Toward optimization of the second aryl substructure common to transthyretin amyloidogenesis inhibitors using biochemical and structural studies

Article abstract of DOI:10.1021/jm801347s

Transthyretin (TTR) amyloidogenesis inhibitors are typically composed of two aromatic rings and a linker. We have previously established optimal structures for one aromatic ring and the linker. Herein, we employ a suboptimal linker and an optimal aryl-X substructure to rank order the desirability of aryl-Z substructures-using a library of 56 N-(3,5-dibromo-4-hydroxyphenyl) benzamides. Coconsideration of amyloid inhibition potency and ex vivo plasma TTR binding selectivity data reveal that 2,6, 2,5, 2, 3,4,5, and 3,5 substituted aryls bearing small substituents generate the most potent and selective inhibitors, in descending order. These benzamides generally lack undesirable thyroid hormone receptor binding and COX-1 inhibition activity. Three high-resolution TTR ? inhibitor crystal structures (1.31-1.35 ?) provide insight into why these inhibitors are potent and selective, enabling future structure-based design of TTR kinetic stabilizers.

Full text of DOI:10.1021/jm801347s

J. Med. Chem. 2009, 52, 1115–1125
1115
Toward Optimization of the Second Aryl Substructure Common to Transthyretin
Amyloidogenesis Inhibitors Using Biochemical and Structural Studies†
Steven M. Johnson,‡ Stephen Connelly,§ Ian A. Wilson,§ and Jeffery W. Kelly*,‡
Departments of Chemistry and Molecular Biology and The Skaggs Institute for Chemical Biology, The Scripps Research Institute,
10550 North Torrey Pines Road, La Jolla, California 92037
ReceiVed October 24, 2008
Transthyretin (TTR) amyloidogenesis inhibitors are typically composed of two aromatic rings and a linker.
We have previously established optimal structures for one aromatic ring and the linker. Herein, we employ
a suboptimal linker and an optimal aryl-X substructure to rank order the desirability of aryl-Z
substructuressusing a library of 56 N-(3,5-dibromo-4-hydroxyphenyl)benzamides. Coconsideration of amyloid
inhibition potency and ex vivo plasma TTR binding selectivity data reveal that 2,6, 2,5, 2, 3,4,5, and 3,5
substituted aryls bearing small substituents generate the most potent and selective inhibitors, in descending
order. These benzamides generally lack undesirable thyroid hormone receptor binding and COX-1 inhibition
activity. Three high-resolution TTR · inhibitor crystal structures (1.31-1.35 Å) provide insight into why
these inhibitors are potent and selective, enabling future structure-based design of TTR kinetic stabilizers.
Introduction
nearly 100 other rarer TTR mutations leads to familial amyloid
polyneuropathy (FAP), usually presenting with peripheral neu-
ropathy and sometimes autonomic and organ system involve-
ment.17 The much rarer central nervous system selective
amyloidoses (CNSA) result from deposition of highly destabi-
lized TTR variants (e.g., D18G and A25T) in the brain, but not
in the periphery. This is because the liver, which secretes TTR
into the blood, detects these variants as misfolding prone and
degrades them, unlike the choroid plexus, which is a more
permissive secretor of misfolding-prone variants into the
brain.18-24 Without treatment, the TTR amyloidoses are fatal.
An aging-associated decline in proteostasis capacity can lead
to aggregation-linked gain-of-toxic-function protein misfolding
diseases such as the amyloidoses, especially when proteome
maintenance is further challenged by the inheritance of mutant
misfolding-prone proteins or by environmental factors.1-6
Transthyretin (TTRa) is one of more than 30 human amy-
loidogenic proteins whose misfolding and/or misassembly into
a variety of aggregate structures, including cross-ꢀ-sheet amyloid
fibrils, appears to cause proteotoxicity.7-11 What the TTR toxic
structures are and how toxicity arises are key unanswered
questions.
To become amyloidogenic outside the cell, tetrameric TTR
must first undergo rate-limiting dissociation, allowing the
resulting monomers to partially unfold and misassemble.9
Another possibility is that TTR amyloidogenesis competes with
folding and TTR tetramerization within the cellular secretory
pathway, leading to intracellular proteotoxicity. Thus, proteotox-
icity could have its origins both within and outside the cell and
this issue remains to be resolved. Aggregation of wild-type
transthyretin (WT-TTR), and the resulting proteotoxicity appears
to cause senile systemic amyloidosis (SSA), a cardiac disease
affecting up to ≈15% of the population over age 65.9,12-14
Deposition of the V122I-TTR variant leads to familial amyloid
cardiomyopathy (FAC) in up to 4% of African Americans
carrying at least one V122I-TTR allele, while amyloid-associ-
ated cardiomyopathy linked to the proteotoxicity arising from
other TTR variant aggregates has a lower penetrance.15,16
Amyloidogenesis of V30M-TTR, or the aggregation of one of
The only currently accepted therapeutic strategy to ameliorate
FAP is gene therapy mediated by liver transplantation, wherein
an FAP-associated mutant TTR/WT-TTR liver is replaced by
a WT-TTR/WT-TTR secreting liver, eliminating the presence
of amyloidogenic mutant TTR in the blood.25-27 Unfortunately,
WT-TTR deposition often continues post-transplantation in the
heart, leading to cardiomyopathy, consistent with the hypothesis
that an age-dependent decline in proteostasis contributes to the
etiology of the TTR amyloidoses.1,28 Because liver transplant-
ation must be performed early in the course of the disease to
be effective, and owing to the shortage of livers, the expense
associated with transplantation, and the requirement for life-
long immune suppression, a generally applicable, oral small
molecule therapeutic strategy for all the TTR-based amyloid
diseases is highly desirable.9,29,30
Transthyretin transports the holo-retinol binding protein
complex and thyroxine (T4) in the blood.9,31-34 The four 127-
amino acid ꢀ-sheet rich subunits of transthyretin create two
distinct dimer-dimer interfaces.9,35,36 The energetically weaker
interface forms two concave thyroxine binding pockets. Small
molecule binding to only one or both of the T4 sites differentially
stabilizes the tetrameric ground state over the dissociative
transition state, imposing kinetic stability by precluding dis-
sociation required for amyloidogenesis.9,35-38
†
Atomic coordinates have been deposited in the RCSB Protein Data
Bank (www.pdb.org) and are available under accession codes 3ESN (WT-
TTR in complex with 2d), 3ESO (WT-TTR in complex with 3b), and 3ESP
(WT-TTR in complex with 5d.
* To whom correspondence should be addressed: Phone: 858-784-9605.
Fax: 858-784-9610. E-mail: jkelly@scripps.edu.
‡
Department of Chemistry and The Skaggs Institute for Chemical
Optimism that such small molecule kinetic stabilizers of
TTR24,39-47 will be effective against the TTR amyloidoses is
warranted based on human genetic evidence. T119M subunit
inclusion into tetramers otherwise composed of V30M disease-
associated subunits in compound heterozygotes kinetically
Biology, The Scripps Research Institute.
§
Department of Molecular Biology, and The Skaggs Institute for
Chemical Biology, The Scripps Research Institute.
a Abbreviations: TTR, transthyretin; COX-1, cyclooxygenase-1; WT, wild
type; HBP, halogen binding pocket; PGE2, prostaglandin E2; T4, thyroxine;
T3, triiodothyronine; NSAID, nonsteroidal anti-inflammatory drug.
10.1021/jm801347s CCC: $40.75
2009 American Chemical Society
Published on Web 02/03/2009

1116 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 4
Johnson et al.
