Monoaryl derivatives as transthyretin fibril formation inhibitors: Design, synthesis, biological evaluation and structural analysis

Article abstract of DOI:10.1016/j.bmc.2020.115673

Transthyretin (TTR) is a ?-sheet-rich homotetrameric protein that transports thyroxine (T4) and retinol both in plasma and in cerebrospinal fluid. TTR also interacts with amyloid-β, playing a protective role in Alzheimer's disease. Dissociation of the nat

Full text of DOI:10.1016/j.bmc.2020.115673

Journal Pre-proofs
Monoaryl derivatives as Transthyretin fibril formation inhibitors: design, syn‐
thesis, biological evaluation and structural analysis
Ciccone Lidia, Nencetti Susanna, Tonali Nicolo, Fruchart-Gaillard Carole,
Shepard William, Nuti Elisa, Camodeca Caterina, Rossello Armando,
Orlandini Elisabetta
PII:
DOI:
Reference:
S0968-0896(20)30503-4
https://doi.org/10.1016/j.bmc.2020.115673
BMC 115673
To appear in:
Bioorganic & Medicinal Chemistry
Received Date:
Revised Date:
Accepted Date:
6 May 2020
21 July 2020
24 July 2020
Please cite this article as: Lidia, C., Susanna, N., Nicolo, T., Carole, F-G., William, S., Elisa, N., Caterina, C.,
Armando, R., Elisabetta, O., Monoaryl derivatives as Transthyretin fibril formation inhibitors: design, synthesis,
biological evaluation and structural analysis, Bioorganic & Medicinal Chemistry (2020), doi: https://doi.org/
10.1016/j.bmc.2020.115673
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover
page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version
will undergo additional copyediting, typesetting and review before it is published in its final form, but we are
providing this version to give early visibility of the article. Please note that, during the production process, errors
may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
© 2020 Published by Elsevier Ltd.

Monoaryl derivatives as Transthyretin fibril formation inhibitors: design,
synthesis, biological evaluation and structural analysis
Ciccone Lidia,1,2 Nencetti Susanna,1* Tonali Nicolo,3 Fruchart-Gaillard Carole,4 Shepard William,2
Nuti Elisa,1 Camodeca Caterina,1 Rossello Armando,1,5 Orlandini Elisabetta6,5*
1. Department of Pharmacy, University of Pisa, Via Bonanno 6, 56126, Pisa, Italy.
2. Synchrotron SOLEIL, L’Orme des Merisiers, Saint-Aubin, BP 48, 91192, Gif-sur-Yvette,
France.
3. Université Paris-Saclay, CNRS, BioCIS, rue Jean-Baptiste Clément 5, 92290 Châtenay-
Malabry, France
4. Université Paris Saclay, CEA, INRAE, Département Médicaments et Technologies pour la
Santé (DMTS), SIMoS, 91191 Gif-sur-Yvette, France
5. Research Center “E. Piaggio”, Università di Pisa, Pisa, 56122, Italy
6. Department of Earth Sciences, University of Pisa, Via Santa Maria 53-55, 56100, Pisa, Italy
Abstract
Transthyretin (TTR) is a ß-sheet-rich homotetrameric protein that transports thyroxine (T4) and
retinol both in plasma and in cerebrospinal fluid. TTR also interacts with amyloid-β, playing a
protective role in Alzheimer’s disease. Dissociation of the native transthyretin (TTR) tetramer is
widely accepted as the critical step in TTR amyloids fibrillogenesis, and is responsible for
extracellular deposition of amyloid fibrils. Small molecules, able to bind in T4 binding sites and
stabilize the TTR tetramer, are interesting tools to treat and prevent systemic ATTR amyloidosis. We
report here the synthesis, in vitro evaluation and three-dimensional crystallographic analyses of new
monoaryl-derivatives in complex with TTR. Of the derivatives reported here, the best inhibitor of
TTR fibrillogenesis, 1d, exhibits an activity similar to diflunisal.
Keywords: transthyretin, fibril formation inhibitors, X-ray TTR-ligand complexes, monoaryl
Corresponding authors:
Prof. Susanna Nencetti, E-mail: susanna.nencetti@unipi.it
Prof. Elisabetta Orlandini, E-mail: elisabetta.orlandini@unipi.it

