Modification of amyloid-beta peptide aggregation via photoactivation of strained Ru(ii) polypyridyl complexes†

Alzheimer's disease (AD) is a chronic neurodegenerative disorder characterized by progressive and irreversible damage to the brain. One of the hallmarks of the disease is the presence of both soluble and insoluble aggregates of the amyloid beta (Aβ) peptide in the brain, and these aggregates are considered central to disease progression. Thus, the development of small molecules capable of modulating Aβ peptide aggregation may provide critical insight into the pathophysiology of AD. In this work we investigate how photoactivation of three distorted Ru(ii) polypyridyl complexes (Ru1–3) alters the aggregation profile of the Aβ peptide. Photoactivation of Ru1–3 results in the loss of a 6,6′-dimethyl-2,2′-bipyridyl (6,6′-dmb) ligand, affording cis-exchangeable coordination sites for binding to the Aβ peptide. Both Ru1 and Ru2 contain an extended planar imidazo[4,5-f][1,10]phenanthroline ligand, as compared to a 2,2′-bipyridine ligand for Ru3, and we show that the presence of the phenanthroline ligand promotes covalent binding to Aβ peptide His residues, and in addition, leads to a pronounced effect on peptide aggregation immediately after photoactivation. Interestingly, all three complexes resulted in a similar aggregate size distribution at 24 h, forming insoluble amorphous aggregates as compared to significant fibril formation for peptide alone. Photoactivation of Ru1–3 in the presence of pre-formed Aβ1–42 fibrils results in a change to amorphous aggregate morphology, with Ru1 and Ru2 forming large amorphous aggregates immediately after activation. Our results show that photoactivation of Ru1–3 in the presence of either monomeric or fibrillar Aβ1–42 results in the formation of large amorphous aggregates as a common endpoint, with Ru complexes incorporating the extended phenanthroline ligand accelerating this process and thereby limiting the formation of oligomeric species in the initial stages of the aggregation process that are reported to show considerable toxicity.

a Fluorolog-3 fluorimeter.Default parameters were used for all computational procedures unless stated otherwise.
Photoejection experiments were carried out via UV-Vis using a visible light source with cool white colour (5500 -6000 K) (SOLLA 30W LED).Data were collected from 200-900 nm, and irradiation intervals were as short as 1 min.at early times and after 15 min., data were collected every 5 min.until 60 min. of experiment.The photoejection time was determined when no further spectral changes were observed.Photoejection kinetics were analyzed by plotting the normalized change in absorption at two wavelengths against irradiation time using a published method. 7,8,9,10 Thavelength selected were those within 50 nm of the longest wavelength isosbestic point and exhibited the greatest change in the course of the experiment.

1 H NMR Binding Assay of A1-16 Peptide to Ru(II) Complexes
Deuterated PBS (0.01 M, pH 7.4) buffer was prepared by removal of water by vacuum drying of PBS buffer and dissolving the powder in D2O.A1-16 and Ru(II) stock solutions in DMSO (1 mM) were dissolved in deuterated PBS (0.01 M, pH 7.4) buffer, and the 1 H NMR spectra of A1-16 alone, Ru(II) complexes (200 M -kept in the dark or exposed to light for their respective activation time), and A1-16 plus Ru(II) complexes (1:1 eq.dark or exposed to light for their respective activation time) were collected after solubilization at 0 h and 24 h.

Mass Spectrometry of Binding of A Peptide to Ru(II) Complexes
Samples were analyzed by direct infusion (1 -4 L) of analyte into a mobile phase of 1:1 water:acetonitrile containing 5 mM ammonium acetate (pH unmodified), flowing at 0.3 mL/min and maintained at 30 °C.All components of the mobile phase were MS grade and water was ultrapure grade from MilliQ A-10 system.Nitrogen drying gas was heated to 250 °C and run at 5 L/min with a nebulizing pressure of 15 psig.Voltages were: capillary 3 kV, fragmentor 175 V, skimmer 30 V, octupole 250 V. Samples were prepared as ~1 mg/mL of total protein (A1-16/1-40) in ammonium carbonate (0.02 M, pH 9) buffer with 0 or 1 eq. of Ru(II) complexes (dark and activated).