Figure 1. Schematic depiction of one T4 binding pocket occupied by
a typical small molecule TTR aggregation inhibitor. Y represents a
linker of variable chemical structure (e.g., NH, O, CHdCH, C(O)NH,
etc.) joining the two aryl rings, which typically bear a combination of
alkyl, carboxyl, halide, trifluoromethyl, or hydroxyl substituents (X and
Z). Library 3, the focus of this manuscript, explores the SAR of the
aryl-Z substructure. While the aryl-X substructure was initially envi-
sioned to bind within the inner cavity of the thyroxine binding site,
previous crystallographic data reveal that inhibitors can bind opposite
to the expected orientation.34,51,52,59-61 Thus, it may not be the case
that the aryl-X substructure of an inhibitor will occupy the inner binding
cavity as illustrated. For conceptual simplicity in terms of integrating
the data from the three inhibitor substructure optimization libraries,
this initial schematic has been retained.
stabilizes the tetrameric structure of TTR, preventing amyloido-
genesis and ameliorating FAP.9,37,48-50
To increase the probability of identifying TTR kinetic
stabilizers with desirable pharmacological properties, three small
molecule libraries have recently been designed and synthesized
to optimize the three substructures comprising a typical TTR
amyloidogenesis inhibitor (Figure 1): the two aromatic rings
and the linker joining them.51,52 In each library, one fixed
substructure was purposefully chosen to be nonoptimal to afford
a wider range of TTR amyloidogenesis inhibitor potencies and
plasma TTR binding selectivities that are together used to rank
order the numerous substructures of interest. Data from the first
library, focused on optimization of the aryl-X ring, revealed that
thyroid hormone-like aryl substitution patterns (3,5-X2-4-OH,
where X ) CH3, F, Cl, Br, I) lead to potent and highly selective
TTR amyloidogenesis inhibitors lacking undesirable thyroid
hormone receptor and cyclooxygenase-1 (COX-1) binding
activities.51 The second library, designed to reveal optimal linker
substructures, demonstrates that directly linking two aryls, or
tethering them through nonpolar E-olefin or -CH2CH2-
substructures, generates the most potent and selective TTR
amyloidogenesis inhibitors that also lack the undesirable thyroid
hormone and COX-1 activities outlined above.52 In this study,
we utilize a nonoptimal amide linker and an optimal aryl-X
substructure in all library members to rank order candidate aryl-Z
substructures. The intention of this three-part series of papers
is to generate optimal substructure information that can ulti-
mately be used to predict the structures of potent and selective
TTR kinetic stabilizers that lack the thyroid hormone receptor
binding capacity and COX-1 inhibition activity contraindicated
for cardiomyopathy applications. These activities should increase
the probability of finding a TTR kinetic stabilizer with desirable
pharmacologic properties going forward.
Figure 2. Selection of an appropriate molecular scaffold for structure-
guided drug design to identify favorable aryl-Z substituents and
substitution patterns. (A) N-(3,5-Dibromo-4-hydroxyphenyl)benzamide
(1) was previously found to be a moderately potent and selective
inhibitor of TTR amyloidogenesis, allowing 26% aggregation of WT-
TTR in vitro (7.2 µM inhibitor, 3.6 µM TTR, pH 4.4, 37 °C, 72 h) and
displaying a stoichiometry of 0.41 equiv bound to TTR in human blood
plasma ex vivo.52 Importantly, compound 1 does not target the thyroid
hormone (TH) receptor or inhibit COX-1 activity. (B) Molecular
modeling was performed using the previously determined TTR ·(1)2
crystal structure (PDB accession code 3CN4), in which compound 1
binds in the reverse orientation with its 3,5-dibromo-4-hydroxyphenyl
substructure occupying the outer binding site cavity, projecting the
aryl-Z ring under evaluation into the halogen binding pockets of the
inner cavity.52 Individual TTR monomers in the ribbon diagram
representation have been color coded red, yellow, green, and blue for
differentiation. The zoomed image of one site shows 1 bound in its
two symmetry-related binding modes (orange and gray). The protein
surface is shown colored by electrostatic potential, with the front portion
cut away to allow easier visualization into the pocket.
N-(3,5-dibromo-4-hydroxyphenyl)benzamide inhibitor (1), while
generally avoiding undesirable binding to the thyroid hormone
receptor and COX-1 inhibition. High-resolution X-ray structures
of three TTR ·inhibitor complexes (1.31-1.35 Å) rationalizes
their potency and selectivity.
Herein, we report a 56-member library wherein all members
are composed of an optimal 3,5-dibromo-4-hydroxyphenyl
aryl-X ring attached to the NH of a nonoptimal secondary amide
linker connected to variable aryl-Z rings, which are the subject
of this optimization study (Figure 2). Numerous aryl-Z substit-
uents and substitution patterns dramatically increase inhibitor
potency and binding selectivity in comparison to the parent
Results
Design and Synthesis of the Aryl-Z Optimization
Library. In the recently published linker optimization compo-
nent (part 2) of this three-part study, N-(3,5-dibromo-4-

Second Aryl Substructure in TTR Amyloidogenesis Inhibitors
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 4 1117
Figure 3. Extent of inhibition of TTR aggregation and stoichiometry of inhibitor bound to TTR in human blood plasma. Percent (%) values
represent the extent of in vitro WT-TTR fibril formation in the presence of inhibitor (7.2 µM inhibitor, 3.6 µM TTR, pH 4.4, 37 °C, 72 h) relative
to aggregation in the absence of inhibitor (100%), with the best values shown in red (<25% aggregation; errors are typically less than (5 percentage
points). The average stoichiometries of the most potent aggregation inhibitors bound to TTR in human blood plasma ex vivo are shown in italics
(10.8 µM inhibitor, 1.8-5.4 µM TTR; theoretical maximum binding stoichiometry ) 2). Those exhibiting exceptional binding selectivity to TTR
are boxed (errors are typically less than (0.1).
hydroxyphenyl)benzamide (1) was established to be a moder-
ately potent and selective transthyretin amyloidogenesis inhibitor
(Figure 2A),52 lacking contraindicated thyroid hormone receptor
binding or COX-1 inhibition properties. The amide linker is
isostructural with the most potent 2-arylbenzoxazole and E-olefin
linkers, but is nonoptimal because its polarity is poorly matched
with the hydrophobic properties of the TTR subsite that
surrounds to the linker. Thus the bisarylamide library described
within to optimize the aryl-Z ring should prove to be highly
relevant, especially considering that all library members are
made up of an optimized aryl-X ring identified in the first paper
in this series.51
favorably with HBP-3 and 3′. In addition, polar meta or para
substituents (e.g., amino groups or high pKa phenols) could
enhance binding affinity through hydrogen bonding with the
Ser-117/117′ hydroxyls.
Certain aryl-Z substructures, such as low pKa phenols, could
change the binding orientation, such that the 3,5-Br2-4-hydrox-
yphenyl substructure common to all library members now
occupies the inner binding cavity. If this were to occur, modeling
suggests that the same aryl-Z substructures bearing ortho and
meta alkyl and halide substituents could interact favorably with
the hydrophobic HBP-1 and 1′, while meta or para carboxyl,
amino, or phenolic substituents could make electrostatic interac-
+
The TTR ·(1)2 crystal structure (PDB accession code 3CN4)
solved previously52 enables structure-based design principles
to be used to identify favorable aryl-Z substituents and substitu-
tion patterns, thus enhancing the probability for success.
Compound 1 binds such that the 3,5-Br2-4-hydroxyphenyl
substructure (having a calculated pKa of 6.5) occupies the outer
thyroxine binding site (true of nearly all low pKa phenols
structurally characterized to date) where the phenolate can
tions with the Lys-15 and 15′ ε-NH3 groups or the Glu-54
and 54′ carboxylate groups.
Using structure-based principles as a rough guideline, a library
of 56 bisarylamides was synthesized to evaluate 10 different
aryl-Z substituents (a-j) in eight distinct substitution patterns
(2-9) (Figure 3). Coconsideration of amyloid inhibition and
ex vivo plasma TTR binding selectivity data using a simple
equation allows us to rank order the aryl-Z substructures from
most desirable (highest score) to least desirable using an efficacy
score (Figure 3; see below).
Synthesis of these bisarylamides was accomplished primarily
via coupling of substituted benzoyl chlorides (phthalic anhydride
in the case of 4i) with 4-amino-2,6-dibromophenol (Scheme 1)
to form the amide bond. The amide bond coupling reactions
+
interact with the Lys-15 and 15′ side chain ε-NH3 groups
(Figure 2B).51,52 This orientation projects the aryl-Z ring into
the inner cavity, where its substituents can occupy the symmetric
hydrophobic depressions referred to as halogen binding pockets
3 and 3′ (HBP-3 and 3′). Upon slight repositioning of the aryl-Z
ring, it appears that ortho and meta substituents could interact

1118 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 4
Johnson et al.
