A Nanoscale Cobalt Functionalized Strandberg-Type Phosphomolybdate with β-Sheet Conformation Modulation Ability in Anti-Amyloid Protein Misfolding

Abstract

For decades, amyloid β-peptide (Aβ) misfolding aggregates with β-sheet structures have been linked to the occurrence and advancement of Alzheimer’s disease (AD) development and progression. As a result, modulating the misfolding mode of Aβ has been regarded as an important anti-amyloid protein misfolding strategy. A polyoxometalate based on {Co(H₂O)₄}²⁺ complex and [P₂Mo₅O₂₃]⁶⁻ fragments, K₈{[Co(H₂O)₄][HP₂Mo₅O₂₃]₂}·8H₂O (abbreviated as CoPM), has been synthesized and structurally characterized using elemental analysis, single-crystal X-ray diffraction (SXRD), IR, UV spectra, bond valence sums (Σs) calculation, and powder XRD (PXRD). CoPM’s primary component, as revealed by structural analysis, is a nanoscale polyoxoanion made of [Co(H₂O)₄]²⁺ sandwiched between two [P₂Mo₅O₂₃]⁶⁻ pieces. Notably, it is demonstrated that CoPM efficiently modulates Aβ aggregates’ β-sheet-rich conformation.

1. Introduction

Amyloid protein misfolding is linked to many disorders [1,2], with AD serving as a prevalent model for biochemistry and inorganic chemistry research [3]. The primary histopathological indicator of AD is the formation of extracellular senile plaques in the brain created by β-sheet-rich Aβ [4,5,6]. Additionally, it has been shown that aberrant cerebral metal ions, such as Zn²⁺, Cu²⁺, and others, might encourage the misfolding aggregation of Aβ [7,8]. Furthermore, the Cu²⁺-Aβ aggregates can generate reactive oxygen species (ROS), which is another causative factor in neuronal death [9]. As a result, controlling the detrimental misfolding aggregates may be key to treating these conditions [10,11,12].

Cobalt has long been recognized as a critical component of the human body, and the majority of its constituents, including cobalamin, have been linked to neuro-nutrition [13]. According to biochemical studies, the protein denaturation process may be significantly influenced by the strong coordination interaction between cobalt ions and nitrogenous heterocyclic compounds [14]. These properties may be useful for regulating the misfolding aggregates by substituting the inducing ions [15]. Finding the appropriate ligand is thus crucial for the creation of a complex that boosts both selectivity and steric hindrance to limit the active binding sites of cobalt.

Polyoxometalates (POMs), known as groups of metal–oxygen clusters, have multitudinous structures and intriguing characteristics [16,17,18,19,20,21,22,23]. POMs have several unique, unmatched characteristics that enable them to serve as substantial polydentate ligands for transition-metal ions with various coordination modes, such as oxygen-rich nucleophilic surfaces, nanoscale sites, and sites that form multiple bonds [24,25,26,27,28,29,30,31,32]. Recently, it was shown that some nanoscale POMs, because of their distinct nanoscale size, high negative charge, and oxygen-rich surface, had a remarkable capacity for modifying misfolded β-sheet-rich conformation [33,34,35,36]. As a result, combining cobalt complexes with nanoscale POM clusters to create a single hybrid molecule may not only preserve but also advantageously express the desirable properties of all individual components.

In the present work, we have synthesized and studied a cobalt complex sandwiched phosphomolybdate, K₈{[Co(H₂O)₄][HP₂Mo₅O₂₃]₂}·8H₂O (abbreviated as CoPM). The new molecular was fabricated from [P₂Mo₅O₂₃]⁶⁻ partnered with {Co(H₂O)₄}²⁺. As anticipated, CoPM can block Aβ’s β-sheet transformation. Furthermore, the decrease of Cu²⁺-induced misfolded aggregates may reduce the formation of ROS.