Gel Electrophoresis and Western Blotting
Lyophilized A1-42 was dissolved in 1:1 DMSO/ddH2O to obtain a stock solution with a concentration of approximately 250 M.The A1-42 stock solution was diluted to 25 M in PBS (0.01 M, pH 7.4) then incubated at 37 °C with continuous agitation at 200 rpm to form aggregates in the presence of activated and non-activated Ru(II) complexes.For the first set of experiments the peptide was incubated for a total of 24 h, in the presence of different concentrations of Ru(II) complexes (0.10, 0.25, 0.50, 1.0, 1.5, and 2.0 eq.).For the second set of experiments, 1.0 eq. of Ru(II) complexes was also incubated for a total of 24 h, but aliquots were collected at different time points (0 h and 24 h).Electrophoresis separation of peptide aggregates was completed using 8-16% Mini-PROTEAN TGX Precast Gels from Bio-Rad, at 100 V for 100 min in running buffer (25 mM Tris, 192 mM glycine, 0.1% SDS).The gels were then transferred to a nitrocellulose membrane for 1 h at 100 V at 4 °C, followed by blocking of the membrane in a 3% BSA solution in Tris-buffered saline (TBS) (0.02 M Tris, 0.15 M NaCl, 0.003 M KCl) for 1 h.The membrane was incubated in a solution (1:2000 dilution) with a primary antibody that recognizes A, 6E10, (Biolegends) overnight.After washing 5 x 5 min with TBS, the membrane was incubated in a solution containing the secondary antibody (Horseradish peroxidase, Caymen Chemicals) for 3 h.A Thermo Scientific SuperSignal West Pico Chemiluminescent Substrate kit was used to visualize the A species using a Bio-Rad ChemiDoc MP imaging system.

Transmission Electron Microscopy (TEM)
TEM grids were prepared from the 1:1 Ru(II)/A samples from the Western blot assay after 0 h and 24 h incubation for the monomeric form of A1-42 and 0 h and 24 h incubation at 37 °C after 96 hours of peptide incubation for the fibrillar form of A1-42.TEM grids were prepared following previously reported methods. 11,12 n order to increase hydrophilicity of the Ultrathin Carbon Film 400-mesh grids (Ted Pella), the grids were glow discharged in a vacuum for 10 seconds.Drops of samples (10 L) were placed onto a sheet of parafilm and the TEM grid was placed on the drop for 5 min.The grid was then placed on top of syringe-filtered 5% uranyl acetate for 1 min.Excess uranyl acetate was removed using a tissue between drops.The grid was allowed to air-dry for at least 15 min.Bright field images were obtained on a FEI Tecnai Osiris STEM at 200 kV.

Binding Constant (Kd)
A1-42 film was dissolved in 1:1 DMSO/ddH2O, and the stock solution was diluted to a final concentration of 10 M in PBS (0.01 M, pH 7.4) buffered solution and incubated for a period of 96 h at 200 rpm at 37 °C.Separate A solutions were prepared with the Ru(II) complexes (0.10, 0.25, 0.50, 0.75, 1.0, 1.25, 1.5, and 2.0 eq., and up to 8 eq. for Ru3) and the fluorescence intensity was measured immediately (ex/em = 275/310 nm) to minimize the effects of ligand dissociation / covalent binding.Dissociation constants were determined using a single-site binding model as reported. 13Analysis of the ThT fluorescence response (Figure S15D) by the same single-site binding model (ex/em = 275/480 nm) affords Kd = 10.6  1.0 M which is in agreement with the same analysis of the Tyrosine response Kd = 9.8  1.4 M.

BCA Assay
A1-42 (60 M) was incubated in PBS buffer (0.01 M, pH 7.4) with and without the Ru(II) complexes (1 eq.) for a period of 24 h for the BCA assay.The samples were centrifuged at 14,000 g for 5 min., and aliquots were taken at 0 h and kept in the freezer.In a 96 well plate, 20 L of solution were added in triplicate for each time point, and 200 L of working reagent from the Thermo Fisher BCA Protein Assay® kit was added to each well.The plate was then incubated for 30 min.at 37 °C, and the concentration of peptide in the supernatant was analyzed by measuring the absorbance at 562 nm.

