Modulation of the fibrillogenesis inhibition properties of two transthyretin ligands by halogenation

Article abstract of DOI:10.1021/jm401061w

The amyloidogenic protein transthyretin (TTR) is thought to aggregate into amyloid fibrils by tetramer dissociation which can be inhibited by a number of small molecule compounds. Our analysis of a series of crystallographic protein-inhibitor complexes has shown no clear correlation between the observed molecular interactions and the in vitro activity of the inhibitors. From this analysis, it emerged that halogen bonding (XB) could be mediating some key interactions. Analysis of the halogenated derivatives of two well-known TTR inhibitors has shown that while flufenamic acid affinity for TTR was unchanged by halogenation, diflunisal gradually improves binding up to 1 order of magnitude after iodination through interactions that can be interpreted as a suboptimal XB (carbonyl Thr106: I..O distance 3.96-4.05 ?; C - I..O angle 152-156) or as rather optimized van der Waals contacts or as a mixture of both. These results illustrate the potential of halogenation strategies in designing and optimizing TTR fibrillogenesis inhibitors.

Full text of DOI:10.1021/jm401061w

Article
pubs.acs.org/jmc
Modulation of the Fibrillogenesis Inhibition Properties of Two
Transthyretin Ligands by Halogenation
Ellen Y. Cotrina,† Marta Pinto,∥,‡ Lluís Bosch,§ Marta Vila,† Daniel Blasi,⊥ Jordi Quintana,⊥
̀
Nuria B. Centeno,‡ Gemma Arsequell,§ Antoni Planas,† and Gregorio Valencia*,§
†Laboratory of Biochemistry, Bioengineering Department, Institut Químic de Sarria,
Group, Research Programme on Biomedical Informatics (GRIB), Department of Experimental and Health Sciences, Universitat
̀
Universitat Ramon Llull, ‡Pharmacoinformatics
§
Pompeu Fabra, IMIM (Hospital del Mar Medical Research Institute)-Universitat Pompeu Fabra, Institut de Química Avanca
̧
da de
⊥
Catalunya (IQAC−CSIC), Drug Discovery Platform, Parc Científic de Barcelona, Barcelona, Spain
S
* Supporting Information
ABSTRACT: The amyloidogenic protein transthyretin
(TTR) is thought to aggregate into amyloid fibrils by tetramer
dissociation which can be inhibited by a number of small
molecule compounds. Our analysis of a series of crystallo-
graphic protein-inhibitor complexes has shown no clear
correlation between the observed molecular interactions and
the in vitro activity of the inhibitors. From this analysis, it
emerged that halogen bonding (XB) could be mediating some
key interactions. Analysis of the halogenated derivatives of two
well-known TTR inhibitors has shown that while flufenamic
acid affinity for TTR was unchanged by halogenation,
diflunisal gradually improves binding up to 1 order of
magnitude after iodination through interactions that can be
interpreted as a suboptimal XB (carbonyl Thr106: I...O distance 3.96−4.05 Å; CI...O angle 152−156°) or as rather optimized
van der Waals contacts or as a mixture of both. These results illustrate the potential of halogenation strategies in designing and
optimizing TTR fibrillogenesis inhibitors.
between E and F strands (Figure 1).40−43 The association of
INTRODUCTION
■
two monomers (A and B) leads to a dimer (AB), and the two
dimers (AB and A′B′) assemble to form the functional tetramer
that features a central channel at the dimer−dimer interface
where the two chemically equivalent hormone binding sites,
named A-A′ and B-B′, respectively, are located (Figure 1).
A relevant feature of the TTR binding sites is the presence of
three symmetry-related depressions termed halogen binding
pockets (HBPs),16 which accommodate the iodine atoms of the
thyroid hormones in their complexes with TTR.44 The
innermost pockets are HBP3 and HBP3′, and the outermost
are HBP1 and HBP1′ (Figure 2).
Structural information on TTR is extensive as exemplified by
the nearly 200 TTR-related entries45 that are currently
collected at the Protein Data Bank (http://www.rcsb.
org).46With the aim to devise new strategies for the design of
more effective TTR tetramer stabilizing ligands, we have
sampled and computationally analyzed this PDB information
on TTR. Attention has been paid to the in silico study of
ligand−HBPs interactions and to experimentally evaluate the
effects of introducing homologous series of halogen atoms into
Mutations in the plasma protein transthyretin (TTR) are the
cause of a series of rare but serious amyloid diseases.1−5
Numerous studies have focused on showing that protein
stabilization by low molecular weight compounds can be an
effective therapeutic alternative for the TTR-related amyloid
diseases,6−11 which include familial amyloid polyneuropathy
(FAP), familial amyloid cardiomyopathy (FAC), senile
systemic amyloidosis (SSA), and central nervous system
selective amyloidoses (CNSA).12,13 This concept comes from
the observation that thyroxine (T4), the endogenous ligand of
transthyretin, is able to inhibit TTR aggregation by stabilizing
the tetrameric structure of the protein.4,14−18 Many different
families of compounds are known to stabilize TTR effectively,
preventing its aggregation in vitro.19−24 Within this family of
compounds, only a benzoxazole derivative (Tafamidis)25−28has
reached approval for clinical use in Europe, while a few others
(diflunisal, doxycycline/TUDCA) are still under clinical
testing.29−34Other potential therapeutic strategies are also
being explored but are less developed, including immunother-
apy,35,36 gene therapy with small interfering RNAs, antisense
oligonucleotides, and single-stranded oligonucleotides.37−39
TTR is a 55 kDa homotetrameric protein containing 127
residues per monomer that are organized in eight β strands,
named from A to H and connected by loops and a short α helix
Received: July 15, 2013
Published: October 23, 2013
© 2013 American Chemical Society
9110
dx.doi.org/10.1021/jm401061w | J. Med. Chem. 2013, 56, 9110−9121

Journal of Medicinal Chemistry
Article
Figure 1. (a) β-Sheet topology of each TTR monomer. (b) Tetrameric structure of TTR (two views at 90° rotation on the y-axis) showing the two
symmetric T4 binding sites.
which is iododiflunisal (13), a drug candidate developed by
us.56
In Vitro Inhibition Activity. To obtain homogeneous
inhibition data to conduct the SAR studies of interest, all of
these 13 compounds were tested in a turbidimetric in vitro
assay to evaluate their potential activity as TTR fibril
inhibitors.57,58 The protocol uses a highly amyloidogenic
TTR variant (Y78F) which enables kinetic monitoring of
protein aggregation in short time under acid-induced fibrillo-
genesis conditions. One of the parameters that can be assessed
by this method is the IC50 value, which is the concentration of
inhibitor at which the initial rate of fibril formation is half than
in the absence of inhibitor. A second parameter is the
percentage of reduction in fibril formation rate (RA%) at a
high concentration of test compound relative to the rate
obtained in the absence of the compound. RA values of 100%
indicate that the inhibitor is able to fully prevent the formation
of fibrils. Both values are reported in Table 1, which shows that
the polychlorinated biphenyl 9 followed by 13, 7, 2, 4, and 6
are the best inhibitors. Conversely, 3 and 5 exhibited the
poorest activity.
Structural Analysis of TTR−Ligand Interactions. After
processing the 3D structure of these 13 TTR crystallographic
Figure 2. Schematic representation of the three chemical regions in
complexes (see Supporting Information Comment 1), one of
which TTR binding sites can be divided, showing also the halogen
the first observations was that the amino acid side chains
binding pockets (HBPs) location.
protruding into the two chemically equivalent binding sites of
TTR show great flexibility. Thus, we have found different
residues that are able to adopt various rotameric conformations
ligands to see if new sites of molecular recognition mediated by
halogen bonding could be created.47−55
within the same crystal structure. This is the case of the TTR−
(5) complex, where two different conformers for residues
Leu17, Ser117, and Thr119 exist. We have also found evidence
for ligand-dependent conformational arrangements upon bind-
ing for the side chains of Lys15, Glu54, Thr106, Ser115,
Ser117, and Thr119. Among them, Lys 15 is the residue which
exhibits the highest conformational variability, which is driven
to maximize in each case hydrophobic interactions between the
methylene groups of the lysine and the aromatic or hydro-
phobic moieties of the ligand placed at the entry of the cavity.
Furthermore, it allows the formation of hydrogen bonds
between the ε-NH2 and any other hydrophilic group on the
ligand (see Supporting Information Comment 2).
RESULTS AND DISCUSSION
■
Ligand Selection. The most noticeable feature of the TTR
crystallographic complexes deposited at the Protein Data Bank
is the wide diversity of chemical structures (more than 40
different scaffolds) that binds TTR. From these diverse
structures, we selected 13 (Figure 3), including the endogenous
ligand of TTR, thyroxine (1) and a close derivative (2),
therapeutic drugs such as flurbiprofen (3), flufenamic acid (4),
and diclofenac (5), natural products like retinoic acid (6) and
resveratrol (7), and six synthetic compounds (8−13), one of
9111
dx.doi.org/10.1021/jm401061w | J. Med. Chem. 2013, 56, 9110−9121

Journal of Medicinal Chemistry
Article
Figure 3. Selected ligands from TTR−ligand complexes available at the PDB: (1) thyroxine or T4; ( 2) 3,3′,5,5′-tetraiodothyroacetic acid or T4Ac;
(3) flurbiprofen; (4) flufenamic acid; (5) diclofenac; (6) retinoic acid; (7) resveratrol; (8) diethylstilbestrol; (9) 3,5,3′,5′-tetrachlorobiphenyl-4,4′-
diol; (10) 2,4-dinitrophenol; (11) 3′,5′-difluorobiphenyl-4-carboxylic acid; (12) 2′,4′-dichloro-4-hydroxy-1,1′-biphenyl-3-carboxylic acid, and (13)
iododiflunisal. (See Figure S0 in Supporting Information for more details on the related PDB structures.)
Table 1. Inhibition of Acid-Induced Fibrillogenesis by Ligands (IC50 and RA%) Measured by the Kinetic Turbidimetric Assay
a
and Structural Properties of the TTR−Ligand Crystallographic Complexes
X-ray features of TTR−ligand complexes
fibrillogenesis
inhibition
occupation of HBP pockets by the ligands
binding
mode
ligand
IC50 (μM) RA (%)
PDB
HBP1
−I
−I
−I
HBP1′
HBP2
HBP2′
HBP3
HBP3′
1
2
10.5
4.8
80
93
2ROX
1Z7J
F
F
R
R
F
R
F
F
F
R
R
R
R
F
−I
−I
−I
−I
−I
−I
−I
−I
−I
−CH−
−CH−
−CH−
−CF3
−COOH
−CH3
−CH−
−CH−
−Cl
−CH−
−CH−
−COOH
−F
3
>100
6.2
26
35
89
1DVT
1BM7
1DVX
1TYR
1DVS
1TT6
2G5U
2B15
−F
4
5
74
−CH−
−CH−
−CH3
−CH−
−CH−
−Cl
−NO2
−CH−
−CH−
−F
6
8.1
4.5
100
89
−CH2−
−OH
−CH−
−Cl
7
−OH
−CH−
−Cl
8
16.6
2.1
67
−CH2−CH3
−CH2−CH3
9
100
77
10
11
12
13
20
10.8
12.8
4.5
89
2B9A
2B77
75
94
1Y1D
−OH
−I
a
See Supporting Information Comment 1 for more details on the PDB structures.
Upon ligand binding, other important conformational and
groups (such as compounds 5 or 12). The conformation
adopted by Ser117 and Thr119 facilitates the binding of the
ligands in the so-called forward (F) and reverse (R) binding
modes. In the forward mode, the hydrophilic group of the
ligand is close to the binding site entrance, favoring electrostatic
interactions between the hydrophilic groups located at the
entry of the cavity and the residues Lys15 and/or Glu54 of one
structural rearrangements suffered by TTR also involve the side
chains of Ser117 and Thr119 (see Supporting Information
Comment 3), modifying the surface polarity of the HBP3/3′
pockets. Consequently, these pockets are able to accommodate
a wide variety of chemical groups (Table 1), varying from
hydrophobic (such as compounds 4 or 6) to negatively charged
9112
dx.doi.org/10.1021/jm401061w | J. Med. Chem. 2013, 56, 9110−9121

