Reactivity of Diphenylpropynone Derivatives Toward Metal-Associated Amyloid-β Species

Abstract

In Alzheimer’s disease (AD), metal-associated amyloid-β (metal–Aβ) species have been suggested to be involved in neurotoxicity; however, their role in disease development is still unclear. To elucidate this aspect, chemical reagents have been developed as valuable tools for targeting metal–Aβ species, modulating the interaction between the metal and Aβ, and subsequently altering metal–Aβ reactivity. Herein, we report the design, preparation, characterization, and reactivity of two diphenylpropynone derivatives (DPP1 and DPP2) composed of structural moieties for metal chelation and Aβ interaction (bifunctionality). The interactions of these compounds with metal ions and Aβ species were confirmed by UV-Vis, mass spectrometry, and docking studies. The effects of these bifunctional molecules on the control of in vitro metal-free and metal-induced Aβ aggregation were investigated and monitored by gel electrophoresis and transmission electron microscopy (TEM). Both DPP1 and DPP2 showed reactivity toward metal–Aβ species over metal-free Aβ species to different extents. In particular, DPP2, which contains a dimethylamino group, exhibited greater reactivity with metal–Aβ species than DPP1, suggesting a structure-reactivity relationship. Overall, our studies present a new bifunctional scaffold that could be utilized to develop chemical reagents for investigating metal–Aβ species in AD.

Introduction

Alzheimer’s disease (AD) is a fatal neurodegenerative disease that affects over 5 million people in the United States.^{1, 2} A pathological hallmark of the diseased brain is an accumulation of misfolded amyloid-β (Aβ) aggregates.^2–8 Monomeric Aβ peptides, generated from the proteolytic cleavage of the transmembrane amyloid precursor protein, can further aggregate to produce oligomers, protofibrils, and eventually fibrils. It is still not completely understood which conformation of Aβ is associated with AD neuropathogensis; however, recent evidence has proposed that soluble oligomers might be the toxic species due to their ability to interrupt neurotransmission.^{4, 5, 9} In addition to Aβ species, elevated concentrations of transition metals, such as Fe, Cu and Zn, have been observed within the deposits of Aβ aggregates.^{2, 7, 8, 10–15} The possible relationship between metal-associated Aβ species (metal–Aβ species) and neurotoxicity has been suggested based on observations that upon binding to Aβ, metal ions facilitate peptide aggregation as well as enhance oxidative stress caused by overproduction of reactive oxygen species ;^{2, 4, 5, 7, 8, 12–21} however, this connection has not been clearly revealed.

To gain a better understanding of the involvement of metal–Aβ species in AD pathogenesis, recent advancements in the development of chemical reagents to specifically target metal–Aβ species and modulate their interaction and reactivity have been made.^{2, 7, 8, 21–33} Among them, rationally-designed small molecules with both metal chelation and Aβ interaction properties (defined as bifunctionality) have been devised. Some of the compounds have been fashioned based on the incorporation approach (Figure 1), where a metal chelation site is directly inserted into a known Aβ imaging agent with minimal structural modifications, along with consideration of criteria for possible brain uptake.^{8, 21–25, 27–35} These molecules have been shown to be able to control the interactions and reactivity of metal–Aβ species in vitro and/or in living cells, suggesting that the incorporation approach (Figure 1) could be considered as a promising design strategy to fashion suitable chemical reagents for uncovering the potential role of metal–Aβ species in AD development.

Figure 1.

Incorporation approach (top) and structures of small molecules (bottom). Left to right: 3-(4-(dimethylamino)phenyl)-1-(4-iodophenyl)-2-propyn-1-one; DPP1 = 3-phenyl-1-(pyridin-2-yl)prop-2-yn-1-one; DPP2 = 3-(4-(dimethylamino)phenyl)-1-(pyridin-2-yl)-2-propyn-1-one.

Herein, we report the design and preparation of a new class of bifunctional small molecules (DPP1 and DPP2, Figure 1), composed of a diphenylpropynone framework, as chemical reagents for targeting and regulating metal–Aβ species, the characterization of their interactions with metal ions and Aβ species, as well as the investigation of their in vitro reactivity with metal–Aβ species. From our overall results and observations, the diphenylpropynone scaffold could be a structural feature for constructing chemical reagents for investigating metal–Aβ species. Moreover, the dimethylamino functionality in DPP2 was observed to be an important moiety for potentially enhancing Aβ interaction of the compound, as reported previously,^{29, 31, 36} offering noticeable reactivity with metal–Aβ species.