2. Results and Discussion

2.1. X-ray Single Crystal Structures

In this study, the CoPM structure was determined using the SXRD method. Table S1 provides a list of bond lengths. Comprehensive data has been submitted to the Cambridge Crystallographic Data Centre (CDCC) under the CCDC number 2129591. The structural analysis results are shown in Figure 1a, which shows that the CoPM unit cell includes eight K⁺ counter ions, eight crystallized water molecules, and one {Co(H₂O)₄}²⁺ complex sandwiched polyoxoanion [P₂Mo₅O₂₃]⁶⁻ (Figure 1a,b).

As shown in Figure 1c–g, the geometry of Strandberg-type [P₂Mo₅O₂₃]⁶⁻ cluster can be deemed as a puckered ring of five nearly coplanar corner-sharing/edge-sharing distorted MoO₆ octahedra [Mo–O: 1.679–2.496 Å] with two capping PO₄ tetrahedra [P–O: 1.510–1.586 Å] on both poles of the {Mo₅O₂₁} ring centers. Furthermore, two such [P₂Mo₅O₂₃]⁶⁻ fragments were connected by a distorted octahedral {Co(H₂O)₄}²⁺ complex [Co–O: 2.035–2.179 Å] to form the polyanion structure of CoPM.

According to a retrieval of the CCDC database, CoPM’s molecular structure is new. Hitherto, there are two similar POMs deposited in the CCDC database, [H₈(H₂O)₁₆][Co(H₂O)₄(HP₂Mo₅O₂₃)₂] (C1) [37] and (H₂en)₆{[Co(H₂O)₄](P₂Mo₅O₂₃)}₃·11H₂O (C2) [38]. However, there are some differences among them: First of all, they have different molecular formulas. Second, CoPM is a salt rather than an acid. Third, CoPM possesses an isolated structure rather than a one-dimensional extended structure.

Additionally, the bond valence sums (Σs) in CoPM were computed using the following formula [39]:

$$V_i={\displaystyle \sum _j}{s}_{ij}={\displaystyle \sum _j\exp }\left(\frac{{r_0}^{\prime }-r_{ij}}{B}\right)$$

(1)

In Formula (1), r_ij is the discovered bond distance provided in Table S1, and r₀′ is the theoretical one between two atoms. The values for r₀′(Mo⁶⁺–O) (1.903 Å), r₀′(P⁵⁺–O) (1.624 Å), and r₀′(Co²⁺–O) (1.698 Å) are from the literature [40,41]. B is defined as having a value of 0.349 [40]. As a consequence, Co, Mo, and P have average valence state sums (Σs) of 1.988, 5.956, and 4.925 in CoPM, respectively.

POMs are readily protonated because they have a lot of basic surface O atoms and strongly negatively charged pieces [42]. The 50 oxygen atoms in CoPM can be divided into bridging O_µ2, O_µ3, and O_µ4 types, respectively, as well as terminal O_t. As illustrated in Figure 2 and

Table 1. Bond valence and Σs of Mo, Co, and P in CoPM.

