Insights on the Interaction between Transthyretin ... - ACS Publications

Article pubs.acs.org/jmc

Insights on the Interaction between Transthyretin and Aβ in Solution. A Saturation Transfer Difference (STD) NMR Analysis of the Role of Iododiflunisal Ana Gimeno,† Luis M. Santos,#,∇ Mobina Alemi,#,∇,○ Josep Rivas,⊥ Daniel Blasi,⊥ Ellen Y. Cotrina,∥ Jordi Llop,◆ Gregorio Valencia,∥ Isabel Cardoso,#,∇ Jordi Quintana,*,⊥ Gemma Arsequell,*,∥ and Jesús Jiménez-Barbero*,†,‡,§,⊥ †

CIC bioGUNE, Bizkaia Technology Park, Building 801A, 48170 Derio, Spain Ikerbasque, Basque Foundation for Science, Maria Diaz de Haro 13, 48009 Bilbao, Spain § Departament of Organic Chemistry II, Faculty of Science and Technology, University of the Basque Country, 48940 Leioa, Bizkaia, Spain ∥ Institut de Química Avançada de Catalunya (I.Q.A.C.-C.S.I.C.), 08034 Barcelona, Spain ⊥ Plataforma Drug Discovery, Parc Científic de Barcelona (PCB), Baldiri Reixac 10, 08028 Barcelona, Spain # IBMCInstituto de Biologia Celular e Molecular, Campo Alegre 823, 4150 Porto, Portugal ∇ i3SInstituto de Investigaçaõ e Inovaçaõ em Saúde, Universidade do Porto, Alfredo Allen, 4200-135 Porto, Portugal ○ Faculdade de Medicina, Universidade do Porto, Alameda Prof. Hernani Monteiro, 4200-319 Porto, Portugal ◆ Radiochemistry and Nuclear Imaging Group, CIC biomaGUNE, Paseo Miramon 182, 20009 Donostia-San Sebastian, Spain ‡

S Supporting Information *

ABSTRACT: Several strategies against Alzheimer disease (AD) are directed to target Aβ-peptides. The ability of transthyretin (TTR) to bind Aβ-peptides and the positive effect exerted by some TTR stabilizers for modulating the TTR−Aβ interaction have been previously studied. Herein, key structural features of the interaction between TTR and the Aβ(12−28) peptide (3), the essential recognition element of Aβ, have been unravelled by STD-NMR spectroscopy methods in solution. Molecular aspects related to the role of the TTR stabilizer iododiflunisal (IDIF, 5) on the TTR−Aβ complex have been also examined. The NMR results, assisted by molecular modeling protocols, have provided a structural model for the TTR−Aβ interaction, as well as for the ternary complex formed in the presence of IDIF. This basic structural information could be relevant for providing light on the mechanisms involved in the ameliorating effects of AD symptoms observed in AD/TTR± animal models after IDIF treatment and eventually for designing new molecules toward AD therapeutic drugs.

■

INTRODUCTION The amyloid hypothesis of Alzheimer disease (AD)1 has guided different drug development efforts. Unfortunately, little success has been achieved, the solanezumab phase III trial being one of the most recent failures.2 Despite this fact, Aβ peptides and their aggregates are still considered prominent molecular targets in the AD drug discovery arena. Thus, several approaches, based on the modulation of Aβ production, inhibition of Aβ aggregation, and enhancement of Aβ degradation, are being pursued.3 © 2017 American Chemical Society

In this context, the studies of the role of the plasma protein transthyretin (TTR) on AD constitute a different alternative. In the past few years, different experimental evidence has shown the ability of transthyretin (TTR) to protect against Aβ peptide-induced neurotoxicity.4 TTR is a 55 kDa homotetrameric protein responsible for the transport of thyroid hormones (T4) and retinol in blood and cerebrospinal fluid (CSF). It has Received: March 23, 2017 Published: June 6, 2017 5749

DOI: 10.1021/acs.jmedchem.7b00428 J. Med. Chem. 2017, 60, 5749−5758

Journal of Medicinal Chemistry

Article

Figure 1. (a) STD-NMR spectrum (on-resonance frequency δ −0.2 ppm) and 1H NMR reference spectrum (off-resonance frequency δ 100 ppm) of a sample containing 1.2 mM 1 and 30 μM TTR at 40 °C (600 MHz). (b) STD-NMR spectrum (on-resonance frequency δ −0.2 ppm) and 1H NMR reference spectrum (off-resonance frequency δ 100 ppm) of a sample containing 0.9 mM 2 and 30 μM TTR at 40 °C (600 MHz). (c) STD-NMR spectrum (on-resonance frequency δ −0.2 ppm) and 1H NMR reference spectrum (off-resonance frequency δ 100 ppm) of a sample containing 1.35 mM 3 and 30 μM TTR at 40 °C (600 MHz).

process even reverses the cognitive impairment observed in those AD model animals treated with IDIF. Herein, we have employed NMR spectroscopy methods, assisted by computational protocols, to analyze the molecular mechanisms underpinning these molecular recognition processes. We have used STD-NMR to examine the interaction between TTR and different Aβ peptide sequences. The NMR results have revealed the specific interaction between TTR and the main recognition element of Aβ, the Aβ(12−28) peptide. These NMR-based findings have also been supported by a preliminary computational model of the TTR−Aβ complex. The NMR protocol has also been applied to study the molecular recognition process in the presence of T4 ligands, particularly of our lead compound, IDIF, allowing us to propose a molecular model for the TTR−IDIF−Aβ interaction. This model provides a structural view of the involvement of TTR and T4 channel-binder molecules in AD, potential key information for designing compounds useful in the AD arena.