Journal of Medicinal Chemistry
Article
or both A and A′ monomers (Table 1). In contrast, other
ligands such as 5, 11, or 12 bind to TTR in the reverse mode,
favoring the formation of a hydrogen bond between the
hydrophilic groups positioned near the tetramer center and the
residues Ser117 and/or Thr119 of one or both monomers
(Table 1). However, no relationship could be found between
the binding mode of the studied ligands and their activity (see
Table 1 and Supporting Information Comment 4).
XBs with two neighboring subunits. Both interaction geo-
metries between the iodine atoms and the backbone carbonyls
of Leu109 residues (equivalent to human Ala 109) are close to
optimal [I...O distances: 3.29 and 3.35 Å; σ-hole angles: 162.3
and 169.7°]. These XBs are also orthogonal with hydrogen
bonds between the backbone carbonyls of the same Leu109
and the backbone amide NH of Leu17 [αX...O...H(N) angles of
88 and 90°].51
Here, we are further addressing experimentally the halogen
bonding issue on TTR−ligand complex formation with the aim
to propose and evaluate a predictive modeling approach for the
optimization of TTR ligand affinity by XB formation. Thus, we
have carried out the synthesis and biochemical study of two
homologous series of halogenated derivatives of flufenamic acid
and diflunisal, two NSAIDs which are well-known TTR
ligands.68,69
One consequence of the discussion above is that ligands are
always distributed along the binding sites adopting multiple
conformations and even showing different orientations, which
makes it very hard to find equivalent functional groups in the
same region of the binding site between the studied complexes.
This, in turn, precludes the identification of the complementary
structural and chemical features on the ligands that may allow
formulation of a hypothetic pharmacophoric site on the
protein. In addition, the lack of a clear correlation between
the ligand structure and its biological activity makes QSAR
studies very difficult (Supporting Information Comment 5).
Role of TTR Halogen Binding Pockets. A remarkable and
unique feature of TTR is the presence of three paired halogen
binding pockets (HBPs) on the binding site. Although these
pockets were initially described for their ability to accom-
modate halogen atoms, our modeling analysis has revealed that
similar to HBP3 and HBP3′, HBP1/1′ and HBP2/2′ can also
accommodate a wide variety of chemical groups due to the
conformational flexibility of their constitutive residues (Table
1).
From previous studies,70 we have experimental evidence that
iodination at the C-4 position of flufenamic acid completely
abolishes the TTR fibrillogenesis inhibition of the parent
compound 4. Similarly, addition of an iodine atom at position
4′ of flurbiprofen (compound 3) did not improve its potency.
However, iodination of diflunisal at the 5 position (see below)
significantly increased the potency of this compound. Taken
together, these results suggested that the incorporation of an
iodine (halogen) atom does not always improve the potency of
a compound. In other words, not any position of the ligand
seems favorable for halogenation.
In this scenario, in order to explore in a predictive fashion the
use of halogen bonding interactions for the optimization of the
fibrillogenesis inhibition potency of this ligand, the most
energetically favored halogenation position that induces the
best fit of the halogen on the HBPs was calculated. As discussed
above, it was assumed that HBPs are the most suitable places
evolved by nature on TTR to interact with halogen atoms.
Accordingly, the GRID software71 was used to predict which
positions of flufenamic acid are the most favorable for
halogenation. From these contour maps, it was shown that
the most energetically favored position of (4) for halogenation
is the C-5′ position, and that the interaction energy of the
halogen located at this position and the protein decreases from
iodine to fluorine. Interestingly, all these ligands are
characterized by presenting the 3′- and 5′-halogenated groups
into the HBP3 and HBP3′ pockets, respectively.
The number of occupied HBPs in the study complexes varies
from 2 to 4 with the exception of (2) bound in reverse mode
and (8), which have 5 and 6, respectively. However, there is no
relationship between the number of HBP pockets occupied by
the ligand and its activity, although the most active compounds
are characterized by the presence of a halogen atom in the
HBP1/1′ and/or HBP3/3′ (Table 1).
Experimental Examination for Halogen Bonding
Interactions. Halogen bonding interactions (XB)59 are
widespread in biological systems,60 but only recently they
have raised considerable interest in medicinal chemistry.47−55
Of particular interest to the present work are the attempts to
use XB as a design feature to stabilize protein structures.61,62
This type of interaction occurs when a halogen forms direct
close contacts of the type R−X···Y, where the halogen X acts as
a Lewis acid and Y can be any kind of electron donor (Lewis
base).63,64 It is driven by the σ-hole, a positively charged region
on the hind side of X along the R−X bond axis caused by
anisotropy of electron density on the halogen.51 The possibility
that XB formation may play a role in stabilizing TTR−ligand
tetrameric complexes has been previously examined in the
literature by analyzing the crystallographic complexes between
TTR and halogenated ligands: 3,5-diiodosalicylic acid (PDB
entry 3B56),65 two brominated stilbenes (PDB entries 3IMU
and 3IMV),66 and diclofenac (PDB entry 1DVX).6 It was found
that one iodine atom of the diiodosalicylic acid binds to the
backbone carbonyl group of Ser117 [I(1)...O distance: 3.50 Å;
C−I(1)...O angle: 148.3°]. Similarly, the bromine atoms of the
stilbenes bind to the same Ser117 carbonyl [Br...O distances:
3.33 and 3.39 Å; C−Br...O angles: 145.9 and 143.5°]. In the
case of diclofenac, there is also the possibility that the Cl(4)
atom may interact with the hydroxyl oxygen atom of Thr119
forming an XB.47 Also, the crystal structure of the piscine TTR
complex with thyroxine (PDB-ID: 1SN0)67 has been examined
for XB. Thyroxine (1) binds to TTR, forming two equivalent
The homologous series of halogenated derivatives at C-5′
(18−21) (Figure 4) was synthesized and, along with the parent
Figure 4. Halogenated flufenamic acid (18−21) and diflunisal (15−17
and 13) analogues in this study.
compound (4), the compounds were tested in the kinetic
turbidity assay described above (see Chemistry section in
Supporting Information). Their fibrillogenesis inhibition
properties expressed by the parameters IC50 and RA (%) are
reported in Table 2. Moreover, binding of the ligands to the
Y78F-TTR variant was determined by isothermal titration
calorimetry (ITC), which provides the thermodynamic
parameters Kd, ΔH, and TΔS of binding under conditions
where no protein aggregation occurs (pH = 7). Although the
9113
dx.doi.org/10.1021/jm401061w | J. Med. Chem. 2013, 56, 9110−9121