Bond	Valence	Bond	Valence	Bond	Valence	Atom	Σs
Mo(1)-O(1)	1.764	Mo(1)-O(15)	0.997	Mo(1)-O(16)	0.298
Mo(1)-O(6)	1.689	Mo(1)-O(11)	0.910	Mo(1)-O(21)	0.262	Mo(1)	5.919
Mo(2)-O(2)	1.804	Mo(2)-O(12)	0.920	Mo(2)-O(19)	0.468
Mo(2)-O(7)	1.709	Mo(2)-O(11)	0.912	Mo(2)-O(16)	0.183	Mo(2)	5.997
Mo(3)-O(8)	1.743	Mo(3)-O(13)	1.012	Mo(3)-O(19)	0.295
Mo(3)-O(3)	1.719	Mo(3)-O(12)	0.905	Mo(3)-O(17)	0.268	Mo(3)	5.942
Mo(4)-O(4)	1.769	Mo(4)-O(13)	0.963	Mo(4)-O(20)	0.360
Mo(4)-O(9)	1.719	Mo(4)-O(14)	0.879	Mo(4)-O(18)	0.252	Mo(4)	5.941
Mo(5)-O(5)	1.734	Mo(5)-O(14)	0.994	Mo(5)-O(18)	0.381
Mo(5)-O(10)	1.694	Mo(5)-O(15)	0.867	Mo(5)-O(21)	0.310	Mo(5)	5.979
Co(1)-O(23)	0.408	Co(1)-O(1W)	0.308	Co(1)-O(2W)#5	0.278
Co(1)-O(23)#5	0.408	Co(1)-O(1W)#5	0.308	Co(1)-O(2W)	0.278	Co(1)	1.988
P(1)-O(17)	1.308	P(1)-O(18)	1.231	P(1)-O(16)	1.272
P(1)-O(22)	1.100					P(1)	4.911
P(2)-O(23)	1.331	P(2)-O(21)	1.156	P(2)-O(20)	1.301
P(2)-O(19)	1.151					P(2)	4.939

, according to the calculation of Σs, we can infer the distribution of protons on different oxygen atoms. Given that the oxidation state of O is −2, we can derive the equation Σs + ΣH = 2. Hence, the O atoms have delocalized protons with Σs values between 0 and 1.60, making them suitable as proton donors. Whereas O atoms with Σs values between 1.90 and 2.00 have dense electron populations. The numerous protons are often stated as being delocalized across the whole polyoxoanion, which is a common phenomenon in POM chemistry and has been extensively studied in the past, for example, [Ni(enMe)₂]₃[H₆Ni₂₀P₄W₃₄(OH)₄O₁₃₆(enMe)₈(H₂O)₆]·12H₂O [39], [H₃W₁₂O₄₀]⁵⁻ [42], [Cu(en)₂][Cu(en)₂H₂O]₂{[Cu(en)₂][Cu₆(en)₂(H₂O)₂(SiW₉O₃₄)₂]}·8H₂O [43], and [Na(H₂O)₅]₂[Ni(H₂O)(en)₂]₂[Ni(H₂O)₃(en)][H₄Ni₆(OH)₂(en)_2.5(B-α-As^VW₉O₃₄)]₂[W₄O₁₆][Ni₄ (H₂O)₂(en)₂]₂[Ni(H₂O)(en)][(α-As^VW₆O₂₆)]₂·3H₂O [44].

2.2. PXRD, IR and UV-Visible Spectrum

By comparing the actual PXRD pattern with the simulated pattern from a single SXRD, the phase purity of CoPM was verified (see Figure 3).

The IR spectrum of CoPM exhibits similar asymmetric those observed in other species containing [P₂Mo₅O₂₃]⁶⁻ [45]. As shown in Figure 4, four characteristic spectral lines are assigned to v(P–O_µ4), v(Mo–O_t), v(Mo–O_µ2), v(Mo–O_µ3) and v(Mo–O_µ4), respectively, and are situated at 1035, 933, 885, and 762~716 cm⁻¹, respectively [45]. Additionally, the stretching and bending vibrations of the –OH bond may be attributed to the corresponding spectral lines at 3460 and 1623 cm⁻¹, respectively [46]. The IR data were consistent with the structural elucidation from SXRD.

Figure 5 illustrates two UV-vis spectral absorption peaks in the wavelength range of 190~400 nm, one at 205.2 nm and the other with a broad shoulder adsorption center at 231.1 nm. These two peaks may be assigned to the O_t→Mo and O_µ→Mo charge transfer transitions, respectively [47].

Utilizing UV-vis spectroscopy, the impact of pH value on the stability of CoPM has been further investigated. Within a pH range of roughly 6.0–8.0, insignificant variations in UV-vis absorption intensity are observed; the absorption peak intensities at 205.2 and 231.1 nm eventually alter outside of this range, suggesting the beginning of skeletal collapse inside the CoPM. Therefore, it may be concluded that the pH range for CoPM stability is between 6.0 and 8.0. The stability of CoPM in Cu²⁺/Zn²⁺ containing solution or in Tris-buffer is shown in Figure S1, Figure S2, and Figure S3, respectively.