Scheme 1. General Synthesis of the Bisarylamide Inhibitors
inhibitor (10.8 µM) was incubated in human blood plasma in
the dark at 37 °C for 24 h. Transthyretin, with any bound
inhibitor, was then captured by a resin-conjugated anti-TTR
antibody and any unbound material was washed away (including
weakly or nonspecifically bound inhibitors). The captured
TTR · (inhibitor)n complex was then dissociated from the
antibody under alkaline conditions, and TTR and inhibitor
stoichiometry was quantified by RP-HPLC. Results represent
the average stoichiometry of inhibitor bound to TTR in blood
plasma (Figure 3, lower black italicized values), the maximum
value being 2 owing to the presence of the two thyroxine binding
sites in each tetramer. Of the 41 inhibitors evaluated, 31 display
average binding stoichiometries exceeding 1 equiv bound per
TTR tetramer, 14 of which are exceptionally selective, display-
ing >1.5 equiv bound (boxed values). Six bisarylamides display
modest yet likely sufficient binding stoichiometries (0.5-1.0
equiv bound), while the remaining four exhibit minimal binding
selectivity to TTR (<0.5 equiv).
under Evaluationa
Evaluating the Potent TTR Amyloidogenesis Inhibitors
for COX-1 Enzymatic Inhibition and Binding to the
Thyroid Hormone Nuclear Receptor. The 41 most potent TTR
aggregation inhibitors were further evaluated for their ability
to inhibit COX-1 enzymatic activity and for their binding to
the thyroid hormone nuclear receptor. These analyses were
contracted out to the Cerep laboratories in Redmond, WA (refer
to the Experimental Section for a detailed description of the
assay protocols).51,52,54,55 For the COX-1 inhibition analyses,
results represent the % inhibition of arachidonic acid conversion
to PGE2 due to competitive binding of test compound (10 µM)
to COX-1 (Figure 4, lower, black values). Of the 41 compounds
evaluated, all but four display <5% inhibition of COX-1
activity: compounds 4e, 4h, 7i, and 8c display slight to modest
(10-33%) COX-1 inhibition. For the thyroid hormone receptor
binding analyses, the % displacement of [125I]-labeled triiodo-
thyronine (T3, the primary thyroid hormone) was determined
from competitive binding of test compound (10 µM) to the
thyroid hormone receptor (Figure 4, red, italicized values). Of
the 41 compounds evaluated, five display >10% displacement
of T3 from the thyroid hormone nuclear receptor (3a, 4a, 5e,
6e, and 7a), the maximum activity observed being 28% for
compound 6e.
a (a) Ar-COCl, THF (20-95%); (b) Ar-CO2H, SOCl2, heat, concentrate,
then 4-amino-2,6-dibromophenol, THF (9-76%); (c) phthalic anhydride,
THF (4i, 66%); (d) BBr3, CH2Cl2 (19-87%); (e) LiOH ·H2O, THF, MeOH,
H2O (7h, 70%; 9h, 64%); (f) Sn, HCl/AcOH (4f, 93%; 7f, 41%; 9f, 86%).
occurred rapidly in THF and, after 1 h of mixing, the resultant
bisarylamides were precipitated by dilution with water. In many
cases, simple filtration and rinsing of the solid with saturated
sodium bicarbonate (no basic rinse in the case of 4i) resulted
in compounds of acceptable purity for biophysical evaluation
(>95% purity as determined by RP-HPLC and 1H NMR).
Compounds not meeting this criterion were further purified by
column chromatography over silica. Several compounds required
further substituent conversions (i.e., anisole or ester deprotection
or NO2 reduction to NH2), accomplished using standard
procedures (refer to Scheme 1 and the Experimental Section
and Supporting Information). In some cases, synthesis of the
appropriate benzoyl chloride building block was required,
accomplished by reacting the commercially available benzoic
acids with thionyl chloride at elevated temperatures, followed
by concentration in vacuo.
Small Molecule-Mediated Inhibition of WT-TTR Amyloido-
genesis. The ability of the bisarylamide library members to
inhibit WT-TTR aggregation was evaluated using the acid-
mediated aggregation assay used previously to identify optimal
aryl-X and linker-Y substructures.51,52 Briefly, physiologically
relevant concentrations of WT-TTR (3.6 µM) were incubated
with a candidate inhibitor (7.2 µM) prior to subjecting the
solution to partial denaturation conditions that enable ≈90%
of WT-TTR to aggregate in the absence of an inhibitor after
72 h (pH 4.4, 37 °C). The extent of TTR aggregation (Figure
3) was monitored by measuring sample turbidity in the presence
of the test compound, expressed as a percentage of TTR
aggregation observed in inhibitor-free samples (assigned to be
100% fibril formation). Of the 56 new compounds evaluated,
41 display increased inhibitor potencies relative to the parent
compound 1 (<25% fibril formation, values in red). Twenty-
eight of the benzamides completely inhibit TTR aggregation
(<5% fibril formation) within the 72 h time frame of the assay.
Binding Selectivity of Amyloidogenesis Inhibitors to
TTR in Human Blood Plasma. The 41 compounds displaying
increased inhibitor potencies relative to 1 (i.e., <25% fibril
formation) were further evaluated for their ability to bind
selectively to TTR in blood plasma using an ex vivo TTR
plasma binding selectivity assay.51-53 Briefly, the candidate
X-ray Crystallographic Analysis of WT-TTR Bound to
Inhibitors 2d, 3b, and 5d. Crystal structures of compounds
2d, 3b, and 5d in complex with WT-TTR were determined to
1.35, 1.31, and 1.35 Å resolution, respectively (Table 1). In all
three structures, the observed electron density allowed unam-
biguous placement of the kinetic stabilizer. The 3,5-Br2-4-
hydroxyphenyl substructure in 2d, 3b, and 5d bound in the outer
thyroid hormone binding pocket, as was the case for the parent
TTR·(1)2 structure (Figure 5).52 As anticipated, the bromine
substituents extend into HBP-1 and 1′, while the putative
phenolate makes electrostatic interactions with the Lys-15 and
+
15′ ε-NH3 groups. In all cases, at least one of the aryl-Z
substituents of 2d, 3b, and 5d positioned in the inner binding
cavity projected their methyl or chloro substituents into HBP-3
and 3′.
In the 2d and 3b structures, an ordered water molecule is
located between the Ser-117 hydroxyls of each T4 binding site,
which is not observed in the TTR ·(1)2 structure (Figure 5, red
dot(s)). In the TTR ·(5d)2 structure, this ordered water molecule
is absent and instead the 4-phenolic substituent creates a
hydrogen bonding network that bridges the Ser-117 and 117′
hydroxyls on adjacent TTR subunits. While additional ordered

Second Aryl Substructure in TTR Amyloidogenesis Inhibitors
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 4 1119
Figure 4. Thyroid hormone receptor T3 displacement and COX-1 inhibition for the most potent aggregation inhibitors (those displaying <25%
aggregation). The extent that the test compounds (10 µM) competitively bind to the thyroid hormone receptor and displace T3 is shown in red italics
(errors are typically less than (2 percentage points). COX-1 inhibition results are shown below in black, with values representing the % inhibition
by the test compounds (10 µM) of COX-1-mediated conversion of arachidonic acid to prostaglandin-E2 (errors are typically less than (6 percentage
points).
water molecules have been modeled elsewhere, these do not
make significant interactions with any of the bound inhibitors.
Taking the positioning of the parent ligand in the TTR ·(1)2
cocrystal structure as a reference, compound 2d appears to bind
∼0.4 Å deeper within the pocket, while compounds 3b and 5d
bind ∼0.3 Å and ∼1.4 Å shallower, respectively. The nonco-
planarity of the aryl-X and aryl-Z rings observed in the TTR ·(1)2
cocrystal structure was also observed for these new kinetic
stabilizers (46-51° for 2d, 41-44° for 3b, and 43-46° for 5d,
compared to 42-45° for 1).
into % inhibition. The PSave values are adjusted by adding 1 in
order to be able to differentiate inhibitors with plasma binding
stoichiometries of 0 from each other based on % inhibition data.
Division by 300% restricts the efficacy scores to a range between
0 and 1, the latter being the best score. Coconsideration of efficacy
and binding selectivity using eq 1 affords clear rank-orderings of
the different aryl-Z substituents and substitution patterns (Figure
3). Nearly all of the 2,6, 2,5, 2, 3,4,5, and 3,5 substitution patterns
afforded excellent TTR amyloidogenesis inhibitors that exhibited
selective binding to TTR in plasma ex vivo, in descending order.
The hydroxy, Cl, Br, CH3, F, and NH2 substituents generally
afforded excellent TTR amyloidogenesis inhibitors that also
exhibited selective TTR binding in plasma, in descending order.
Strictly analogous ranking of aryl-X substituents and substitution
patterns as well as optimal linkers now allows highly efficacious
and selective inhibitor structures to be predicted.
Discussion
Optimal Aryl-Z Substructures for Creating Potent
TTR Amyloidogenesis Inhibitors that also Exhibit High
TTR Binding Selectivity in Blood. To use the previously
employed substructure ranking calculation52 that coconsiders
inhibition efficacy and plasma TTR binding selectivity, we first
averaged the percent fibril formation (% FFave) and plasma TTR
binding stoichiometry (PSave) values across the columns, reflect-
ing different substitution patterns, and across rows, reflecting
the influence of a given substituent in several contexts.
Compounds that were not tested in the plasma selectivity assay
because they allowed >25% aggregation of TTR were assigned
plasma stoichiometry values of 0 for averaging purposes, as
experience demonstrates that poor aggregation inhibitors typi-
cally display little, if any, binding to TTR in blood plasma. To
obtain an “efficacy score” for each aryl-Z substituent and
substitution pattern, the % FFave and PSave values were then
input into eq 1.