Journal of Medicinal Chemistry
Article
their effective van der Waals radii for phenyl substitution77 (H:
1.20 Å; F: 1.47 Å; Cl: 1.77 Å; Br: 1.92 Å; I: 2.06 Å). The
stabilization provided by halogenation reaches more than 1
order of magnitude in terms of the affinity constant for
iododiflunisal (13), which translates to a gain in free energy of
− ΔΔG = 1.5 kcal mol−1. Indeed, taking also into account the
crystallographic information available from the complex of (13)
and TTR (PDB 1Y1D),56 these results could be interpreted by
a new protein−ligand interaction of XB type arising after the H
for X exchange. As shown in Figure 5, in this complex the
Table 2. Inhibition of Acid-Induced TTR Fibrillogenesis
(Turbidity Assay) and Ligand Binding (Isothermal Titration
Calorimetry) of Halogenated Flufenamic Acid and Diflunisal
a
Derivatives to TTR Y78F Variant
inhibition
(turbidity
assay)
binding (isothermal titration calorimetry)
Kd ΔH TΔS ΔG
IC50 RA
No.
X
(μM) (%) (nM) (kcal/mol) (kcal/mol) (kcal/mol)
flufenamic acid
derivatives
4
H
F
6.2
6.5
6.0
5.9
6.4
89
91
86
91
94
109
91
86
64
85
−10.7
−17.7
−14.8
−12.9
−10.1
−1.2
−8.1
−4.0
−3.1
−0.47
−9.5
−9.6
−10.8
−9.8
18
19
20
21
Cl
Br
I
−9.6
diflunisal
derivatives
14
15
16
17
13
H
F
18
85
87
89
88
94
1160
1120
375
194
87
−8.16
−8.45
−5.93
−8.65
−7.52
−0.05
−0.32
2.8
−8.1
−8.1
−8.8
−9.2
−9.66
7.6
7.5
5.6
4.5
Cl
Br
I
0.51
2.1
a
Turbidity assay: 0.4 mg/mL TTR-Y78F, 0−40 μM inhibitor, pH 4.2,
37 °C. Isothermal titration calorimetry: 20−30 μM TTR-Y78F, 0−120
μM ligand, pH 7.6, 25 °C.
two symmetrical binding sites of TTR show negative
cooperativity for T4 binding,72,73 a simple binding model
with two identical sites gave the best fitting of the calorimetric
data (Table 2).
Figure 5. TTR−iododiflunisal (13) complex. Residues Thr106,
Ala108, Thr119, and Val121 shape the halogen binding pocket
establishing van der Waals contacts with the iodine atom. Also, a
suboptimal XB interaction with the carbonyl group of Thr106 is
shown.
As seen in Table 2, flufenamic acid (4) is a good TTR
fibrillogenesis inhibitor for which halogenation at C-5′ does not
significantly increase the inhibition potency of the parental
compound. Likewise, the ligand binding affinity data (Kd) of
the compounds between 100 and 60 nM loosely correlate with
the van der Waals radii of the halogen atoms and with the
increasing affinity predicted by the GRID calculations,
suggesting that no stabilizing halogen bonding interactions
seem to be at play in this series of halogenated compounds.
On the other hand, previous studies74−76 have experimentally
evidenced that iodination at the 5 position of diflunisal led to a
much efficient TTR fibrillogenesis inhibitor, namely, iododi-
flunisal (13) (Table 1). To seek if this enhanced potency could
be caused by an XB that arises by the presence of the iodine
atom, we have synthesized the three remaining compounds of
the homologous series of halogen-substituted diflunisal
derivatives having F, Cl, and Br at the 5 position. The synthesis
of the series (15, 16, 17) was effected by conventional aromatic
electrophilic halogenation methods (Figure 4 and Supporting
Information Chemistry section).
The whole series of halogenated diflunisal derivatives and the
parent compound (13−17) were tested for inhibitory activity
and affinity toward Y78F-TTR (Table 2). The inhibition values
of this series clearly indicate that the introduction of a halogen
atom at C-5 of diflunisal always improves its potency. This
improvement almost reaches the lowest measurable values of
IC50 because the assay is run at 7.2 μM concentration of TTR,
the concentration at which the corresponding amount of ligand
is required to semisaturate the protein. More interesting is the
fact that gradual and large differences in the binding constants
(Kd) are observed along the series. Such decreases of Kd
correlate with increasing halogen volume as measured by
iodine atom of (13) occupies the HBP1 pocket of TTR
through a series of contacts with residues Thr106, Ala108,
Thr119, and Val121. The interaction of the iodine with the
carbonyl group of Thr106 takes place through a bonding
geometry I...O distance of 3.96−4.05 Å and σ-angle of 152−
156° that, following the criteria established by Boeckler,78 can
be regarded as a suboptimal XB. Using the model data provided
by these authors, the stabilization energy contributed by this
XB may not be more than 50% of the one with an optimal XB
geometry I...O distance of 3 Å and σ-angle of 180°.
Nevertheless, other explanations such as optimized van der
Waals interactions generated by the increasing volume of the
halogen atom in the series which progressively fills the halogen
binding pocket and results in better complex stabilization
cannot be ruled out. Moreover, the possible effect of an
increasing acidity of the phenolic group of (13) due to the
gradual electron-withdrawing effect of the halogen atoms seems
not to be at play (see Figure S7 in Supporting Information).
In summary, both inhibitors, flufenamic acid and diflunisal,
show different effects upon halogenation. Flufenamic acid is a
good inhibitor with a Kd of 109 nM, and halogenation does not
significantly improve binding, meaning that no XB interaction
occurs or, if any, its contribution is very small. However,
diflunisal is a moderate inhibitor and has a high Kd of 1160 nM,
1 order of magnitude higher than flufenamic acid. In this case,
halogenation improves binding significantly through inter-
actions that can be interpreted as suboptimal XB or optimized
van der Waals contacts or simultaneous contributions of both,
9114
dx.doi.org/10.1021/jm401061w | J. Med. Chem. 2013, 56, 9110−9121

Journal of Medicinal Chemistry
Article
resulting in an 1 order of magnitude lower Kd (for
iododiflunisal), at values similar to those obtained for the
flufenamic acid series.
General Procedure A: Pd(0)-Catalyzed Reaction
between 3-Halo-5-(trifluoromethyl)anilines and Methyl
2-bromobenzoate. A round-bottom flask was charged with
Cs2CO3 (1.20 mmol), rac-BINAP (0.05 mmol), and Pd2(dba)3
(0.025 mmol). The mixture was gently purged with argon,
suspended in anhydrous toluene (8 mL), and refluxed for 5
min. Then, a solution of 3-halo-5-(trifluoromethyl)aniline (1.00
mmol) and methyl 2-bromobenzoate (1.10 mmol) in
anhydrous toluene (2 mL) was added dropwise. The reaction
mixture was refluxed overnight and then diluted with 50 mL of
ethyl acetate, filtered over Celite, and evaporated to dryness.
The residue was then purified by silica gel column
chromatography (hexane/ethyl acetate 85:15).
General Procedure B: Saponification of Methyl Esters.
To a stirred solution of the corresponding anthranilic methyl
ester (1.00 mmol) in 1,4-dioxane (5 mL), was added a solution
of LiOH (5.00 mmol) in water (1 mL). The mixture was then
heated at 80 °C for 4 h, acidified with 0.25 M HCl (20 mL),
and diluted with ethyl acetate (50 mL). The phases were
separated, and the aqueous phase was back extracted with more
ethyl acetate (25 mL). The organic layer was separated and
washed with brine, dried over anhydrous Na2SO4, and
concentrated. The residue was purified by silica gel column
chromatography (hexane/ethyl acetate 1:1) to afford the pure
acid.
Methyl N-(3-fluoro-5-(trifluoromethyl)phenyl)-
anthranilate (18a) General Procedure A. The title
compound was produced from 3-fluoro-5-(trifluoromethyl)-
aniline (361 mg, 2.01 mmol) and methyl 2-bromobenzoate
(337 μL, 2.4 mmol). Silica gel column chromatography
afforded 18a (534 mg, 85%) as a white solid. RP-HPLC-
(GEN1): RT = 23.85 min. 1H NMR (500 MHz, CD3COCD3):
3.91 (s, 3H), 6.98 (ddd, J = 16.5, 11.0, 3.5, 1H), 7.04−7.12 (m,
1H), 7.34 (dt, J = 18.0, 3.5, 1H), 7.41 (m, 1H), 7.45−7.47 (m,
2H), 8.00 (ddd, J = 13.5, 2.5, 1.1, 1H), 9.63 (br s, 1H). 13C
NMR (75 MHz, CD3COCD3): 53.3, 107.1 (dq, J = 25.1, 3.5),
111.1 (d, J = 24.8), 114.0 (dq, J = 3.3, 3.3), 116.5, 117.7, 121.6,
125.1 (qd, J = 267.4, 5.2), 133.4, 133.8 (qd, J = 34.2, 12.3),
136.0, 146.5 (d, J = 10.7), 146.5, 165.1 (d, J = 244.5), 169.8.
HRMS (ESI+): calcd for [C15H11F4NO2 + H]+, 314.0799;
found, 314.0806.
CONCLUSION
■
In conclusion, experimental evidence is provided that
halogenation of TTR ligands may improve their affinity and
fibrillogenesis inhibition properties. In the case of diflunisal,
these may take place through new although suboptimal XB
interactions and/or by optimized van der Waals contacts.
Regardless of their nature, such interactions may serve to
optimize the tetramer stabilization properties of TTR ligands.
Prediction of the optimal halogenation positions on the ligand
is the crucial step and may be guided by directing the desired
halogen interactions toward the HBPs of TTR.
EXPERIMENTAL SECTION
■
Chemistry. TLC analyses were carried out on Merck
Kieselgel 60 F254 silica gel plates. Visualization was by UV light
(254 nm). Chromatographic purification was carried out on
silica gel columns 60 (70−200 μm, 70−230 mesh) from SDS.
Solvents for synthesis purposes were used at GPR grade.
Reverse-phase HPLC analyses were performed on a Merck-
Hitachi (D-6000) HPLC system with an UV L-4000 detector
(λ = 214 nm), a L-6200 pump, and an AS-2000 automatic
injector, using a Merck LiChroCART 250-4 LiChrospher 100-5
RP-C18 column (250 mm × 4.6 mm) with injection volume at
10 μL and sample concentrations at 1−2 mg/0.5 mL in 50:50
(v/v) acetonitrile/water; the sample was detected at single
wavelength of 214 nm with a mobile phase system composed of
a mixture of acetonitrile/water, each containing 0.1% of TFA at
1 mL/min flow. The column was maintained at room
temperature. The gradient program (GEN1) used was as
follows: from an initial (50:50) mixture of A/B to a (10:90)
mixture of A/B in 25 min, then back to (50:50) in 15 min. All
compounds reported here exhibited spectral data consistent
with their proposed structures and had HPLC purities in excess
of 95%. Chemical names were generated using ChemBioDraw
1
version by CambridgeSoft. H and 13C NMR spectra were
recorded at 400 MHz and 100.62 MHz, respectively, in CDCl3
or (CD3)2SO on a Varian Mercury-400 spectrometer. Chemical
shifts are reported in parts per million (ppm) relative to the
residual signal of the deuterated solvent (δ 7.27 for 1H; δ 77.23
for 13C) or deuterated or 13C labeled dimethyl sulfoxide (δ 2.50
for 1H, δ 39.53 for 13C) as internal standards. 1H NMR data are
tabulated in the following order: multiplicity as follows (s,
singlet; d, doublet; t, triplet; m, multiplet; br, broad), coupling
constants (J) are reported in Hertz (Hz), and the number of
protons. HRMS spectra (UPLC-TOF/MS) were recorded on a
Waters ACQUITY UPLC System with a Waters LCT Premier
XE mass spectrometer operating either in the positive ion
electrospray mode or in negative electrospray mode. Water and
acetonitrile were used as carrier solvents.
N-(3-Fluoro-5-(trifluoromethyl)phenyl)anthranilic
acid (18). The title compound was produced by general
procedure B from 18a (200 mg, 0.64 mmol). Silica gel column
chromatography afforded 18 (190 mg, 89%) as a white solid.
1
RP-HPLC (GEN1): RT = 14.93 min. H NMR (500 MHz,
DMSO-d6): 6.99 (ddd, J = 9.0, 7.0, 1.0, 1H), 7.13 (dt, J = 8.5,
1.5, 1H), 7.30−7.38 (m, 2H), 7.40 (dd, J = 8.0, 1.0, 1H), 7.50
(ddd, J = 10.0, 6.5, 1.5,1H), 7.94 (dd, J = 8.0, 2.0, 1H), 9.65 (s,
1H). 13C NMR (100 MHz, CDCl3): 104.8 (dq, J = 25.5, 3.8),
108.8 (d, J = 24.7), 111.4, 116.3, 116.7, 120.2, 123.3 (qd, J =
267.1, 3.4), 131.7 (qd, J = 34.2, 12.9), 131.8, 134.0, 143.9,
144.8, (d, J = 11.1), 162.8 (d, J = 243.4), 169.1. HRMS (ESI−):
calcd for [C14H9F4NO2 − H]−, 298.0497; found, 298.0482.
Methyl N-(3-chloro-5-(trifluoromethyl)phenyl)-
anthranilate (19a) General Procedure A. The title
compound was produced from 3-chloro-5-(trifluoromethyl)-
aniline (215 mg, 1.10 mmol) and methyl 2-bromobenzoate
(143 μL, 1.00 mmol). Silica gel column chromatography
afforded 19a (106 mg, 32%) as an oil. Colorless oil. Rf (hexane/
Compound Characterization. All compounds were
obtained in milligram quantities. Details of the synthesis and
characterization of each compounds is described below.
Selected spectra are included in the Supporting Information.
1
Structures of all reported compounds were confirmed by H
and 13C NMR and their mass confirmed by HRMS. Purity was
determined by HPLC analysis. All reported compounds had
>95% purity.
1
EtOAc 70:30) = 0.65. H NMR (400 MHz, CDCl3): 3.92 (s,
3H), 6.89 (ddd, J = 8.1, 7.2, 1.2, 1H), 7.24 (s, 1H), 7.31−7.36
(m, 2H), 7.38−7.45 (m, 2H), 8.01 (dd, J = 8.0, 1.6, 1H), 9.65
9115
dx.doi.org/10.1021/jm401061w | J. Med. Chem. 2013, 56, 9110−9121