been described that TTR is involved in different amyloidosis diseases, due to the tendency of a large number of its natural mutants to aggregate, forming amyloid deposits in various vital organs, including the central as well as the peripheral nervous system.5 Accordingly, extensive efforts have been made to investigate the aggregation pathway of TTR and to discover powerful compounds that stabilize the folded tetrameric structure of the protein by docking on its T4 binding pocket and prevent its aggregation.6 In addition, TTR has been shown to play a wide neuroprotective role in CNS by being involved in ischemia, neuroregeneration, and memory performance.7 The importance of TTR stability is likely to be extensive to other processes as suggested by the finding that genetic stabilization of TTR (i.e., T119M TTR, a very stable tetramer and related to protection in TTR amyloidosis) has been associated with decreased risk of cerebrovascular disease and with increased life expectancy in the general population.8 Noteworthy, more recently, attention has been concentrated on the protective effect of TTR against Aβ toxicity. TTR is considered one of the major Aβ-binding-proteins in human CSF.9 In AD patients, its concentration in CSF is abnormally low and this fact has been considered as a biomarker of the disease.10 In fact, different biophysical studies have demonstrated that TTR binds Aβ monomers as well as low- and high-molecular mass oligomers, preventing their aggregation and deposition and thus decreasing the toxicity of Aβ-peptides.11 The use of different TTR mutants has permitted the proposal that the stability of the tetrameric structure of TTR plays a key role in its interaction with Aβ-peptides,12 although none of these studies have provided yet insightful evidence, at the molecular and atomic level, that could lead to a structural model useful for the design of potential drugs for AD. We have earlier demonstrated, using in vitro and in vivo studies, that TTR−Aβ interactions can be enhanced by employing tetramer-stabilizing compounds, such as iododiflunisal (IDIF).12b,13 This putative stabilization

■

RESULTS AND DISCUSSION Ligand-based STD-NMR experiments have been successfully applied to provide high resolution structural data on a variety of ligand−protein molecular recognition processes, especially for small ligands with affinities in mM to μM range.14 Thus, we decided to employ STD-NMR to analyze the interaction between TTR and Aβ peptides. Given the difficulties dealing with the complete Aβ sequence in solution, we especially focused our attention on the central hydrophobic domain of Aβ. It is considered to be the essential recognition element of the whole peptide and to be directly involved in the selfassembly process of Aβ.15 We have also analyzed the binding features of the hydrophilic N-terminus and the central loop of Aβ, regions that are found to play a key role in the aggregation process.16 Thus, STD-NMR experiments were performed for mixtures of TTR and three peptides, Aβ(1−11) (1), Aβ(10− 5750



Article

Figure 2. (a) Scheme of the TTR structure where the surface pocket, the retinol binding protein (RBP)-binding pocket, and the T4 central channel binding pocket are indicated. (b) Sequence alignment between TTR and the Aβ peptide using ClustalW software. The central hydrophobic core of Aβ (residues 17−21, LVFFA, purple color) provides a proper alignment with the hydrophobic patch of the surface pocket of TTR (residues 93−97, VVFTA). The following color code was used for describing the TTR amino acid sequence: orange for the retinol binding protein (RBP) binding regions, yellow for the T4 binding pocket residues, and green for the TTR surface pocket (dark green for the main surface pocket residues and pale green for the external residues of the same pocket). The color code for Aβ amino acid sequence is purple for the central hydrophobic cluster and pale purple for the highly hydrophobic region located at the C-terminus.

it has been described that this peptide 3 displays amyloid properties that resemble those of full length Aβ.17 These ligand-based NMR data identify the structural elements of Aβ(12−28) required for the binding with TTR, although obviously they do not shed light directly on the amino acids of the protein involved in the interaction. As an initial strategy to decipher this aspect, we used computational protocols. First, we carried out a computational prediction of the putative aggregating regions in the TTR and Aβ sequences, using the TANGO software, which provided information on the contribution of each key residue to the amyloidogenicity capacity of the whole molecules. The alignment between the two protein/peptide sequences of TTR and Aβ was also scrutinized by using molecular properties similarity, with the ClustalW software. Although the examination of the computational results (Figure 2) did not exclude other patterns of interaction between TTR and the full sequence of Aβ, a preferred interaction was suggested between residues 17−21 at the central hydrophobic core of Aβ with the hydrophobic patch formed by residues 93−97 in the TTR sequence (VVFTA), located at the surface pocket of the protein. Fittingly, the described peptide sequence contains the major NMR-based epitope motif described above for the interaction of 3 with TTR.

20) (2), and Aβ(12−28) (3), under similar experimental conditions (Figure 1). No STD response was observed for the TTR:1 and TTR:2 mixtures, even though experiments were acquired at many different temperatures and protein:ligand molar ratios. Conversely, clear STD signals were observed in the spectrum registered for the TTR:3 (1:45 molar ratio) mixture. This result indicates the exclusive recognition of this peptide fragment 3 by TTR. Furthermore, analysis of the STD profile allowed us to identify the peptide residues involved in the interaction. Relevant signals for the aromatic protons belonging to F19 and F20 were detected. Accordingly, signals for the aliphatic protons of F19-20 and for the methyl group of the nearest V18 also showed up in the spectrum. Thus, STD-NMR experiments were able to detect the interaction and indicated the close proximity of the VFF fragment of the peptide 3 to TTR, highlighting this epitope as the major structural motif for the recognition process. These results are in good agreement with dot-blot epitope mapping experiments with Aβ(1−42), which identified the region 18−21 as the binding site of TTR on Aβ.11e Noteworthy, the lack of STD response for peptide 2, also containing the VFF moiety, suggests that specific properties of the peptide fragment 3 are required for the interaction. In fact, 5751



Article

Figure 3. (a) X-ray crystal structure for the TTR−4 complex (PDB code 3NG5) indicating both different binding modes. (b) 1H NMR reference spectrum (off-resonance frequency δ 100 ppm) and STD spectrum (on-resonance frequency δ 0.16 ppm) of a sample containing 1.5 mM 4 and 30 μM TTR at 25 °C (600 MHz). (c) Relative STD-AF (amplification factor) for H1−H6 protons is indicated by colors.

To experimentally confirm the location of the TTR binding site for peptide 3, competition STD-NMR experiments were performed, based on the displacement of the TTR-bound peptide by specific TTR ligands with a well-established binding mode.18 Small compounds able to bind to TTR have been extensively screened as drug candidates against TTR-related amyloidosis diseases.6 Therefore, a broad data bank of TTR ligands is available and, in some cases, key binding features have been reported.19 The described TTR ligands include molecules that are buried at the interface between TTR monomers, while others bind at the TTR surface. We initially focused on ligands that are known to bind at the surface of the protein.20 In particular, (−)-epigallocatechin gallate (EGCG, 4) has been reported as a surface binding ligand with ability to stabilize the tetrameric structure of TTR.21 The deposited X-ray data (PDB code 3NG5) for the complex between the V30M TTR variant and 4 described three different binding sites, all located at the protein surface. Although the strongest electron density of 4 was found at the RBP binding pocket located between two dimers (binding mode A), a second binding site at the surface pocket (binding mode B), where 4 binds to TTR monomers, was also found (Figure 3a). Thus, we first analyzed this interaction by means of STD-NMR experiments. Accordingly, STD experiments performed on a 1:50 molar ratio of TTR and 4 (Figure 3b) provided evident STD signals for all protons of the ligand. Moreover, the group epitope mapping analysis (Figure 3c) showed the lowest STD intensity for protons H1, H2, and H3. This result is compatible with both binding modes (A and B) reported by the X-ray diffraction studies, where these protons of 4 are oriented away from the protein surface and exposed to the solvent. Corcema-ST calculations22 were performed to unmask possible subtle differences between the two binding modes,23 by comparing the experimental STD intensity detected for every proton of the ligand with that predicted for each binding mode (Figure 4a). The major differences between A and B binding modes were predicted for protons H5 and H5′ (Figure 4b). For the binding at the surface pocket (mode B), proton