1.
Introduction
Proteins misfolding and aggregating into various toxic aggregate structures are associated with
many neurodegenerative diseases known as amyloid disorders. Transthyretin (TTR) is an
amyloidogenic protein responsible for transthyretin-related amyloidosis (ATTR) caused by the
deposition of fibrillar aggregates in extracellular tissue. ATTR is a systemic amyloidosis in which
amyloid fibrils can be found in most organs thus interfering with their normal function.[1] The disease
can exist in an autosomal dominant form, due to an inherited mutation in the TTR gene, or a non-
hereditary form due to native TTR that is itself amyloidogenic. More than 120 single point mutations
are described in the literature; the majority of these mutations increase the intrinsic amyloidogenic
potential of TTR and are linked to TTR amyloid diseases such as familial amyloid polyneuropathy
(FAP) and familial amyloid cardiomyopathy (FAC). Due to the loss of stability and function of the
mutant TTR, amyloids deposit in the peripheral and autonomic nerves (FAP), or in the heart (FAC).
Non-hereditary transthyretin amyloidosis (senile systemic amyloidosis; SSA), characterized by a late
onset, is due to native TTR and affects about 20% of people over the age of 70, with a predominance
in men.[2]
TTR is produced mainly by the liver and in minor amounts by retinal pigment epithelial cells and in
the choroid plexus of the brain.[3][4][5] The main physiological function of TTR is the transport of
thyroxine (T4) and retinol through the interaction with retinol-binding protein (RBP) in plasma, and
of T4 in cerebrospinal fluid (CSF).[4] TTR also interacts with amyloid-β (Aβ) and plays a protective
role in Alzheimer’s disease (AD).[6][7][8] In AD patients, the level of several metals such as copper,
iron and zinc drastically increases in neuritic plaque and amygdala.[9] The high metal concentrations
acts as modulator of protein conformation,[10] and it has been reported that TTR in the presence of
some metals, in particular Cu2+ and Fe2+, undergoes a conformational change.[11][12] Moreover, an
in vitro experiment has shown that in the presence of Cu2+ TTR-Aβ binding affinity increases from a
micro- to nanomolar range, suggesting that Cu2+ has a relevant role as a Aβ scavenger.[12]
TTR is a 55 kDa homo-tetrameric protein composed of four identical subunits of 127 amino acid
residues, β‑sheet-rich, named A, A’, B and B’. The four monomers are assembled according to a 222
symmetry to form two stable dimers (AB and A’B’) associated back to back to generate the tetramer.
At the dimer-dimer interface, the tetramer is crossed by a long channel coincident with one of the
two-fold symmetric axes that defines two identical binding sites for the natural TTR ligand T4. Each
site has a small inner and a larger outer binding subsite. The crystal complex between TTR and its
natural ligand T4 shows three pairs of symmetric hydrophobic depressions, which are occupied by
the iodine atoms named halogen binding pockets (HBP1, HBP2, and HBP3).[13]

The wild-type TTR (wt-TTR) is potentially amyloidogenic because of its high degree of β-sheet
structure. Biophysical studies demonstrated that wt-TTR is inclined to dissociation and through a
multistep dynamic process fluctuates between the tetrameric structure and the folded monomeric
form.[14][15] The intrinsic propensity towards β-aggregation, enhanced by amyloidogenic mutations,
destabilizes the TTR tetramer and leads to its disassembly into folded monomers that easily unfold
becoming amyloidogenic intermediates; these can spontaneously misassemble to form amyloid fibrils
in the target tissue.[16] Dissociation of the TTR tetramer into its component subunits is the rate-
limiting step in the amyloids formation process.[17] Under physiological conditions, only small
amounts of T4 in the plasma is bound to TTR, the majority of TTR T4 binding sites are unoccupied
(90-85%) and available to bind ligands to prevent tetramer dissociation.[18] Stabilization of the
tetrameric state of TTR by small molecules binding to the TTR tetramer is a promising strategy to
treat TTR amyloidosis. In fact, the binding of a small molecule into the T4-binding sites should
stabilize the tetramer, slowing its conversion into monomers and amyloid fibrils. The first-in-class
drug discovered with this strategy and directed to the treatment of FAP is Tafamidis meglumine
(Vyndaqel ®)[19] that reached the European drug market in November 2011. Studies in vitro, in vivo
and clinical trials for this drug confirmed that stabilization of the native tetrameric form of TTR is a
promising approach to prevent or delay disease progression.[20][21]
The examination of the chemical structure of TTR amyloid inhibitors studied over the past decade
[22], shows that the majority of TTR ligands discovered using structure-based design, possess the
biaryl scaffold. To explore new and more simple chemical structures far from the typical halogenated
biaryl scaffold of the non-steroidal anti-inflammatory drugs NSAIDs, we took into consideration a
series of carboxylic acid derivatives presenting only one aromatic portion.[23][24][25]
Our previous work reported some -aminooxypropionic acids as inhibitors of TTR fibril formation
with a flexible linker that allows a good interaction with TTR binding sites.[23] Among these
monoaryl compounds, the 2-methoxy derivative (lead compound) showed a good inhibition of TTR
fibril formation (Figure 1).
OMe
type A (
n=1)
R = 2-OH
3-OH
N
COOH
N
COOH
O
O
n
R
H
H
4-OH
2,4-OH
3,4-OH
Lead compound
3,5-OH
2-OH,3,5-Br
type B (
R = o-,m-,p-OH
n=0)
2-OH,3,5-Br

