Molecular Cobalt(II) Complexes for Tau Polymerization in Alzheimer's Disease.

Tau is an axonal protein known to form abnormal aggregates and is the biomarker of Alzheimer's disease. Metal-based therapeutics for inhibition of Tau aggregation is limited and rarely reported in contemporary science. Here, we report the first example of rationally designed molecular cobalt(II)-complexes for effective inhibition of Tau and disaggregation of preformed Tau fibrils. The mechanistic studies reveal that prevention of Tau aggregation by cobalt-based metal complexes (CBMCs) is concentration-dependent and Tau seldom exhibits conformational changes. Interestingly, CBMCs play dual role in causing disassembly of preformed aggregates as well as inhibition of complete Tau aggregation. Furthermore, CBMCs were nontoxic and maintained the tubulin network intact. CBMCs also prevented okadaic acid-induced toxicity in SH-SY5Y cells thus, preventing hyperphosphorylation of Tau. We believe that this unprecedented finding by the newly developed molecular complexes has a potential toward metal-based therapeutics for Alzheimer's disease.

Introduction

Alzheimer’s disease (AD) is a neurodegenerative disorder characterized by progressive cognitive and behavioral impairment. Worldwide, 44 million people are known to have AD and its related dementia; moreover the available therapeutic aid to overcome AD is limited.[1,2] Abnormal protein deposits in the brain, such as extracellular amyloid plaques and intracellular neurofibrillary tangles (NFTs), characterize AD. The microtubuleassociated protein Tau, plays a key role in several neurodegenerative diseases-like AD, frontotemporal dementia with parkinsonism-17 and Parkinson’s disease and so forth.[3] The axonal Tau is expressed in an adult human brain as six different isoforms. Because of alternative splicing, two N-terminal inserts and the second repeat (R2) in the C-terminal MT-binding domain can be present or absent (Figure A). Upon hyperphosphorylation, Tau disassembles from MTs and self-assembles to form NFTs which consist of paired helical filaments (PHFs).[4] Several factors are responsible that trigger the PHF formation under pathological conditions, such as post-translational modifications, oxidative stress, truncation, and metal ions and so forth.[5,6] Molecules derived from metal complexes, natural products, and peptides were screened for inhibition of Tau aggregation or to disaggregate the preformed fibrils of Tau.[7,8] Recently, methylene blue was identified as an inhibitor of Tau aggregation and subjected to phase III clinical trials.[9] Indeed, there is a strong need for the discovery of new potential therapeutics.[10,11] In this regard, significant research is being devoted in recent times to identify active compounds[12−15] with novel scaffolds that may have potential properties for the treatment of AD by inhibiting Tau aggregation.[16,17]

Figure 1

(A) Domain organization of full-length Tau. The longest isoform of Tau is composed of 441 amino acids, with four repeats toward the C-terminal that is crucial in both physiological processes and in AD pathology. In physiological conditions, it interacts with the tubulin dimer and helps in the assembly of MTs; whereas in pathological conditions they act as main nucleating centers and form the core of aggregates. The four repeats, R1–R4 comprises hexapeptides at the beginning of R2 and R3 which are signature motifs responsible for Tau aggregation. (B) Chemical structure of cobalt-based metal complexes. (C) X-ray crystal-structure analysis of NNNL2CoCl2 with 50% probability of thermal ellipsoids. Selected bond length [Å] and angle [°]: Co(1)–N(1) 2.0265, Co(1)–N(2) 2.3215, Co(1)–N(3) 2.4530, N(1)–Co(1)–N(2) 76.97, N(1)–Co(1)–N(3) 73.51, Cl(1)–Co(1)–Cl(2) 113.56.

The importance of different metals in AD is well studied, which elucidates the critical role of metal ions in pathology.[18] Metal ions are mostly involved in the physiological functions and are also known to interact with proteins, leading to their aggregation.[19,20] Metal ions such as copper(II), zinc(II), iron(III), and aluminium(III) are well studied for their protein aggregation properties. It was reported that copper and zinc interact with amyloid-β and promote their aggregation.[21,22] The mode of binding of these metal complexes with amyloid-β was studied by Raman spectroscopy and the results revealed importance of the three histidines present at the N-terminus. Later, the role of iron(III) in aggregation of amyloid-β was studied by NMR experiments, which demonstrated the importance of first 16 amino acids in the formation of iron coordination.[23] The interaction of copper(II) with Tau was analyzed by NMR. The amino acid sequences adjacent to the hexapeptide motif in the second and third repeat of Tau were involved in interacting with copper. Copper bound to Tau via oxidized cysteine residues caused aggregation of Tau.[24] In an AD brain, Tau hyperphosphorylates and aggregates to form PHFs, aluminium(III) ions specifically interact with these phosphorylated epitopes.[25] These findings suggest that phosphorylated Tau is triggered for aggregation in the presence of aluminium(III). Further, the effect of aluminium maltolate administration was reported for neurodegeneration in rabbits.[26] However, the effect of zinc(II) on Tau fibrillization under physiological conditions was contradicting to that of copper(II) and aluminium(III). The lower levels of zinc(II) would enhance Tau aggregation by decreasing the expression of Tau or by reducing the ability of Tau to interact with MTs. Sequestration of zinc(II) by extracellular senile plaques led to decreased intracellular zinc levels, which ultimately elevated NFT formation. Importantly, the presence of higher intracellular zinc levels led to hyperphosphorylation of Tau at S214, which eventually resulted in aggregates.[27] Furthermore, ferric iron led to Tau aggregation in vitro, and its reduced form caused hyperphosphorylation of Tau.[28,29]

In the present studies, we screened the effect of rationally designed molecular cobalt-based metal complexes (CBMCs; Figure B and Chart ) on Tau aggregation and found that these complexes are effective in preventing the formation of a toxic population of Tau. The CBMCs were effective in inhibiting the polymerization of Tau in a concentration-dependent manner. The efficacy of these metal complexes was analyzed by fluorescence assay and the morphology of aggregates was observed by microscopic analysis. The conformational changes in Tau were analyzed using spectroscopic techniques and the toxicity studies were performed in neuronal cells. The nontoxic nature of CBMCs and their ability to maintain tubulin networks suggests the lead role of these complexes. CBMCs effectively protected SH-SY5Y cells from okadaic acid (OA)-induced toxicity, which supports them as a potent molecule for AD.