Experimental Section

Materials and Procedures

All reagents were purchased from commercial suppliers and used as received unless otherwise stated. The compound, 3-phenyl-l-(pyridin-2-yl)prop-2-yn-l-one (DPP1) was prepared following previously reported methods.^37–39 Aβ_1–40 (DAEFRHDSGYEVHHQKLVFFAEDVGSNKGAIIGLMVGGVV) was purchased from AnaSpec (Fremont, CA). An Agilent 8453 UV-Visible (UV-Vis) spectrophotometer was used to measure the optical spectra. Transmission electron microscopy (TEM) images were recorded with a Philips CM-100 transmission electron microscope. A SpectraMax M5 microplate reader (Molecular Devices, Sunnyvale, CA) was employed for the measurement of absorbance for 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) and parallel artificial membrane permeability (PAMPA) assays. Nuclear magnetic resonance (NMR) spectra of small molecules and for Zn²⁺ binding studies were obtained by a Varian 400 MHz NMR spectrometer. Mass spectrometric measurements for compounds were conducted by a Micromass LCT Electrospray Time-of-Flight mass spectrometer. Mass spectrometric studies for investigating the interaction of DPP1 and DPP2 with the peptide were carried out on a Waters Synapt G2 ion-mobility mass spectrometer (Milford, MA).

Preparation of 3-(4-(dimethylamino)phenyl)-l-(pyridin-2-yl)-2-propyn-l-one (DPP2)

DPP2 was synthesized by slight modifications to a previously reported procedure.³⁹ To a solution of 4’-dimethylaminophenyl acetylene (0.42 g, 2.9 mmol) in dry tetrahydrofuran (THF, 5 mL) was added n-butyllithium (0.72 mL, 2.9 mmol, 2.5 M solution in hexanes) dropwise by a syringe over 5 min at −40 °C. The reaction mixture was allowed to stir at −40 °C for 10 min and then warmed to −15 °C. After 30 min, the mixture was cooled to −78 °C (dry ice/acetone) and a solution of Weinreb’s amide (0.40 g, 2.4 mmol in 9 mL of dry THF) was introduced through a syringe. The reaction mixture was allowed to react at −78 °C for 10 min and then warmed to room temperature and followed by 1 h stirring. The reaction was quenched by adding saturated aqueous NaHCO₃ (2 mL), diluted with EtOAc (10 mL), and washed with brine (2 × 25 mL). The aqueous layer was extracted with EtOAc (2 × 15 mL), and the combined organic solutions were dried over MgSO₄ and filtered, followed by removal of the solvent in vacuo. The crude product was purified by silica gel chromatography (CH₂Cl₂:EtOAc = 9:1) to yield an orange product (202 mg, 0.81 mmol, 28%). ¹H NMR (400 MHz, CDCl₃)/δ (ppm): 8.77 (d, J = 4.8 Hz, 1 H), 8.14 (d, J = 7.6 Hz, 1 H), 7.82 (td, J = 7.6, 1.6 Hz, 1 H), 7.57 (d, J = 8.4 Hz, 2 H), 7.44 (m, 1 H), 6.59 (d, J = 8.4 Hz, 2 H), 2.98 (s, 6 H). ¹³C NMR (100 MHz, CDCl₃)/δ (ppm): 177.6, 153.7, 151.8, 149.7, 136.8, 135.6, 126.9, 123.3, 111.4, 105.5, 100.2, 89.3, 39.9. HRMS: Calcd for [M+H]⁺, 251.1179; Found, 251.1176.

Parallel Artificial Membrane Permeability Assay Adapted for Blood-Brain Barrier (PAMPA-BBB)

PAMPA-BBB experiments were carried out using the PAMPA Explorer kit (pION Inc., Billerica, MA) with modification to previously reported protocols.^{29, 40–42} Each stock solution was diluted with pH 7.4 Prisma HT buffer (pION) to a final concentration of 25 µM (1% v/v final DMSO concentration) and 200 µL were added to the wells of the donor plate (number of replicates = 12). BBB-1 lipid formulation (5 µL pION) was used to coat the polyvinylidene fluoride (PVDF, 0.45 µM) filter membrane on the acceptor plate. The acceptor plate was placed on top of the donor plate forming a -sandwich- and the brain sink buffer (BSB, 200 µL, pION) was added to each well of the acceptor plate. The sandwich was incubated for 4 h at ambient temperature without stirring. UV-Vis spectra of the solutions in the reference, acceptor, and donor plates were measured using a microplate reader. The PAMPA Explorer software v. 3.5 (pION) was used to calculate the −logP_e values for the compounds. CNS+/− designations were assigned by comparison to compounds that were identified in previous reports.^{29, 40–42}