Figure 1. Formulas of the lead compound and of the new type A and B compounds
The crystal structure of lead compound bound to TTR (PDB 3GS7) shows that the carboxylic acid
group is placed in the outer binding pocket and oriented towards residues Lys15 and Lys15’ while
the 2-methoxy substituent of the aromatic ring protrudes into halogen binding pockets 3 and 3’. This
type of interaction is defined forward binding mode. This compound interacts with both T4 hormone
binding pockets adopting two conformations with equal occupancy. [23] The lead compound shows
to possess the key structural elements to bind the TTR binding site and turning out to be a new
intriguing chemotype for the design of structurally new TTR fibril formation inhibitors, and it was
selected as starting point for further investigation (Figure 1).
The observation that many natural polyphenols such as curcumin, resveratrol, genistein and
epigallocatechin gallate (EGCG) are able to stabilize the TTR tetramer [26][27][28][29][30],
prompted us to functionalize the aromatic portion of mono aryl compound A with one- or two
hydroxyl groups in various positions of the aromatic ring or with a combination of hydroxyl group
and halogen atoms to obtain new arylidene(aminooxy)propionic acid derivatives of type A (Figure
1). In the newly synthesized compounds the hetero atomic linker (-CH=N-O-) and the carboxylic acid
moiety, that is essential for the interaction with keys amino acids in the T4 binding site (Lys15/15',
Ser117/117’, Thr119/119’), were both maintained.
With the aim to evaluate the effect of a different linker length on the activity, we synthesized also the
arylidene(aminooxy)acetic acid derivatives of type B. This modification resulted to be of interest on
the basis of the observation that in the monoaryl 9-fluorenyl-acetic acid derivative, previously studied
by us (PDB 4TQH), the activity was maintained compared to its propionic acid analogue (PDB 4TQI)
(Figure 1) [24].
In the present work were synthesized new mono aryl compounds of type A, 1a-g (propionic acid
derivatives) (Scheme 1) and of type B, 2a-d (acetic acid derivatives) (Scheme 2), and their interaction
with TTR was investigated using a turbidimetric UV assay. The ligands with a low percentage of
fibril formation were also tested using a Thioflavin T binding assay. Finally, to explain the
biochemical results, we performed X-ray crystallographic analyses of the most active compounds in
complex with TTR.
2. Chemistry

The (E) 3-(((mono-hydroxy- (1a-c), 3-(((3,5-dibromo-2-hydroxy- (1d), 3-(((di-hydroxy- (1e-g)
benzylidene)amino)oxy)propionic acids were prepared as shown in Scheme 1. Mitsunobu reaction
between N-hydroxyphthalimide (4) and tert-butyl 3-hydroxypropanoate (5) yielded the tert-butyl 3-
((1,3-dioxoisoindolin-2-yl)oxy)propanoate (6). Compound 6 after hydrolysis with 7M methanolic
ammonia gave the 3-(aminooxy)propionic acid that was crystallised as hydrochloride (7). Compound
7
was condensed with the opportune hydroxy aldehydes to obtain the (E)-3-
((hydroxybenzylidene)amino)oxy)propanoic acid derivatives 1a-g.
Scheme 1
O
O
O
O
i
ii
+
HO
O
N O
O
O
N OH
O
6
4
5
R1
R1
R4
O
R2
R3
R
H
R2
R3
R
H
iii
HCl .
+
H2NO
OH
O
N
COOH
O
R4
8a-g
1a-g
7
a
b
c,
d,
, R=OH, R1=R2=R3=R4=H; , R1=OH, R=R2=R3=R4=H; R2=OH, R=R1=R3=R4=H; R=OH, R1=R3=Br,
e, f, g,
R2=R4=H; R=R2=OH, R1=R3=R4=H; R1=R2=OH, R=R3=R4=H; R1=R3=OH, R=R1=R3=H
Reaction conditions i, ii iii
:
P(Ph3),DEAD,dryTHF,r.t. 72h; , NH3/MeOH 7M, CHCl3,r.t. 2h; , CHCl3/H2O, 40 °C.
Scheme 1. Synthesis of arylidene(aminooxy)propionic acid derivatives 1a-g.
The
(E)-2-((hydroxybenzylidene)amino)oxy)acetic
acids
(2a-d)
and
the
(E)-2-
(((carboxymethoxy)benzylidene)amino)oxy)acetic acid (3i-l) derivatives were prepared as described
in Scheme 2. By reaction of the opportune hydroxyaldehyde 9a-d with hydroxylamine hydrochloride
the corresponding oximes 10a-d of (E) configuration were obtained. The reaction of 10a-d with ethyl
2-bromoacetate, yielded the corresponding desired mono esters 11a-d or di-esters 12e-h due to O-
acylation. The esters 11a-d and 12e-h were hydrolysed using mild conditions with a stoichiometric
amount of 1M NaOH solution to yield the corresponding acids 2a-d and 3i-l, respectively.
Dicarboxylic acids 3i-l, obtained as products from the alkylation of phenolic groups with ethyl 2-
bromoacetate, were also tested to assess their ability to inhibit TTR fibril formation.

Scheme 2
R
R
i
N
O
OH
H
H
10a-d
9a-d
ii
R
R1
+
N
N
iii
O
COOEt
O
COOEt
H
H
11a-d
12e-h
iii
R
R1
N
N
O
COOH
O
COOH
H
H
2a-d
3i-l
a
e
b,
c,
d
,
,
R=2-OH;
R=3-OH;
R=4-OH;
R=2-OH,3,5-diBr;
h,
f
g
, R1= 2-OCH2COOEt; , R1 3-OCH2COOEt; , R1= 4-OCH2COOEt;
i
j
R1= 2-OCH2COOEt, 3,5-diBr-; , R1= 2-OCH2COOH; , R1= 3-
k
l
OCH2COOH; , R1= 4-OCH2COOH; , R1= 2-OCH2COOH, 3,5-diBr-
.
Reaction conditions i,
ii
,
:
NH2OH HCl,DMSO/H2O, r.t. 4h;
iii
BrCH2COOEt, CH3CN, r.t. 3h; , NaOH,THF/H2O, r.t. 3h;
Scheme 2. Synthesis of arylidene(aminooxy)acetic acid derivatives 2a-d and 3i-l.
3. Results and discussion
3.1 Turbidimetric assay
The propionic acid derivatives (1a-g) and the acetic acids derivatives (2a-d) and (3i-l) were tested in
vitro against the amyloid fibril formation of TTR using the turbidimetric assay as previously
described [24]. Results determined as a percentage of fibril formation are shown in Figure 2. The
ability of the newly synthesized compounds to reduce fibril formation was compared to diflunisal,
used as a readily available reference drug (positive control), and TTR without inhibitor, used as the
negative control. The inhibitor concentrations used in the turbidimetric assay was 7.2 µM, twice the
TTR concentration in plasma (3.6 µM) and the values were acquired at 72 h.