Chart 1

NNN-Pincer Type Cobalt Complexes

Results and Discussion

CBMCs, As a Barrier for Tau Aggregation

Tau protein is one of the major MT-associated proteins in neuronal axons that mainly functions to stabilize and assemble MTs.[30] Tau is a soluble protein, and adopts natively unfolded structure in solution.[31] The repeat domains of Tau and the flanking proline-rich regions confer the properties of MT binding and assembly (Figure A). The repeat domains of Tau represents the core of PHFs and the hexapeptide motifs in repeat 2 and 3 form the basic motif for aggregation.[32] Under in vitro conditions, heparin is used as an inducer to enhance Tau aggregation.[32] Heparin binds to the positively charged residues of the flanking repeats 2 and 3, and thus leading to charge neutralization. Further, the binding of heparin also induces a change in the conformation to β-sheet, which serves as a nucleation center for aggregation. In our present study, Tau protein was diluted in assembly buffer, incubated for the formation of aggregates in the presence or absence of CBMCs and the extent of aggregates formation was monitored by ThS fluorophore. Our interest in metal-based therapeutics in AD began with the discovery of novel Co(II) complexes. Cobalt is relatively abundant and biorelevant, which leads us to find its therapeutics in AD.[33] The reaction of the tridentate ligand with CoCl2 in methanol at 65 °C incubated for 4 h resulted in the corresponding cobalt-pincer complexes with an excellent yield (see Figures S1–S7). All the complexes were well characterized and the structure of NNN-L2 CoCl2 was confirmed by a single-crystal X-ray diffraction study (Figure C). The Co(II)-complexes (CBMCs) were incubated with a constant Tau concentration of 0.91 mg mL–1, with increasing concentrations of metal complexes (0.01, 0.025, 0.05, and 0.1 mg mL–1). The aggregation of full-length Tau with CBMCs at higher concentration substantially decreased the ThS fluorescence (Figure A–C), which indicates the prevention of aggregate formation. Further, we have quantified inhibition, which revealed that L2 is more potent with 92.5% inhibition at lowest concentrations of 0.025 mg mL–1, whereas, L3 and L1 showed 85.5 and 73.9% inhibition, respectively. Similar observations were obtained with CoCl2, where Tau aggregation inhibition was observed and the quantification of ThS fluorescence at the end of the assay that is 72 h, indicated about 79 to 82% inhibition (Figure S17A,B). At the highest concentration of 0.1 mg mL–1, they showed up to 93% inhibition (Figure D). The higher order species formed upon Tau aggregation were observed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) analysis (Figure E–G). At 0 h, higher order aggregates are observed by SDS-PAGE, which was not comparable with ThS. These early oligomers induced by these complexes were consistently observed in two sets of experiments and being SDS-stable they are observed by SDS-PAGE but not by ThS fluorescence. In the presence of metal complexes at higher concentrations the higher order aggregates faded away over the time. These data were in agreement with size-exclusion chromatography (SEC) analysis, which was evidenced by the retention volume (Figure S12A–D).

Figure 2

Cobalt-based complexes inhibit Tau aggregation. (A–C) The aggregation inhibition of full-length Tau in the presence of CBMCs monitored by ThS fluorescence. The aggregation was induced from full-length Tau in the presence of heparin as inducer and assembly buffer. The typical full-length Tau reaches its highest propensity of aggregation, but the incorporation of the cobalt complex completely decreases its aggregation with increasing concentration of metal complexes. (D) Although all the three complexes are compelling in the inhibiting aggregation process, the lower concentration of L2 is more effective in comparison with L1 and L3. At higher concentrations of 0.05 and 0.1 mg mL–1 L2 and L3 show maximum inhibition of about 97%. L1 at its highest concentration of 0.1 mg mL–1 has an inhibition of 97%, which clearly suggests the proficiency of these metal complexes in inhibiting Tau aggregation. (E) Tau was analyzed by SDS-PAGE at 0 h, where the higher order species were observed in the highest concentration of CBMCs. (F) The higher order species were reduced at 24 h. (G) When Tau was further analyzed at 72 h these higher order species completely faded away.

Native Conformation of Tau Maintained by CBMCs

The changes in Tau conformation due to metal complexes were studied by circular dichroism (CD) spectroscopy. Tau has no secondary structure and exists as a random coil protein and shows an absorbance at around 190 nm. The aggregated full-length Tau shows few nanometers shift towards the higher wavelength, which signifies the transition of the random coil to β-sheet conformation.[34,35] Similarly, the control Tau (red line) exhibits a shift in the wavelength indicating β-sheet conformation. On the other hand the soluble Tau (black line) wavelength falls below 200 nm, indicating random coil conformation. In the presence of CBMCs, the typical random coil conformation of full-length Tau was not altered (Figure A–C), signifying the effect of these complexes in preventing aggregate formation. Tau protein in in vitro forms fibrillar morphology in the presence of heparin, which was observed as a long filamentous structure. The extent of aggregate formation by Tau in the presence of L1, L2, and L3 were studied by transmission electron microscopy (TEM), which discloses the vigor of CBMCs in completely preventing Tau aggregation (Figure D–G). Further, the energy dispersive X-ray revealed a negligible amount of CBMCs to be associated with Tau, which denotes that Tau was devoid of conjugation with CBMCs (data not shown). X-ray photon electron spectroscopy analysis showed that the oxidation state of CBMCs remains unchanged during Tau aggregation (data not shown). CBMCs formed higher order Tau aggregates at 0 h, which led us to study its role in interaction with soluble Tau. However, surprisingly no higher order aggregates were observed upon incubation at 37 °C (Figure S9A) and this manifested the inability of complexes to form toxic Tau species, when compared to the control (Figure S9B). The initial Tau conformation plays a critical role in order to form the nucleating species, which further accelerates the formation of aggregates. Here, we showed that these molecular Co(II)-complexes did not induce conformational changes in monomer Tau and maintained its native conformation (Figure S10A–C). To support the data, we further studied the effect of CBMCs on monomer full-length Tau (Figure S11A–C) for their aggregation. Tau aggregates were not observed, which suggest that Co(II)-complexes did not lead to the formation of aggregates during the initial stages of Tau assembly.