Figure 4. (a) Epitope mapping of 4 when bound to TTR with the alternative binding geometries dubbed as A and B. For each geometry, the protons of the ligand are colored as a function of the CORCEMAcalculated saturation. Protons H5 and H5′ are indicated by arrows. (b) Graphic representation of the relative saturations of each proton of 4. The calculated Corcema-ST values for binding modes A and B are plotted in blue and red, respectively. The experimental values are given in green.

H5 would receive much lower saturation than H5′, while similar STD intensities would be expected for the binding at the RBP binding pocket (mode A). Fittingly, the expected theoretical STD profile for binding mode A perfectly matched with the experimental STD-NMR data, strongly suggesting the major contribution in solution of this binding mode. Indeed, 5752



Article

molecular model of the TTR−Aβ complex was generated (Figure 6). A molecular docking protocol was employed, using the software package MOE, affording a TTR−Aβ complex in which the Aβ peptide mainly interacts at the TTR surface. Although our STD-NMR competition experimental data have been obtained for the short Aβ(12−28) peptide, they are in satisfactory agreement with this proposed model, with the FFA triad providing clear contacts at the 93−97 amino acid sequence. Obviously, the full peptide expands along a large portion of the TTR surface and basically occupies one of the sides of the protein. The intrinsic flexibility of the peptide would provide a large occupancy of the conformational space around the protein. This model also supports previous investigations, based on SPOT binding analysis and mutation experiments, which have identified residues located in the E/F helix and in the loop at the surface of the protein as the key recognition motifs. However, binding to amino acids in the inner β-sheet of TTR was also postulated in these studies.24 As a further step, we decided to analyze the possible modulation of the interaction in the presence of other tetramer stabilizing compounds. As mentioned in the Introduction, we have shown that IDIF is a good TTR tetramer stabilizer,25 which decreases brain Aβ levels and deposition when administered to AD/TTR± mouse models.13 Thus, we explored the interaction between TTR and IDIF by STD-NMR and compared the obtained data in solution to the available X-ray results. 2 6 The water-soluble IDIF meglumine salt ([IDIF]−[MEG]+) (5) was used for the STD-NMR experiments. Clear STD signals for the aromatic protons of the negatively charged IDIF ligand were observed when mixtures of TTR and 5 were employed (between 1:20 to 1:50 molar ratios). In contrast, no STD intensities where detected for the counterion signals, indicating that the megluminium ion was not implicated in the binding event (Figure 7). Similar results were obtained for the noniodinated analogue compound diflunisal (DIF·meglumine salt, 6) (see Supporting Information for additional details). The existence of interactions was further confirmed by trNOESY experiments27 acquired for a 1:4 TTR:5 mixture (Figure 8). A blank NOESY spectrum of 5 was recorded in the absence of TTR (Figure 8a). Positive and negative cross peaks were observed in the 2D-NOESY spectrum for the free ligand. Inspection of the NOE signals evidenced the interconversion between the cis- and trans-rotational conformers (interresidual H2−H6′ and H6−H6′ cross peaks), indicating the existence of a low energy barrier for the rotation around the inter-ring C−C bond. Total conversion to negative NOESY cross peaks was observed for the [IDIF]− signals in the presence of TTR (Figure 8b). This change, due to an increase in the effective rotational correlation time of the ligand in the presence of the protein, supported the existence of binding between the negatively charged [IDIF]− moiety and TTR. Importantly, in the presence of TTR, the H6′−H2 cross peaks in the NOESY spectrum disappeared indicating the existence of a conformational selection process during the binding event, in favor of the cis-type conformer (Figure 8c). These results are in very good agreement with the X-ray data (PDB code 1Y1D) and clearly illustrate the molecular architecture around the binding site. The biphenyl molecule is buried between TTR subunits, thereby increasing the energy barrier for the rotation around the inter-ring C−C bond of the bound ligand.28

these results were also in agreement with reported point mutation experiments,21b where alteration of selected amino acids located at the surface and RBP binding pockets evidenced the importance of the binding at the RBP site for TTR−EGCG interaction. The accurate structural information provided by the NMR experiments about the binding event between TTR and 4 allowed us to use this ligand as competitor for Aβ-peptide displacement studies. Thus, addition of the surface binding ligand 4 to a 1:45 mixture of TTR:3 produced a noticeable reduction of the STD intensities of the signals of peptide 3, whereas signals for 4 appeared, indicating the binding of both compounds to TTR in a competitive manner (Figure 5).

Figure 5. (a) STD-NMR spectra (600 MHz) of the TTR:3 mixture (1:45 molar ratio) in the presence of increasing amounts of 4. The STD signals of 4 are indicated as asterisks (∗). (b) Representation of the decrease of the relative STD intensities of the HAr signals of peptide 3 as a function of added equivalents of 4.

Hence, the NMR competition experiments demonstrate that ligand 4 competes with peptide 3 for the same TTR binding site, thus allowing for unequivocal ascertaining of the binding site for peptide 3. Peptide 3 binds at the surface of TTR as 4, specifically at the RBP binding pocket located between the TTR dimers. Furthermore, the progressive reduction of the STD signal intensities of the aromatic protons of Phe19 and Phe20 of 3 after addition of 4 (Figure 5b) provided the estimation of the binding affinity of TTR for peptide 3, which is in the μM range. The Kd for the binding of 3 to TTR was estimated as ∼2 μM, based on the Kd value of 0.4 μM reported for the interaction of 4 with TTR. In order to provide a structural perspective for the interaction between TTR and the full Aβ monomer, a putative 5753



Article

Figure 6. TTR−Aβ model obtained through molecular docking. TTR and Aβ-peptide are represented as solid ribbons. The interacting residues, including the 93−97 fragment (TTR) and 18−20 (Aβ), are highlighted as CPK. There are van der Waals stabilizing contacts between Thr96-Val18 and Thr96-Phe20, Thr96-Ph21 CH−π interactions. PDB entry 1DVQ was used as starting point for building TTR geometry.