Generally the Most Potent and Selective TTR Amyloido-
genesis Inhibitors Interact Minimally with the COX-1 Enzyme
and the Thyroid Hormone Receptor. Because many of the
potent TTR amyloidogenesis inhibitors in this aryl-Z optimiza-
tion library bear resemblance to some common nonsteroidal anti-
inflammatory drugs (NSAIDs), their ability to inhibit the COX-1
enzyme was evaluated. COX-1 binding is undesirable, especially
for familial amyloid cardiomyopathy patients, as NSAID activity
can further compromise renal blood flow.56 Of the 41 com-
pounds evaluated, four display some COX-1 inhibition (4e, 4h,
7i, and 8c).
Because these compounds all contain the thyroid hormone-
like 3,5-Br2-4-OH aryl-X ring, we evaluated the 41 most potent
kinetic stabilizers for their ability to bind to the thyroid hormone
nuclear receptor. Five compounds, the majority bearing OH or
F substitutents in distinct substitution patterns, displace >10%
of T3 from the thyroid hormone receptor (3a, 4a, 5e, 6e, and
(100% - %FFave) × (1 + PSave
)
efficacy score )
(1)
300%
Subtraction of the % FFave values from 100% converts the data

1120 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 4
Johnson et al.
Table 1. Data Collection and Refinement Statistics for the Crystal Structures of WT-TTR in Complex with Inhibitors 2d, 3b, and 5d
WT-TTR/2d
WT-TTR/3b
WT-TTR/5d
data collection
beamline
wavelength (Å)
resolution (Å)
space group
SSRL 9-2
SSRL 9-2
SSRL 9-2
0.9795
1.35 (1.35-1.36)
P21212
0.9795
0.9795
1.35 (1.35-1.40)a
P21212
1.31 (1.31-1.36)
P21212
a, b, c (Å)
42.65, 84.62, 64.94
2
42.50, 85.00, 64.74
2
353372 (16464)
54365 (3829)
6.5 (4.3)
95.1 (68.0)
3.4 (45.2)
39.1 (2.3)
42.58, 85.34, 64.28
2
358657 (16884)
55178 (4221)
6.5 (4.0)
96.4 (75.1)
4.0 (33.1)
36.6 (3.6)
no. of molecules in a.u.
no. of observations
no. of unique reflections
redundancy
346443 (22334)a
51708 (4558)a
6.7 (64.9)a
98.4 (88.1)a
3.7 (50.1)a
38.3 (2.3)a
completeness (%)
Rsym (%)b
average I/σ
refinement statistics
resolution (Å)
64.96-1.35
49039 (3100)a
2629 (173)a
64.68-1.31
51574 (2573)
2744 (142)
17.7
64.28-1.31
52337 (2930)
2794 (156)
17.7
no. of reflections (working set)
no. of reflections (test set)
Rcryst (%)a,c
17.3a c
,
Rfree (%)a,d
18.8a d
20.4
20.0
,
no. of protein/ligand/water atoms
891884/38/215
891884/38/208
891884/44/209
average B values
protein
ligand
16.6
19.0
20.8
15.2
19.3
18.4
14.9
19.4
17.5
Wilson B value
Ramachandran plot
most favored (%)
additionally allowed (%)
generously allowed (%)
disallowed (%)
92.1
7.9
0
92.1
7.9
0
92.1
7.9
0
0
0
0
rms deviations
bond lengths (Å)
angles (deg)
0.018
1.73
0.018
1.78
0.019
1.84
a Numbers in parentheses are for highest resolution shell of data. b Rsym ) ∑hkl |I - I | ∑hkl I. c Rcryst ) ∑hkl |Fo - Fc| ∑hkl Fo. d Rfree is the same as Rcryst
but for 5% of data excluded from the refinement.
,
7a). The random distribution of substitution patterns suggests
that this undesirable characteristic could be prevented by
employing different linkers and by minimizing the use of the
3,5-Br2-4-OH aryl-X ring.52 It might be prudent to pursue future
inhibitors bearing hydroxyl and fluoro substituents in meta
positions with caution.
in this study reveal that the aryl-X and aryl-Z rings are oriented
out-of-plane with respect to one another.9,34,38,57-62 With ortho
substituents, unbound inhibitors may have a steric predisposition
to the nonplanar bound conformations, thereby displaying more
favorable binding energetics.
Para-substitution, reflected in the 8 and 9 series, generally
exhibits decreased potency and selectivity, likely due to the
inability of these aryl-Z rings to engage the HBPs and because
of steric clashing with the Ser-117/117′ side chains when
positioned in the inner binding cavity. The inability of many of
these substituents to hydrogen bond with the Ser-117/117′
Structural Characterization of Potent and Selective
TTR Amyloidogenesis Inhibitors with Optimal Aryl-Z
Substitution Patterns. Structures of kinetic stabilizers 2d, 3b,
and 5d in complex with WT-TTR (Figure 5) support the
hypothesis that low pKa phenols, including 3,5-Br2-4-hydrox-
yphenyl substructures, bind in the outer cavity of the thyroxine
binding site where their phenolates interact with the Lys-15 and
+
hydroxyls in the inner binding cavity or the Lys-15/15′ ε-NH3
and/or Glu-54/54′ carboxyl groups when positioned in the outer
binding cavity could also reduce ligand binding affinity and thus
inhibitor potency. This hypothesis is consistent with the
observation that hydrogen bonding capable substituents such
as phenols (9a and 5b-e) and carboxyls (9i) are well tolerated
in this position.
It is clear that alkyl and halide substituents on the aryl-Z ring
produce the most potent and selective inhibitors, likely owing
to their hydrophobic and charge transfer interactions with the
halogen binding pockets, as described in more detail in the first
study and elsewhere.51,63,64 The hydroxyl substituent on the
aryl-Z ring also contributes to high potency and selectivity over
a range of substitution patterns, likely owing in large part to its
ability to hydrogen bond.
+
15′ side chain ε-NH3 groups, whereas high pKa phenols like
3,5-(CH3)2-4-hydroxyphenyl prefer to occupy the inner binding
cavity, H-bonding with the Ser-117 and 117′ hydroxyl groups.51,52
That 5d is one of the most potent and selective TTR amyloido-
genesis inhibitors discovered to date is not surprising, as it
incorporates both preferred subsite binding aryls. On the basis
of previous structural data51,52 and that obtained herein, it is
likely that the remaining compounds bind with their 3,5-Br2-
4-OH aryl ring occupying the outer binding cavity.
Even though the aryl-Z substituents under evaluation are fairly
diverse, every one is capable of increasing inhibitor potency
and selectivity if displayed in the proper substitution pattern.
Bisarylamides bearing ortho aryl-Z substituents are generally
more potent than their counterparts bearing meta and especially
para substituents, almost independent of the nature of the
substituent. Consistent with previously published TTR ·(inhib-
itor)2 X-ray cocrystal structures, the three structures determined
Concluding Remarks
The systematic optimization of the three substructures
comprising a typical transthyretin kinetic stabilizer (i.e., the two

Second Aryl Substructure in TTR Amyloidogenesis Inhibitors
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 4 1121
Figure 5. X-ray crystallography structures and schematic representations of key features of homotetrameric WT-TTR cocrystallized with inhibitors
1, 2d, 3b, and 5d (PDB accession codes 3CN4,52 3ESN, 3ESO, and 3ESP, respectively). The zoomed images show the respective inhibitors bound
in one binding site in their two symmetry-related binding modes (orange and gray), with the protein surface shown colored by electrostatic potential
(nb: the front portion has been cut away to allow easier visualization into the pocket). Schematic representations of each bound inhibitor are
presented as two-dimensional topology diagrams (generated using MOE (2006.08), Chemical Computing Group, Montreal, Canada). Inhibitors are
shown in only one of their two symmetry-related binding modes for clarity. Important hydrogen bonding and electrostatic interactions between
protein side chains and the ligands are indicated by arrows, with distances shown in Å. A graphical legend for interpretation of key binding site
characteristics is displayed below the TTR tetramer ribbon diagram. Ligand exposure indicates specific portions of the ligand structures that are
solvent accessible (i.e., not completely buried within the binding pocket). Residue exposure indicates those amino acids whose side chains and/or
peptide backbones are partially solvent accessible.
aromatic rings and the linker joining them) has enabled the rank
ordering of substructures that can be combined to produce potent
and selective inhibitors. Because most of these TTR kinetic
stabilizers are devoid of thyroid hormone receptor binding and
COX-1 inhibition activity, it is likely that kinetic stabilizers
derived from combinations of the best aryl and linker substruc-
tures will be as well, mitigating toxicity in the envisioned long-
term use of these drugs.51,52
The high-resolution X-ray crystallographic data obtained from
this three-part series supports the hypothesis that thyroid-
hormone-like, 3,5-X2-4-hydroxyphenyl aryl substructures (X )
halide) prefer to bind in the outer binding site when combined

1122 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 4
Johnson et al.
was added to a stirring mixture of the respective anisole (0.1-0.35
mmol, 1 equiv) in anhydrous CH2Cl2 (5-10 mL), and the reaction
was stirred at room temperature under an argon atmosphere. After
18 h, the reaction was quenched with MeOH (5 mL), extracted
into EtOAc (50 mL), and washed with 1 N HCl (25 mL), sat.