2.3. Modulation of Conformation

Transmission electron microscopy (TEM) was used to examine Zn²⁺- or Cu²⁺-treated Aβ40 with or without CoPM. Under self- or induction of Zn²⁺/Cu²⁺ conditions, certain fibrils can be detected in the groups of Aβ40, Aβ40 + Zn²⁺, and Aβ40 + Cu²⁺ (Figure 6a–c), which is indicative of β-sheet-rich misfolding protein shape [48]. Additionally, the fibrils in the Aβ40 + Zn²⁺ group are stronger after treatment with Zn²⁺ than those in the Aβ40 + Cu²⁺ or Aβ40 alone, which suggests that Zn²⁺ may exacerbate the β-sheet-relate conformational shift [49]. These findings show that the incubation fluids include abundant soluble “β-sheet-rich Aβ.” Aβ40 and Aβ40 were treated with Zn²⁺ or Cu²⁺, however, great morphological changes when incubated with CoPM (Figure 6d–f). It is interesting to note that the shape of the Aβ40 + CoPM group differs from that of Aβ40 alone, suggesting that CoPM may also be able to prevent Aβ from misfolding on itself. According to reports, mono-functional chelators have no direct interactions with the Aβ peptide and can only stop metal ions from causing conformational misfolding. Therefore, they have no impact on the Aβ peptide’s self-misfolding [49]. These results may indicate that CoPM can serve as an interfering agent in the formation of β-sheets rather than a chelation agent.

The Thioflavin T (ThT) assay was conducted to further examine the conformational modulation impact of CoPM on the fibrils of Aβ that are rich in β-sheets [50]. According to the amount of β-sheet-rich amyloid aggregates, the ThT may particularly attach to the β-sheet fibrils and cause a considerable enhancement in fluorescence, which has frequently been utilized to identify β-sheet content in incubation fluid [51]. As depicted in Figure 7, after one day of incubation at 37 °C with Zn²⁺, an Aβ40 solution (20 µM) exhibits significant fluorescence, demonstrating that the Aβ40 primarily occurs in the β-sheet-rich conformation [51]. However, a steady drop in fluorescence intensity was seen when CoPM was present. Since CoPM possesses no ability to quench the fluorescence of ThT (shown in Figure S4), it would mean that CoPM can prevent the development of β-sheet-rich fibrils that Zn²⁺ induces. The same phenomena were also observed in another two groups, which is consistent with morphology experiment results.

It has been established that the misfolding Cu²⁺-Aβ aggregates can be extremely effective in catalyzing ROS production [52]. Therefore, the DCF fluorescence test was conducted to further investigate the inhibitory impact of CoPM on ROS generation mediated by those Cu²⁺-Aβ species. Non-fluorescent 2′, 7′-dichlorofluorescein (DCFH) is transformed into DCF by interacting with ROS in the presence of horseradish peroxidase (HRP), a fluorescent probe that can measure the system’s total ROS output [53]. According to Figure 8, the DCF fluorescence intensity of Cu²⁺-Aβ and CoPM seems to be less intense than that of Cu²⁺-Aβ incubation alone, suggesting that the formation of ROS with CoPM is much lower than that without CoPM. These findings demonstrate that CoPM effectively inhibits Cu²⁺-Aβ complex ROS production [54]. Figure 8 shows that the Cu²⁺-Aβ plus Co²⁺ group creates a significant amount of ROS, suggesting that the Co²⁺ did not impact the experimental group’s HRP activity. It is interesting to note that the H₆P₂Mo₅O₂₃ (abbreviated as PM) Strandberg-type cluster is likewise capable of preventing ROS from modifying β-sheet-rich aggregates. We conclude that CoPM has a greater inhibitory impact than PM, suggesting that the unique structure of CoPM makes it have a higher modulation capacity.