Determination of Solution Speciation for DPP1, DPP2, and the Cu²⁺-DPP2 Complex

The pK_a values for DPP1 and DPP2 were determined by UV-Vis variable-pH titrations as previously reported.^{27, 29–33, 43} To establish the pK_a values, a solution (100 mM NaCl, 10 mM NaOH, pH 12) of DPP1 (40 µM) or DPP2 (20 µM) was titrated with small aliquots of HC1. At least 30 spectra were recorded in the range of pH 2–10. Similarly, a solution containing DPP2 (20 µM) and CuCl₂ in a ratio of 2:1 was titrated with small additions of HC1 and at least 30 spectra were recorded over the range of pH 2–7. The acidity and stability constants were calculated by using the HypSpec program (Protonic Software, UK).⁴⁴ Speciation diagrams for DPP1, DPP2, and Cu²⁺–DPP2 complexes were modeled by the HySS2009 program (Protonic Software).⁴⁵

Metal Binding Studies

The interaction of DPP1 and DPP2 with Cu²⁺ and Zn²⁺ was determined by UV-Vis and ¹H NMR spectroscopy, respectively, based on previously reported procedures.^{28–33, 46} A solution of ligand (20 µM in EtOH) was prepared, treated with 1 to 20 equiv of CuCl₂, and incubated at room temperature for 2.5 h (for DPP1) or 5 min (for DPP2). The optical spectra of the resulting solutions were measured by UV-Vis. The interaction of DPP1 or DPP2 with ZnCl₂ was observed by ¹H NMR. ZnCl₂ (1 equiv) was added to a solution of DPP1 or DPP2 (4 mM) in CD₃CN. The metal selectivity of both compounds was investigated by measuring the optical changes upon addition of 1 equiv of CuCl₂ to a solution of ligand (DPP1 = 40 µM; DPP2 = 20 µM in EtOH) containing 1 or 25 equiv of a divalent metal chloride salt (MgCl₂, CaCl₂, MnCl₂, FeCl₂, CoCl₂, NiCl₂, or ZnCl₂). The Fe²⁺ samples were prepared anaerobically (all solutions were purged with N₂). Quantification of the metal selectivity was calculated by comparing and normalizing the absorption values of metal-ligand complexes at 360 nm (for DPP1) or 580 nm (for DPP2) to the absorption at these wavelengths before and after the addition of CuCl₂ (A_M/A_Cu). Cu²⁺ binding of DPP2 in the presence of Aβ was examined by UV-Vis. Aβ (25 µM) was treated with CuCl₂ (25 µM in 20 mM HEPES (2-[4-(2-hydroxyethyl)piperazin-l-yl]ethanesulfonic acid), pH 6.6, 150 mM NaCl for 2 min at room temperature. DPP2 (50 µM)was added to the resulting solution, followed by 0.5–4 h incubation. For comparison, the optical spectra of DPP2 (50 µM)were measured in the absence and presence of CuCl₂ (25 µM; 0.5–4 h incubation) without Aβ at pH 6.6.

Aβ Interaction of DPP1 and DPP2 by Mass Spectrometry

The interaction of DPP1 or DPP2 with Aβ_1–40 was investigated by nano-electrospray ionization-mass spectrometry (nESI-MS) that was carried out on a Synapt G2 quadrupole-ion mobility-mass spectrometry system. Samples were prepared by mixing stock solutions of DPP1 or DPP2 (prepared in DMSO) and Aβ_1–40 (dissolved in 100 mM ammonium acetate, pH 6.8) to generate desired final concentrations of the peptide and the compound. Mixtures were incubated on ice or at room temperature for 2 or 4 h, respectively and then analyzed. To produce protein complex ions, an aliquot of the sample (ca. 5 µL) was sprayed from the nESI emitter using a capillary voltage of 1.4 kV, with the source operating in positive ion mode and the sample cone operated at 50 V. In order to normalize nESI-MS data for non-specific and electrospray artifact interactions which could occur at high concentrations, data were acquired for Aβ samples containing thioflavin-T (ThT), a compound known to have no affinity for soluble forms of the Aβ peptide,⁴⁷ under identical concentration conditions as our DPP1 and DPP2 experiments. Any ThT binding observed was assumed to be due to either non-specific binding or the electrospray process, and subtracted from the intensities of the DPP1 and DPP2 interactions observed.⁴⁸ This procedure was performed over a broad range of concentrations. The mass spectra were acquired with the following settings and tuned to avoid ion activation and to preserve non-covalent protein-ligand complexes:⁴⁹ backing pressure, 7.3 mbar; IMS pressure reading, 3.09 mbar; ToF analyzer pressure, 1.14 × 10⁻⁶ mbar.