NaHCO3 (25 mL), and brine (25 mL). The organics were then dried
over Na2SO4, filtered, and concentrated. Compounds were purified
by flash chromatography over silica, employing a hexanes:EtOAc
elution system. All compounds were characterized by 1H NMR and
RP-HPLC and were >95% in purity. See the Supporting Informa-
tion for specific synthetic procedures and characterization data.
Representative Methyl Ester Hydrolysis Procedures. LiOH ·H2O
(0.5-0.85 mmol, ∼4 equiv) was added to a stirring mixture of the
respective methyl ester (0.13-0.2 mmol, 1 equiv) in H2O/MeOH/
THF (0.5/0.5/1.5 mL), and the reaction was stirred at room
temperature. After 18 h, the reaction was acidified with 1 N HCl
and extracted into EtOAc (50 mL), washed with H2O (2 × 25 mL)
and brine (25 mL), dried over Na2SO4, filtered, and concentrated.
with most other substituted aromatics, including phenols
exhibiting high pKas.51,52 These structural insights enable us to
move beyond simply screening collections of compounds to now
being able to confidently employ structure-based drug design
for the development of diverse, potent, and highly selective
transthyretin amyloidogenesis inhibitors. Use of the structural
data, along with the rankings of the three substructural elements
enabled by this three-part series, is currently motivating the
synthesis of inhibitors that should be exceptionally potent and
selective and thus potential clinical candidates. Upon the
identification of clinical candidates, it will be highly desirable
to evaluate their cellular and organismal toxicity.
Experimental Section
General Synthetic Methods. Unless otherwise stated, all
chemicals were purchased from commercial suppliers and used
without further purification. Reaction progress was monitored by
thin-layer chromatography on silica gel 60 F254 coated glass plates
(EM Sciences). Flash chromatography was performed using 230-400
mesh silica gel 60 (EM Sciences). NMR spectra were recorded on
a Bruker 500 MHz spectrometer. 1H NMR spectra were calibrated
to the DMSO-d6 solvent peak at 2.49 ppm. Reverse phase high
performance liquid chromatography (RP-HPLC) was performed
using a Waters 600 E multisolvent delivery system employing a
Waters 486 tunable absorbance detector and 717 autosampler.
Samples were chromatographically separated using a ThermoHy-
persil-Keystone Betabasic-18 column (model 71503-034630, 150
Å pore size, 3 µm particle size), eluting with a H2O:CH3CN gradient
solvent system. Linear gradients were run from either 100:0, 80:
20, or 60:40 A:B to 0:100 A:B (A ) 95:5 H2O:CH3CN, 0.25%
trifluoroacetic acid (TFA); B ) 5:95 H2O:CH3CN, 0.25% TFA).
Final compound purities were additionally evaluated under distinct
RP-HPLC conditions by chromatographically separating samples
using a Vydac-C4 column (model 214TP5415, 300 Å pore size, 5
µm particle size), eluting with a H2O:MeOH gradient solvent
system. Linear gradients were run from 100:0 to 0:100 C:D (C )
99.75% H2O, 0.25% TFA; D ) 100% MeOH). Representative
synthetic procedures are presented below, while experimental details
for compounds synthesized analogously are presented in the
Supporting Information.
Representative Amide Coupling Procedures: Method A.
4-Amino-2,6-dibromophenol (∼0.3-0.7 mmol, 1.0 equiv) was
mixed with the respective substituted benzoyl chloride (∼0.3-0.7
mmol, ∼1-1.2 equiv) in THF (∼0.5-2.0 mL) at ambient temper-
ature. After ∼1 h, the reaction mixtures were diluted with H2O
(∼20 mL), sonicated, and the resulting precipitates were filtered,
rinsed with H2O, collected, sonicated with sat. NaHCO3 (∼10 mL),
filtered, rinsed with H2O, and collected. When necessary, com-
pounds were further purified by flash chromatography over silica,
employing a hexanes:EtOAc elution system. All compounds were
characterized by 1H NMR and RP-HPLC and were >95% in purity.
See the Supporting Information for specific synthetic procedures
and characterization data.
1
Compounds were characterized by H NMR and RP-HPLC and
were >95% in purity. See the Supporting Information for specific
synthetic procedures and characterization data.
Representative NO2 to NH2 Reduction Procedures. Tin powder
(0.6-2.2 mmol, 4.0-4.7 equiv) was added to a stirring mixture of
the respective nitro compound (0.13-0.55 mmol, 1 equiv) in an
HCl/AcOH mixture (0.2/2.0 mL). After 18 h, the reaction was
diluted with H2O (25 mL), neutralized with NaHCO3, extracted
into EtOAc (50 mL), and the organics were washed with H2O (25
mL) and brine (25 mL), dried over Na2SO4, filtered, and concen-
trated. Compounds were purified by flash chromatography over
silica, employing a hexanes:EtOAc elution system. All compounds
were characterized by 1H NMR and RP-HPLC and were >95% in
purity. See the Supporting Information for specific synthetic
procedures and characterization data.
Evaluating Small Molecule-Mediated Inhibition of WT-TTR
Amyloidogenesis. WT-TTR was purified from an Escherichia coli
expression system as described previously.65 To a 495 µL aliquot
of 0.4 mg/mL WT-TTR (7.2 µM, 10 mM phosphate, pH 7.0, 100
mM KCl, 1 mM EDTA) in a disposable cuvette was added 5.0 µL
of the desired small molecule in DMSO (1.44 mM) and the sample
was vortexed briefly. After incubating for 30 min at ambient
temperature, the pH of the sample was lowered to 4.4 by addition
of 500 µL of acidic buffer (100 mM acetate, pH 4.2, 100 mM KCl,
1 mM EDTA), and the solution was briefly vortexed and incubated
in the dark for 72 h at 37 °C (final protein and inhibitor
concentrations were 3.6 and 7.2 µM, respectively). The sample was
then vortexed to evenly distribute any precipitate, and the turbidity
was measured at 400 nm in a Hewlett-Packard model 8453 UV-vis
spectrophotometer. The extent of WT-TTR aggregation (% fibril
formation, FF; % inhibition ) 100 - %FF) was determined by
comparing the sample turbidity in the presence of small molecule
as a percentage of control WT-TTR sample incubated with 5.0 µL
of pure DMSO (representing 100% aggregation, 0% inhibition).
All samples were performed in at least quintuplicate, with average
values obtained presented in Figure 3 (errors are typically less than
(5 percentage points).
Representative Amide Coupling Procedures: Method B. Sub-
stituted benzoic acids (∼0.3-0.7 mmol, ∼1-1.2 equiv) were stirred
in SOCl2 (∼2.0 mL) at ∼70-80 °C. After ∼1 h, the reactions were
concentrated and then the resulting acid chlorides were mixed with
4-amino-2,6-dibromophenol (∼0.3-0.7 mmol, 1.0 equiv) in THF
(∼0.5-2.0 mL) at ambient temperature. After ∼1 h, the reaction
mixtures were diluted with H2O (∼20 mL), sonicated, and the
resulting precipitates were filtered, rinsed with H2O, collected,
sonicated with sat. NaHCO3 (∼10 mL), filtered, rinsed with H2O,
and collected. When necessary, compounds were further purified
by flash chromatography over silica, employing a hexanes:EtOAc
elution system. All compounds were characterized by 1H NMR and
RP-HPLC and were >95% in purity. See the Supporting Informa-
tion for specific synthetic procedures and characterization data.