3. Materials and Methods

All of the reagents used in the present investigation were analytically pure and used exactly as they were given to us. DCFH-DA (2′, 7′-dichlorofluorescein diacetate), HRP, tris(hydroxymethyl)aminomethane (Tris) were acquired from Sigma-Aldrich Inc. (Shanghai, China), whereas ascorbic acid, KCl, CuCl₂, ZnCl₂, CoCl₂·2H₂O, Na₂MoO₄·2H₂O, and Na₂HPO₄·12H₂O were bought from J & K Scientific Inc. (Beijing, China). Milli-Q water (Millipore, Burlington, MA, USA) was used to prepare all of the solutions, and a Millipore filter (0.22 µm) was used for all filtrations.

The Nicolet (Thermo Fisher Nicolet Inc., Waltham, MA, USA) 170 SX FTIR spectrometer was used to analyze the CoPM samples’ spectra while scanning from 4000 to 400 cm⁻¹. A UV-3600 spectrometer was used to record UV spectra between 190 and 400 nm. DCF fluorescence was done using a Thermo (Thermo Fisher Scientific Inc., Singapore) Scientific Varioskan Flash microplate reader. The Edinburgh (Edinburgh instruments Inc., Edinburgh, UK) Raman spectrometer RM5 was used to test the fresh and recrystallized samples of CoPM before and after incubation with Tris buffer solution scanning from 200 to 2000 cm⁻¹ (λ_ex = 785 nm). Graphite monochromatized Mo Kα radiation (λ = 0.71073 Å) at 296 K was utilized to measure CoPM single crystal intensity using a Bruker (Bruker Inc., Saarbrücken, Saarland, Germany) Apex-2 diffractometer with the aid of a CCD detector.

3.1. Synthesis

Solutions A and B were prepared separately. Solution A: Na₂MoO₄·2H₂O (2.416 g, 10.00 mmol) and Na₂HPO₄·12H₂O (2.399 g, 6.70 mmol) were dissolved in water (30 mL) under stirring. Solution B: CoCl₂ (1.300 g, 10.00 mmol) and KCl (1.50 g, 0.20 mmol) were added to water (30 mL) under stirring. After 10 min, the resulting mixture of B was added to solution A. After 10 min, the resulting mixture of B was introduced into solution A. Before adjusting the pH to 5.0 with 6 mol·L⁻¹ HCl, the mixture was stirred for 10 min at room temperature (RT). The solution was then deposited in a Teflon reaction kettle at 150 °C for four days. After 5 °C/h of programmed cooling, 28% of K₈{[Co(H₂O)₄][HP₂Mo₅O₂₃]₂}·8H₂O crystals were obtained from Na₂MoO₄·2H₂O.

3.2. X-ray Crystallography

Utilizing graphite monochromatized Mo Kα radiation (λ = 0.71073 Å) at 296 K, a single crystal was put in an Apex-2 diffractometer (Bruker) with a CCD detector. For data integration, Bruker’s SAINT software (Version 6.02A) suite was used. Corrections for polarization and Lorentz were applied according to protocol [55]. The SADABS software (Version 5.624) suite (Bruker) was used to perform adsorption adjustments using the multi-scan method [56]. The full-matrix least-squares approach enhanced F2′s structure after being directly solved. This same refinement was performed successively along with Fourier syntheses for the remaining atoms. SHELXL-97 (Georg-August-Universität Göttingen, 2014, University of Göttingen, Göttingen, Niedersachsen, Germany) was used for calculations [57]. The locations of any hydrogen atoms connected to the water molecule were not shown on the Fourier difference map. Hydrogen atoms coupled to C and N were placed geometrically. All of the hydrogen atoms underwent isotropic refinement using the riding model and SHELXL’s default settings. The crystal information and structural refinement processes for CoPM are listed in

Table 2. Crystal data and structure refinements for CoPM.