Figure 7. (a) STD spectrum (on-resonance frequency δ 0.16 ppm) and 1H NMR reference spectrum (off-resonance frequency δ 100 ppm) of a sample containing 2.5 mM ligand 5 and 50 μM TTR at 25 °C (600 MHz). (b) Relative STD-AF (amplification factor) for IDIF protons is indicated by colors. (c) X-ray structure for the TTR−IDIF complex (PDB code 1Y1D).

Next, we investigated the interaction between TTR and peptide 3 in the presence of 5 by using STD-NMR, trying to provide a structural basis to the positive effects exerted by this compound in the AD features observed in animal experiments.13 Addition of increasing amounts of 5 over a 1:45 mixture of TTR:3 yielded the progressive appearance of the signals of ligand 5 in the STD spectra (Figure 9). These data reflected the persistence of the interaction between 5 and TTR, strongly suggesting that the tetrameric structure of the protein was preserved in the presence of this Aβ-peptide. However, no variation of the STD response of 3 was observed, even in the presence of large excess of ligand 5 (an amount up to 225 equiv was added). These results showed that the TTR−3 interaction was not precluded by 5, although this compound has been described as a high-affinity T4-binder. According to the evidence, no significant changes in the kinetics or affinity of the TTR−3 interaction caused by the presence of 5 could be detected by this method under these experimental conditions. Thus, the obtained NMR data strongly suggest that both molecules (5 and 3) are able to simultaneously bind to TTR, although at different binding sites. While peptide 3 binds to TTR at the surface and may be displaced by ligand 4, ligand 5 is

exclusively located at the T4 channel, as also described in the Xray structure. Thus, the homotetrameric TTR stabilized by T4 ligands is still capable of properly interacting with the peptide 3, indirectly supporting the demonstrated ability of T4 ligands to improve the TTR−Aβ interaction.

■

CONCLUSION A combination of NMR techniques and modeling protocols have provided a structural model for the TTR−Aβ interaction and for the ternary complex formed in the presence of the T4binding ligand 5. STD-NMR experiments detected the interaction between TTR and peptide 3, the central hydrophobic core of Aβ, where the VFF (18−20) epitope is the main structural motif for the recognition. STD competition experiments assisted by modeling revealed the specific regions of TTR that are involved in this interaction. The binding of the Aβ-peptide is likely to occur at the surface of the protein, coincident with the surface binding region of compound 4. In contrast, other molecules such as ligand 5 bind within the T4 binding pocket generated between protein dimers. The NMR study has revealed that TTR can simultaneously recognize the peptide 3 and the T4-binding ligand 5 at different binding sites. 5754



Article

Figure 8. A conformational selection process takes place in the molecular recognition event of 5 by TTR. (a) Aromatic region of the 2D-NOESY NMR spectrum (left, 600 ms mixing time, 298 K) recorded at 600 MHz for the IDIF meglumine salt 5 (0.2 mM). (b) The same region in the 2DtrNOESY NMR spectrum (right, 200 ms mixing time, 298 K) obtained for the mixture containing TTR (50 μM) and IDIF meglumine salt (0.2 mM). (c) Representation of the cis and trans isomers of [IDIF]−, which coexist in solution in the free state, as deduced from the analysis of the 2DNOESY spectrum. The obtained cross peaks in the 2D-trNOESY spectrum indicate that only the cis-rotamer is bound.

■

EXPERIMENTAL SECTION

Transthyretin Production and Purification. Human recombinant wild-type (WT) TTR (WT TTR) was produced in a bacterial expression system, using Escherichia coli BL2 and purified as previously described.29 Briefly, after growing the bacteria, the protein was isolated and purified by preparative gel electrophoresis after ion exchange chromatography. Protein concentration was determined by the Bradford method (Bio-Rad), using bovine serum albumin (BSA) as standard. Ligands. Peptides 1−3 were purchased from Bachem (Switzerland). Compound 4 [(−)-epigallocatechin gallate, EGCG] and the NSAID drug diflunisal were purchased from Sigma-Aldrich and used without further purification. The water-soluble meglumine salts of the NSAID diflunisal and of iododiflunisal (IDIF), compounds 5 and 6, respectively, were synthesized as described previously.13 IDIF was prepared at IQAC as previously described.26 Purity of all compounds was proved to be ≥95% by means of HPLC and NMR techniques. NMR. The 1H NMR resonances of the ligands and peptides were assigned through standard TOCSY (60 and 90 ms mixing times), NOESY (300 and 500 ms mixing times), and HSQC experiments using Bruker AVANCE 2 600 and 800 MHz spectrometer equipped with standard triple-channel probe (600 MHz) and a cryoprobe (800 MHz). Typical concentrations and buffer conditions were 0.5−1 mM and H2O:D2O 90:10 for the peptides 1−3, whereas 1−2 mM concentrations and D2O were used for small aromatic compounds 4− 6. STD-NMR. The samples for the saturation transfer difference (STD) experiments were prepared in phosphate buffered saline (pH 6.0) using ligand/protein ratios varying from 1:20 to 1:50, with protein concentration of 20−50 μM and temperatures between 298 and 313 K. The molar ratios and temperatures were optimized in each specific case. In all cases, the on-resonance frequency was set at the aliphatic region (between δ −0.2 and 0.16 ppm), while the off-resonance frequency was set at δ 100 ppm. Protein saturation was achieved by using a series of 20 ms PC9 pulses with a total saturation time of the protein envelope signals of 2 s, with or without water suppression, using the Bruker AVANCE 2 600 MHz spectrometer. A spin-lock filter (50 ms) was used to remove the NMR signals of the protein background. Proper control experiments were performed with the

Figure 9. STD-NMR spectra (800 MHz) of TTR:3 (1:45 molar ratio) at 313 K in the presence of increasing amounts of IDIF·meglumine salt 5. The STD signals of 3 are not affected by the presence of 5. The STD signals of the aromatic moiety of 5 are indicated as asterisks (∗).