Figure 2. In vitro acid denaturation-mediated of WT-TTR. Bars represent data of type A (propionic acid
derivatives) and B (acetic acid derivatives) compounds. The fibril formation in absence of inhibitor was
assigned to be 100%. The error in the fibril formation assay is ±5%
An examination of the results obtained for the propionic acids of type A that possess only one
phenolic group (1a-c) shows that all compounds possess the ability to stabilize the TTR tetramer.
In particular, among the monohydroxy substituted compounds, the 4-OH-substituted derivative (1c)
is the one that shows the best inhibitory effect on fibrillogenesis with a percentage of fibril formation
of 17%. Instead, a hydroxyl group in 2 (1a) or 3 (1b) position of the aromatic ring decreases the
inhibitory activity with a percentage of fibril formation of 38% and 42 %, respectively. An interesting
result is obtained by the introduction of two bromine atoms in 3 and 5 position of the aromatic ring
of the 2-OH substituted compound (1a). This modification leads to an increase of the ability to
stabilize the TTR tetramer with a decrease in the of percentage of TTR fibril formation from 38 %
with 1a to 16% with the 3,5-dibromo-2-hydroxysubstitured derivative 1d.
The inhibitory activity is moderately affected by the introduction of a second hydroxyl group: the
catechol derivative (1f) presents a moderate ability to stabilize the TTR tetramer with 34% of fibril
formation, while the 2,4-hydroxy (1e) and the 3,5-hydroxy substituted (1g) derivatives partially
stabilize the TTR tetramer with 38% and 39% of fibril formation, respectively.
As regards to the ability of (E)-2-((hydroxybenzylidene)amino)oxy)acetic acid derivatives (2a-d) in
inhibiting fibrillogenesis, only the 4–hydroxy substituted compound 2c shows an inhibitory activity
even if it is very moderate (59% fibril formation), while compounds 2a and 2b are devoid of any
activity. As already seen for the arylidene(aminooxy)propionic acid series, the formal introduction of

a bromine atom at the 3 and 5 positions of the 2-hydroxy acetic acid derivative converts the inactive
compound 2a into a moderately active compound 2d (35% of fibril formation).
The (carboxymethoxy)benzylidene)amino)oxy)acetic acids (3i-l) show a very low (3k, 3l) or no
ability to inhibit the TTR fibril formation (3i, 3j).
These new monoaryl-acetic acid derivatives (2a-d), contrary to what has been highlighted by our
previous study on monoaryl 9-fluorenyl-acetic acid derivative [24], efficient in stabilizing the
tetramer, are less active than propionic acid ones. We can hypothesize that this result is probably due
to a weaker interaction in the active site of the TTR of the aryl portion with respect to the fluorenyl
one that is able to occupy an alternative hydrophobic area in the TTR binding site. [23]
3.2 Thioflavin T binding assay
We speculated that our compounds could stabilize the tetramer structure and suppress amyloid
fibril formation of wild-type TTR as could be observed by a Thioflavin T (ThT) binding assay. For
this reason, starting from the turbidimetric result, we decided to further investigate the propionic acid
derivatives using the ThT binding assay with the aim to have a global view on the effect of the
inhibitor compounds on the complex amyloid aggregation process. In the amyloid field, often a
combination of techniques is appropriate to get more information. In particular, we selected the mono-
ortho, meta and para substituted propionic acid derivatives 1a-1c, the 3,5-dibromo-2-
hydroxysubstitured derivative 1d, which showed the lowest percentage of fibril formation (16%), and
among the di-substituted propionic acid compounds 1e-g (fibril formation ~ 34-39%), the derivative
1e. Compound 1e was chosen because, according to our reference lead compound (o-OCH3, PDB
3GS7), it is substituted in ortho position, and in addition, the second hydroxyl group is located in the
meta position as the second most active compound 1c (fibril formation 17%). We incubated TTR
samples with the indicated concentrations of inhibitors (10 µM) and EGCG, as a positive control, in
acetate buffer (pH 4.4). We analysed them for amyloid fibril formation by registering the Thioflavin
T fluorescence intensity at the end of the aggregation process (Figure 3). The presence of EGCG at
10 µM decreased the level of amyloidogenesis of wild-type TTR by at least 53% compared to the
sample with TTR alone.[29] A slight decrease of the fluorescence was observed for compounds 1a
and 1b, which can be considered less active than EGCG in the inhibition of TTR aggregation.
Conversely, compounds 1c and 1d showed a similar inhibitory activity to EGCG, with 1c a little more
active than 1d. Surprisingly, compound 1e resulted as the best inhibitor of the series, with an effect
on the ThT fluorescence intensity decrease even better than EGCG (68% of inhibition and a
significantly difference at p < 0.001, compared to the EGCG control). The slight discrepancy between
the turbidimetric and ThT assay results for compound 1e could be ascribed to the fact that ThT can
bind only to β-sheet-rich aggregates. It seems that in ThT assay compound 1e is more active than