Journal of Medicinal Chemistry
Article
(br s, 1H). 13C NMR (100 MHz, CDCl3): 52.1, 113.9, 115.7,
115.7 (quad, J = 3.8), 119.2 (quad, J = 3.9), 119.3, 123.1 (quad,
J = 1.0), 123.2 (quad, J = 272.9), 131.9, 133.0 (quad, J = 32.9),
134.3, 135.7, 143.2, 145.5, 168.7. HRMS (ESI+): calcd for
[C15H11ClF3NO2 + H]+, 330.0504, 332.0527; found, 330.0528,
332.0507.
8.0, 7.0, 1.3, 1H), 7.32−7.44 (m, 2H), 7.46 (s, 1H), 7.53 (s,
1H), 7.57 (s, 1H), 8.01 (dd, J = 8.1, 1.6, 1H), 9.71 (br s, 1H).
Methyl N-(3-iodo-5-(trifluoromethyl)phenyl)-
anthranilate (21b). Compound 21a (73 mg, 0.126 mmol)
was dissolved in acetonitrile, and NaI (21 mg, 0.140 mmol) and
N-chlorosuccinimide (17 mg, 0.140 mmol) were added. The
mixture was stirred for 30 min at rt and diluted with ethyl
acetate (20 mL), water (10 mL), and 5% sodium thiosulfate
solution (3 mL). Phases were separated, and the aqueous phase
was extracted again with ethyl acetate (10 mL). The organic
layer was dried over Na2SO4 anhydrous, filtered, and
evaporated to dryness. Purification by silica gel column
chromatography (hexane/ethyl acetate (85:15)) afforded 21b
(42 mg, 79%) as an oil. Yellowish wax. Rf (hexane/EtOAc
85:15) = 0.40. 1H NMR (400 MHz, CDCl3): 3.92 (s, 3H), 6.88
(ddd, J = 8.2, 7.1, 1.1, 1H), 7.30 (dd, J = 8.5, 1.1, 1H), 7.39−
7.45 (m, 2H), 7.58 (s, 1H), 7.75 (s, 1H), 8.02 (dd, J = 8.0, 1.7,
1H), 9.61 (br s, 1H). HRMS (ESI+): calcd for [C15H11F3INO2
+ H]+, 421.9859; found, 421.9873.
N-(3-Chloro-5-(trifluoromethyl)phenyl)anthranilic
acid (19) General Procedure B. The title compound was
produced from 19a (65 mg, 0.197 mmol). Silica gel column
chromatography afforded 19 (41 mg, 66%) as a white solid. RP-
HPLC(GEN1): RT = 17.49 min. Rf (hexane/EtOAc 50:50) =
1
0.16. H NMR (400 MHz, DMSO-d6): 6.97 (ddd, J = 8.1, 7.2,
1.1, 1H), 7.31 (m, 1H), 7.36 (dd, J = 8.4, 1.1, 1H), 7.45−7.48
(m, 1H), 7.50 (ddd, J = 7.2, 7.0, 1.6, 1H), 7.51−7.53 (m, 1H),
7.92 (dd, J = 7.9, 1.6, 1H), 9.58 (s, 1H), 13.21 (br s, 1H). 13C
NMR (100 MHz, CDCl3): 113.9 (quad, J = 3.7), 116.6, 116.9,
117.4 (quad, J = 4.0), 120.5, 121.6, 123.3 (quad, J = 273.0),
131.5 (quad, J = 32.5), 132.0, 134.2, 134.9, 143.9, 144.4, 169.2.
HRMS (ESI−): calcd for [C14H9ClF3NO2 − H]−, 314.0201,
316.0171; found, 314.0220, 316.0161.
Methyl N-(3-bromo-5-(trifluoromethyl)phenyl)-
anthranilate (20a) General Procedure A. The title
compound was produced from 3-bromo-5-(trifluoromethyl)-
aniline (240 mg, 1.000 mmol) and methyl 2-bromobenzoate
(351 μL, 2.50 mmol). Silica gel column chromatography
afforded 20a (106 mg, 32%) as a yellowish wax. RP-HPLC
(GEN1): RT = 25.17 min. Rf (hexane/EtOAc 85:15) = 0.40.
1H NMR (400 MHz, CDCl3): 3.92 (s, 3H), 6.89 (ddd, J = 8.1,
7.1, 1.2, 1H), 7.32 (dd, J = 8.5, 1.1, 1H), 7.39 (dd, J = 1.8, 0.7,
1H), 7.43 (dddd, J = 8.4, 7.2, 1.7, 0.5, 1H), 7.55 (tt, J = 1.9, 0.5,
1H), 8.01 (ddd, J = 8.0, 1.7, 0.4, 1H), 9.64 (s, 1H). 13C NMR
(100 MHz, CDCl3): 52.1, 113.8, 114.8, 115.8 (quad, J = 3.8),
119.3 (quad, J = 1.0), 121.9 (quad, J = 3.8), 123.2 (quad, J =
272.9), 123.3, 126.0, 133.1 (quad, J = 32.9), 131.9, 134.3, 143.2,
145.5, 168.7. HRMS (ESI+): calcd for [C15H11BrF3NO2 + H]+,
373.9999, 375.9978; found, 374.0025, 375.9985.
N-(3-Iodo-5-(trifluoromethyl)phenyl)anthranilic acid
(21) General Procedure B. The title compound was
produced from 21b (42 mg, 0.100 mmol). Silica gel column
chromatography afforded 21 (17 mg, 41%) as a white solid. Rf
(hexane/EtOAc 70:30) = 0.15. RP-HPLC (GEN1): RT =
1
19.27 min. H NMR (400 MHz, CDCl3): 6.89 (ddd, J = 8.1,
7.1, 1.2, 1H), 7.24−7.27 (m, 1H), 7.35−7.41 (m, 2H), 7.51 (s,
1H), 7.70 (s, 1H), 8.01 (dd, J = 8.0, 1.6, 1H), 9.68 (br s, 1H).:
1H NMR (400 MHz, DMSO-d6): 6.98 (dt, J = 7.4, 1.1, 1H),
7.32 (dd, J = 12.9, 1.0, 1H), 7.45−7.48 (m, 1H), 7.48 (dd, J =
3.6, 1.7, 1H), 7.52−7.55 (m, 1H), 7.77−7.80 (m, 1H), 7.94 (t, J
= 2.0, 1H), 9.67 (br s, 1H). 13C NMR (100 MHz, DMSO-d6):
95.9, 114.7 (quad, J = 4.1), 116.4, 120.4, 123.0 (quad, J =
272.9), 125.8 (quad, J = 3.7), 130.6, 131.8 (quad, J = 32.1),
131.9, 134.0, 143.9, 144.0, 169.2. HRMS (ESI−): calcd for
[C14H9F3INO2 − H]−, 405.9557; found, 405.9558.
3-Fluoro-5-iodosalicylic acid (15b). To a stirred solution
of 3-fluorosalicylic acid (15a) (158 mg, 1.000 mmol) in CH2Cl2
(15 mL) was added, in small portions, IPy2BF4 (409 mg, 1.100
mmol). The mixture was stirred at rt for 2 h and then diluted
with ethyl acetate (50 mL), washed with 0.5 M HCl (2 × 10
mL), with brine (15 mL) and dried over anhydrous Na2SO4,
filtered, and evaporated to dryness. The residue was further
purified by silica gel column chromatography (CH2Cl2/
MeOH/AcOH (90:10:0.5)) to afford pure 15b (264 mg,
94%) as a white solid. Rf (CH2Cl2/MeOH/AcOH 90:10:0.5) =
0.13. 1H NMR (400 MHz, DMSO-d6) δ (ppm): 7.78−7.82 (m,
3H). 13C NMR (100 MHz, DMSO-d6) δ (ppm): 78.8 (d, J =
7.0), 118.0 (d, J = 3.4), 129.2 (d, J = 19.7), 133.8 (d, J = 3.7),
149.2 (d, J = 13.0), 151.0 (d, J = 249.3), 169.8 (d, J = 3.4).
HRMS (ESI−): calcd for [C7H4FIO3 − H]−, 280.9116; found,
280.9119.
N-(3-Bromo-5-(trifluoromethyl)phenyl)anthranilic
acid (20) General Procedure B. The title compound was
produced from 20a (100 mg, 0.27 mmol). Silica gel column
chromatography afforded 20 (83 mg, 86%) as a white solid. RP-
HPLC(GEN1): RT = 18.23 min. Rf (hexane/EtOAc 70:30) =
1
0.15. H NMR (400 MHz, DMSO-d6): 6.97 (ddd, J = 7.4, 7.3,
1.1, 1H), 7.30 (dd, J = 13.0, 1.1, 1H), 7.39−7.42 (m, 1H), 7.44
(dd, J = 3.6, 1.7, 1H), 7.45−7.48 (m, 1H), 7.60−7.63 (m, 1H),
7.92 (t, J = 2.0, 1H), 9.70 (br s, 1H). 13C NMR (100 MHz,
DMSO-d6): 114.1 (quad, J = 4.3), 116.7, 119.9 (quad, J = 3.8),
120.2, 120.3, 122.9, 123.0 (quad, J = 267.7), 124.4, 131.8 (quad,
J = 32.0), 131.9, 133.8, 143.9, 144.4, 169.2. HRMS (ESI−):
calcd for [C14H9BrF3NO2 − H]−, 357.9696, 359.9675; found,
357.9671, 359.9658.
Methyl N-(3-(tributylstannyl)-5-(trifluoromethyl)-
phenyl)anthranilate (21a). A round-bottom flask was
charged with LiCl (31 mg, 0.720 mmol), Pd(PPh3)4 (50 mg,
0.043 mmol), and 20a (54 mg, 0.144 mmol). The flask was
purged with argon, and anhydrous 1,4-dioxane (1.5 mL) was
added. Then, (SnBu3)2 (144 μL, 0.288 mmol) was added, and
the solution was refluxed for 4 h. The crude reaction mixture
was adsorbed on silica gel and purified by silica gel column
chromatography (hexane/ethyl acetate 7:3) to afford 21a (40
mg, 50%) as an oil. Colorless oil. Rf (hexane/EtOAc 70:30) =
0.51. 1H NMR (400 MHz, CDCl3): 0.92 (t, J = 7.3, 9H), 1.22−
1.42 (m, 12H), 1.59−1.70 (m, 6H), 3.93 (s, 3H), 6.86 (ddd, J =
2′,4′,5-Trifluoro-4-hydroxy-[1,1′-biphenyl]-3-carbox-
ylic acid (15). Under Ar atmosphere, degassed water (2.5 mL)
was added over a mixture of (2,4-difluorophenyl)boronic acid
(45 mg, 0.285 mmol), sodium carbonate (82 mg, 0.774 mmol),
Pd(OAc)2 (0.6 mg, 1 mol %), and compound 15b (73 mg,
0.258 mmol). The suspension was stirred at rt for 3 h and then
diluted with ethyl acetate (50 mL) and water (10 mL). The pH
was adjusted to 2−3 with 0.5 M HCl, and the phases were
separated. The aqueous layer was back-extracted with more
ethyl acetate (2 × 10 mL), and the combined organic phases
were dried over Na2SO4 anhydrous, filtered, and evaporated to
dryness. The residue was purified by silica gel column
9116
dx.doi.org/10.1021/jm401061w | J. Med. Chem. 2013, 56, 9110−9121