These findings provide additional fundamental knowledge on the molecular recognition processes that probably govern the mechanisms behind the AD ameliorating effect of IDIF observed in AD animal models. They could also be relevant for the design of new compounds toward achieving therapeutic drugs based on the TTR−Aβ molecular recognition event. 5755



Article

ligands in the presence and absence of TTR in order to optimize the frequency for protein saturation and to ensure that the ligand signals were not affected. trNOESY. Samples for NOESY (free ligand) and trNOESY (in the presence of TTR) experiments were prepared in phosphate buffered saline at pH 6.0. trNOESY experiments were performed with protein/ ligand molar ratios of 1:4 (based on a 50 μM protein concentration). The concentration of compound 5 was 0.2 mM in both cases. Spectra were obtained with the standard NOESY sequences using mixing times of 600 ms for the free ligand and 200 ms for the trNOESY experiment. The 2D spectra were acquired with 2K data points in the F2 dimension and 256 data points in the F1 dimension. The residual water signal was suppressed by presaturation. STD Competition Experiments. Competition experiments were performed recording sequential STD experiments at distinct concentrations of the employed competitors (EGCG compound 4 and IDIF compound 5) at 313 K with a constant protein concentration of 30 μM and a constant concentration of 1.35 mM Aβ(12−28) 3. The interference caused by the competitor 4, (Kd = 0.4 μM) was quantified by assessing the decrease of STD intensity of the aromatic protons of the Phe residues of peptide 3. In this manner, the Kd value of peptide 3 could be estimated.30 No interference was observed when IDIF 5 was employed. Corcema-ST Calculation. Corcema-ST MATLAB scripts22 were applied to the TTR−EGCG complex structures in both binding modes to deduce the binding mode of 4. The following input parameters were used in the calculations: 2 s saturation time; amino acids in a radius of 10 Å around the ligand as deduced from the X-ray structure of TTR (PDB code 3NG5); direct irradiation on the methyl groups of Ile, Leu, and Val (as an approximation to the 0.16 ppm experimental irradiation frequency); experimental Kd of 0.4 μM; experimental concentration conditions of 1.5 mM for 4 and 0.03 mM for TTR. A kon of 108 L mol−1 s−1 was used, assuming a diffusion controlled kinetic model. Correlation times of 0.5 and 30 ns for the ligand in the free and bound forms, respectively, were estimated following the typical empirical approximation31 and considering a 55 kDa tetrameric form for TTR. In Silico Prediction of TTR and Aβ Aggregation Regions. TANGO calculations were performed via http://tango.switchlab.org/ for both TTR and Aβ peptide, at a range of pHs from 4.2 to 7 (finally, only pH 7 was found to be consistent with the rest of the computational studies), 298.15 K, and 0.02 M ionic strength, without N- and C-terminal protection in both systems. This statistical mechanics based algorithm predicts the aggregation level for peptide sequences.32 For each residue in the sequence under study, TANGO computes the percent occupancy of the β-aggregated conformation. Proteins or peptides containing segments of at least five residues populating the β-aggregated conformation (each with scores higher than 5%) are considered to have a tendency to aggregate. Alignment Method. The alignment between TTR and the Aβ peptide sequences was performed using ClustalW33 via http://www. genome.jp/tools/clustalw/. For this protocol, a fast/approximate pairwise alignment method was selected. The employed parameters were 1 for K-tuple size, 5 for window size, 3 for gap penalty, 5 as number of top diagonals, and finally, percent was used as scoring method. TTR Crystallographic Structure Used in the Computational Studies. Three dimensional coordinates of the TTR protein structure (PDB entry 1DVQ)34 used in the present work were obtained from the structural information available at the Protein Data Bank (PDB) (www.rcsb.org). Hydrogen Atoms Refinement and Energy Minimization. The Protonate 3D package implemented in MOE 2015.10 was employed to add hydrogen atoms to TTR and Aβ which were further submitted to an energy-minimization process.35 Partial charges were obtained by computing the electrostatic potentials around the optimized structures. The energy minimization step was carried out using a distancedependent dielectric constant and a cutoff distance of 10 Å for the van der Waals interactions. Finally, the refinement was accomplished using 1000 cycles of steepest descents followed by conjugate gradients until

the maximum gradient of the energy was smaller than 0.05 kcal/(mol Å2). Docking Experiments. The MOE 2015.10 package was used to perform the docking studies between the refined TTR X-ray crystallography structure (as template) and Aβ peptide (as ligand). Alpha Triangle was used as placement method, Alpha HB as score function, and AMBER10 as force field in the first refinement step of the docking solutions. A rescoring step was also implemented in this computational pipeline, using London DG as function score and again AMBER10 as force field for the last refinement step of this docking study.

■

ASSOCIATED CONTENT

S Supporting Information *

. The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jmedchem.7b00428. Assigned 1H NMR spectra for ligands 1−6, additional details of Corcema-ST calculations, Kd determination by STD competition experiments, and binding analysis of TTR−ligand 6 (PDF) Molecular formula strings (CSV) Coordinates information for structure representation (PDB)

■

AUTHOR INFORMATION

Corresponding Authors

*J.Q.: e-mail, [email protected]. *G.A.: e-mail, [email protected]. *J.J.-B.: e-mail, [email protected]. ORCID

Jesús Jiménez-Barbero: 0000-0001-5421-8513 Author Contributions

The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The work was supported by a grant from the Fundació Marató de TV3 (Neurodegenerative Diseases Call, Project Reference 20140330-31-32-33-34, http://www.ccma.cat/tv3/marato/en/ projectes-financats/2013/212/). The group at CIC bioGUNE also acknowledges MINECO (Spain) for funding through Grant CTQ2015-64597-C2-1-P and a Juan de la Cierva contract to A.G. The group at IBMC-i3S also acknowledges funding through Grant Norte-01-0145-FEDER-000008-Porto Neurosciences and Neurologic Disease Research Initiative at I3S, supported by Norte Portugal Regional Operational Programme (NORTE 2020), under the PORTUGAL 2020 Partnership Agreement, through the European Regional Development Fund (FEDER). I.C. works under the Investigator FCT Program which is financed by national funds through FCT and cofinanced by ESF through HPOP, Type 4.2, Promotion of Scientific Employment. M.A. is currently a recipient of a Research Fellowship (BIM) funded by the project of Fundació Marató de TV3, Spain, and L.M.S. is currently a recipient of a fellowship from Norte 2020. IQAC-CSIC acknowledges a contract to E.Y.C. funded by the project of Fundació Marató de TV3, Spain. 5756