observed in the turbidimetric test, probably because the decreased fluorescence intensity in the ThT
assay is correlated to a reduced number of β-sheet-rich fibrils. One possible explanation is that 1e is
able to interfere with the β-sheet-rich morphology of the TTR fibrils but is not efficient enough to
avoid the formation of amorphous aggregates that cannot be detected by the fluorescent dye (ThT),
but are responsible of the turbidity.
Figure 3. Percentage of TTR fibril formation assessed by ThT fluorescence assay (10 µM ThT in 50 mM
glycine buffer; TTR final concentration at 0.045 µM). Amyloid fibril ratio (%) = 100 x (fluorescence TTR +
compound / fluorescence without compound). Bars are representative of 11 measurements from the same
experiment. Values represent the mean ± the standard error. Statistical significance was calculated by an
ANOVA test (* P < 0.05, ** P < 0.01, *** P < 0.001, n.s. not significantly different).
3.3 Crystallographic Studies
In order to investigate the structure-activity relationships at an atomic level, crystal structures of the
inhibitors 1c (PDB 6TJN) and 1d (PDB 6TI9) in complex with TTR have been obtained according to
[24][31]
previously described procedures.
We selected these compounds because they show a high
percentage of fibril formation inhibition both in turbidimetric and in ThT assays, even if they are
differently functionalized.
The crystals grown belong to the orthorhombic space group P21212 with unit cell constants (Table 1)
comparable to other TTR depositions in the PDB database. There are two monomers (A and B) in

asymmetric unit, and by a two-fold crystallographic symmetry operation, monomer A and B form a
tetramer (AB/A’B’) in which the two T4-binging sites are located along the crystallographic 2-fold
axis, see Figure 4.
In both crystal complexes, additional electron density is visible in both T4 binding pockets,
confirming the presence of inhibitors 1c and 1d, respectively, see Figures 4 and 5. Ligands 1c and 1d
bind along the two pockets with the same orientation: in forward mode, so that the carboxylic acid
groups point towards Lys15 located at the entrance of the cavity, Figure 4. The occupancy of each
ligand is fixed at 50% because they are located along the two-fold axis of the tetramer.
The two TTR monomers A and B display only minor differences between their protein structures.
The root-main-square deviations (RMSD) calculated on Cα positions of the two crystal structures,
6TJN and 6TI9, is 0.38 Å, suggesting that there are no significant structural differences.
Structure
PDB code
TTR-1c
6TJN
TTR-1d
6TI9
Data Collection
Source
SOLEIL Proxima2a
SOLEIL Proxima2a
Space group
Unit-cell
P21212
P21212
a (Å)
b (Å)
43.32
85.99
43.14
85.82
c (Å)
64.09
63.84
Molec./asym.
Wavelength (Å)
Resolution (Å)
Highest resolution bin (Å)
No. of reflections
No. of unique reflections
Completeness (%)
I/σ (I)
2
2
0.980
0.919
50-1.45
1.46-1.45
569880
42800
99.9 (99.8)
16.29 (2.01)
0.999 (0.75)
8.1 (107.4)
7.8(101.9)
13.31 (10.02)
50-1.71
1.72-1.71
344252
26756
99.8 (99.9)
22.90 (1.85)
1.0 (0.74)
6.0 (110)
5.8(105.5)
13.03 (12.68)
CC1/2
Rmerge (%)
Rp.i.m. (%)
Multiplicity
Refinement
R-work
R-free
0.16 (0.174)
0.17 (0.192)
0.161 (0.155)
0.188 (0.182)
R.m.s. deviations
Bond lengths (Å)
Bond angles (°)
Ramachandran
Favored (%)
Outliers (#)
0.12
1.3
0.008
0.95
98.26
0
98.68
0
Table 1. Statistics for data collection, processing and refinement on TTR–ligand complexes.