Docking Studies

Flexible ligand docking studies using AutoDock Vina⁵⁰ for DPP1 and DPP2 were conducted against the Aβ_1–40 monomer from the previously determined aqueous solution NMR structure (PDB 2LFM).⁵¹ Ten conformations were selected from 20 conformations within the PDB file (1, 3, 5, 8, 10, 12, 13, 16, 17, and 20). The MMFF94 energy minimization in ChemBio3D Ultra 11.0 was used to optimize the structures of the ligands for the docking studies. The structural files of DPP1, DPP2, and the peptide were generated by AutoDock Tools and imported into PyRx,⁵² which were used to run AutoDock Vina. The search space dimensions were set to contain the entire peptide. The exhaustiveness for the docking runs was set at 1024. Docked poses of the ligands were visualized with Aβ using Pymol.

Amyloid-β (Aβ) Peptide Experiments

Aβ_1–40 was used in all Aβ experiments. Aβ_1–40 peptide (1 mg) was dissolved with ammonium hydroxide (NH₄OH, 1% v/v, aq), aliquoted, lyophilized, and stored at −80 °C. A stock solution (ca. 200 µM) was prepared by redissolving Aβ with NH₄OH (1% v/v, aq, 10 µL) followed by dilution with ddH₂O. All Aβ solutions were prepared following previously reported procedures.^28–32 The buffered solutions (20 µM HEPES, pH 6.6 (for Cu²⁺ samples) or pH 7.4 (for metal-free and Zn²⁺ samples), 150 µM NaCl) were used for both inhibition and disaggregation studies. For the inhibition experiment, Aβ (25 µM) was first treated with a metal chloride salt (CuCl₂or ZnCl₂, 25 µM) for 2 min followed by addition of a compound (DPP1 or DPP2, 50 µ in DMSO, 1% v/v final DMSO concentration). The resulting samples were incubated at 37 °C for 4, 8, or 24 h with constant agitation. For the disaggregation experiment, Aβ and a metal chloride salt (CuCl₂ or ZnCl₂) were initially incubated at 37 °C for 24 h with steady agitation. The compound was added afterwards followed by additional 4, 8, or 24 h incubation at 37 °C with constant agitation.

Native Gel Electrophoresis/Sodium Dodecyl Sulfate-Polyacrylamide Gel Electrophoresis (SDS-PAGE) with Western Blotting

The Aβ peptide experiments (as described above) were analyzed by both native gel electrophoresis and SDS-PAGE with Western blotting using an anti-Aβ antibody (6E10).^{28–32, 53} Each sample containing 25 µM Aβ (10 µL) was separated using either a 10–20% gradient Tris-tricine gel (Invitrogen, Grand Island, NY) or SDS gel (4% stacking gel; 10% resolving gel; non-reducing conditions). The gel was transferred to a nitrocellulose membrane and blocked overnight with bovine serum albumin (BSA, 3% w/v, Sigma, St. Louis, MO) dissolved in Tris-buffered saline (TBS, Fisher, Pittsburgh, PA) containing 0.1% Tween-20 (TBS-T, Sigma). The membrane was treated with 6E10 (1:2,000; 2% BSA in TBS-T, Covance, Princeton, NJ) for 4 h at room temperature. The membrane was probed with a horseradish peroxidase-conjugated goat anti-mouse secondary antibody (1:5,000; Cayman Chemical, Ann Arbor, MI) in 2% BSA in TBS-T solution for 1 h at room temperature. The protein bands were visualized using the Thermo Scientific Supersignal West Pico Chemiluminescent Subtrate (Rockford, IL).

Transmission Electron Microscopy (TEM)

Samples for TEM were prepared following a previously reported method.^{28–31, 33, 53} Glow-discharged grids (Formar/Carbon 300-mesh, Electron Microscopy Sciences, Hatfield, PA) were treated with samples from either inhibition or disaggregation experiments (5 µL) for 2 min at room temperature. Excess sample was removed with filter paper and washed with ddH₂O five times. Each grid was stained with uranyl acetate (1% w/v, ddH₂O, 5 µL) for 1 min. Uranyl acetate was blotted off and grids were dried for 15 min at room temperature. Images of samples were taken by a Philips CM-100 transmission electron microscope (80 kV, 25,000x magnification).

Cytotoxicity (MTT Assay)

The murine neuro-2a (N2a) neuroblastoma cell line was purchased from the American Type Culture Collection (ATCC, Manassas, VA). The cell line was maintained in media containing 45% Dulbecco’s modified Eagle’s medium (DMEM) and 50% OPTI-MEM (GIBCO, Grand Island, NY), supplemented with 5% fetal bovine serum (FBS, Sigma), 2 mM glutamine, 100 U/ml penicillin, and 100 mg/ml streptomycin (GIBCO). The cells were grown and maintained at 37 °C in a humidified atmosphere with 5% CO₂. Cell viability upon treatment of compounds was determined using a MTT assay (Sigma). N2a cells were seeded in a 96 well plate (15,000 cells in 100 µL per well) and treated with various concentrations of DPP1 and DPP2 (2.5–50 µM, final 1% v/v DMSO). After 24 h incubation at 37 °C, 25 µL MTT (5 mg/mL in phosphate buffered saline, PBS, pH 7.4, GIBCO) was added to each well and the plates were incubated for 4 h at 37 °C. Formazan produced by the cells was dissolved overnight at room temperature by addition of a solution (100 µL) containing N,N-dimethylformamide (DMF, 50% v/v, aq, pH 4.5) and sodium dodecyl sulfate (SDS, 20% w/v). A microplate reader was used to measure the absorbance (A₆₀₀).