Figure 3

Conformation of full-length Tau measured by CD spectroscopy. (A–C) The full-length Tau in the native state unveils its random coil conformation, but upon aggregation it attains β-sheet conformation, represented in red. On addition of CBMCs in increasing concentrations of 0.01 and 0.1 mg mL–1 Tau exhibits random coil conformation, indicating the effective role of CBMCs in preventing β-sheet formation by Tau and thus, preventing its aggregation. (D) The morphology of Tau upon incubation with an inducer alone and the typical morphology of Tau fibrils were observed. (E–G) The presence of CBMCs stipulated their ability to prevent aggregate formation. The insets in each micrograph represent the morphology of Tau aggregates at a magnification of 0.5 μm.

Destabilization of Preformed Tau Fibrils

We further investigated the role of CBMCs in the disassembly of Tau PHFs. This would illustrate both the properties of aggregation inhibition and disaggregation of Tau. The preformed Tau aggregates were incubated with various concentrations of L1, L2, and L3; it was observed that L3 was more effective in disaggregating Tau, when compared to L1 and L2 (Figure A–C). At a concentration of 0.5 mg mL–1, L3 showed about 77.4% of inhibition but, L1 and L2 showed 73.5 and 71.9% inhibition, respectively (Figure D). Furthermore, SDS-PAGE analysis showed decrease in aggregates by CBMCs in a time-dependent manner (Figure E–G). Unlike ThS fluorescence, SDS-PAGE did not show a decrease in the intensity of higher order aggregates of Tau. However, at a concentration of 0.5 mg mL–1, L3 exhibited a decrease in Tau aggregates, whereas L3 showed a decrease in aggregates formation at 24 h. At 120 h of incubation, CBMCs effectively disintegrated the Tau aggregates as ascertained by the lower intensity of higher molecular weight (around 250 kDa) Tau (indicated by red arrow). Initially, at 0 h of incubation, no changes were observed in the morphology of aggregates in the presence of CBMCs (Figure S8B–D), when compared to untreated Tau aggregates (Figure S8A). However, at the end of 120 h, there was a definite decrease in the formation of PHFs (Figure A–D). Overall, these results suggest that the efficacy of CBMCs in preventing formation of Tau aggregates that could be toxic and hence, can be a potent therapeutic agent against AD.

Figure 4

CBMCs disintegrate full-length Tau aggregates. (A–C) The effect of disaggregation by L1, L2, and L3 conjugated to CoCl2 was analyzed by ThS fluorescence assay. Tau aggregates were incubated with CBMCs in an increasing concentration of 0.01–0.5 mg mL–1. (D) It was observed that the highest concentration of L1 showed 73.5% disaggregation, whereas L2 and L3 showed 71.9 and 77.4% disaggregation, respectively. (E–G) The presence of higher order aggregates were monitored by SDS-PAGE at different time points of incubation, at 0 (E) 24 (F) and 120 h (G) where L3 was more efficient amongst the three in disintegrating Tau aggregates.

Figure 6

Disaggregation of Tau PHFs by CBMCs. (A) Aggregates of full-length Tau was observed by TEM after 120 h of incubation in the absence of metal complexes. (B–D) Upon prolonged incubation with L1, L2, and L3 the aggregates were destructed into shorter filaments. The insets in each micrograph represent the morphology of disaggregation of Tau at a magnification of 0.5 μm. (E) Cytotoxicity assay. The MTT assay showed the dose-dependent cell toxicity with increasing concentrations of CBMCs. At the higher concentrations of 50 and 100 μg mL–1 of CBMCs cell viability was decreased to 50 and 30%, respectively. (F) Tau aggregates (5 μM) reduced the viability of SH-SY5Y cells to less than 40%, which was found to be reversed when incubated with CBMCs. (G) The phase contrast images of SH-SY5Y cells showed the morphology of healthy cells at 10× magnifications. The cells were treated with vehicle control, that is, 2% DMSO did not show any toxicity. CBMCs did not affect the morphology of the cells at 10 μg/mL, but the morphology of the cells were changed at 100 μg mL–1, suggesting their toxicity. L3 was observed to be the most toxic amongst three as the morphology of cells were completely lost. (H) In comparison with the cell control, Tau aggregate-treated cells showed the change in the morphology of the cells indicating their toxicity. The vehicle control-treated cells were observed to be healthy with neurite extensions suggesting the nontoxic nature of the vehicle used. The cells were treated with 10 μg mL–1 of CBMCs suggested the morphological changes, which could be due to Tau aggregates. This effect is pronounced at 100 μg mL–1 concentration of CBMCs, L3 was found to be more toxic as the majority of the cells lost their morphology; L1 and L2 were moderately toxic. The significance was calculated using Tukey’s method and ***, **, and * indicated p value <0.001, <0.01, and <0.05, respectively. ns is not significant where, p = >0.05.