Figure 4. TTR-inhibitor crystal complexes: electron density and binding mode for propionic acid inhibitors
1c and 1d. (a) TTR tetramer in complex with compound 1c (PDB 6TJN). The dimer formed by monomer A
and B is represented in yellow, its symmetry light-orange. Ligand 1c and its symmetric (orange and light
orange respectively) bind both binding sites. Electron density (weighted 2Fo-Fc, contoured at 0.30, 0.8, 1.6,
2.5 σ) represented as multi-wire meshes. (b) Superposition of ligands 1c, 1d and their symmetric show minor
differences in the accommodation of TTR binding sites. (c) Inhibitor 1d in complex with TTR tetramer. Dimer
A-B colored teal and the symmetry operation in light-orange. There is clear electron density (weighted 2Fo-
Fc, contoured at 0.30, 0.8, 1.6, 2.5 σ) for ligand 1d in both TTR binding sites. (d) Ligand 1c binds in both TTR
binding sites in forward mode. The binding modes are similar but not identical. H-bonds are showed with
yellow dashes. (e) Ligand 1d binds both T4 binding sites in forward mode.
The single crystal X-ray diffraction data of the TTR-1c crystal complex (6TJN) was collected to 1.7
Å resolution. The phenolic moiety of compound 1c is located in the inner cavities of both T4 binding
sites. In the binding pocket A/A’, the 4-hydroxyl group of the aromatic portion forms two hydrogen
bonds, one with Ser117 (-OH – O Ser117 distance of 3.5 Å) and another with a water molecule
(distance 3.1 Å), Figure 4 (d), while the carboxylic acid group of the propionic linker points towards
Lys15/15’ (COOH – NH2 of Lys15 distance 3.5 Å), see Figure 4 (a), (d). In the second binding pocket
B/B’, the polar interaction between the COOH and Lys 15 is preserved (COOH – NH2 of Lys15
distance 3.5 Å), and a slight torsion of the aromatic portion orients the aromatic ring towards the
hydrophobic pockets HBP3/HBP3’, Figure 4 (a) (b) (d).

Figure 5. Electron density for the TTR-1d crystal complex. (a) Surface representation of the TTR tetramer
shown with the ligand and its electron density sandwiched between two monomers (B-B’). The two bromine
atoms are placed in HBP3 and HBP3’. (b) Ligand 1d in complex with TTR. The figure was obtained rotating
the dimer by 180° (B’-B). Red dashes show polar interaction between the carboxylic group and the Lys15. (c)
View of the cavity B-B'. Hot-pink dashes display the interaction between the ligand and Lys15. (d)
Experimental map shown in COOT for crystals grown in the presence of 1d. The Fo-Fc Fourier map contoured
at 3σ in green, 2Fo-Fc Fourier map is contoured at 1σ in blue mesh. Anomalous difference Fourier map of
bromine atoms in purple contoured at 5σ.
The single crystal X-ray data for TTR in complex with (E)-3-(((3,5-dibromo-2-
hydroxybenzylidene)amino)oxy)propionic acid (1d) (6TI9) has been collected to 1.45 Å resolution
(Table 1). Ligand 1d binds to both T4 binding sites in forward mode, with the two bromine atoms
placed into HBP3/HBP3’ and the carboxylic acid moiety oriented towards Lys15/15’ (COOH – NH2
Lys15/Lys15’ distance 3.4 Å) Figure 4 (b, c, e), and Figure 5 (a-d). To confirm the ligand binding
mode, the data-set for TTR in complex with 1c was collected just above the K absorption edge of
bromine (0.919 Å) and a phased anomalous difference Fourier (DANO) map was calculated using
the refined model of TTR. Two high anomalous electron density peaks were identified in both of the
inner T4 binding sites, HBP3 and HBP3’. This anomalous signal allows us to detect unambiguously
the presence and the position of the two bromine atoms, Figure 5 (d).

The X-ray analysis of compounds 1c and 1d suggests that, for this type of monoaryl derivatives, the
propionic linker has the correct length to strengthen the interaction between the -COOH moiety of
the ligand and the -NH2 group of Lys15 increasing the tetramer stabilization. Therefore, we can
hypothesize that the acetic linker is not long enough to establish a strong and stable interaction
between the ligand and binding pocket.
4. Conclusions
In the present work we have explored the effect produced by the decoration of a mono-aryl scaffold
previously reported with hydroxyl and hydroxyl/halogen (Br) groups on the activity and the binding
mode to TTR. The results of turbidimetric and thioflavin T binding assays presented here show that
the (E)-3-((arylideneamino)oxy)propionic acids (Type A, 1c and 1d, propionic acid derivatives) have
the best activity with a percentage of inhibition of fibril formation values comparable to diflunisal
and EGCG at the tested concentration. Moreover, this study shows that a single atom in the linker
between the (arylideneamino)oxy and acidic moieties (Type B compounds, acetic acid derivatives) is
not suitable to display any suitable inhibition of TTR fibril formation. The X-ray crystallographic
studies have highlighted that for the propionic acid derivatives of type A, both the halogen atoms (1d)
and the 4-hydroxyl group (1c) drive the aromatic ring into the inner cavity, allowing the interaction
of compounds in the forward binding mode. In the future, the decoration of mono-aryl moiety with a
combination of groups, halogen/hydroxyl, in various positions, will be useful to better define the
optimal fit of the propionic acid derivatives in the T4 binding site.
Considering that transthyretin stabilization is a key point for the inhibition of transthyretin
amyloidosis, the biological results and the structural data presented here, confirm the
(arylidenamino)oxypropionic acid structure as a promising scaffold for the design of effective TTR
stabilizers.
5. Experimental
5.1. Material and methods
Analytical grade reagents and solvent were purchased from Sigma-Aldrich (St. Louis, MO), and were
used as supplied. Solvents were dried according to standard methods. Melting points were determined
on a Köfler hot-stage apparatus and are uncorrected. 1H-NMR spectra of all synthesized compounds
were determined with a Bruker UltrashieldTM 400 MHz (Fällander, Switzerland). Coupling constants
J are reported in Hertz. The following abbreviations are used: singlet (s), doublet (d), doublet of
doublet (dd), doublet of doublet of doublets (ddd), triplet (t), doublets of triplet (dt), broad (br) and