Results and Discussion

Design Consideration, Preparation, and Characterization of Diphenylpropynone Derivatives for Targeting and Modulating Metal–Aβ Species

The diphenylpropynone scaffold was shown to have high binding affinity (ca. 6 nM) for Aβ aggregates; thus, it has been utilized for the design of an Aβ plaque imaging probe (Figure 1).⁵⁴ Based on the incorporation approach, a nitrogen donor atom was installed into this framework to generate a metal chelation site with an oxygen donor atom of the carbonyl group, which afforded two bifunctional molecules (DPP1 and DPP2, Figure 1). Furthermore, a minor structural difference between DPP1 and DPP2 (i.e., a dimethylamino functionality, suggested to be critical for Aβ interaction)^{29, 31, 36} was made to understand a structure-reactivity relationship. DPP1 was synthesized as previously established.^37–39 The new compound, DPP2, was prepared by slight modifications to a previously reported method (Scheme 1).³⁹

Scheme 1.

Synthesis of DPP2.

In order to predict potential drug-like and BBB penetration properties of the structural scaffolds of DPP1 and DPP2, values of Lipinski’s rules and logBB were calculated.^{7, 34, 35, 40, 41} As shown in Table 1, the theoretical values indicate that both compounds are drug-like and possibly BBB permeable. To verify this prediction for BBB penetration of compounds, in vitro PAMPA-BBB was performed following a previously reported procedure.^{29, 40–42} Permeability values (−logP_e were measured to be 4.2 (±0.1) for both DPP1 and DPP2 (Table 1). Based on empirical classification of BBB-permeable molecules (i.e., verapamil),^{29, 40–42} DPP1 and DPP2 would also be likely to cross the BBB.

Table 1.

Values (MW, clogP, HBA, HBD, PSA, logBB, and −logP_e) of DPP1 and DPP2.

Calculationa	DPP1	DPP2	Lipinski’s rules & others
MW	207	250	≤ 450
clogP	2.47	2.63	≤ 5.0
HBA	2	3	≤ 10
HBD	0	0	≤ 5
PSA	30.0	33.2	≤ 90 Å2
logBB	0.0620	0.0384	> 0.3 (readily cross the BBB) <−1.0 (poorly distributed in the brain)
−logPeb	4.2 ± 0.1	4.2 ± 0.1
CNS+/− predictionc	CNS+	CNS+	−logPe < 5.4 (CNS+) −logPe > 5.7 (CNS−)

^aMW: Molecular weight; clogP: Calculated logarithm of the octanol-water partition coefficient; HBA: Hydrogen-bond acceptor atoms; HBD: Hydrogen-bond donor atoms; PSA: Polar surface area; logBB = −0.0148 × PSA + 0.152 × clogP + 0.130.

^bThe values of −logP_e were measured by the Parallel Artificial Membrane Permeability assay (PAMPA).

^cCompounds categorized as CNS+ have the ability to permeate through the BBB and target the CNS. In the case of compounds assigned as CNS−, they have poor permeability through the BBB and therefore, their bioavailability into the CNS is considered to be minimal.

Along with BBB permeability, the solution speciation of DPP1 and DPP2 was determined through UV-Vis variable-pH titration experiments (I = 0.10 M NaCl; room temperature; pH 2 – 10).^{27, 29–33, 43} As summarized in Figure 2, titration results indicated a single acidity constant (pK_a) for DPP1 (pK_a = 2.035(5)) and two pK_a values for DPP2 (pK_al = 7.106(1) and pK_a2 = 2.959(4)). These pK_a values suggest that monoprotonated, diprotonated, and neutral species exist in solution of DPP2 depending on pH (from 2 to 10), while the monoprotonated and neutral forms of DPP1 are present in this pH range (Figure 2). In addition, the generated solution speciation diagrams of DPP1 and DPP2 exhibit that their neutral forms are relatively predominant at physiologically relevant pH (i.e., 7.4) (DPP1, 100%; DPP2, ca. 65%).

Figure 2.