In recent years, several drugs failed in the clinical trials, which emphasize the need to develop and screen natural as well as synthetic molecules to target AD pathology. Although AD has multifaceted effectors like kinases, proteases, oxidative stress, and so forth, the structure of protein plays an important role in the interaction, and further change in its conformation can lead to either oligomers or aggregates formation. The oligomeric precursors are substantially more toxic than Tau fibers. The amyloid fibrils are likely to play an important role, either as reservoirs or sinks to toxic oligomers. Hence, a detailed screening of the compounds could help in discovering potential drugs that prevent protein aggregation in AD. The role of metal ions in interacting with proteins and leading to their aggregation is a well-known phenomenon.[36] Metal ions such as copper(II), zinc(II), aluminium(III), and iron(III) cause protein aggregation, but our results suggested that molecular cobalt(II) complexes have a paradoxical effect. Previously, Rajendran et al., have reported the deposition of metals such as iron, copper, and zinc in the AD brain, which revealed increased metal accumulation in comparison with healthy brain tissue.[37] Aluminium is known to play a key role in the accumulation of Tau in cultured neurons and AD pathology.[38] Animal studies revealed that injecting PHFs with and without aluminium salts led to distinct effects. Aluminium salts caused the deposition of amyloid-β and several other proteins along with Tau.[39] Moreover, the in vitro studies showed that aluminium formed higher order aggregates, as examined by SDS-PAGE. Aluminium caused aggregation in phosphorylated Tau. Aluminium affects Tau aggregation by reducing the activity of protein phosphatase 2 (PP2A), an enzyme essential for dephosphorylation of Tau; and activating kinases such as cyclin-dependent kinase 5 (cdk5) and glycogen synthase kinase 3 beta (GSK3β).[40,41] Kawahara et al., showed that prolonged exposure to aluminium led to conformational changes in amyloid-β and its aggregation.[42]

In our present observations, the biochemical studies revealed that potency of CBMCs in preventing Tau polymerization. Copper interacts with both Tau and amyloid-β, driving AD pathology. The oxidative properties of copper leading to Tau aggregation via cysteine residues is well understood.[24] Studies in 3XTg-AD mice showed that treatment with copper led to the activation of cdk5/p25 which caused hyperphosphorylation of Tau.[43] Voss et al., elucidated that copper led to hyperphosphorylation of Tau in SH-SY5Y cells, in an amyloid-β-independent manner.[44] Copper was also known for reactive oxygen species production in the presence of amyloid-β, leading to cytotoxicity.[45,46] Copper interacted with amyloid-β and reduced the content of β-sheet conformation that further formed amorphous aggregates.[47] In our current studies, CBMCs did not drive conformational changes in full-length Tau. The monomer Tau manifested the signature of typical random-coil conformation in the presence of metal complexes, which indicated that CBMCs did not drive conformational changes in Tau. Tau usually accumulates to form filamentous aggregates, but aluminium led to the formation of amorphous aggregates.[48] Complementary to CD spectroscopy of soluble Tau, TEM analysis also suggested the absence of toxic Tau species. The SEC analysis clearly indicated that the higher order species of Tau aggregates were prevented by CBMCs (Figure S12).

Direct Interaction of CBMCs with Tau

Isothermal titration calorimetry (ITC) is the direct method to obtain free energy (ΔG), enthalpy (ΔH), and entropy (ΔS) changes along with dissociation constant (KD) and number of binding sites (N) for ligands on the protein. Soragni et al., and Zhu et al., in individual studies, employed ITC to analyze the binding of copper and lead, respectively.[24,34] The ITC titration yielded differential power values during interaction of Tau with L2 (Figure A). These values were integrated to obtain exothermic peaks (Figure B). The negative ΔG value, −7.41 revealed the spontaneity of binding. The titration suggested binding of L2 to Tau with a dissociation constant (KD) of 6.03 ± 2.35 μM, a weak interaction. The fitting by one set of sites resulted in the N value of 0.403 ± 0.057. The binding indicated small enthalpy changes (ΔH) with a negative entropy of −0.654 ± 0.226 and −4.47 kcal/mol, respectively. L2 was also titrated with N,N-bis(2-hydroxyethyl)-2-aminoethanesulfonic acid (BES) buffer alone to obtain the isotherm, which suggested that L2 exhibited less heat changes when compared to the heat changes measured as shown in Figure (Figure S15).

Figure 5

Isothermal titration microcalorimetric analysis for CBMCs binding to Tau. (A) The differential power in terms of μcal/s was plotted for each injections of L2. (B) The raw data was integrated as kcal/mol of injection vs the molar ratio of ligand, L2. The fitting of integrated points using one set of sites revealed the interaction of L2 with Tau, with a KD of 6.03 ± 2.35 μM. The N value of 0.403 ± 0.057 suggests binding of one L2 to two Tau molecules.

Tau-Mediated Cytotoxic Studies of CBMCs in SH-SY5Y

Metals are essential components of the brain at physiological levels but their accumulation at higher concentration would be detrimental. Metals are known to exert toxicity in either of the following ways, by inhibiting PP2A or activating kinases such as cdk5/p25, GSK3β, and so forth. The neurons isolated from the hippocampus were treated with iron, this led to oxidative stress, which ultimately caused Tau hyperphosphorylation via the cdk5/p25 complex.[49] Similarly, studies in rat primary cortical neurons showed that zinc caused phosphorylation of Tau at Ser262, which was marked as an early pathological event in AD.[50] Hence, the toxicity of these compounds was analyzed by treating SH-SY5Y cells with CBMCs. As shown in Figure E, maximum toxicity was observed with 100 μg mL–1 of the compounds used. SH-SY5Y cells treated with 5 μM Tau aggregates alone showed 59% cell toxicity. However, this toxicity was arrested in the presence of CBMCs up to 50% (Figure F) at lower doses suggesting the potential role of CBMCs to rescue the cells from Tau-induced cytotoxicity even at much lower concentrations. The cell morphology of SH-SY5Y in the absence of CBMCs or Tau aggregates, that is, cell control showed extended neurite outgrowths (Figure G,H). The cells treated with dimethyl sulfoxide (DMSO) vehicle exhibited no toxicity, indicating that the vehicle had no role in inducing toxicity to the cells. Further the cell morphology was lost at 100 μg mL–1 of L3. L1 and L2 showed moderate toxicity at this concentration.

