysfunction via Aβ ligation to β1-integrin conformers. Cell Death Differ. 2015, 22, 921−934. (38) Ohnishi, T.; Yanazawa, M.; Sasahara, T.; Kitamura, Y.; Hiroaki, H.; Fukazawa, Y.; Kii, I.; Nishiyama, T.; Kakita, A.; Takeda, H.; et al. Na, K-ATPase α3 is a death target of Alzheimer patient amyloid-β assembly. Proc. Natl. Acad. Sci. U. S. A. 2015, 112, E4465−E4474. (39) Tanzi, R. E. The synaptic Aβ hypothesis of Alzheimer disease. Nat. Neurosci. 2005, 8, 977−979. (40) De Felice, F. G.; et al. Protection of synapses against Alzheimer’s-linked toxins: Insulin signaling prevents the pathogenic binding of Aβ oligomers. Proc. Natl. Acad. Sci. U. S. A. 2009, 106, 1971−1976. (41) Chen, Y.; Liu, L. Modern methods for delivery of drugs across the blood-brain barrier. Adv. Drug Delivery Rev. 2012, 64, 640−665. (42) Li, M.; Howson, S. E.; Dong, K.; Gao, N.; Ren, J. S.; Scott, P.; Qu, X. G. Chiral metallohelical complexes enantioselectively target amyloid-β for treating Alzheimer’s disease. J. Am. Chem. Soc. 2014, 136, 11655−11663. (43) Rodriguez-Rodriguez, C.; et al. Design, selection, and characterization of thioflavin-based intercalation compounds with metal chelating properties for application in Alzheimer’s disease. J. Am. Chem. Soc. 2009, 131, 1436−1451. (44) Storr, T.; Thompson, K. H.; Orvig, C. Design of targeting ligands in medicinal inorganic chemistry. Chem. Soc. Rev. 2006, 35, 534−544. (45) Oladepo, S. A.; Xiong, K.; Hong, Z.; Asher, S. A.; Handen, J.; Lednev, I. K. UV resonance Raman investigations of peptide and protein structure and dynamics. Chem. Rev. 2012, 112, 2604−2628. (46) Lee, B. I.; et al. Photoexcited porphyrins as a strong suppressor of β-Amyloid aggregation and synaptic toxicity. Angew. Chem., Int. Ed. 2015, 54, 11472−11476. (47) Hou, L. M.; Zagorski, M. G. NMR reveals anomalous copper(II) binding to the amyloid Aβ peptide of Alzheimer’s disease. J. Am. Chem. Soc. 2006, 128, 9260−9261. (48) Del Bene, J. E.; Jordan, M. J. T. Vibrational spectroscopic and NMR properties of hydrogen-bonded complexes: Do they tell us the same thing? J. Am. Chem. Soc. 2000, 122, 4794−4797. (49) Trott, O.; Olson, A. J. AutoDock Vina: improving the speed and accuracy of docking with a new scoring function, efficient optimization, and multithreading. J. Comput. Chem. 2009, 31, 455− 461. (50) Frost, B.; Diamond, M. I. Prion-like mechanisms in neurodegenerative diseases. Nat. Rev. Neurosci. 2010, 11, 155−159. (51) Ehrnhoefer, D. E.; Wong, B. K.; Hayden, M. R. Convergent pathogenic pathways in Alzheimer’s and Huntington’s diseases: shared targets for drug development. Nat. Rev. Drug Discovery 2011, 10, 853−867. (52) Chiti, F.; Dobson, C. M. Protein misfolding, functional amyloid, and human disease. Annu. Rev. Biochem. 2006, 75, 333−366.

(53) Tanaka, H.; Yamamoto, Y.; Chuman, H. Determination of the distribution coefficients of phenol derivatives and others based on a volume ratiometry coupled with a shake-flask method. Anal. Sci. 2002, 18, 485−487. (54) Brown, I. D.; Altermatt, D. Bond-valence parameters obtained from a systematic analysis of the inorganic crystal-structure database. Acta Crystallogr., Sect. B: Struct. Sci. 1985, 41, 244−247. (55) Yang, T.; Wang, X. H.; Zhang, C. L.; Ma, X.; Wang, K.; Wang, Y. Q.; Luo, J.; Yang, L.; Yao, C.; Wang, X. Y. Specific self-monitoring of metal-associated amyloid-β peptide disaggregation by a fluorescent chelator. Chem. Commun. 2016, 52, 2245−2248.

K

DOI: 10.1021/acs.inorgchem.8b02115 Inorg. Chem. XXXX, XXX, XXX−XXX