Solution speciation studies of DPP1 and DPP2. Top: UV-Vis spectra of DPP1 (40 µM, left) and DPP2 (20 µM, right) in the range of pH 2–10. Middle: Solution speciation diagrams for DPP1 (left) and DPP2 (right) (F_L = fraction of compound with given protonation). Bottom: Acidity constants (pK_a) of L (L = DPP1 or DPP2). Charges are omitted for clarity. ^a Error in the parentheses is shown in the last digit. Conditions: I = 0.10 M NaCl; room temperature.

Metal Binding Properties of DPP1 and DPP2

Metal binding properties of DPP1 and DPP2 (specifically, Cu²⁺ and Zn²⁺) were studied by UV-Vis and NMR spectroscopy. Upon the addition of CuCl₂ (1–20 equiv) to a solution of DPP1 and DPP2 in EtOH, new optical bands were observed, indicative of Cu²⁺ binding to the ligand (Figure 3a). In particular, in the presence of Cu²⁺, DPP2, which has a dimethylamino group, showed a distinguishable optical shift from 407 to 525 nm. NMR was employed to investigate the interaction of DPP1 or DPP2 with Zn²⁺. When 1 equiv of ZnCl₂ was introduced in a solution of DPP1 or DPP2 (in CD₃CN), noticeable downfield chemical shifts of the pyridyl protons were recorded (Figure 3b), demonstrating that Zn²⁺ binding to the ligand occurred through the pyridyl N-donor atom.⁴⁶ Overall, both UV-Vis and NMR studies confirmed Cu²⁺ or Zn²⁺ binding to DPP1 and DPP2.

Figure 3.

Cu²⁺ or Zn²⁺ binding of DPP1 and DPP2. (a) UV-Vis spectra of DPP1 (left) and DPP2 (right) with CuCl₂ (1–20 equiv) in EtOH at room temperature (incubation for 2.5 h (for DPP1) and 5 min (for DPP2)). (b) ¹H NMR spectra of DPP1 (left, black) or DPP2 (right, black) with ZnCl₂ (red) in CD₃CN at room temperature ([compound] = 4 mM; [ZnCl₂] = 4 mM).

To further identify binding stoichiometry and affinity, the solution speciation investigation of the Cu²⁺–DPP2 complexes was carried out through UV-Vis variable-pH titration experiments (1:2 [Cu²⁺]/[DPP2]; I = 0.10 M NaCl, room temperature). Based on the pK_a values of DPP2 and these titration results, stability constants for the Cu²⁺–DPP2 complexes were obtained (Figure 4, M + LH ⇌ M(LH) (logβ₁ = 12.99(9)); M + L ⇌ ML (logβ₂ = 5.85(3)); M = Cu²⁺, L = DPP2). A solution speciation diagram was generated from these stability constants, suggesting that the major species at pH 7 are a mixture of Cu(LH) and CuL complexes in a ratio of 3:2. Free Cu²⁺ was shown up to pH 7, indicating pCu = 6.6 at pH 6.6 (pCu = −log[Cu_unchelated]) (Figure 4).^{27, 29, 31–33, 43} The pCu value suggests the approximate dissociation constant (K_d) of Cu²⁺–DPP2 to be ca. high nanomolar. When compared to the reported K_d values of Cu²⁺–Aβ (picomolar to nanomolar),^{2, 5, 7, 8, 12, 14, 21} DPP2 may interact with Cu²⁺ from soluble Cu²⁺–Aβ species. In order to test if Cu²⁺ binding of ligand occurs in the presence of Aβ, a solution containing DPP2 with Cu²⁺-treated Aβ was monitored by UV-Vis. The new spectral features coincided with those of the Cu²⁺–DPP2 complex without Aβ were observed, suggesting an interaction of DPP2 with Cu²⁺ in the presence of Aβ (Figure S1). Cu²⁺ binding of DPP2 with Aβ occurred more slowly than that without Aβ, proposing that Aβ might interfere with metal binding to the ligand. Taken together, our spectroscopic studies present the capability of DPP1 and DPP2 to chelate Cu²⁺ and Zn²⁺, as well as the potential interaction of DPP2 with Cu²⁺ in the presence of Aβ species, which may be associated with its noticeable reactivity toward metal–Aβ species (vide infra).

Figure 4.

Solution speciation investigation of the Cu²⁺–DPP2 complexes. Top left: UV-Vis spectra (pH 2–7) for the Cu²⁺–DPP2 complexes ([Cu²⁺]/[L] = 1:2; [Cu²⁺]_total = 10 µM; 7 h incubation with ligand (L) prior to pH titration, L = DPP2; room temperature). Top right: Solution speciation diagram of the Cu²⁺–DPP2 complexes (F_Cu = fraction of free Cu and Cu complexes). Bottom: Stability constants (logβ) of the Cu²⁺–DPP2 complexes. Charges are omitted for clarity. ^aError in the parentheses is shown in the last digit. ^b The species containing CuL₂ was introduced into the calculation model yielding a good fit to the data.