Representative Anisole Deprotection Procedures. Boron tri-
bromide (0.5-3.5 mmol of 1 M BBr3 in hexanes, 5-10 equiv)
Evaluating the Binding Selectivity of Amyloidogenesis
Inhibitors to TTR in Human Blood Plasma. The procedure for
the antibody capture method to evaluate the stoichiometry of
inhibitor bound to TTR in human blood plasma has been described
in detail elsewhere.53 Briefly, 7.5 µL of a 1.44 mM DMSO solution
of test compound was incubated with human blood plasma (1.0
mL) in a 2 mL microfuge tube at 37 °C for 24 h on a rocker (30
rpm). Then, 125 µL of a 1:1 v/v slurry of unfunctionalized sepharose
resin in TSA (10 mM Tris, pH 8.0, 140 mM NaCl) was added and
the mixture was incubated for another hour at 4 °C on a rocker (18
rpm). The sample was then centrifuged and the supernatant divided
into 2 aliquots of 400 µL each, which were each added to 200 µL
of a 1:1 v/v slurry of sepharose resin conjugated to an anti-TTR
antibody in TSA, and the sample was incubated again at 4 °C for
20 min on a rocker (18 rpm). The sample was then centrifuged,
the supernatant removed, and the TTR bound resin was washed

Second Aryl Substructure in TTR Amyloidogenesis Inhibitors
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 4 1123
three times (10 min each wash) with 1 mL of TSA/0.05% saponin,
then twice (10 min each wash) with 1 mL of TSA at 4 °C on a
rocker (18 rpm). After centrifugation and removal of the final
supernatant, dissociation of the TTR and bound test compound from
the resin-bound antibody was achieved through addition of 155 µL
of aqueous triethylamine (100 mM, pH ∼ 11.5) and rocking (18
rpm) at 4 °C for 30 min. The suspension was then centrifuged and
145 µL of the supernatant containing both TTR and test compound
was removed, neutralized with 0.84 mL of glacial acetic acid (to
prevent peak bleeding during HPLC analysis), and then analyzed
by RP-HPLC to determine the stoichiometry of small molecule
bound to TTR (the test compound-TTR complex dissociates and
the small molecule and protein are chromatographically separable
under the HPLC buffer conditions). Quantification of test compound
and TTR is achieved by comparing the integrated peak areas to
standard curves: the ratio of the amount of test compound to TTR
yields the binding stoichiometry, of which a theoretical maximum
value of 2 is possible owing to the two T4 binding sites per TTR
tetramer. Analyses were performed in duplicate from at least two
different blood plasma samples (i.e., at least four analyses), with
average values obtained presented in Figure 3 (errors are typically
less than (0.1).
Evaluating the Inhibition of COX-1 Enzymatic Activity by
the Potent TTR Amyloidosis Inhibitors. The evaluation of the
most potent TTR aggregation inhibitors (displaying <25% fibril
formation) at inhibiting COX-1 activity was contracted out to the
Cerep laboratories in Redmond, WA. Compound analyses were
performed using assay catalog reference no. 777-1hr, which uses
procedures developed by Glaser et al.54 A brief experimental
protocol as provided by Cerep is outlined below. The enzyme (∼2
µg) was preincubated in the absence (water control) or presence of
test compound (10.0 µM) for 20 min at 22 °C in 250 µL of buffer
(100 mM Tris-HCl, pH 8, 2 mM phenol, 1 µM hematine).
Arachidonic acid (4 µM) was then added to initiate the reaction
(no arachidonic acid added for basal control measurements). After
incubation at 22 °C for 10 min, the reaction was quenched by
addition of 2 M HCl and 1 M Tris-HCl (pH 7.8) and cooling at 4
°C. Prostaglandin-E2 (PGE2) quantification was performed using
an EIA detection kit with measurements made using a microplate
reader. Values of analyses performed in singleton are presented in
Figure 4 (black values), which represent the % inhibition of
arachidonic acid conversion to prostaglandin-E2 due to competitive
binding of test compound to COX-1 (previous studies typically
show representative errors in the data of less than (6 percentage
points).51,52 Control analyses were performed analogously, with
the standard inhibitory reference compound diclofenac tested at
several concentrations to obtain an inhibition curve from which its
IC50 is calculated (12 nM).
Evaluating the Binding of the Potent TTR Amyloidosis
Inhibitors to the Thyroid Hormone Nuclear Receptor. The
evaluation of the most potent TTR aggregation inhibitors (displaying
<25% fibril formation) at binding to the thyroid hormone receptor
was contracted out to the Cerep laboratories in Redmond, WA.
Compound analyses were performed using assay catalog reference
no. 855, which uses procedures developed by Inoue et al.55 The
experimental protocol as provided by Cerep is briefly outlined
below. Liver membrane homogenates (100 µg protein) were
incubated for 18 h at 4 °C with 0.1 nM 125I-labeled triiodothyronine
([125I]T3, the primary thyroid hormone) in the absence or presence
of test compound (10.0 µM) in 500 µL of buffer (20 mM Tris-
HCl, pH 7.6, 50 mM NaCl, 2 mM EDTA, 10% glycerol, and 5
mM ꢀ-mercaptoethanol). The samples were then vacuum filtered
through glass fiber filters (GF/B, Packard), rinsed several times with
ice-cold buffer (50 mM Tris-HCl and 150 mM NaCl), and the filters
were dried and counted for radioactivity in a scintillation counter
(Topcount, Packard) using a scintillation cocktail (Microscint 0,
Packard). Nonspecific binding, determined in the presence of 1 µM
T3, was subtracted from the [125I]T3 binding results. Values of
analyses performed in singleton are presented in Figure 4 (red
italicized values), which represent the % displacement of [125I]T3
due to competitive binding of test compound to the thyroid hormone
receptor (previous studies typically show representative errors in
the data of less than (2 percentage points).51,52 Control analyses
were performed analogously with T3 tested at several concentrations
to obtain a competition curve from which its IC50 is calculated (0.38
nM).
X-ray Crystallographic Analysis of WT-TTR Bound to Inhibi-
tors 2d, 3b, and 5d. WT-TTR was purified from an E. coli
expression system as described previously.65 The protein was
concentrated to 4 mg/mL in 10 mM NaPi, 100 mM KCl, at pH 7.6
and cocrystallized at room temperature with various inhibitors at a
5 molar excess using the vapor-diffusion sitting drop method.
Crystals were grown from 1.395 M sodium citrate, 3.5% v/v
glycerol at pH 5.5. The crystals were cryoprotected with inhibitor-
free 1.395 M sodium citrate, 10% v/v glycerol at pH 5.5. Data
were collected on beam line 9-2 at the Stanford Synchrotron
Radiation Laboratory (SSRL) at a wavelength of 0.9795 Å. Data
sets were integrated and scaled using HKL2000.66 The crystals were
indexed in orthorhombic space group P21212 with two subunits per
asymmetric unit (see Table 1 for full data collection and refinement
statistics). The three crystal structures were determined by molecular
replacement using the 2FBR67 coordinates in the program Phaser.68
Further model building and refinement were completed using Coot69
and Refmac.70 For all structures, hydrogens were added during
refinement and anisotropic B values calculated. Final models were
validated using the JCSG quality control server (http://jcsgsrv2/
QC) incorporating Molprobity,71 ADIT (http://rcsb-deposit.rut-
gers.edu/validate) WHATIF,72 Resolve,73 and Procheck.74 Full data
collection and refinement statistics are presented in Table 1.
Acknowledgment. We are grateful for the financial support
of the NIH DK46335 (J.W.K), CA58896 and AI42266 (I.A.W),
as well as the Skaggs Institute for Chemical Biology and the
Lita Annenberg Hazen Foundation. Assistance with the thyroid
hormone receptor and COX-I binding assays by the Cerep
laboratories (Redmond, WA, and France) and Richard Labau-
diniere of FoldRx Pharmaceuticals (Boston, MA), as well as
the technical expertise of Ted Foss, M. T. Dendle, and Mike
Saure, are also greatly appreciated. X-ray diffraction data were
collected at beam line 9-2 at the Stanford Synchrotron
Radiation Laboratory (SSRL), a national user facility operated
by Stanford University on behalf of the U.S. Department of
Energy, Office of Basic Energy Sciences. The SSRL Structural
Molecular Biology Program is supported by the Department of
Energy, Office of Biological and Environmental Research, and
by the National Institutes of Health, National Center for
Research Resources, Biomedical Technology Program, and the
National Institute of General Medical Sciences. We also thank
Drs. Xiaoping Dai, Andre Schiefner, and Xiaojin Xu in the
Wilson laboratory for assistance with data collection.
Supporting Information Available: Detailed synthesis and
characterization data on additional compounds prepared analogously
to the general synthetic procedures presented in the Experimental
Section, tabulation of compound purity as determined by C4 and
C18 RP-HPLC, and calculated phenolic pKa values. This material
is available free of charge via the Internet at http://pubs.acs.org.
References
(1) Balch, W. E.; Morimoto, R. I.; Dillin, A.; Kelly, J. W. Adapting
proteostasis for disease intervention. Science 2008, 319, 916–919.
(2) Zhang, Q.; Powers, E. T.; Nieva, J.; Huff, M. E.; Dendle, M. A.;
Bieschke, J.; Glabe, C. G.; Eschenmoser, A.; Wentworth, P., Jr.;
Lerner, R. A.; Kelly, J. W. Metabolite-initiated protein misfolding may
trigger Alzheimer’s disease. Proc. Natl. Acad. Sci. U.S.A 2004, 101,
4752–4757.
(3) Cohen, E.; Bieschke, J.; Perciavalle, R. M.; Kelly, J. W.; Dillin, A.