CBMCs Maintained Tubulin Network in Neuronal Cells

MTs are one of the cytoskeletal elements essential for the maintenance of cell shape, integrity, and cellular tracks for the transport of cargos. Axonal Tau is involved in polymerizing tubulin to MTs. The loss of Tau would lead to MT disassembly and thus neuronal degeneration. CBMCs did not alter the levels of tubulin and Tau in SH-SY5Y cells (Figures A,C and S13).[49,51] The tubulin distribution in SH-SY5Y suggested the maintenance of the tubulin network by L1 and L2. The quantification suggested no change in the tubulin levels after CBMCs treatment; however, L3 led to a change in the cell morphology as observed by the retraction of cellular extensions (Figure B). Further, the levels of Tau phosphorylation at Ser212 and Thr214 was probed by AT100 antibody. AT100 is a conformational antibody that interacts with the pTau present as aggregates. In our present study the AT100 epitope in Tau has been induced in SH-SY5Y cells by using OA. OA is known to induce Tau hyperphosphorylation by inhibiting protein phosphatase 1 and PP2A. OA induces Tau phosphorylation at various sites such as Ser202, Ser205, Ser212, Thr214, Ser262, Ser356, and Ser404.[52,53] The SH-SY5Y treated with OA showed a change in the morphology where cellular extensions were retracted and exhibited round-shaped cells, indicating the toxicity of OA. The treatment with L1 rescued the cells from OA-induced cytotoxicity and showed decreased levels of AT100 epitopes (Figures and S14). L2 and L3 had no effect in reducing the AT100 levels. L2 however, prevented the change in the morphology of SH-SY5Y cells in the presence of OA (Figure S14). On the other hand, OA-induced toxicity was observed in the presence of L3. Quantifying the levels of pTau suggested the efficacy of L1 in reducing AT100 levels; L2 showed no change, whereas L3 increased the levels of pTau and total Tau (Figure S16). These results indicate that L1 and L2 prevented the cytotoxic effects of OA in contrary to L3.

Figure 7

Tau levels are unaltered by CBMCs. (A) SH-SY5Y cells were treated with CBMCs at a concentration of 25 μg/mL, and were probed for the cytoskeletal network and levels of Tau by using pan Tau antibody (K9JA). The immunofluorescence studies revealed that L1 and L2 maintained the cytoskeletal network as probed by the β-tubulin antibody. L3-treated cells showed a rounded morphology of the cells, indicating the loss of the tubulin structure. (B) Quantification of tubulin levels in SH-SY5Y as in (A) suggested that CBMCs do not change tubulin levels. (C,D) Tau levels remain unaffected in SH-SY5Y cells on treatment with L1 and L2. In comparison with the cell control the Tau levels were observed to be increased in L3 treatment groups. Quantification of Tau levels in SH-SY5Y as in (C), in terms of mean intensity per square micrometer. The significance was calculated by SigmaPlot and ***, **, and * indicated p value <0.001, <0.01, and <0.05, respectively. ns is not significant where, p = >0.05.

Figure 8

pTau levels mapped by AT100 in an OA-induced manner. In SH-SY5Y cells Tau hyperphosphorylation was induced by OA. The AT100 epitope that corresponds to phosphorylation at Ser212 and Thr214 was studied by the immunofluorescence. L1 effectively reduced the levels of pTau when compared to L2 and L3. L1 and L2 protected SH-SY5Y cells from OA-induced toxicity, as the cell morphology was not altered. L3 failed to prevent OA-induced toxicity and the cells exhibited rounded morphology due to the loss of neurite extensions.

In summary, the newly developed molecular Co(II)-complexes (CBMCs) showed a significant role in the inhibition and disaggregation of Tau. The cytotoxicity studies on human SH-SY5Y cells revealed that CBMCs are nontoxic in nature and helped in maintaining the tubulin network at optimal concentrations under in vitro conditions. Furthermore, CBMCs also prevented OA-induced toxicity in SH-SY5Y thus preventing cytotoxicity due to hyperphosphorylation of Tau. We firmly believe that the present molecular Co(II)-complexes can be a potential therapeutic agent for Alzheimer’s disease and related neurodegenerative diseases.

Methods

Materials or Chemicals

MES, heparin, BES, bovine serum albumin, bicinchoninic acid (BCA), and ThS were purchased from Sigma. Isopropyl β-d-1-thiogalactopyranoside (IPTG) and dithiothreitol (DTT) were purchased from Calbiochem. Other chemicals such as ampicillin, NaCl, ethylene glycol-bis(β-aminoethyl ether)-N,N,N′,N′-tetraacetic acid (EGTA), and so forth were purchased from MP. Cell culture related chemicals and plastics were purchased from Sigma and Thermo Scientific Pvt. Ltd.