The metal selectivity of DPP1 and DPP2 was also determined by competitive reactions with Cu²⁺ over biologically relevant divalent metal ions (Mg²⁺, Ca²⁺, Mn²⁺, Fe²⁺, Co²⁺, Ni²⁺, and Zn²⁺), which was monitored by UV-Vis. As depicted in Figure S2, DPP1 and DPP2 displayed selectivity for Cu²⁺ over Mg²⁺, Ca²⁺, Mn²⁺, and Zn²⁺. Binding of both ligands to Co²⁺ and Ni²⁺ was also observed. Considering the lower abundance of Co²⁺ and Ni²⁺ than Cu²⁺ in biological systems,^{55, 56} the overall metal selectivity of DPP1 and DPP2 may be sufficient to be used for targeting and interacting with Cu²⁺–Aβ species in heterogeneous biological environments like the brain.

Aβ Interaction with DPP1 and DPP2 Studied by MS and Docking Studies

The interaction of DPP1 and DPP2 with Aβ_1–40 in the absence of metal ions was probed by electrospray ionization-mass spectrometry (ESI-MS), tuned to preserve non-covalent protein-ligand interactions.⁴⁹ At a low Aβ concentration (10 µM), a small signal corresponding to the interaction between DPP2 (30 µM) and the Aβ monomer in the 3+ charge state could be detected, while no interaction between DPP1 (60 µM) and the peptide was observed under these conditions (Figure S3). At high concentrations of the peptide (100 µM) and compounds (600 µM) both DPP1 and DPP2 interacted with Aβ species to different extents (Figure 5a). Data for DPP1 indicated that the molecule interacted broadly with Aβ monomers and oligomers in 1:1, 2:1, and 3:1 Aβ to ligand ratios. In the case of DPP2, a stronger preference toward larger Aβ oligomers was shown, but with similar stoichiometries as DPP1. The total bound intensities recorded from MS data, and those from individual oligomeric species were shown in Figure 5b and Table S1. The intensities shown were normalized for both non-specific interactions and artifactual complexes formed during the electrospray process using Aβ:ThT binding data as a control, and ion mobility separation was used to separate oligomers that overlapped in m/z.^{6, 49, 57} From these data, it was clear that, at high concentrations, a higher proportion of DPP1 was bound to Aβ species than DPP2, but that both could be classified as having weak Aβ affinity in solution (low mM K_d). Therefore, a weak Aβ/compound interaction was captured by MS. Normalized intensity MS data suggest that DPP2 binding was almost exclusively driven through Aβ multimer interactions (Figure 5b). Overall, our MS results suggest that while both compounds could interact with Aβ species at high concentrations, DPP2 was able to bind Aβ species at both low and high concentrations.

Figure 5.

Aβ interactions of DPP1 and DPP2. (a) MS data for the complexes of Aβ_1–40 and DPP1 or DPP2 ([Aβ] =100 µM; [compound] = 600 µM; M = monomer, D = dimer, and T = trimer). Many binding stoichiometries were detected, including 1:1 (star), 2:1 (square), and 3:1 (triangle), (b) A histogram showing the total bound MS signal intensity, normalized for non-specific interactions and ESI-MS artifacts, for each binding stoichiometry observed in (a), (c) Docking studies of DPP1 (orange) and DPP2 (green) with Aβ_1–40 (PDB 2LFM) by AutoDock Vina. Poses for both compounds were overlapped in this conformation (other conformations, Figure S4). The helical region of Aβ (H13-D23) is highlighted in color (tan) in both the cartoon (left) and surface (right) representations.

In order to visualize the potential interaction between Aβ and DPP1/DPP2, docking studies by AutoDock Vina⁵⁰ were performed using the previously determined NMR structure (PDB 2LFM)⁵¹ of Aβ_1–40 monomer. Typically, both compounds were positioned between the α-helix and the unstructured N-terminal side of Aβ (Figures 5c and S4). Most docked structures showed a non-specific orientation of the ligand with respect to the surface features of Aβ. Our preliminary docking studies support the potential interaction of the compounds with Aβ monomer.

Effects of DPP1 and DPP2 on Metal-free and Metal-induced Aβ Aggregation In Vitro

Confirming metal binding and Aβ interaction properties (bifunctionality) of DPP1 and DPP2, their influence on in vitro metal-free and metal-induced Aβ aggregation were studied.^26–33 Two different experiments (inhibition and disaggregation) were performed to investigate whether DPP1/DPP2 can control the formation of metal-free and metal-induced Aβ aggregates (inhibition, Figure 6) or transform preformed metal-free and metal-induced Aβ aggregates (disaggregation, Figure S5). Various-sized Aβ species from both studies were monitored by native gel electrophoresis and SDS-PAGE followed by Western blotting with an anti-Aβ antibody (6E10), while morphological changes were identified by transmission electron microscopy (TEM).^{26, 28–33, 53}

Figure 6.