Docking
All molecular mechanics methods were performed in the Molecular Operating Environment version 2015-19 (MOE, Chemical Computing Group, Montreal, Canada) using the Born solvation model (LJ 12-6, dielectrics 78.6 and 4.0).All DFT calculations were performed in Gaussian 16 (G16RevC.01) 14using the polarizable continuum model (PCM, water) for solvation.Docking was performed on the whole surface of each PDB structure.Trios of sidechains of the fibril structures form channels perpendicular to the axis of growth, and each of these channels was defined as a binding site for docking.For each ligand, 1, 000 initial binding poses were generated per channel, with grid-based energy minimization (GRIDMIN).The London_dG scoring function was used to select the best 300 poses for energy minimization (RMSG=0.01,with fixed backbone atoms and unrestrained side chains).Results were ranked based on the forcefield interaction energy score.Ligands were not restricted from moving across different channels.
Prior to undertaking docking, our force fields were investigated for their ability to reproduce correct ligand structures.Due to its ubiquity in biological processes, Fe is considerably better parameterized than Ru.We thus looked at both the Fe(II) and Ru(II) compounds, as the properties of the ligand complexes would be largely identical within the limits of molecular mechanics.We compared AMBER-AM1-BCC with MMFF94x in a Born solvation model, using DFT (B3LYP/LANL2DZ in PCM continuum solvent model) as our reference.
The AMBER force field was originally developed to reproduce the thermodynamic properties of proteins, nucleic acids, and some small molecules, and as such is a good representation of the geometry and electrostatics of the protein and especially non-bonded interactions. 15It has the additional advantage of being compatible with the semi-empirical quantum mechanics method known as AM1-BCC for small molecules, and as such has the capacity to reasonably well describe electrostatics of small molecules. 16We additionally looked at the MMFF94x force field.A force field updated from the original MMFF94 and developed to describe drug-like molecules, the specific implementation within MOE has the additional benefit of being able to reasonably guess the properties of small molecules and metals which it wasn't originally parameterized for. 17Thus, MMFF94x has the advantage of potentially being more accurate in reproducing the geometry and electrostatics of small molecules, but the disadvantage of not being directly developed for describing protein interaction energies.Even so, the use of MMFF94x for docking purposes has previously been described and benchmarked. 18,19,20 Fre S25 demonstrates that MMFF94x performs considerably better at reproducing the DFT structures for both metal centres.We therefore performed our docking using the Ru(II) complexes and the MMFF94x force field..4) at 24 h incubation with agitation at 37 °C, using anti-A antibody 6E10.Lane 1: A1-42; lane 2: A1-42 + 0.10 eq.Ru complex; lane 3 A1-42 + 0.25 eq.Ru complex; lane 4: A1-42 + 0.50 eq.Ru complex; lane 5: A1-42 + 1.0 eq.Ru complex; lane 6: A1-42 + 1.5 eq.Ru complex; lane 7: A1-42 + 2.0 eq.Ru complex..4) at 0 h and 24 h incubation with agitation at 37 °C, using anti-A antibody 6E10.Lane 1: A1-42; lane 2: A1-42 + 0.10 eq.Ru complex; lane 3 A1-42+ 0.25 eq.Ru complex; lane 4: A1-42+ 0.50 eq.Ru complex; lane 5: A1-42+ 1.0 eq.Ru complex; lane 6: A1-42+ 1.5 eq.Ru complex; lane 7: A1-42+ 2.0 eq.Ru complex.S23).We excluded sites with a ligand interaction (docking score) below 5 kcal/mol (See Table S1 for values).

Figure S2 .
Figure S2.ESI-MS of unactivated Ru1 (A), Ru2 (B), and Ru3 (C) in NH4CO3 buffer (20 mM, pH 9.0) showing the stability of Ru1-2 when kept in the dark and ligand exchange for Ru3.Zoomed in regions show the isotope pattern for Ru complexes, red shows the theoretical isotope pattern expected for the complexes.

Figure S5 .
Figure S5.ESI-MS of activated Ru1 (A), Ru2 (B), and Ru3 (C) in NH4CO3 buffer (20 mM, pH 9.0) showing the release of the 6,6'-dmb ligand and new ligands occupying the vacant sites of the complexes.Zoomed in regions show the isotope pattern for Ru complexes, red shows the theoretical isotope pattern expected for the complexes.

Figure S6 .
Figure S6.(A) 1 H NMR of the red precipitate from photoactivation of Ru1 in 5% DMSO-d and PBS buffer (0.01 M, pH 7.4).(B) ESI-MS of the red precipitate from photoactivation of Ru1 showing the release of the 6,6'-dmb ligand and new ligands occupying the vacant sites of the complexes.Zoomed in regions show the isotope pattern for Ru complexes, red shows the theoretical isotope pattern expected for the complexes.