Journal of Medicinal Chemistry
Article
chromatography (CH2Cl2/ethyl acetate/AcOH 50:50:0.5) to
afford pure 15 (71 mg, 100%). White solid. Rf (CH2Cl2/MeOH
85:15) = 0.13. RP-HPLC (GEN1): RT = 10.27 min. 1H NMR
(400 MHz, DMSO-d6) δ (ppm): 7.19 (tdd, J = 8.6, 2.6, 1.0,
1H), 7.37 (ddd, J = 11.6, 9.3, 2.6, 1H), 7.62 (td, J = 9.0, 6.5,
1H), 7.70 (ddd, J = 11.9, 2.3, 1.3, 1H), 7.74 (dt, J = 2.5, 1.4,
1H). 13C NMR (100 MHz, DMSO-d6) δ (ppm): 104.6 (t, J =
26.0), 112.2 (dd, J = 21.2, 3.8), 115.8 (d, J = 3.9), 121.3 (dd, J =
19.3, 2.4), 122.5−123.1 (m), 124.4 (d, J = 6.7), 125.5 (d, J =
3.0), 131.7 (dd, J = 9.7, 4.4), 149.0 (d, J = 13.1), 150.7 (d, J =
243.6), 159.1 (dd, J = 248.7, 15.5), 161.7 (dd, 260.0, 12.3),
170.9 (d, J = 3.6). HRMS (ESI−): calcd for [C13H7F3O3 − H]−,
268.0347; found, 267.0295.
diluted with dicloromethane, and the organic phase was
successively washed with HCl 1N and a 0.1N sodium
thiosulfate solution. The organic phase was dried over
MgSO4 and after evaporation yielded a residue (>98% purity),
which was further purified by column chromatography on silica
gel using Cl3CH/MeOH gradient mixture or HPLC. White
1
solid. HPLC (GEN1): RT = 14.32 min. H NMR (500 MHz;
DMSO-d6) δ (ppm): 7.12−7.16 (m, 1H). 7.27−7.31 (m, 1H),
7.57 (ddd, J = 9.0, 9.0, 6.5, 1H), 7.92−7.93 (m, 1H), 8.09−8.10
(m, 1H). 13C NMR (125.7 MHz, DMSO-d6) δ (ppm): 171.2,
161.6 (dd, J = 12.6, 247.6), 159.7, 158.9 (dd, J = 12.3, 248.5),
144.1, 131.5 (dd, J = 4.5, 9.5), 130.4, 126.8, 122.4 (dd, J = 3.6,
13.3), 113.2, 111.9 (d, J = 20.9), 104.3 (t, J = 26.4), 85.9.
HRMS (ESI−): calcd for [C13H7F2IO3 − 1], 375,9408; found,
374,9306.
Protein and Inhibitors. The human TTR variant Y78F
protein was recombinant expressed in E. coli and purified as
already reported.57All assays were performed in buffers
containing a final 5% (v/v) DMSO concentration for
solubilization of the ligands.
Kinetic Turbidimetric Assay. Inhibition of fibrillogenesis
was determined by the kinetic turbidimetric assay previously
reported.57,58 Briefly, in seven different wells of a 96-well
microplate, 20 μL of a 4 mg/mL TTRY78F solution in 20 mM
potassium phosphate buffer, 100 mM KCl, and 1 mM EDTA at
pH 7.6, was mixed with an 80 μL solution of inhibitor prepared
by mixing different volumes of a stock solution of the
compound in H2O/DMSO (1:1) to give a range of final
compound concentrations of 0−40 μM. DMSO content was
adjusted to a final 5% (v/v), where all ligands tested are soluble.
After 30 min incubation at 37 °C with 15 s shaking every
minute, 100 μL of 400 mM KAcO, 100 mM KCl, and 1 mM
EDTA buffer at pH 4.2 were added to each well. The final
mixture, containing 0.4 mg/mL TTR, 0 to 40 μM ligand, and
5% DMSO, was incubated at 37 °C with 15 s shaking every
minute. Absorbance at 340 nm was monitored for 1.5 h at 1
min intervals. Initial rates of protein aggregation (v0) were
obtained from the linear plot absorbance versus time. The
dependence of v0 on inhibitor concentration is defined as:
5-Chloro-2′,4′-difluoro-4-hydroxy-[1,1′-biphenyl]-3-
carboxylic acid (16). To a stirred solution of diflunisal (50
mg, 0.200 mmol) in acetonitrile (2 mL) was added a solution of
N-chlorosuccinimide (56 mg, 0.420 mmol) in acetonitrile (1.5
mL). Next, 1 drop of concentrated HCl was added, and the
mixture was stirred for 1 h at rt The mixture was then diluted
with CH2Cl2 (20 mL), washed with 5% solution of Na2SO3 (5
mL) and brine (5 mL), and dried over anhydrous Na2SO4,
filtered, and evaporated to dryness. Further purification by silica
gel column chromatography (CH2Cl2/MeOH/AcOH 95:5:0.5)
afforded pure 16 (27 mg, 47%) as a white solid. Rf (CH2Cl2/
MeOH/AcOH 95:5:0.5) = 0.11. RP-HPLC (GEN1): RT =
1
12.46 min. H NMR (400 MHz, CDCl3/CD3OD 95:5) δ
(ppm): 6.86−7.00 (m, 2H), 7.31−7.42 (m, 1H), 7.71 (s, 1H),
7.96 (s, 1H), 9.14 (br s, 1H). 13C NMR (100 MHz, CDCl3/
CD3OD 95:5) δ (ppm): 104.5 (dd, J = 25.4, 26.5), 111.7 (dd, J
= 21.2, 3.8), 113.9, 122.1, 123.1 (dd, J = 13.6, 4.0), 126.2, 129.3
(d, J = 2.6), 131.0 (dd, J = 9.5, 4.6), 135.9 (d, J = 3.2), 157.0,
159.6 (dd, J = 250.4, 11.9), 162.4 (dd, J = 249.7, 11.8), 178.0.
HRMS (ESI−): calcd for [C13H7ClF2O3 − H]−, 282.9979;
found, 282.998.
5-Bromo-2′,4′-difluoro-4-hydroxy-[1,1′-biphenyl]-3-
carboxylic acid (17). To 600 mg of the polymer-bound
brominating agent (pyridine hydrobromideperbromide poly-
mer-bound, PVPHP, ∼1 mmol Br3 per g of resin, 3 equiv) in
CH2Cl2 (5 mL), a solution of diflunisal(14) (50 mg, 0.200
mmol) in CH2Cl2 (1 mL) was added and the mixture was
shaken o/n. After removing the resin, the filtrate was diluted
with more CH2Cl2 (20 mL), and washed with 5% aq solution of
Na2SO3 (5 mL) and brine (5 mL) and dried over anhydrous
Na2SO4, filtered and evaporated to dryness. Further purification
by silica gel column chromatography (CH2Cl2/MeOH/AcOH
95:5:0.5) afforded pure 17 (34 mg, 52%) as a white solid. RP-
HPLC (GEN1): RT = 13.32 min. 1H NMR (400 MHz,
DMSO-d6) δ (ppm): 7.17 (ddd, J = 8.5, 8.4, 2.7, 1H), 7.35
(ddd, J = 11.6, 9.3, 2.7, 1H), 7.61 (ddd, J = 9.0, 9.0, 6.6, 1H),
7.92 (t, J = 1.9, 1H), 7.97 (t, J = 1.9, 1H). 13C NMR (100 MHz,
DMSO-d6) δ (ppm): 104.6 (dd, J = 26.9, 25.9), 110.6, 112.2
(dd, J = 21.1, 3.7), 114.5, 122.6 (dd, J = 13.3, 3.7), 126.1, 129.8
(d, J = 3.5), 131.8 (dd, J = 9.7, 4.4), 138.2 (d, J = 2.3), 157.4,
159.0 (dd, J = 248.4, 11.6), 161.8 (dd, J = 259.7, 11.6), 171.3.
HRMS (ESI−): calcd for [C13H7BrF2O3 − H], 326.9473,
328.9453; found, 326.9463, 328.9445.
v0 = A + B·exp−C[I]
(1)
where v0 is the initial rate of fibril formation (in absorbance
units per hour, AU·h−1) and [I] the concentration of the
inhibitor (μM).
From the adjustable parameters, the IC50 (inhibitor
concentration at which the initial rate of protein aggregation
is half than that in absence of inhibitor) and RA(%)
(percentage reduction of amyloidosis at high inhibitor
concentration) were calculated.
Isothermal Titration Calorimetry Assay. Dissociation
constants for the tested compounds at 25 °C were obtained
using a Microcal iTC200 isothermal titration microcalorimeter
(Microcal, Inc., Northampton, MA). The titrant solution
containing 500 μM compound in 100 mM potassium
phosphate, 100 m MKCl, 1 mM EDTA, and 5% (v/v)
DMSO at pH 7.6 was placed in the calorimeter syringe. The
cell was filled with 20−30 μM TTR solution (Y78F mutant)
prepared in the same previous buffer, filtered through a 0.45
μm diameter pore membrane, and degassed for 5 min. Protein
concentration was determined measuring the absorbance at 280
nm using molar absorptivity of 17 395 M−1·cm−1 for Y78F
monomers, determined by the Bradford assay. An initial 0.5−1
2′,4′-Difluoro-4-hydroxy-5-iodo-[1,1′-biphenyl]-3-car-
boxylic acid or Iododiflunisal (13). To a solution of 200 mg
(0.80 mmol) of diflunisal (14) in 5 mL of dichloromethane,
357 mg (1.2 mmol) of IPy2BF4 was added to obtain a
substrate/iodinating reagent ratio of (1:1.5) equiv. The reaction
was left under stirring at rt and monitored by HPLC until full
conversion to the iodo derivative. After 1.5 h, the solution was
9117
dx.doi.org/10.1021/jm401061w | J. Med. Chem. 2013, 56, 9110−9121