■

Article

(8) Hornstrup, L. S.; Frikke-Schmidt, R.; Nordestgaard, B. G.; Tybjaerg-Hansen, A. Genetic stabilization of transthyretin, cerebrovascular disease, and life expectancy. Arterioscler., Thromb., Vasc. Biol. 2013, 33, 1441−1447. (9) Schwarzman, A. L.; Gregori, L.; Vitek, M. P.; Lyubski, S.; Strittmatter, W. J.; Enghilde, J. J.; Bhasin, R.; Silverman, J.; Weisgraber, K. H.; Coyle, P. K.; Zagorki, M. G.; Talafous, J.; Eisenberg, M.; Saunders, A. M.; Roses, A. D.; Goldgaber, D. Transthyretin sequesters amyloid beta protein and prevents amyloid formation. Proc. Natl. Acad. Sci. U. S. A. 1994, 91, 8368−8372. (10) (a) Gloeckner, S. F.; Meyne, F.; Wagner, F.; Heinemann, U.; Krasnianski, A.; Meissner, B.; Zerr, I. Quantitative analysis of transthyretin, tau and amyloid-beta in patients with dementia. J. Alzheimer's Dis. 2008, 14, 17−25. (b) Poulsen, K.; Bahl, J. M.; Simonsen, A. H.; Hasselbalch, S. G.; Heegaard, N. H. Distinct transthyretin oxidation isoform profile in spinal fluid from patients with Alzheimer’s disease and mild cognitive impairment. Clin. Proteomics 2014, 11, 12−21. (c) Ribeiro, C. A.; Santana, I.; Oliveira, C.; Baldeiras, I.; Moreira, J.; Saraiva, M. J.; Cardoso, I. Transthyretin decrease in plasma of MCI and AD patients: investigation of mechanisms for disease modulation. Curr. Alzheimer Res. 2012, 9, 881−889. (11) (a) Tsuzuki, K.; Fukatsu, R.; Yamaguchi, H.; Tateno, M.; Imai, K.; Fujii, N.; Yamauchi, T. Transthyretin binds amyloid beta peptides, Abeta1-42 and Abeta1-40 to form complex in the autopsied human kidney - possible role of transthyretin for abeta sequestration. Neurosci. Lett. 2000, 281, 171−174. (b) Buxbaum, J. N.; Ye, Z.; Reixach, N.; Friske, L.; Levy, C.; Das, P.; Golde, T.; Masliah, E.; Roberts, A. R.; Bartfai, T. Transthyretin protects Alzheimer’s mice from the behavioral and biochemical effects of Ab toxicity. Proc. Natl. Acad. Sci. U. S. A. 2008, 105, 2681−2686. (c) Costa, R.; Gonçalves, A.; Saraiva, M. J.; Cardoso, I. Transthyretin binding to A-beta peptide − Impact on Abeta fibrillogenesis and toxicity. FEBS Lett. 2008, 582, 936−942. (d) Li, X.; Masliah, E.; Reixach, N.; Buxbaum, J. N. Neuronal production of transthyretin in human and murine Alzheimer’s disease: is it protective? J. Neurosci. 2011, 31, 12483−12490. (e) Li, X.; Zhang, X.; Ladiwala, A. R. A.; Du, D.; Yadav, J. K.; Tessier, P. M.; Wright, P. E.; Kelly, J. W.; Buxbaum, J. N. Mechanisms of transthyretin inhibition of β-amyloid aggregation in vitro. J. Neurosci. 2013, 33, 19423−19433. (12) (a) Du, J.; Murphy, R. M. Characterization of the interaction of β-amyloid with transthyretin monomers and tetramers. Biochemistry 2010, 49, 8276−8289. (b) Ribeiro, C. A.; Saraiva, M. J.; Cardoso, I. Stability of the transthyretin molecule as a key factor in the interaction with A-beta peptide-relevance in Alzheimer’s disease. PLoS One 2012, 7, e45368. (c) Yang, D. T.; Joshi, G.; Cho, P. Y.; Johnson, J. A.; Murphy, R. M. Transthyretin as both a sensor and a scavenger of βamyloid Oligomers. Biochemistry 2013, 52, 2849−2861. (13) Ribeiro, C. A.; Oliveira, S. M.; Guido, L. F.; Magalhaes, A.; Valencia, G.; Arsequell, G.; Saraiva, M. J.; Cardoso, I. Transthyretin stabilization by iododiflunisal promotes amyloid-β peptide clearance, decreases its deposition, and ameliorates cognitive deficits in an Alzheimer’s disease mouse model. J. Alzheimer’s Dis. 2014, 39, 357− 370. (14) (a) Meyer, B.; Peters, T. NMR spectroscopy techniques for screening and identifying ligand binding to protein receptors. Angew. Chem., Int. Ed. 2003, 42, 864−890. (b) Bhunia, A.; Bhattacharjya, S.; Chatterjee, S. Applications of saturation transfer difference NMR in biological systems. Drug Discovery Today 2012, 17, 505−13. (15) Takahashi, T.; Mihara, H. Peptide and protein mimetics inhibiting amyloid beta-peptide aggregation. Acc. Chem. Res. 2008, 41, 1309−1318. (16) (a) Nguyen, P. H.; Tarus, B.; Derreumaux, P. Familial Alzheimer A2V mutation reduces the intrinsic disorder and completely changes the free energy landscape of the Aβ1−28 monomer. J. Phys. Chem. B 2014, 118, 501−510. (b) Das, P.; Murray, B.; Belfort, G. Alzheimer’s protective A2T mutation changes the conformational landscape of the Aβ1−42 monomer differently than does the A2V mutation. Biophys. J. 2015, 108, 738−747. (c) Nguyen, P. H.; Sterpone, F.; Campanera, J. M.; Nasica-Labouze, J.; Derreumaux, P.