Empirical Formula	H24CoK8Mo10O58P4
Formula weight	2407.20
Crystal system	Triclinic
Space group	P-1
a/Å	9.486(3)
b/Å	10.270(3)
c/Å	15.597(4)
α/deg	94.275(4)
β/deg	97.073(4)
γ/deg	114.840(4)
V/Å3	1355.0(6)
Z	1
Dc/g cm−3	2.950
μ/mm−1	3.391
T/K	296(2)
Limiting indices	−11 ≤ h ≤ 11
	−9≤ k ≤ 12
	−18 ≤ l ≤ 16
Measured reflections	6895
Independent reflections	4721
Rint	0.0149
Data/restraints/parameters	4721/0/376
GOF on F2	1.051
Final R indices [I > 2σ(I)]	R1 = 1.051
	wR2 = 0.0686
R indices (all data)	R1 = 0.0264
	wR2 = 0.0698

.

3.3. ThT Fluorescence Assay

In Tris buffer solution (20 mM Tris-HCl/150 mM NaCl, 990 μL), Aβ40 (20 μM) was incubated with Zn(OAc)₂ (4 μL, 10 mM). The samples were subsequently incubated with CoPM (20 μM) or DMSO (1.5 μL) at 37 °C for 24 h. A Corning Costar Corp. flat-bottomed 96-well black plate received each 300 μL sample. Each well received 2 μL of 5 mM ThT solution in the darkness and subjected to incubation for 1 h at 37 °C. The Thermo Scientific Varioskan Flash microplate reader was used to detect fluorescence intensity (λ_ex = 415 nm).

3.4. Morphological Analysis

Identical to the ThT fluorescence test, samples were prepared. On the 300-mesh copper grids with carbon coatings, 10 μL of the solution was seen at RT. The extra solution was drained after 2 min. After 2 min uranyl acetate (10 μL, 1%, w/v) staining, the grids were rinsed with 10 μL ofMilli-Q water. A JEOL JEM-2100 LaB6 (HR) TEM was used to analyze the samples.

3.5. Inhibition of ROS Generation

Following the instructions in the report [58], the HRP stock solution (4 µM) and the DCF stock solution (1 mM) were prepared using a Tris buffer (20 mM Tris-HCl/150 mM NaCl, pH 7.4). CuCl₂ (40 μM) and Aβ40 (20 μM) samples were incubated at 37 °C with or without CoPM (20 μM, with a final DMSO concentration of 1.5 μL). Each sample was then added with 10 μM of ascorbate solution, which was then incubated for 10 min at 37 °C. The 96 wells of a black plate with a flat bottom received the samples (200 µL) via injection. After that, each solution received injections of DCF (100 μM) and HRP (0.04 μM), which were subsequently incubated at 37 °C in the dark. Thermo Scientific’s Varioskan Flash microplate reader was used to detect the fluorescence intensity (λ_ex = 485 nm).

4. Conclusions

Misfolded protein accumulation is a critical factor involved in the onset and progression of AD. The toxic species’ essential structure is an Aβ in a β-sheet form created by the misfolding aggregation process. Furthermore, ROS generated by toxic aggregates is a crucial neurodegenerative factor. This study describes the synthesis and characterization of a novel cobalt complex functionalized phosphomolybdate K₈{[Co(H₂O)₄][HP₂Mo₅O₂₃]₂}·8H₂O (CoPM). CoPM displays a sandwich-type structure, with two Strandberg-type fragments [P₂Mo₅O₂₃]⁶⁻ linked by {Co(H₂O)₄}²⁺ in the solid state. Because POM fragments and Co complexes interact synergistically, CoPM can modulate conformation. As a conformational modulator, CoPM can prevent Aβ from aggregating, thus inhibiting Cu²⁺-Aβ species to produce ROS. Due to its novel structure and advantageous features, CoPM is predicted to have a wide variety of potential applications in coordination chemistry and bio-inorganic chemistry investigations.