Journal of Medicinal Chemistry
Article
μL initial injection of titrant solution was followed by several
1.5−2 μL injections separated by 150 s. The integration of the
thermograms, after subtracting the dilution blanks, gave the
binding isotherms that were fitted to a simple binding model
with two identical sites by a nonlinear least-squares method
using the Origin v.7.0 software provided by Microcal.
Dissociation constants were obtained with an 1−5% error.
Crystallographic Complexes. Three-dimensional atomic
coordinates of the apo-TTR (PDB entry 1BMZ)79 and the 13
TTR−ligand complexes studied in the present work (PDB
entries 1DVT,6 2B9A,68 2B77,68 1Y1D,56 2G5U,80 1TYR,81
2B15,82 2ROX,83 1Z7J,84 1TT6,85 1DVS,6 1BM7,79 and
1DVX6) were obtained from the structural information
available in the Protein Data Bank (PDB).
Before performing the computational analysis of the different
TTR−ligand complexes, a previous processing of the PDB files
was needed. Considering that the asymmetric crystal unit of all
studied complexes is formed by the dimer AB, two ligand
molecules (one for each binding site) and water molecules,
coordinates for the tetrameric form of TTR were obtained by
applying the crystallographic symmetry transformations
described in the corresponding PDB file. For residues having
multiple conformations, we only considered the one having the
highest occupation factor. In the case of having all conformers
resolved with the same occupation, we then considered the first
one listed in the PDB file. The same procedure was applied for
ligands presenting more than one conformation bound to the
protein.
Hydrogen atoms refinement was accomplished using 5000
cycles of the steepest descents followed by a conjugate gradient
until the maximum gradient of the AMBER energy was smaller
than 0.001 kcal/mol Å2.
Proposal of Halogenated Derivatives. The GRID v.22a
program was also used to propose which positions of flufenamic
acid and diflunisal are susceptible to halogenation. To this end,
the ligands and water molecules were removed from the
crystallographic complexes. The molecular interaction fields
(MIF) were generated into the binding site with the GRID
halogen probes using the same box dimensions as previously
described. Then, superposition of the crystallographic ligands
to the corresponding MIFs allowed us to suggest which
positions of the ligand are more favorable for halogen
substitution.
ASSOCIATED CONTENT
* Supporting Information
■
S
Added comments to the article sections, full description of the
synthesis and physicochemical characterization of reported
compounds, that is, halogenated flufenamic acids (18−21) and
halogenated diflunisal analogues (15−17 and 13), and selected
spectra (HPLC, 1H NMR, 13C NMR, HRMS) of target
compounds. This material is available free of charge via the
Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
*E-mail: gregorio.valencia@iqac.csic.es. Ph.: +34934006113
■
Water Molecule Placement. Application of the symmetry
operations indicated in the corresponding PDB file to obtain
the biologically functional tetramer of TTR implies that the
distribution of water molecules is exactly the same in both
dimers. Owing that this is not the case for nonsymmetric
ligands, it was necessary to confirm crystallographic-reported
water molecules and propose additional water molecule sites.
This strategy allowed us to avoid a symmetrical distribution of
water molecules around the ligand, which is an unrealistic
description for ligands not possessing 2-fold symmetry.
The GRID v.22a program was applied for water molecule
placement in the free and ligand-bound forms of TTR and to
validate the ones experimentally determined, as has been
previously reported.71,86−88 To this end, all water molecules
present in the crystallographic structures were first removed.
The grid box was defined to include the ligand and all the side
chains projected into the binding site, yielding a box of
dimensions 44 × 40 × 33 Å3. The molecular interaction fields
(MIFs) of a standard water probe were calculated into the box
using a grid spacing of 0.5 Å. The MINIM program
implemented in the GRID package was used to locate the
position of the minima in the interaction field, using an energy
cutoff of −5 kcal/mol and without interpolation. Finally, these
minima were populated with simulated annealing by using the
FILMAP utility included in the GRID program package.
Hydrogen Atoms Refinement. Added hydrogen atoms
were energy minimized by using the parm94 force field
implemented in AMBER v.6.89,90 Ligand partial charges were
obtained by computing the electrostatic potentials around the
optimized structures at the RH/6-31G* level using Gaussian
9491 and then fitting the charges by means of the RESP
method.92
Present Address
∥Pharmacoinformatics Research Group, Department fur
Medizinische, Pharmazeutische Chemie, Universitat Wien,
Wien, Austria.
Author Contributions
Overall research design and writing of the manuscript: J.Q.,
N.B.C., G.A., A.P., G.V. Chemistry experiments: L.B. Biological
experiments: E.Y.C., M.V. Computational studies: M.P., D.B.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
We thank Dr. Xevi Biarnes
S7 and analysis of XB geometries. This work was supported by
■
́
for the generation of Figures 5 and
a Grant No. 080530/31/32 from the Fundacio
Barcelona, Spain.
́ ́
Marato de TV3,
ABBREVIATIONS
■
AcOH, acetic acid; IPy2BF4, bis(pyridine)iodonium(I) tetra-
fluoroborate; NCS, N-chlorosuccimide; F-TEDA, 1-Chloro-
methyl-4-fluoro-1,4-diazoniabicyclo[2.2.2]octane bis-
(tetrafluoroborate); DMSO, dimethylsulfoxide; EtOAc, ethyl
acetate; HRMS, high-resolution mass spectrometry; HPLC,
1
high-performance liquid chromatography; H NMR, proton
nuclear magnetic resonance; PVPHP, pyridine hydrobromide
p e r b r o m i d e p o l y m e r - b o u n d ; P d ( P P h 3 ) 4 ,
Tetrakis(triphenylphosphine)palladium(0); (Pd2(dba)3),
Tris(dibenzylideneacetone)dipalladium(0); RT, retention time
REFERENCES
■
Minimization was carried out using a distance-dependent
dielectric constant of 1r to screen for electrostatic interactions
and a cutoff distance of 12 Å for van der Waals interactions.
(1) Saraiva, M. J.; Magalhaes, J.; Ferreira, N.; Almeida, M. R.
Transthyretin deposition in familial amyloidotic polyneuropathy. Curr.
Med. Chem. 2012, 19, 2304−2311.
9118
dx.doi.org/10.1021/jm401061w | J. Med. Chem. 2013, 56, 9110−9121