ABBREVIATIONS USED AD, Alzheimer disease; TTR, transthyretin; CSF, cerebrospinal fluid; IDIF, iododiflunisal; EGCG, (−)-epigallocatechin gallate; RBP, retinol binding protein; STD, saturation transfer difference

■

REFERENCES

(1) Selkoe, D. J.; Hardy, J. The amyloid hypothesis of Alzheimer’s disease at 25 years. EMBO Mol. Med. 2016, 8, 595−608. (2) Amanatkar, H. R.; Papagiannopoulos, B.; Grossberg, G. T. Analysis of recent failures of disease modifying therapies in Alzheimer’s disease suggesting a new methodology for future studies. Expert Rev. Neurother. 2017, 17, 7−16. (3) (a) Citron, M. Alzheimer’s disease: strategies for disease modification. Nat. Rev. Drug Discovery 2010, 9, 387−398. (b) Karran, E.; Mercken, M.; De Strooper, B. The amyloid cascade hypothesis for Alzheimer’s disease: an appraisal for the development of therapeutics. Nat. Rev. Drug Discovery 2011, 10, 698−712. (c) Mullard, A. Alzheimer amyloid hypothesis lives on. Nat. Rev. Drug Discovery 2016, 16, 3−5. (d) Nasica-Labouze, J.; Nguyen, P. H.; Sterpone, F.; Berthoumieu, O.; Buchete, N.-V.; Coté, S.; DeSimone, A.; Doig, A. J.; Faller, P.; Garcia, A.; Laio, A.; Li, M. S.; Melchionna, S.; Mousseau, N.; Mu, Y.; Paravastu, A.; Pasquali, S.; Rosenman, D. J.; Strodel, B.; Tarus, B.; Viles, J. H.; Zhang, T.; Wang, C.; Derreumaux, P. Amyloid β protein and Alzheimer’s disease: when computer simulations complement experimental studies. Chem. Rev. 2015, 115, 3518−3563. (e) Doig, A. J.; Derreumaux, P. Inhibition of protein aggregation and amyloid formation by small molecules. Curr. Opin. Struct. Biol. 2015, 30, 50− 56. (4) (a) Stein, T. D.; Johnson, J. A. Lack of neurodegeneration in transgenic mice overexpressing mutant amyloid precursor protein is associated with increased levels of transthyretin and the activation of cell survival pathways. J. Neurosci. 2002, 22, 7380−7388. (b) Buxbaum, J. N.; Ye, Z.; Reixach, N.; Friske, L.; Levy, C.; Das, P.; Golde, T.; Masliah, E.; Roberts, A. R.; Bartfai, T. Transthyretin protects Alzheimer’s mice from the behavioral and biochemical effects of Aβ toxicity. Proc. Natl. Acad. Sci. U. S. A. 2008, 105, 2681−2686. (c) Stein, T. D.; Anders, N. J.; De Carli, C.; Chan, S. L.; Mattson, M. P.; Johnson, J. A. Neutralization of transthyretin reverses the neuroprotective effects of secreted amyloid precursor protein (APP) in APPSw mice resulting in tau phosphorylation and loss of hippocampal neurons: support for the amyloid hypothesis. J. Neurosci. 2004, 24, 7707−7717. (5) (a) Ando, Y.; Ueda, M. Recent advances in transthyretin amyloidosis therapy. Curr. Med. Chem. 2012, 19, 2312−2323. (b) Sekijima, Y. Recent progress in the understanding and treatment of transthyretin amyloidosis. J. Clin. Pharm. Ther. 2014, 39, 225−233. (6) (a) Johnson, S. M.; Wiseman, R. L.; Sekijima, Y.; Green, N. S.; Adamski-Werner, S. L.; Kelly, J. W. Native state kinetic stabilization as a strategy to ameliorate protein misfolding diseases: a focus on the transthyretin amyloidosis. Acc. Chem. Res. 2005, 38, 911−921. (b) Connelly, S.; Choi, S.; Johnson, S. M.; Kelly, J. W.; Wilson, I. A. Structure-based design of kinetic stabilizers that ameliorate the transthyretin amyloidoses. Curr. Opin. Struct. Biol. 2010, 20, 54−62. (c) Nencetti, S.; Orlandini, E. TTR fibril formation inhibitors: is there a SAR? Curr. Med. Chem. 2012, 19, 2356−2379. (d) Obici, L.; Merlini, G. An overview of drugs currently under investigation for the treatment of transthyretin-related hereditary amyloidosis. Expert Opin. Invest. Drugs 2014, 23, 1239−1251. (7) (a) Santos, S. D.; Lambertsen, K. L.; Clausen, B. H.; Akinc, A.; Alvarez, R.; Finsen, B.; Saraiva, M. J. CSF transthyretin neuroprotection in a mouse model of brain ischemia. J. Neurochem. 2010, 115, 1434−1444. (b) Fleming, C. E.; Saraiva, M. J.; Sousa, M. M. Transthyretin enhances nerve regeneration. J. Neurochem. 2007, 103, 831−839. (c) Sousa, J. C.; Marques, F.; Dias-Ferreira, E.; Cerqueira, J. J.; Sousa, N.; Palha, J. A. Transthyretin influences spatial reference memory. Neurobiol. Learn. Mem. 2007, 88, 381−385. 5757