Experiments were carried out using standard Schlenk techniques. All solvents were of reagent grade or better. Deuterated solvents were used as received. Acetonitrile was refluxed over P2O5 and freshly distilled under an argon atmosphere. Metal complexes (CoCl2) and other chemicals used in the reactions were used without additional purification. Thin layer chromatography was performed using silica gel precoated glass plates, which were visualized under UV light at 254 nm or under iodine. Column chromatography was performed with SiO2 [SiliCycle SiliaFlash F60 (230–400 mesh)]. 1H NMR (400 or 500 MHz), 13C{1H} NMR (100 MHz) spectra were recorded on an NMR spectrometer. Deuterated chloroform was used as the solvent and chemical shift values (δ) were reported in parts per million relative to the residual signals of this solvent [δ 7.26 for 1H (chloroform-d), δ 77.2 for 13C{1H} (chloroform-d)]. Abbreviations used in the NMR follow-up experiments: br, broad; s, singlet; d, doublet; t, triplet; q, quartet; and m, multiplet. Mass spectra were obtained on GCMS-QP 5000 instruments with an ionization voltage of 70 eV (Scheme ). High-resolution mass spectra (HRMS) were obtained by fast atom bombardment using a double focusing magnetic sector mass spectrometer and electron impact (EI) ionization technique (magnetic sector–electric sector double focusing mass analyzer).

Scheme 1

Synthesis of Cobalt Complexes (L1–L3)

Synthesis of Ligands

2,6-Bis(4-methylpiperazine-1-yl-methyl) Pyridine (NNN-L1)

A solution of 2,6-bis(bromomethyl) pyridine (0.8 g, 3.0 mmol) in acetonitrile (45 mL) was added dropwise to a solution of 1-methylpiperazine (0.669 g, 6.0 mmol) and K2CO3 (1.249 g, 9.0 mmol) in CH3CN (20 mL), the resulting reaction mixture was allowed to stir for 14 h at 80–85 °C, then cooled to room temperature, and the solvent was removed under reduced pressure. Finally, to the reaction mixture, water was added and the product was extracted with chloroform. The organic layer was collected and evaporated in vacuum under reduced pressure, afforded yellow oil. Yield (0.82 g; 89%). IR (KBr) ν: 2945 (s), 2520 (m), 2042 (m), 1452 (s), 1029 (s), 651 (m). 1H NMR (500 MHz, chloroform-d): δ 7.59 (s, 1H), 7.28 (s, 2H), 3.66 (s, 4H), 2.54 (s, 8H), 2.46 (s, 8H), 2.28 (s, 6 H). 13C NMR (126 MHz, chloroform-d): δ 158.0, 136.6, 121.3, 77.3, 76.7, 64.4, 55.1, 53.2, 46.1. HRMS (EI) m/z: calcd for C17H30N5, 304.2501; found, 304.2496.

2,6-Bis(piperazin-1-yl-methyl) Pyridine (NNN-L2)

Step-1: Synthesis of bocNNN

Solution of 2,6-bis(bromomethyl) pyridine (1 g, 3.77 mmol) in acetonitrile (40 mL) was added dropwise to a solution of 1-boc-peprazine (1.4037 g, 7.54 mmol) and K2CO3 (1.56 g, 1.13 mmol) in CH3CN (20 mL), the resulting reaction mixture was allowed to stir for 14 h at 85 °C; after being cooled to room temperature, the reaction mixture was extracted in chloroform. The organic fractions were combined and dried over anhydrous and evaporated in vacuum, which afforded yellow oil. Yield (0.8 g; 80%). IR (KBr) ν: 3383 (w), 2843 (m), 2077 (m), 1639 (s), 1431 (m), 1273 (m), 1014 (m), 559 (m) cm–1. HRMS (ESI): calcd for C25H41N5O4 [M + H]+, 475.62; found, 476.3244.

Step-2: Synthesis of NNN-L2

bocNNN (0.8 g) was dissolved in MeOH (15 mL) and followed by addition of 1 N HCl (8 mL), then the mixture was allowed to stir for 6 h at 60 °C. After cooling the reaction mixture to room temperature, it was neutralized with an aqueous solution of 5% NaHCO3, then the solvent was evaporated in vacuum. The resulting sticky product was further dissolved in ethanol and filtered. The filtrate was concentrated in vacuum, which afforded a yellow oil product. Yield (0.75 g, 75%). 1H NMR (300 MHz, CDCl3): δ 7.68–759 (t, 1H), 7.45–7.29 (d, 2H), 3.92–3.64 (s, 4H), 3.15–2.72 (m, 8H), 2.71–2.42 (s, 8H), 2.43–2.21 (s, 2H). HRMS (ESI): calcd for C15H25N5 [M + H]+, 275.39; found, 276.21.

Synthesis of 2,6-Bis(morpholinomethyl) Pyridine (NNN-L3)

A solution of 2,6-bis(bromomethyl) pyridine (0.3 g, 1.13 mmol) in acetonitrile (30 mL) was added dropwise to a solution of morpholine (0.197 g, 2.26 mmol) and K2CO3 (0.468 g, 3.39 mmol) in CH3CN (15 mL), the resulting reaction mixture was allowed to stir for 14 h at 80–85 °C, then cooled to room temperature, and subsequently the reaction mixture was extracted with chloroform and water. The organic layer was collected, and evaporated in vacuum under reduced pressure to afford a colorless solid. Yield (0.282 g, 90%). IR (KBr) ν: 2800 (m), 1575 (m), 1454 (m), 1298 (m), 1111 (s), 906 (m). 1H NMR (500 MHz, chloroform-d): δ 7.65–7.52 (m, 1H), 7.31 (d, J = 7.6 Hz, 2H), 3.84–3.69 (m, 8H), 3.66 (s, 4H), 2.51 (s, 8H). 13C NMR (126 MHz, chloroform-d): δ 157.7, 136.7, 121.4, 77.3, 76.7, 66.9, 64.8, 53.7.