Opposing activities protect against age-onset proteotoxicity. Science
2006, 313, 1604–1610.
(4) Mu, T. W.; Fowler, D. M.; Kelly, J. W. Partial restoration of mutant
enzyme homeostasis in three distinct lysosomal storage disease cell
lines by altering calcium homeostasis. PLoS Biol. 2008, 6, e26.

1124 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 4
Johnson et al.
(5) Morley, J. F.; Brignull, H. R.; Weyers, J. J.; Morimoto, R. I. The
threshold for polyglutamine-expansion protein aggregation and cellular
toxicity is dynamic and influenced by aging in Caenorhabditis elegans.
Proc. Natl. Acad. Sci. U.S.A 2002, 99, 10417–10422.
(29) Tojo, K.; Sekijima, Y.; Kelly, J. W.; Ikeda, S. Diflunisal stabilizes
familial amyloid polyneuropathy-associated transthyretin variant tet-
ramers in serum against dissociation required for amyloidogenesis.
Neurosci. Res. 2006, 56, 441–449.
(30) Sekijima, Y.; Dendle, M. A.; Kelly, J. W. Orally administered diflunisal
stabilizes transthyretin against dissociation required for amyloidogen-
esis. Amyloid 2006, 13, 236–249.
(31) Blake, C. C.; Geisow, M. J.; Oatley, S. J.; Rerat, B.; Rerat, C. Structure
of prealbumin: secondary, tertiary and quaternary interactions deter-
mined by Fourier refinement at 1.8 A. J. Mol. Biol. 1978, 121, 339–
356.
(6) Gidalevitz, T.; Ben-Zvi, A.; Ho, K. H.; Brignull, H. R.; Morimoto,
R. I. Progressive disruption of cellular protein folding in models of
polyglutamine diseases. Science 2006, 311, 1471–1474.
(7) Dobson, C. M. Protein folding and misfolding. Nature 2003, 426, 884–
890.
(8) Selkoe, D. J. Folding proteins in fatal ways. Nature 2003, 426, 900–
904.
(9) 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 amyloidoses. Acc. Chem. Res. 2005, 38, 911–921.
(32) Hornberg, A.; Eneqvist, T.; Olofsson, A.; Lundgren, E.; Sauer-
Eriksson, A. E. A comparative analysis of 23 structures of the
amyloidogenic protein transthyretin. J. Mol. Biol. 2000, 302, 649–
669.
(10) Cohen, F. E.; Kelly, J. W. Therapeutic approaches to protein-misfolding
(33) 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.
(34) 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.
diseases. Nature 2003, 426, 905–909.
(11) Kelly, J. W.; Colon, W.; Lai, Z.; Lashuel, H. A.; McCulloch, J.;
McCutchen, S. L.; Miroy, G. J.; Peterson, S. A. Transthyretin
quaternary and tertiary structural changes facilitate misassembly into
amyloid. AdV. Protein Chem. 1997, 50, 161–181.
(12) Westermark, P.; Bergstrom, J.; Solomon, A.; Murphy, C.; Sletten, K.
Transthyretin-derived senile systemic amyloidosis: clinicopathologic
and structural considerations. Amyloid 2003, 10 (Suppl 1), 48–54.
(13) Westermark, P.; Sletten, K.; Johansson, B.; Cornwell, G. G., III. Fibril
in senile systemic amyloidosis is derived from normal transthyretin.
Proc. Natl. Acad. Sci. U.S.A. 1990, 87, 2843–2845.
(14) Buxbaum, J. N.; Tagoe, C. E. The genetics of the amyloidoses. Annu.
ReV. Med. 2000, 51, 543–569.
(15) Jiang, X.; Buxbaum, J. N.; Kelly, J. W. The V122I cardiomyopathy
variant of transthyretin increases the velocity of rate-limiting tetramer
dissociation, resulting in accelerated amyloidosis. Proc. Natl. Acad.
Sci. U.S.A. 2001, 98, 14943–14948.
(35) Foss, T. R.; Kelker, M. S.; Wiseman, R. L.; Wilson, I. A.; Kelly,
J. W. Kinetic stabilization of the native state by protein engineering:
Implications for inhibition of transthyretin amyloidogenesis. J. Mol.
Biol. 2005, 347, 841–854.
(36) Foss, T. R.; Wiseman, R. L.; Kelly, J. W. The pathway by which the
tetrameric protein transthyretin dissociates. Biochemistry 2005, 44,
15525–15533.
(37) Hammarstrom, P.; Wiseman, R. L.; Powers, E. T.; Kelly, J. W.
Prevention of transthyretin amyloid disease by changing protein
misfolding energetics. Science 2003, 299, 713–716.
(38) Wiseman, R. L.; Johnson, S. M.; Kelker, M. S.; Foss, T.; Wilson,
I. A.; Kelly, J. W. Kinetic stabilization of an oligomeric protein by a
single ligand binding event. J. Am. Chem. Soc. 2005, 127, 5540–5551.
(16) Jacobson, D. R.; Pastore, R. D.; Yaghoubian, R.; Kane, I.; Gallo, G.;
Buck, F. S.; Buxbaum, J. N. Variant-sequence transthyretin (isoleucine
122) in late-onset cardiac amyloidosis in black Americans. N. Engl.
J. Med. 1997, 336, 466–473.
(39) Green, N. S.; Foss, T. R.; Kelly, J. W. Genistein, a natural product
from soy, is a potent inhibitor of transthyretin amyloidosis. Proc. Natl.
Acad. Sci. U.S.A 2005, 102, 14545–14550.
(17) Plante-Bordeneuve, V.; Said, G. Transthyretin related familial amyloid
polyneuropathy. Curr. Opin. Neurol. 2000, 13, 569–573.
(18) Gambetti, P.; Russo, C. Human brain amyloidoses. Nephrol. Dial.
Transplant. 1998, 13 (Suppl 7), 33–40.
(19) Sekijima, Y.; Hammarstrom, P.; Matsumura, M.; Shimizu, Y.; Iwata,
M.; Tokuda, T.; Ikeda, S.; Kelly, J. W. Energetic characteristics of
the new transthyretin variant A25T may explain its atypical central
nervous system pathology. Lab. InVest. 2003, 83, 409–417.
(20) Hammarstrom, P.; Sekijima, Y.; White, J. T.; Wiseman, R. L.; Lim,
A.; Costello, C. E.; Altland, K.; Garzuly, F.; Budka, H.; Kelly, J. W.
D18G transthyretin is monomeric, aggregation prone, and not detect-
able in plasma and cerebrospinal fluid: a prescription for central
nervous system amyloidosis. Biochemistry 2003, 42, 6656–6663.
(21) Sekijima, Y.; Wiseman, R. L.; Matteson, J.; Hammarstrom, P.; Miller,
S. R.; Sawkar, A. R.; Balch, W. E.; Kelly, J. W. The biological and
chemical basis for tissue-selective amyloid disease. Cell 2005, 121,
73–85.
(22) Hurshman Babbes, A. R.; Powers, E. T.; Kelly, J. W. Quantification
of the thermodynamically linked quaternary and tertiary structural
stabilities of transthyretin and its disease-associated variants: the
relationship between stability and amyloidosis. Biochemistry 2008, 47,
6969–6984.
(23) Wiseman, R. L.; Powers, E. T.; Buxbaum, J. N.; Kelly, J. W.; Balch,
W. E. An adaptable standard for protein export from the endoplasmic
reticulum. Cell 2007, 131, 809–821.
(24) Hammarstrom, P.; Jiang, X.; Hurshman, A. R.; Powers, E. T.; Kelly,
J. W. Sequence-dependent denaturation energetics: A major determi-
nant in amyloid disease diversity. Proc. Natl. Acad. Sci. U.S.A 2002,
99, 16427–16432.
(40) Petrassi, H. M.; Johnson, S. M.; Purkey, H.; Chiang, K. P.; Walkup,
T.; Jiang, X.; Powers, E. T.; Kelly, J. W. Potent and selective structure-
based dibenzofuran inhibitors of transthyretin amyloidogenesis: Kinetic
stabilization of the native state. J. Am. Chem. Soc. 2005, 127, 6662–
6671.
(41) Razavi, H.; Powers, E. T.; Purkey, H. E.; Adamski-Werner, S. L.;
Chiang, K. P.; Dendle, M. T.; Kelly, J. W. Design, synthesis, and
evaluation of oxazole transthyretin amyloidogenesis inhibitors. Bioorg.
Med. Chem. Lett. 2005, 15, 1075–1078.
(42) Miller, S. R.; Sekijima, Y.; Kelly, J. W. Native state stabilization by
NSAIDs inhibits transthyretin amyloidogenesis from the most common
familial disease variants. Lab. InVest. 2004, 84, 545–552.