Article

Impact of the A2V mutation on the heterozygous and homozygous Aβ1-40 dimer structures from atomistic simulations. ACS Chem. Neurosci. 2016, 7, 823−832. (17) (a) Jarvet, J.; Damberg, P.; Bodell, K.; Göran Eriksson, L. E.; Gräslund, A. Reversible random coil to β-sheet transition and the early stage of aggregation of the Aβ(12−28) fragment from the Alzheimer peptide. J. Am. Chem. Soc. 2000, 122, 4261−4268. (b) Narayanan, S.; Reif, B. Characterization of chemical exchange between soluble and aggregated states of β-amyloid by solution-state NMR upon variation of salt conditions. Biochemistry 2005, 44, 1444−1452. (c) Milojevic, J.; Esposito, V.; Das, R.; Melacini, G. Understanding the molecular basis for the inhibition of the Alzheimer’s Aβ-peptide oligomerization by human serum albumin using saturation transfer difference and offresonance relaxation NMR spectroscopy. J. Am. Chem. Soc. 2007, 129, 4282−4290. (18) Wang, Y. S.; Liu, D.; Wyss, D. F. Competition STD NMR for the detection of high-affinity ligands and NMR-based screening. Magn. Reson. Chem. 2004, 42, 485−489. (19) Palaninathan, S. K. Nearly 200 X-Ray crystal structures of transthyretin: What do they tell us about this protein and the design of drugs for TTR amyloidoses? Curr. Med. Chem. 2012, 19, 2324−2342. (20) Ortore, G.; Orlandini, E.; Braca, A.; Ciccone, L.; Rossello, A.; Martinelli, A.; Nencetti, S. Targeting different transthyretin binding sites with unusual natural compounds. ChemMedChem 2016, 11, 1865−1874. (21) (a) Ferreira, N.; Cardoso, I.; Domingues, M. R.; Vitorino, R.; Bastos, M.; Bai, G.; Saraiva, M. J.; Almeida, M. R. Binding of epigallocatechin-3-gallate to transthyretin modulates its amyloidogenicity. FEBS Lett. 2009, 583, 3569−3576. (b) Miyata, M.; Sato, T.; Kugimiya, M.; Sho, M.; Nakamura, T.; Ikemizu, S.; Chirifu, M.; Mizuguchi, M.; Nabeshima, Y.; Suwa, Y.; Morioka, H.; Arimori, T.; Suico, M. A.; Shuto, T.; Sako, Y.; Momohara, M.; Koga, T.; MorinoKoga, S.; Yamagata, Y.; Kai, H. The crystal structure of the green tea polyphenol(−)-epigallocatechin gallate−transthyretin complex reveals a novel binding site distinct from the thyroxine binding site. Biochemistry 2010, 49, 6104−6114. (c) Ferreira, N.; Saraiva, M. J.; Almeida, M. R. Natural polyphenols as modulators of TTR amyloidogenesis: in vitro and in vivo evidences towards therapy. Amyloid 2012, 19, 39−42. (22) (a) Jayalakshmi, V.; Krishna, N. R. Complete relaxation and conformational exchange matrix (CORCEMA) analysis of intermolecular saturation transfer effects in reversibly forming ligand-receptor complexes. J. Magn. Reson. 2002, 155, 106−118. (b) Rama Krishna, N.; Jayalakshmi, V. Complete relaxation and conformational exchange matrix analysis of STD-NMR spectra of ligand-receptor complexes. Prog. Nucl. Magn. Reson. Spectrosc. 2006, 49, 1−25. (23) Zhang, W.; Li, R.; Shin, R.; Wang, Y.; Padmalayam, I.; Zhai, L.; Krishna, N. R. Identification of the binding site of an allosteric ligand using STD-NMR, docking, and CORCEMA-ST calculations. ChemMedChem 2013, 8, 1629−1633. (24) (a) Du, J.; Cho, P. Y.; Yang, D. T.; Murphy, R. M. Identification of beta-amyloid-binding sites on transthyretin. Protein Eng., Des. Sel. 2012, 25, 337−345. (b) Yang, D. T.; Joshi, G.; Cho, P. Y.; Johnson, J. A.; Murphy, R. M. Transthyretin as both a sensor and a scavenger of βamyloid oligomers. Biochemistry 2013, 52, 2849−2861. (25) (a) Almeida, M. R.; Macedo, B.; Cardoso, I.; Alves, I.; Valencia, G.; Arsequell, G.; Planas, A.; Saraiva, M. J. Selective binding to transthyretin and tetramer stabilization in serum from patients with familial amyloidotic polyneuropathy by an iodinated diflunisal derivative. Biochem. J. 2004, 381, 351−356. (b) Cardoso, I.; Almeida, M. R.; Ferreira, N.; Arsequell, G.; Valencia, G.; Saraiva, M. J. Comparative in vitro and ex vivo activities of selected inhibitors of transthyretin aggregation: relevance in drug design. Biochem. J. 2007, 408, 131−138. (26) (a) Mairal, T.; Nieto, J.; Pinto, M.; Almeida, M. R.; Gales, L.; Ballesteros, A.; Barluenga, J.; Pérez, J. J.; Vázquez, J. T.; Centeno, N. B.; Saraiva, M. J.; Damas, A. M.; Planas, A.; Arsequell, G.; Valencia, G. Iodine atoms: a new molecular feature for the design of potent transthyretin fibrillogenesis inhibitors. PLoS One 2009, 4, e4124.

(b) Gales, L.; Macedo-Ribeiro, S.; Arsequell, G.; Valencia, G.; Saraiva, M. J.; Damas, A. M. Human transthyretin in complex with iododiflunisal: structural features associated with a potent amyloid inhibitor. Biochem. J. 2005, 388, 615−662. (27) Unione, L.; Galante, S.; Díaz, D.; Cañada, F. J.; JiménezBarbero, J. NMR and molecular recognition. The application of ligandbased NMR methods to monitor molecular interactions. MedChemComm 2014, 5, 1280−1289. (28) (a) Grein, F. Twist angles and rotational energy barriers of biphenyl and substituted biphenyls. J. Phys. Chem. A 2002, 106, 3823− 3827. (b) Di Pietro, M. E.; Aroulanda, C.; Merlet, D.; Celebre, G.; De Luca, G. Conformational investigation in solution of a fluorinated antiinflammatory drug by NMR spectroscopy in weakly ordering media. J. Phys. Chem. B 2014, 118, 9007−9016. (29) Almeida, M. R.; Damas, A. M.; Lans, M. C.; Brouwer, A.; Saraiva, M. J. Thyroxine binding to transthyretin Met 119. Endocrine 1997, 6, 309−315. (30) Meinecke, R.; Meyer, B. Determination of the binding specificity of an integral membrane protein by saturation transfer difference NMR: RGD peptide ligands binding to integrin αIIbβ3. J. Med. Chem. 2001, 44, 3059−3065. (31) Cantor, C. R.; Schimmel, P. R. Biophysical Chemistry. Part II. Techniques for the Study of Biological Structure and Function; W. H. Freeman and Company: Oxford, U.K., 1980; p 503. (32) Fernandez-Escamilla, A. M.; Rousseau, F.; Schymkowitz, J.; Serrano, L. Prediction of sequence-dependent and mutational effects on the aggregation of peptides and proteins. Nat. Biotechnol. 2004, 22, 1302−1306. (33) Larkin, M. A.; Blackshields, G.; Brown, N. P.; Chenna, R.; McGettigan, P. A.; McWilliam, H.; Valentin, F.; Wallace, I. M.; Wilm, A.; Lopez, R.; Thompson, J. D.; Gibson, T. J.; Higgins, D. G. Clustal W and Clustal X version 2.0. Bioinformatics 2007, 23, 2947−2948. (34) Klabunde, T.; Petrassi, H. M.; Oza, V. B.; Raman, P.; Kelly, J. W.; Sacchettin, J. C. Rational design of potent human tranthyretin amyloid disease inhibitors. Nat. Struct. Biol. 2000, 7, 312−321. (35) Molecular Operating Environment (MOE), version 2013.08; Chemical Computing Group Inc. (1010 Sherbooke St. West, Suite No. 910, Montreal, QC, Canada), 2016.

5758


Insights on the Interaction between Transthyretin ... - ACS Publications

Recommend Documents