Synthesis of (NNN-L1)CoCl2

Cobalt chloride hexahydrate (0.312 g, 1.34 mmol) in methanol (15 mL) was added dropwise to a solution of NNN-L1 (0.408 g, 1.34 mmol) in MeOH (15 mL) with stirring. The resulting reaction mixture was allowed to stir for 3 h at room temperature. The resulting solution was evaporated under vacuum which afforded a blue color solid; the solid was washed with diethyl ether and dried in air. Yield (0.54 g; 93%). IR (KBr) ν: 2924 (s), 2314 (m), 1612 (m), 1462 (s), 1207 (m), 972 (m). HRMS (EI) m/z: calcd for C17H30N5Cl2Co, 433.1210; found, 433.1205.

Synthesis of (NNN-L2)CoCl2

Cobalt chloride hexahydrate (0.086 g, 0.36 mmol) was added to a solution of NNN-L1 (0.1 g, 0.36 mmol) in MeOH (10 mL) with stirring. The resulting reaction mixture was allowed to stir for 3 h at room temperature. The resulting solution was concentrated in vacuum which afforded a blue color solid; the solid was washed with diethyl ether and dried in air. Yield (0.132 g; 90%). IR (KBr) ν: 3446 (w), 1633 (s), 1460 (m), 1165 (m), 989 (m), 613 (m). HRMS (ESI): calcd for C16H28Cl2CoN5 [M + Na]+, 405.23; found, 429.24.

Synthesis of (NNN-L3)CoCl2

Cobalt chloride hexahydrate (0.129 g, 0.54 mmol) in methanol (8 mL) was added dropwise to a solution of NNN-L3 (0.151 g, 0.54 mmol) in MeOH (10 mL) with stirring. The resulting reaction mixture was allowed to stir for 3 h at room temperature. The resulting solution was evaporated under vacuum which afforded a blue colored solid, and the solid was washed with diethyl ether and dried in air. Yield (0.21 g, 95%). IR (KBr) ν: 2958 (s), 2841 (m), 1610 (s), 1450 (m), 1290 (m), 1111 (s), 999 (m), 869 (s), 815 (m). HRMS (EI) m/z: calcd for C17H28N3Cl2Co, 403.0992; found, 403.0987.

Expression and Purification of Tau

The recombinant full-length Tau was expressed in the BL21* strain of Escherichia coli. Cells were induced with 0.5 mM IPTG after the optical density (OD) at A600 reached 0.5–0.6. The cells were allowed to grow at 37 °C postinduction and were harvested by pelleting at 4000 rpm, at 4 °C for 10 min. The pellet was resuspended in 50 mM MES buffer pH 6.8 containing 1 mM EGTA, 2 mM MgCl2, 5 mM DTT, 1 mM phenylmethylsulfonyl fluoride, and a protease inhibitor cocktail and was lysed by using a constant cell disruption system. The purification was done as described previously with minor changes.[54] The concentration was estimated using the BCA method.

Preparation of Tau Aggregates

Tau was induced to aggregate as described previously with minor modifications.[54] In the presence of heparin (17 500 Da) at a ratio of 4:1, Tau was polymerized in an assembly buffer containing 20 mM BES, pH 7.4, 25 mM NaCl, 1 mM DTT, 0.01% NaN3, and a protease inhibitor cocktail. The reaction mixture was incubated at 37 °C and the aggregates formation was monitored by ThS fluorescence, SDS-PAGE, and TEM at 0 and 120 h.[55,56] Tau protein was allowed to assemble in the absence and presence of compounds in increasing concentration with constant Tau concentrations of 0.91 mg mL–1.

Disaggregation Assay

The potency of the metal complexes in disaggregating the preformed Tau aggregates was analyzed. Soluble Tau was incubated at 37 °C, at a concentration of 4.58 mg mL–1 for the PHF assembly. The formation of aggregates was analyzed by ThS fluorescence assay and SDS-PAGE. Thus, the formed aggregates were diluted to 0.91 mg mL–1 final concentration of 20 mM BES buffer, pH 7.4, and further, the mixture was incubated with increasing concentration of the metal complex as discussed earlier.

Thioflavin S Fluorescence Assay

5 μL of reaction mixture was diluted with 45 μL of 8 μM ThS in 50 mM ammonium acetate, pH 7.0 and added to 384 well plates in triplicates. Subsequently a blank was also prepared for subtracting the background fluorescence. The plate was incubated for 20 min in the dark before measuring ThS fluorescence, at an emission wavelength of 521 nm by exciting it at 440 nm in a Tecan Infinite 200 PRO multimode microplate reader.

SDS-PAGE Analysis for Tau Aggregates

The effect of the compounds on inhibiting the aggregates formation by Tau was observed by SDS-PAGE.[57,58] The reaction mixture incubated with and without the compound were collected at different time intervals of 0, 24, and 72 h (end point) and resolved in 10% SDS-PAGE using a miniVE vertical electrophoresis system from GE Healthcare.

Soluble Tau Assay

The soluble Tau was studied in the presence of metal complexes alone to analyze the conformational change occurring due to the compound. 0.91 mg mL–1 of Tau was incubated for 1 h at 37 °C with and without different concentrations of 0.01, 0.025, 0.05, and 0.1 mg mL–1 of metal complexes. At the end of 1 h the samples were analyzed by SDS-PAGE, TEM, and CD spectroscopy to monitor the formation of aggregates and the change in Tau conformation, respectively.

CD Spectroscopy

The conformational change in Tau was analyzed by CD spectroscopy in the far UV region. Tau is a random coiled protein and upon aggregation it acquires β-sheet conformation. The impact of the compounds on preventing the formation of the β-sheet structure was studied by CD spectroscopy. The spectra was collected as described previously, in a Jasco J-815 spectrometer, by using a cuvette with 1 mm path length.[59] The measurements were performed in the range of 250–190 nm, with a data pitch of 1.0 nm, and scanning speed of 100 nm min–1. All the spectra were obtained at 25 °C. The reaction mixture was diluted to 0.13 mg mL–1 in 50 mM phosphate buffer, pH 6.8. The effect of the compound on soluble Tau was studied by incubating Tau along with compounds alone at 37 °C and the spectra was read at 25 °C.