Journal of Medicinal Chemistry
Article
(2) Plante-
Lancet Neurol. 2011, 10, 1086−1097.
(3) Connors, L. H.; Lim, A.; Prokaeva, T.; Roskens, V. A.; Costello,
C. E. Tabulation of human transthyretin (TTR) variants. Amyloid
2003, 10, 160−184.
(4) Saraiva, M. J. Hereditary transthyretin amyloidosis: molecular
basis and therapeutical strategies. Expert Rev. Mol. Med. 2002, 4, 1−11.
(5) Adams, D.; Lozeron, P.; Lacroix, C. Amyloid neuropathies. Curr.
Opin. Neurol. 2012, 25, 564−572.
́
Bordeneuve, V.; Said, G. Familial amyloid polyneuropathy.
(24) Ferreira, N.; Saraiva, M. J.; Almeida, M. R. Epigallocatechin-3-
gallate as a potential therapeutic drug for TTR-related amyloidosis: ″in
vivo″ evidence from FAP mice models. PLoS ONE 2012, 7 (1),
e29933.
(25) Clinical trials: safety and efficacy study of Fx-1006A in patients
with familial amyloidosis. http://clinicaltrials.gov/ct2/show/
NCT00409175. An extension of study Fx-005 evaluating long-term
safety and clinical outcomes of Fx-1006A in patients with transthyretin
amyloid polyneuropathy. http://clinicaltrials.gov/ct2/show/
NCT00791492/.
(6) Klabunde, T.; Petrassi, H. M.; Oza, V. B.; Raman, P.; Kelly, J. W.;
Sacchettini, J. C. Rational design of potent human transthyretin
amyloid disease inhibitors. Nat. Struct. Biol. 2000, 7, 312−321.
(7) 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.
(8) 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.
(9) Ando, Y.; Ueda, M. Diagnosis and therapeutic approaches to
transthyretin amyloidosis. Curr. Med. Chem. 2012, 19, 2312−2323.
(10) Sacchettini, J. C.; Kelly, J. W. Therapeutic strategies for human
amyloid diseases. Nat. Rev. Drug Disc. 2002, 1, 267−275.
(11) Ong, D. S.; Kelly, J. W. Chemical and/or biological therapeutic
strategies to ameliorate protein misfolding diseases. Curr. Opin. Cell
Biol. 2011, 23, 231−238.
(12) Kapoor, P.; Thenappan, T.; Singh, E.; Kumar, S.; Greipp, P. R.
Cardiac amyloidosis: a practical approach to diagnosis and manage-
ment. Am. J. Med. 2011, 124, 1006−1015.
(13) Benson, M. D. Amyloidosis. In The Metabolic Basis of Inherited
Disease, Scriver, C. R., Beaudet, A. L., Sly, W. S., Valle, D., Childs, B.,
Kinzler, K. W., Vogelstein, B., Eds.; McGraw-Hill: Columbus, OH,
2001; pp 5345−5378.
(14) Saraiva, M. J. Transthyretin amyloidosis: a tale of weak
interactions. FEBS Lett. 2001, 498, 201−203.
(15) Hou, X.; Aguilar, M. I.; Small, D. H. Transthyretin and familial
amyloidotic polyneuropathy. Recent progress in understanding the
molecular mechanism of neurodegeneration. FEBS J. 2007, 274,
1637−1650.
(16) de la Paz, P.; Burridge, J. M.; Oatley, S. J.; Blake, C. C. F.
Multiple modes of binding of thyroid hormones and other
iodothyronines to human plasma transthyretin. In The Design of
Drugs to Macromolecular Targets; Beddell, C. R. Ed.; John Wiley and
Sons: New York, 1992; pp 119−172.
(17) Andrea, T. A.; Cavalieri, R. R.; Goldfine, I. D.; Jorgensen, E. C.
Binding of thyroid hormones and analogues to the human plasma
protein prealbumin. Biochemistry 1980, 19, 55−63.
(18) Bartalena, L.; Robbins, J. Thyroid hormone transport proteins.
Clin. Lab. Med. 1993, 13, 583−598.
(19) Almeida, M. R.; Gales, L.; Damas, A. M.; Cardoso, I.; Saraiva, M.
J. Small transthyretin (TTR) ligands as possible therapeutic agents in
TTR amyloidoses. Curr. Drug Targets: CNS Neurol. Disord. 2005, 4,
587−596.
(20) Nencetti, S.; Orlandini, E. TTR fibril formation inhibitors: is
there a SAR? Curr. Med. Chem. 2012, 19, 2356−2379.
(26) Coelho, T.; Maia, L. F.; Martins da Silva, A.; Waddington, Cruz,
M.; Plante-
́
Bordeneuve, V.; Lozeron, P.; Suhr, O. B.; Campistol, J. M.;
e,
Conceicao, I. M.; Schmidt, H. H.; Trigo, P.; Kelly, J. W.; Labaudinier
̧
̀
̃
R.; Chan, J.; Packman, J.; Wilson, A.; Grogan, D. R. Tafamidis for
transthyretin familial amyloid polyneuropathy: a randomized, con-
trolled trial. Neurology 2012, 79, 785−792.
(27) de Lartigue, J. Tafamidis for transthyretin amyloidosis. Drugs
Today 2012, 48, 331−337.
(28) Bulawa, C. E.; Connelly, S.; Devit, M.; Wang, L.; Weigel, C.;
Fleming, J. A.; Packman, J.; Powers, E. T.; Wiseman, R. L.; Foss, T. R.;
̀
Wilson, I. A.; Kelly, J. W.; Labaudiniere, R. Tafamidis, a potent and
selective transthyretin kinetic stabilizer that inhibits the amyloid
cascade. Proc. Natl. Acad. Sci. U.S.A. 2012, 109, 9629−9634.
(29) Clinical trials: the effect of diflunisal on familial amyloidosis.
http://clinicaltrials.gov/ct2/show/NCT00294671.
(30) Castano, A.; Helmke, S.; Alvarez, J.; Delisle, S.; Maurer, M. S.
̃
Diflunisal for ATTR cardiac amyloidosis. Congestive Heart Failure
2012, 18, 315−319.
(31) Berk, J. L.; Suhr, O. B.; Sekijima, Y.; Yamashita, T.; Heneghan,
M.; Zeldenrust, S. R.; Ando, Y.; Ikeda, S.; Gorevic, P.; Merlini, G.;
Kelly, J. W.; Skinner, M.; Bisbee, A. B.; Dyck, P. J.; Obici, L. Familial
amyloidosis consortium. The diflunisal trial: study accrual and drug
tolerance. Amyloid 2012, 19 (Suppl 1), 37−38.
(32) Obici, L.; Cortese, A.; Lozza, A.; Lucchetti, J.; Gobbi, M.;
Palladini, G.; Perlini, S.; Saraiva, M. J.; Merlini, G. Doxycycline plus
tauroursodeoxycholic acid for transthyretin amyloidosis: a phase II
study. Amyloid 2012, 19 (Suppl 1), 34−36.
(33) Cardoso, I.; Martins, D.; Ribeiro, T.; Merlini, G.; Saraiva, M. J.
Synergy of combined doxycycline/TUDCA treatment in lowering
Transthyretin deposition and associated biomarkers: studies in FAP
mouse models. J. Transl. Med. 2010, 8, 74.
(34) Clinical trials: safety, efficacy and pharmacokinetics of
doxycycline plus tauroursodeoxycholic acid in transthyretin amyloi-
dosis. http://clinicaltrials.gov/ct2/show/NCT01171859/.
(35) Terazaki, H.; Ando, Y.; Fernandes, R.; Yamamura, K.; Maeda, S.;
Saraiva, M. J. Immunization in familial amyloidotic polyneuropathy:
counteracting deposition by immunization with a Y78F TTR mutant.
Lab. Invest. 2006, 86, 23−31.
(36) Su, Y.; Jono, H.; Torikai, M.; Hosoi, A.; Soejima, K.; Guo, J.;
Tasaki, M.; Misumi, Y.; Ueda, M.; Shinriki, S.; Shono, M.; Obayashi,
K.; Nakashima, T.; Sugawara, K.; Ando, Y. Antibody therapy for
familial amyloidotic polyneuropathy. Amyloid 2012, 19 (Suppl 1), 45−
46.
(37) Benson, M. D.; Pandey, S.; Witchell, D.; Jazayeri, A.; Siwkowski,
A.; Monia, B.; Kluve-Beckerman, B. Antisense oligonucleotide therapy
for TTR amyloidosis. Amyloid 2011, 18 (Suppl 1), 60.
(38) Benson, M. D.; Kluve-Beckerman, B.; Zeldenrust, S. R.; Siesky,
A. M.; Bodenmiller, D. M.; Showalter, A. D.; Sloop, K. W. Targeted
suppression of an amyloidogenic transthyretin with antisense
oligonucleotides. Muscle Nerve 2006, 33, 609−618.
(39) Nakamura, M.; Ando, Y. Applications of gene therapy for
familial amyloidotic polyneuropathy. Expert Opin. Biol. Ther. 2004, 4,
1621−1627.
(40) Hamilton, J. A.; Benson, M. D. Transthyretin: a review from a
structural perspective. Cell. Mol. Life Sci. 2001, 58, 1491−1521.
(41) Monaco, H. L.; Rizzi, M.; Coda, A. Structure of a complex of
two plasma proteins: transthyretin and retinol-binding protein. Science
1995, 268, 1039−1041.
(21) Alhamadsheh, M. M.; Connelly, S.; Cho, A.; Reixach, N.;
Powers, E. T.; Pan, D. W.; Wilson, I. A.; Kelly, J. W.; Graef, I. A.
Potent kinetic stabilizers that prevent transthyretin-mediated cardio-
myocyte proteotoxicity. Sci. Transl. Med. 2011, 3 (97), 97ra81
DOI: 10.1126/scitranslmed.3002473.
(22) ECGC: 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 (Suppl 1), 39−42.
(23) Kristen, A. V.; Lehrke, S.; Buss, S.; Mereles, D.; Steen, H.;
Ehlermann, P.; Hardt, S.; Giannitsis, E.; Schreiner, R.; Haberkorn, U.;
Schnabel, P. A.; Linke, R. P.; Rocken, C.; Wanker, E. E.; Dengler, T. J.;
̈
Altland, K.; Katus, H. A. Green tea halts progression of cardiac
transthyretin amyloidosis: an observational report. Clin. Res. Cardiol.
2012, 101, 805−813.
9119
dx.doi.org/10.1021/jm401061w | J. Med. Chem. 2013, 56, 9110−9121