Inhibition experiment (scheme, top). Analysis of various-sized Aβ species by (a) native gel electrophoresis and (b) SDS-PAGE (non-reducing conditions) with Western blot using an anti-Aβ antibody (6E10). (c) TEM images of the 24 h incubated samples. Conditions: [Aβ] = 25 µM; [CuCl₂ or ZnCl₂] = 25 µM; [compound] = 50 µM; pH 6.6 (for Cu²⁺ samples) or 7.4 (for metal-free and Zn²⁺ samples); 4, 8, or 24 h incubation; 37 °C; constant agitation.

The inhibition studies, as shown in Figure 6, demonstrated a different reactivity of DPP1 or DPP2 toward metal-induced Aβ species over metal-free Aβ species. In the case of metal-involved Aβ aggregation, Aβ species with a wide range of MW were indicated with DPP2 for both Cu²⁺- and Zn²⁺-treated samples upon longer incubation (Figure 6a, lanes 6 and 9). In the samples containing DPP1 (lanes 5 and 8), less intense gel bands were detected across the longer incubation time, suggesting that further Aβ aggregation may have occurred. The reaction of DPP1 or DPP2 with metal-free Aβ also exhibited a different distribution of various-sized Aβ species (Figure 6a, lanes 2 and 3). Aβ species formed with compounds in both metal-mediated and metal-free conditions were not completely denatured by SDS implying that these molecules may generate different Aβ assemblies (Figure 6b). TEM images of metal-induced Aβ species incubated with DPP2 for 24 h revealed smaller amorphous aggregates compared to DPP1; some of the metal-free Aβ species treated with DPP1 and DPP2 presented similar morphology to those untreated with compounds (Figure 6c). Overall, DPP1 and DPP2 displayed their ability to recognizably modulate metal-induced Aβ aggregation to different extents.

Furthermore, for the disaggregation experiment (Figure S5), DPP2-treated metal-triggered Aβ aggregates presented different-sized Aβ species than DPPl-treated samples, indicating that DPP2 could alter the properties of preformed Aβ aggregates to a greater extent than DPP1 (Figure S5a). In the metal-free conditions, more various-sized Aβ species were indicated in the presence of DPP2 than DPP1 (in particular, at 4 h, Figure S5a, lanes 2 and 3). The Aβ species generated with compounds in both metal-triggered and metal-free conditions were relatively stable in the presence of SDS (Figure S5b). As shown in Figure S5c, DPP2 was able to reorganize preformed structured metal–Aβ aggregates to amorphous species more noticeably than DPP1. Taken together, the results from both the inhibition and disaggregation experiments presented that DPP1 and DPP2 could regulate metal-involved Aβ aggregation over metal-free aggregation in vitro differently. Moreover, the structural variation (i.e., dimethylamino functionality) may enhance the contact with Aβ species via hydrogen bonding and/or hydrophobic interactions,^{29, 31, 36} which may afford greater reactivity toward metal–Aβ species.

Summary and Perspective.

Following the incorporation approach, we have developed two bifunctional small molecules (DPP1 and DPP2) composed of a metal chelation site and a diphenylpropynone framework (for Aβ interaction), which could possibly serve as chemical reagents to target and modulate metal–Aβ species in vitro. Their bifunctionality (metal chelation and Aβ interaction) was confirmed by physical methods and preliminary docking studies. Biochemical and TEM studies revealed that DPP1 and DPP2 could modulate metal-induced Aβ aggregation in vitro. Notably, DPP2, which has a dimethylamino group, exhibited more apparent reactivity toward metal–Aβ species, compared to DPP1. This suggests that the interaction and reactivity of molecules with metal–Aβ species can be tuned by such structural variations, proposing a structure-interaction-reactivity relationship. DPP1 and DPP2, however, would be limited in their biological applications since they displayed cytotoxicity in living cells at low micromolar concentrations (Figure S6). Overall, the promising in vitro reactivity of these potentially BBB-permeable molecules toward metal–Aβ species warrants pursuit of structural modifications that would improve the viability of diphenylpropynone derivatives in biological settings, followed by more detailed characterization by MS and molecular modeling. This study has demonstrated the capability of two diphenylpropynone derivatives to target metal–Aβ species and modulate their interaction and reactivity in vitro, which can be further optimized toward the development of future chemical reagents for investigating metal–Aβ species in biological systems.