Figure S9 .
Figure S9.(A) ESI-MS of unactivated Ru2 + A1-16 showing no evidence of adduct formation.(B) ESI-MS of photoactivated Ru2 + A1-16 showing evidence of adduct formation.Zoomed regions exhibit the isotopic pattern of the adducts detected, and in red the theoretical isotopic pattern for the corresponding adduct.Samples were prepared in NH4CO3 buffer (20 mM, pH 9.0) and data was collected after 12 min.of activation.

Figure S10 .
Figure S10.(A) ESI-MS of unactivated Ru3 + A1-16 showing evidence of adduct formation.(B) ESI-MS of photoactivated Ru3 + A1-16 also showing evidence of adduct formation.Zoomed regions exhibit the isotopic pattern of the adducts detected, and in red the theoretical isotopic pattern for the corresponding adduct.Samples were prepared in NH4CO3 buffer (20 mM, pH 9.0) and data was collected after 25 min. of activation.

Figure S11 .
Figure S11.ESI-MS of Ru1 (A), Ru2 (C), and Ru3 (E) and A1-40 in the absence of photoactivation.ESI-MS of Ru1 (B), Ru2 (D), and Ru3 (F) and A1-40 after photoactivation indicating the adduct formation in the zoomed in regions.In red the theoretical isotopic pattern.Samples were prepared in NH4CO3 buffer (20 mM, pH 9.0) and data was collected after the respective times for activation.

Figure S13 .
Figure S13.Dot blot of 25 M A1-42 alone, and 25 M A1-42 in the presence of 1.0 eq. of photoactivated Ru1-3 at 0 h and 24 h of incubation showing that even after interaction with Ru complexes, the peptide is recognized by the 6E10 antibody.

Figure S14. 1 H
Figure S14.1 H NMR spectra of A1-16 (200 M) in the presence of 1.0 eq.Ru(bpy)2CO3 showing no changes of peptide residues after 24 h of incubation.Samples were prepared in PBS buffer (0.01 M, pH 7.4) at 37 °C.* His 6 , His13 and His 14 .† Tyr 10 .Broadening of signals is likely due to precipitation and/or multiple Ru species at the 24 h timepoint.

Figure S15 .
Figure S15.ESI-MS of Ru(bpy)2CO3 and A1-16 indicating adduct formation in the zoomed in region.The theoretical isotopic pattern is shown in red.Samples were prepared in NH4CO3 buffer (20 mM, pH 9.0) and data was collected after 25 min.activation time.

Figure S17 .
Figure S17.Increased magnification TEM image of amorphous aggregates and developing fibrils after incubation of A1-42 alone for 24 h (scale bar = 100 nm).See Figure S19 for further development into mature fibrils at 96 h.

Figure S19 .
Figure S19.TEM images of the change in morphology of A1-42 over time leading to the formation of mature fibrils after 96 h of incubation (scale bar = 200nm).

Figure S21 :
Figure S21: Binding sites found for Ru1-3 on 2MXU.Images are rotated in 3 dimensions to present the whole surface.For clarity, only Ru1 results are shown, as Ru2 is essentialy identical.Zoomed views of each site are provided below (FigureS23).We excluded sites with a ligand interaction (docking score) below 5 kcal/mol (See TableS1for values).

Figure S22 :
Figure S22: Binding sites found for Ru1-3 on 5OQV.Images are rotated in 3 dimensions to present the whole surface.For clarity, only Ru1 results are shown, as Ru2 is essentialy identical.Zoomed views of each site are provided below (FigureS23).As a dimer, binding sites on the other monomer were essentially identical and omitted for clarity.

Figure S23 :
Figure S23: Comparison of Ru1 (purple carbons) and Ru3 (yellow carbons) in each identified binding site in PDB structures 2MXU (sites A -G) and 5OQV (sites H-Q).The docking describes a shallow potential energy surface, with multiple bindings sites effectively contributing.Particular attention is drawn to sites F, I, and O as clearly demonstrating cases where Ru3 appears to bind more strongly due to a more accessible metal center available to carboxylate interactions.Other sites (e.g.B -D, H, J, L, P) show a shift in Ru3 position, but are equal or lower in score.