Journal of Medicinal Chemistry
Article
(42) Buxbaum, J. N.; Reixach, N. Transthyretin: the servant of many
masters. Cell. Mol. Life Sci. 2009, 66, 3095−3101.
(43) Ingbar, S. H. Observations concerning the binding of thyroid
hormones by human serum prealbumin. J. Clin. Invest. 1963, 42, 143−
160.
2013 [Online early access]. DOI: 10.1021/cb400526n. Published
Online: August 30, 2013.
(63) Bent, H. A. Structural chemistry of donor−acceptor interactions.
Chem. Rev. 1968, 68, 587−648.
(64) Hassel, O. Structural aspects of interatomic charge−transfer
bonding. Science 1970, 170, 497−502.
́ ́
(44) Blake, C. C. F.; Geisow, M. J.; Oatley, S. J.; Rerat, B.; Rerat, C.
(65) Gales, L.; Almeida, M. R.; Arsequell, G.; Valencia, G.; Saraiva,
M. J.; Damas, A. M. Iodination of salicylic acid improves its binding to
transthyretin. Biochem. Biophys. Acta 2008, 1784, 512−517.
(66) Choi, S.; Reixach, N.; Connelly, S.; Johnson, S. M.; Wilson, I. A.;
Kelly, J. W. A substructure combination strategy to create potent and
selective transthyretin kinetic stabilizers that prevent amyloidogenesis
and cytotoxicity. J. Am. Chem. Soc. 2010, 132, 1359−1370.
(67) Eneqvist, T.; Lundberg, E.; Karlson, A.; Huang, S.; Santos, C. R.;
Power, D. M.; Sauer-Eriksson, A. E. High resolution crystal structures
of piscine transthyretin reveal different binding modes for triiodothyr-
onine and thyroxine. J. Biol. Chem. 2004, 279, 26411−26416.
(68) Adamski-Werner, S. L.; Palaninathan, S. K; Sacchettini, J. C.;
Kelly, J. W. Diflunisal analogues stabilize the native state of
transthyretin. Potent inhibition of amyloidogenesis. J. Med. Chem.
2004, 47, 355−374.
Structure of prealbumin: secondary, tertiary and quaternary
interactions determined by Fourier refinement at 1.8 Å. J. Mol. Biol.
1978, 121, 339−356.
(45) 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.
(46) Berman, H. M; Westbrook, J.; Feng, Z.; Gilliland, G.; Bhat, T.
N.; Weissig, H.; Shindyalov, I. N.; Bourne, P. E. The protein data bank.
Nucleic Acids Res. 2000, 28, 235−242.
(47) Parisini, E.; Metrangolo, P.; Pilati, T.; Resnati, G.; Terraneo, G.
Halogen bonding in halocarbon-protein complexes: a structural survey.
Chem. Soc. Rev. 2011, 40, 2267−2678.
(48) Metrangolo, P.; Resnati, G. Chemistry. Halogen versus
hydrogen. Science 2008, 321, 918−919.
(49) Metrangolo, P.; Neukirch, H.; Pilati, T.; Resnati, G. Halogen
bonding based recognition processes: a world parallel to hydrogen
bonding. Acc. Chem. Res. 2005, 38, 386−395.
(50) Politzer, P.; Murray, J. S. Halogen bonding: an interim
discussion. ChemPhysChem 2013, 14, 278−294.
(51) Wilcken, R.; Zimmermann, M. O.; Lange, A.; Joerger, A. C.;
Boeckler, F. M. Principles and applications of halogen bonding in
medicinal chemistry and chemical biology. J. Med. Chem. 2013, 56,
1363−1388.
(52) Scholfield, M. R.; Zanden, C. M.; Carter, M.; Ho, P. S. Halogen
bonding (X-bonding): a biological perspective. Protein Sci. 2013, 22,
139−152.
(53) Hardegger, L. A.; Kuhn, B.; Spinnler, B.; aAnselm, L.; Ecabert,
R.; Stihle, M.; Gsell, B.; Thoma, R.; Diez, J.; Benz, J.; Plancher, J. M.;
Hartmann, G.; Banne, D. W.; Haap, W.; Diederich, F. Systematic
investigation of halogen bonding in protein−ligand interactions.
Angew. Chem., Int. Ed. 2011, 50, 314−318.
(54) Lu, Y.; Liu, Y.; Xu, Z.; Li, H.; Liu, H.; Zhu, W. Halogen bonding
for rational drug design and new drug discovery. Expert Opin. Drug
Discov. 2012, 7, 375−383.
(55) Lu, Y.; Shi, T.; Wang, Y.; Yang, H.; Yan, X.; Luo, X.; Jiang, H.;
Zhu, W. Halogen bondinga novel interaction for rational drug
design? J. Med. Chem. 2009, 52, 2854−2862.
(56) 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, 615−621.
(69) Baures, P. W.; Oza, V. B.; Peterson, S. A.; Kelly, J. W. Synthesis
and evaluation of inhibitors of transthyretin amyloid formation based
on the non-steroidal anti-inflammatory drug, flufenamic acid. Bioorg.
Med. Chem. 1999, 7, 1339−1347.
(70) Mairal, T.; Nieto, J.; Pinto, M.; Almeida, M. R.; Gales, L.;
Ballesteros, A.; Barluenga, J.; Perez, J. J.; Vazquez, 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 (1), e4124.
(71) Goodford, P. J. A computational procedure for determining
energetic favourable binding site of biological important macro-
molecules. J. Med. Chem. 1985, 28, 849−857.
(72) Neumann, P.; Cody, V.; Wojtczak, A. Structural basis of negative
cooperativity in transthyretin. Acta Biochim. Pol. 2001, 48, 867−875.
(73) Trivella, D. B.; Sairre, M. I.; Foguel, D.; Lima, L. M.; Polikarpov,
I. The binding of synthetic triiodo L-thyronine analogs to human
transthyretin: molecular basis of cooperative and non-cooperative
ligand recognition. J. Struct. Biol. 2011, 173, 323−332.
(74) 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.
(75) Blasi, D.; Pinto, M.; Nieto, J.; Arsequell, G.; Valencia, G.; Planas,
A.; Centeno, N. B.; Quintana, J. Drug discovery targeted at
transthyretin cardiac amyloidosis: rational design, synthesis, and
biological activity of new transthyretin amyloid inhibitors. Amyloid
2011, 18 (Suppl 1), 55−57.
(57) Dolado, I.; Nieto, J.; Saraiva, M. J.; Arsequell, G.; Valencia, G.;
Planas, A. Kinetic assay for high-throughput screening of in vitro
transthyretin amyloid fibrillogenesis inhibitors. J. Comb. Chem. 2005, 7,
246−252.
(58) Arsequell, G.; Planas, A. Methods to evaluate the inhibition of
TTR fibrillogenesis induced by small ligands. Curr. Med. Chem. 2012,
19, 2343−55.
(76) Pinto, M.; Blasi, D.; Nieto, J.; Arsequell, G.; Valencia, G.; Planas,
A.; Quintana, J.; Centeno, N. B. Ligand-binding properties of human
transthyretin. Amyloid 2011, 18 (Suppl 1), 51−54.
(77) Bondi, A. van der Waals volumes and Radii. J. Phys. Chem. 1964,
68, 441−451.
(59) Desiraju, G. R.; Ho, P. S.; Kloo, L.; Legon, A. C.; Marquardt, R.;
Metrangolo, P.; Politzer, P.; Resnati, G.; Rissanen, K. Definition of the
halogen bond (IUPAC Recommendations 2013)*. Pure Appl. Chem.
2013, 85, 1711−1713.
(60) Auffinger, P.; Hays, F. A.; Westhof, E.; Ho, P. S. Halogen bonds
in biological molecules. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 16789−
16794.
(78) Wilcken, R.; Zimmermann, M. O.; Lange, A.; Zahn, S.; Boeckler,
F. M. Using halogen bonds to address the protein backbone: a
systematic evaluation. J. Comput. Aided Mol. Des. 2012, 26, 935−945.
(79) Peterson, S. A.; Klabunde, T.; Lashuel, H. A.; Purkey, H.;
Sacchettini, J. C.; Kelly, J. W. Inhibiting transthyretin conformational
changes that lead to amyloid fibril formation. Proc. Natl. Acad. Sci. U. S.
A. 1998, 95, 12956−12960.
(61) Wilcken, R.; Liu, X.; Zimmermann, M. O.; Rutherford, T. J.;
Fersht, A. R.; Joerger, A. C.; Boeckler, F. M. Halogen-enriched
fragment libraries as leads for drug rescue of mutant p53. J. Am. Chem.
Soc. 2012, 134, 6810−6818.
(80) Purkey, H. E.; Palaninathan, S. K.; Kent, K. C.; Smith, C.; Safe,
S. H.; Sacchettini, J. C.; Kelly, J. W. Hydroxylated polychlorinated
biphenyls selectively bind transthyretin in blood and inhibit amyloido-
genesis: rationalizing rodent PCB toxicity. Chem. Biol. 2004, 11,
1719−1728.
(81) Zanotti, G.; D’Acunto, M. R.; Malpeli, G.; Folli, C.; Berni, R.
Crystal structure of the transthyretin−retinoic−acid complex. Eur. J.
Biochem. 1995, 234, 563−569.
(62) Fanfrlik, J.; Kolar
́
,
̌
M.; Kamlar, M.; Hurny, D.; Ruiz, F. X.;
J.; Munusamy, E.; Lepsik, M.;
Cousido-Siah, A.; Mitschler, A.; Rezac
Matejicek, P.; Vesely, J.; Podjarny, A. Hobza, P. The modulation of
́
,
̌
aldose reductase inhibition by halogen bond tuning. ACS Chem. Biol.
9120
dx.doi.org/10.1021/jm401061w | J. Med. Chem. 2013, 56, 9110−9121

Journal of Medicinal Chemistry
Article
́
(82) Morais-de-Sa, E.; Neto-Silva, R. M.; Pereira, P. J.; Saraiva, M. J.;
Damas, A. M. The binding of 2,4-dinitrophenol to wild-type and
amyloidogenic transthyretin. Acta Crystallogr., Sect. D: Biol. Crystallogr.
2006, 62, 512−519.
(83) Wojtczak, A.; Cody, V.; Luft, J. R.; Pangborn, W. Structures of
human transthyretin complexed with thyroxine at 2.0 Å resolution and
3′,5′-dinitro-N-acetyl-L-thyronine at 2.2 Å resolution. Acta Crystallogr.,
Sect. D: Biol. Crystallogr. 1996, 52, 758−765.
(84) Neumann, P.; Cody, V.; Wojtczak, A. Ligand binding at the
transthyretin dimer-dimer interface: structure of the transthyretin−
T4Ac complex at 2.2 Angstrom resolution. Acta Crystallogr., Sect. D:
Biol. Crystallogr. 2005, 61, 1313−1319.
́
(85) Morais-de-Sa, E.; Pereira, P. J.; Saraiva, M. J.; Damas, A. M. The
crystal structure of transthyretin in complex with diethylstilbestrol: a
promising template for the design of amyloid inhibitors. J. Biol. Chem.
2004, 279, 53483−53490.
(86) Wade, R. C.; Clark, K. J.; Goodford, P. J. Further development
of hydrogen bond functions for use in determining energetically
favorable binding sites on molecules of known structure. 1. Ligand
probe groups with the ability to form two hydrogen bonds. J. Med.
Chem. 1993, 36, 140−147.
(87) Wade, R. C.; Goodford, P. J. Further development of hydrogen
bond functions for use in determining energetically favorable binding
sites on molecules of known structure. 2. Ligand probe groups with the
ability to form more than two hydrogen bonds. J. Med. Chem. 1993,
36, 148−156.
(88) Pastor, M.; Cruciani, G.; Watson, K. A. A strategy for the
incorporation of water molecules present in a ligand binding site into a
three-dimensional quantitative structure-activity relationship analysis. J.
Med. Chem. 1997, 40, 4089−4102.
(89) Cornell, W.; Cieplak, P.; Bayley, C. I.; Gould, I.; Merz, K. M.;
Ferguson, D. M.; Spellmeyer, D. C.; Fox, T.; Caldwell, J. W.; Kollman,
P. A. A Second Generation Force Field for the Simulation of Proteins,
Nucleic Acids, and Organic Molecules. J. Am. Chem. Soc. 1995, 117,
5179−5197.
(90) Case, D. A.; Pearlman, D. A.; Caldwell, J. W.; Cheatham III, T.
E.; Ross, W. S.; Simmerling, C. L.; Darden, T. A.; Merz, K. M.;
Stanton, R. V.; Cheng, A. L.; Vincent, J. J.; Crowley, M.; Tsui, V.;
Radmer, R. J.; Duan, Y.; Pitera, J.; Massova, I.; Seibel, G. L.; Singh, U.
C.; Weiner, P. K.; Kollman, P. A. AMBER 6; University of California:
San Francisco, CA, 1999.
(91) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Gill, P. M. W.;
Johnson, B. G.; Robb, M. A.; Cheeseman, J. R.; Keith, T.; Petersson,
G.; Montgomery, J. A.; Raghavachari, K.; Al-Laham, M. A.; Zakrzewski,
V. G.; Ortiz, J. V.; Foresman, J. B.; Cioslowski, J.; Stefanov, B. B.;
Nanayakkara, A.; Challacombe, M.; Peng, C. Y.; Ayala, P. Y.; Chen,
W.; Wong, M. W.; Andres, J. L.; Replogle, E. S.; Gomperts, R.; Martin,
R. L.; Fox, D. J.; Binkley, J. S.; Defrees, D. J.; Baker, J.; Stewart, J. P.;
Head-Gordon, M.; Gonzalez, C.; Pople, J. A. Gaussian 94, revision E.1;
Gaussian, Inc.: Pittsburgh, PA, 1996.
(92) Bayly, C. I.; Cieplak, P.; Cornell, W.; Kollman, P. A. A well-
behaved electrostatic potential based method using charge restraints
for deriving atomic charges: the RESP model. J. Phys. Chem. 1993, 97,
10269−10280.
9121
dx.doi.org/10.1021/jm401061w | J. Med. Chem. 2013, 56, 9110−9121