multiplet (m). Reaction were monitored by thin layer chromatography (TLC) on silica gel plates
containing a fluorescent indicator (Merck Silica Gel 60 F254), spots were detected under UV light
(254 nm). Preparative TLC were performed with silica gel plates (2mm, Merck Silica Gel 60 F254).
Chromatographic separations were performed on silica gel columns by flash column chromatography
(Kieselgel 40, 0.040−0.063 mm; Merck). Na2SO4 was always used as the drying agent. Evaporation
was carried out “in vacuo” (rotating evaporator). Elemental analyses were performed by our
analytical laboratory and agreed with theoretical values to within  0.4%.
5.2. Synthetic procedure
5.2.1. Tert-butyl 3-((1,3-dioxoisoindolin-2-yl)oxy)propanoate (6)
To a stirred solution of 3-hydroxypropanoate (5) (1.71 mmol; 250 mg) and anhydrous THF (8.6 ml)
were added under argon P(Ph)3 (1.71 mmol; 448 mg), N-hydroxyphthalimide (4) (1.71 mmol; 279
mg) and DEAD (0.54 ml). The reaction mixture was stirred at room temperature for 72h then
evaporated to dryness. The crude product was purified by flash chromatography on silica gel column,
eluting with a mixture of EtOAc/hexane 1:9 to obtain 6 as on oil (230 mg, 46% yield); 1H-NMR (400
MHz, CDCl3) : 7.84-7.80 (m, 2H); 7.76-7.72 (m, 2H); 4.45 (t, 2H, J=6.8 Hz); 2.75 (t, 2H, J=6.8
Hz); 1.43 (s, 9H).
5.2.2. 3-(aminooxy)propionic acid hydrochloride (7•HCl)
To a solution of 6 (0.789 mmol; 230 mg) in CHCl3 (6.3 ml) was added ammonia in methanol solution
(7M, 3.1 ml). The reaction mixture was stirred at room temperature (2h) until the disappearance of
the starting compound (TLC, hexane/EtOAc 8:2). After evaporation to dryness the solid residue was
dissolved in Et2O and treated with an excess of Et2O•HCl to yield a solid precipitate which was
filtered and triturated with Et2O to give pure 7•HCl (38% yield). 1H-NMR (400 MHz, CD3OD-d4) :
4.28 (t, 2H, J=5.8 Hz); 2.72 (t, 2H, J=5.8 Hz).
5.2.3. General procedure for the synthesis of (E)-3-(((mono-hydroxy (1a-c), 2-hydroxy-3,5-bromo
(1d) or di-hydroxy-(1e-g) benzylidene)amino)oxy)propionic acids
A stirred solution of the appropriate mono-hydroxy (8a-c), 2-hydroxy-3,5-dibromo- (8d) or di-
hydroxy- (8e-g) benzaldehyde (0.96 mmol) in CHCl3 (3 ml) was added 3-(aminooxy)propionic acid
hydrochloride (7•HCl) (1.06 mmol) in H2O (1ml). The resulting mixture was stirred at 40°C until the
disappearance of the starting compound (TLC, hexane/EtOAc 8:2) then diluted with H2O and
extracted with CHCl3. The chloroformic layer were washed (H2O) dried and evaporated to dryness to
yield the crude acids.