(43) Oza, V. B.; Petrassi, H. M.; Purkey, H. E.; Kelly, J. W. Synthesis and
evaluation of anthranilic acid-based transthyretin amyloid fibril inhibi-
tors. Bioorg. Med. Chem. Lett. 1999, 9, 1–6.
(44) Baures, P. W.; Oza, V. B.; Peterson, S. A.; Kelly, J. W. Synthesis
and evaluation of inhibitors of transthyretin amyloid formation based
on the nonsteroidal anti-inflammatory drug, flufenamic acid. Bioorg.
Med. Chem. 1999, 7, 1339–1347.
(45) 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.
(46) Baures, P. W.; Peterson, S. A.; Kelly, J. W. Discovering transthyretin
amyloid fibril inhibitors by limited screening. Bioorg. Med. Chem.
1998, 6, 1389–1401.
(47) Miroy, G. J.; Lai, Z.; Lashuel, H. A.; Peterson, S. A.; Strang, C.; Kelly,
J. W. Inhibiting transthyretin amyloid fibril formation via protein
stabilization. Proc. Natl. Acad. Sci. U.S.A. 1996, 93, 15051–15056.
(25) Holmgren, G.; Ericzon, B. G.; Groth, C. G.; Steen, L.; Suhr, O.;
Andersen, O.; Wallin, B. G.; Seymour, A.; Richardson, S.; Hawkins,
P. N. Clinical improvement and amyloid regression after liver
transplantation in hereditary transthyretin amyloidosis. Lancet 1993,
341, 1113–1116.
(48) Hammarstrom, P.; Schneider, F.; Kelly, J. W. Trans-suppression of
misfolding in an amyloid disease. Science 2001, 293, 2459–2462.
(49) Coelho, T.; Carvalho, M.; Saraiva, M. J.; Alves, I.; Almeida, M. R.;
Costa, P. P. A strikingly benign evolution of FAP in an individual
found to be a compound heterozygote for two TTR mutations: TTR
Met 30 and TTR Met 119. J. Rheumatol. 1993, 20, 179.
(50) Coelho, T.; Chorao, R.; Sausa, A.; Alves, I.; Torres, M. F.; Saraiva,
M. J. Compound heterozygotes of transthyretin Met30 and transthyretin
Met119 are protected from the devastating effects of familial amyloid
polyneuropathy. Neuromusc. Disord. 1996, 6, 27.
(26) Tan, S. Y.; Pepys, M. B.; Hawkins, P. N. Treatment of amyloidosis.
Am. J. Kidney Dis. 1995, 26, 267–285.
(27) Suhr, O. B.; Herlenius, G.; Friman, S.; Ericzon, B. G. Liver
transplantation for hereditary transthyretin amyloidosis. LiVer Trans-
plant. 2000, 6, 263–276.
(28) Olofsson, B. O.; Backman, C.; Karp, K.; Suhr, O. B. Progression of
cardiomyopathy after liver transplantation in patients with familial
amyloidotic polyneuropathy, Portuguese type. Transplantation 2002,
73, 745–751.
(51) Johnson, S. M.; Connelly, S.; Wilson, I. A.; Kelly, J. W. Biochemical
and structural evaluation of highly selective 2-arylbenzoxazole-based
transthyretin amyloidogenesis inhibitors. J. Med. Chem. 2008, 51, 260–
270.

Second Aryl Substructure in TTR Amyloidogenesis Inhibitors
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 4 1125
(52) Johnson, S. M.; Connelly, S.; Wilson, I. A.; Kelly, J. W. Toward
optimization of the linker substructure common to transthyretin
amyloidogenesis inhibitors using biochemical and structural studies.
J. Med. Chem. 2008, 51, 6348–6358.
(53) Purkey, H. E.; Dorrell, M. I.; Kelly, J. W. Evaluating the binding
selectivity of transthyretin amyloid fibril inhibitors in blood plasma.
Proc. Natl. Acad. Sci. U.S.A. 2001, 98, 5566–5571.
(54) Glaser, K.; Sung, M. L.; O’Neill, K.; Belfast, M.; Hartman, D.; Carlson,
R.; Kreft, A.; Kubrak, D.; Hsiao, C. L.; Weichman, B. Etodolac
selectively inhibits human prostaglandin G/H synthase 2 (PGHS-2)
versus human PGHS-1. Eur. J. Pharmacol. 1995, 281, 107–111.
(55) Inoue, A.; Yamakawa, J.; Yukioka, M.; Morisawa, S. Filter-binding
assay procedure for thyroid hormone receptors. Anal. Biochem. 1983,
134, 176–183.
(56) Epstein, M. Nonsteroidal anti-inflammatory drugs and the continuum
of renal dysfunction. J. Hypertens. Suppl 2002, 20 (Suppl.), S17–
S23.
(57) Johnson, S. M.; Petrassi, H. M.; Palaninathan, S. K.; Mohamedmo-
haideen, N. N.; Purkey, H.; Nichols, C.; Chiang, K. P.; Walkup, T.;
Sacchettini, J. C.; Sharpless, K. B.; Kelly, J. W. Bisaryloxime ethers
as potent inhibitors of transthyretin amyloid fibril formation. J. Med.
Chem. 2005, 48, 1576–1587.
(58) Oza, V. B.; Smith, C.; Raman, P.; Koepf, E. K.; Lashuel, H. A.;
Petrassi, H. M.; Chiang, K. P.; Powers, E. T.; Sachettinni, J.; Kelly,
J. W. Synthesis, structure, and activity of diclofenac analogues as
transthyretin amyloid fibril formation inhibitors. J. Med. Chem. 2002,
45, 321–332.
(59) 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.
(62) Petrassi, H. M.; Klabunde, T.; Sacchettini, J. C.; Kelly, J. W. Structure-
based design of N-phenyl phenoxazine transthyretin amyloid fibril
inhibitors. J. Am. Chem. Soc. 2000, 122, 2178–2192.
(63) 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.
(64) 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.
(65) Lashuel, H. A.; Wurth, C.; Woo, L.; Kelly, J. W. The most pathogenic
transthyretin variant, L55P, forms amyloid fibrils under acidic condi-
tions and protofilaments under physiological conditions. Biochemistry
1999, 38, 13560–13573.
(66) Otwinowski, Z.; Minor, W. Processing of X-ray diffraction data
collected in oscillation mode. Methods Enzymol. 1997, 276, 307–326.
(67) Green, N. S.; Palaninathan, S. K.; Sacchettini, J. C.; Kelly, J. W.
Synthesis and characterization of potent bivalent amyloidosis inhibitors
that bind prior to transthyretin tetramerization. J. Am. Chem. Soc. 2003,
125, 13404–13414.
(68) Storoni, L. C.; McCoy, A. J.; Read, R. J. Likelihood-enhanced fast
rotation functions. Acta Crystallogr, Sect. D: Biol. Crystallogr. 2004,
60, 432–438.
(69) Emsley, P.; Cowtan, K. Coot: model-building tools for molecular
graphics. Acta Crystallogr., Sect. D: Biol. Crystallogr. 2004, 60, 2126–
2132.
(70) Murshudov, G. N.; Vagin, A. A.; Dodson, E. J. Refinement of
macromolecular structures by the maximum-likelihood method. Acta
Crystallogr., Sect. D: Biol. Crystallogr. 1997, 53, 240–255.
(71) Lovell, S. C.; Davis, I. W.; Arendall, W. B., III; de Bakker, P. I.;
Word, J. M.; Prisant, M. G.; Richardson, J. S.; Richardson, D. C.
Structure validation by Calpha geometry: phi,psi and Cbeta deviation.
Proteins 2003, 50, 437–450.
(72) Vriend, G. WHAT IF: a molecular modeling and drug design program.
J. Mol. Graphics 1990, 8, 52–56.
(73) Terwilliger, T. C. Automated main-chain model building by template
matching and iterative fragment extension. Acta Crystallogr., Sect.
D: Biol. Crystallogr. 2003, 59, 38–44.
(74) Laskowski, R.; MacArthur, M.; Moss, D.; Thornton, J. PROCHECK:
A program to check the stereochemical quality of protein structures.
J. Appl. Crystallogr. 1993, 26, 283–291.
(60) 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 amyloidogenesis:
rationalizing rodent PCB toxicity. Chem. Biol. 2004, 11, 1719–1728.
(61) Razavi, H.; Palaninathan, S. K.; Powers, E. T.; Wiseman, R. L.; Purkey,
H. E.; Mohamedmohaideen, N. N.; Deechongkit, S.; Chiang, K. P.;
Dendle, M. T.; Sacchettini, J. C.; Kelly, J. W. Benzoxazoles as
transthyretin amyloid fibril inhibitors: synthesis, evaluation, and
mechanism of action. Angew. Chem., Int. Ed. 2003, 42, 2758–2761.
JM801347S