Transmission Electron Microscopy

The extent of aggregates formed in the presence of the metal complexes was analyzed by TEM (Tecnai T-20). The assay mixture was diluted to 0.04 mg mL–1 final concentration, spotted on a carbon coated copper grid of 400 mesh and incubated for 45 s. The excess Tau aggregates were removed by incubating the grid in water for 30 s and this was repeated twice. The grid was further stained by 2% uranyl acetate for 1 min to observe the morphology of aggregates under TEM.

Size-Exclusion Chromatography

The high-molecular weight species formed by Tau polymerization was analyzed by SEC.[60−62] Tau protein was diluted to a concentration of 4.58 mg mL–1 in assembly buffer along with heparin in a ratio of 4:1 and incubated at 37 °C in the presence and absence of 0.1 mg mL–1 of NNN-L2CoCl2. Tau was subjected to SEC using Superdex 75 PG in order to resolve aggregated Tau from the soluble, which was accessed as a decrease in the retention volume at different time points of 0, 2, and 24 h in the presence and absence of NNN-L2CoCl2.

Isothermal Titration Calorimetry

ITC was carried out to understand the thermodynamics behind Tau interaction with CBMCs. Here, the titration was done using 2.3 mg mL–1 of full-length Tau and 0.407 mg mL–1 of L2. The titrations were recorded in MicroCal PEAQ-ITC at 25 °C. The titration was conducted by giving 19 injections, first injection of 0.4 μL was followed by injections of 2 μL each with 240 s interval at a stirring speed of 650 rpm. Tau and L2 were prepared in 20 mM BES containing 50 mM NaCl at pH 7.4. The samples were re-buffered, filtered, and loaded. The sample cell was loaded with 200 μL of full-length Tau and syringe with 40 μL of L2. Similarly, L2 was titrated into BES buffer as a compound control to measure the heat changes caused by the compound alone. The data was analyzed in MicroCal PEAQ-ITC analysis software and fitted to one set of site model. The heat change from buffer was assigned as control and fitting was done using a line mode in analysis software.

Cytotoxicity Assay

SH-SY5Y cells were cultured in Dulbecco’s Modified Eagle Medium (DMEM)-F12 media (Gibco) supplemented with 20% FBS, 100 U/mL penicillin, and 100 U/mL streptomycin. Subconfluent cells were harvested by trypsinization and 25 000 cells/well were seeded in a 96 well plate (100 μL/well). The cells were then incubated overnight at 37 °C. Postincubation, the cells were treated with different concentrations of (10–100 mg mL–1) the compounds to study the toxic effect of CBMCs. In order to study the effect of CBMCs on Tau-induced toxicity, cells were first treated with 100 μL of DMEM containing 5 μM full-length Tau aggregates, followed by the indicated amounts of CBMCs. The full-length Tau aggregates alone were used as the control. After 12 h of incubation at 37 °C, cell viability was evaluated using thiazolyl-blue-tetrazolium-bromide (MTT) assay. Each treatment was performed in triplicates. Briefly, 10 μL of 5 mg mL–1 MTT was added into each well and further incubated for 4 h at 37 °C. Later, 100 μL of DMSO was added into each well and the color intensity was measured using an ELISA reader at 570 nm. The percentage of cell viability was calculated as following. Cells viability (%) = OD (570 nm) in the presence of full-length Tau with or without an inhibitor × 100.

Immunofluorescence Assay

SH-SY5Y cells were grown on coverslips and kept for 24 h after seeding. The cells were treated with CBMCs and OA at a concentration of 25 μg/mL and 25 nM for 24 h. Postincubation the cells were washed with ice cold phosphate buffered saline (PBS) and fixed using chilled methanol for 20 min at −20 °C. The cells were further washed with PBS and permeabilized by 0.2% Triton X-100 for 15 min at room temperature. Triton X-100 was removed by giving three PBS washes, followed by incubation with 2% horse serum as a blocking agent at 37 °C for 1 h. The cells were treated with primary antibody at room temperature for 3 h. pan Tau antibody (K9JA) against Tau and β-tubulin monoclonal antibody (BT7R) were used in dilutions of 1:1000 and 1:500, respectively. Phospho-Tau (Thr212, Ser214) monoclonal antibody (AT100) was used at a dilution of 1:100. After 3 h of incubation, the cells were washed thrice with PBS and incubated with goat anti-rabbit Alexa 488 conjugate and goat anti-mouse Alexa 555 conjugate at the dilutions of 1:1000 and 1:500, respectively. Further cells were washed twice with PBS and counterstained with 4′,6-diamidino-2-phenylindole for 5 min. The coverslips were mounted and allowed to dry. The SH-SY5Y cells were imaged under Zeiss Axio Observer 7.

Quantitative Analysis

Quantification of immunofluorescence studies was performed using image analysis software in Zeiss Axio Observer 7. The cell area was marked to measure the intensities for tubulin, Tau, and pTau levels. The background intensity adjacent to the cells was similarly calculated to obtain the final value. Thus the obtained intensities were normalized with respect to the area of the cell. Using SigmaPlot, plotted the mean intensity per square micrometer of cell.

Statistical Analysis

Data are represented in terms of mean ± s.e.m. The cytotoxicity data was analyzed by Student’s t-test, two-tailed and unpaired at 95% confidence interval. Tukey’s method was used for statistical analysis. The statistical significance was determined as ***, **, and * which indicated p value <0.001, <0.01, and <0.05, respectively. ns indicates not significant data where p value was >0.05.