(E)-3-(((2-hydroxybenzylidene)amino)oxy)propionic acid (1a). The title compound was prepared
from the aldehyde 8a following the general procedure. The crude product was triturated with hexane
to give the acid 1a as white solid (56% yield); m.p. 114-116 °C; 1H NMR (400 MHz, CDCl3)  9.59
(s, 1H, COOH); 8.19 (s, 1H, HC=N); 7.29 (t, 1H, J=7.2 Hz, Ar), 7.16 (d, 1H, J=7.2 Hz, Ar), 6.99 (d,
1H, J=8.0 Hz, Ar), 6.90 (t, 1H, J=7.2 Hz, Ar); 4.48 (t, 2H, J=6 Hz, OCH2); 2.81 (t, 2H, J=6 Hz,
CH2CO);13C NMR (400 MHz, CDCl3) : 176.6 (C=O); 157.40 (C=N); 152.4 (1C, Ar-O); 131.5,
130.9, 119.7, 116.8, 116.1 (5 Ar-C); 69.6 (OCH2); 34.1 (CH2CO); Anal. Calcd for C10H11NO4
(209.20); C, 57.41; H, 5.30; N, 6.70; Found 57.32; H, 5.39; N, 6.83.
(E)-3-(((3-hydroxybenzylidene)amino)oxy)propionic acid (1b). The title compound was prepared
from the aldehyde 8c following the general procedure. The crude product was purified by flash
chromatography on silica gel column eluting with 2:1 n-hexane/EtOAc to give after trituration with
hexane the acid 1b as white solid (22 % yield); m.p. 107-109°C; 1H NMR (400 MHz, DMSO-d6) :
12.28 (s, 1H, COOH); 9.61 (s, 1H, OH); 8.14 (s, 1H, HC=N); 7.21 (t, 1H, J=7.2 Hz, Ar), 7.03-6.80
(m, 3H, Ar) ; 4.28 (t, 2H, J=6.4 Hz, OCH2); 2.60 (t, 2H, J=6.4 Hz, CH2CO). 13C-NMR (400 MHz,
DMSO-d6)  172.8 (C=O); 158.0 (C=N); 149.5 (1C, Ar-O); 133.5, 130.3, 118.7, 117.7, 113.4 (5 Ar-
C); 69.9 (OCH2); 34.5 (CH2CO); Anal. Calcd for C10H11NO4 (209.20); C, 57.41; H, 5.30; N, 6.70;
Found 57.54; H, 5.38; N, 6.81.
5.2.3.3 (E)-3-(((4-hydroxybenzylidene)amino)oxy)propionic acid (1c). The title compound was
prepared from the aldehyde 8c following the general procedure. The crude product was purified by
flash chromatography on silica gel column eluting with 3:1 n-hexane/EtOAc to give after trituration
with hexane the acid 1c as white solid (20% yield); m.p. 144-146 °C; 1H NMR (400 MHz, CD3OD-
d4) : 7.98 (s, 1H, HC=N); 7.42 (d, 2H, J=8.8 Hz, Ar); 6.76 (d, 2H, J=8.8 Hz, Ar); 4.35 (t, 2H, J=6.4
Hz, OCH2). 2.64 (t, 2H, J=6.4 Hz, CH2CO).13C-NMR (400 MHz, CD3OD-d4) : 175.4 (C=O); 160.5
(C=N); 150.2 (1C, Ar-O); 129.7; 124.9; 116.5 (5 Ar); 70.4 (OCH2); 35.4 (CH2CO). Anal. Calcd for
C10H11NO4 (209.20); C, 57.41; H, 5.30; N, 6.70; Found 57.52; H, 5.25; N, 6.56.
(E)-3-(((3,5-dibromo-2-hydroxybenzylidene)amino)oxy)propionic acid (1d). The title compound was
prepared from the aldehyde 8d following the general procedure. The crude product was purified by
flash chromatography on silica gel column eluting with (3:1 n-hexane/EtOAc) to give after trituration
with hexane the acid 1d as white solid (32%yield); m.p.: 129-130 °C; 1H-NMR (400 MHz, CDCl3)
: 10.40 (s, 1H, COOH); 8.08 (s, 1H, HC=N); 7.66 (d, 1H, J=2.4 Hz, Ar); 7.24 (d, 1H, J=2.4 Hz, Ar);
13
4.50 (t, J=6 Hz, 2H, OCH2); 2.80 (t, 2H, J=6 Hz, CH2CO). C-NMR (400 MHz, CDCl3) : 175.3
(C=O); 153.3 (C=N); 150.4 (1C, Ar-O); 136.9, 132.4 (2C, Ar-Br); 118.5, 111.8, 111.4 (3C, Ar); 70.3

(OCH2); 33.9 (CH2CO). Anal. Calcd for C10H9NO4Br2 (366.98); C, 32.72; H, 2.45; N, 3.81; Found
32.71; H, 2.40; N, 3.81.
(E)-3-(((2,4-dihydroxybenzylidene)amino)oxy)propionic acid (1e). The title compound was prepared
from the aldehyde 8e following the general procedure. The crude product was triturated with hexane
to give the acid 1e as white solid (41% yield); m.p. 169-171 °C; 1H-NMR (400 MHz, CD3OD-d4) 
8.17 (s, 1H, HN=C); 7.11 (d, 1H, J=8.4 Hz, Ar); 6.33 (ddd, 2H, J=10, 8, 1.6 Hz, Ar); 4.35 (t, 2H, J=6
Hz, OCH2); 2.68 (t, 2H, J=6 Hz, CH2CO); 13C-NMR (400 MHz, CD3OD-d4) δ 175.1 (C=O); 161.9,
160.2 (2C Ar-O); 152.4 (C=N); 132.7, 110.3, 108.8, 103.7 (4C, Ar); 70.8 (OCH2); 35.2 (CH2CO).
Anal. Calcd for C10H11NO5 (225.19); C, 53.33; H, 4.92; N, 6.22; Found 53.42; H, 5.07; N, 6.34.
(E)-3-(((3,4-dihydroxybenzylidene)amino)oxy)propionic acid (1f). The title compound was prepared
from the aldehyde 8f following the general procedure. The crude product was triturated with hexane
to yield the acid 1f as white solid. (50% yield); m.p.: 163-165 °C; 1H-NMR (400 MHz, DMSO-d6) 
12.28 (s, 1H, COOH); 9.37 (s, 1H, OH); 9.18 (s, 1H, OH); 7.99 (s, 1H, HC=N); 7.05 (s, 1H, Ar), 6.84
(d, 1H, J=7.6 Hz, Ar); 6.73 (d, 1H, J=8.0 Hz, Ar); 4.23 (t, 2H, J=6 Hz, OCH2); 2.59 (t, 2H, J=6 Hz,
CH2CO). 13C-NMR (400 MHz, DMSO- d6) : 172.9 (C=O); 149.5 (C=N); 148.1, 146.0 (2C, Ar-O);
123.6, 120.4, 116.0, 113.5 (4C, Ar); 69.6 (OCH2); 34.5 (CH2CO). Anal. Calcd for C10H11NO5
(225.20); C, 53.33; H, 4.92; N, 6.22; Found 53.45; H, 5.03; N, 6.18.
(E)-3-(((3,5-dihydroxybenzylidene)amino)oxy)propionic acid (1g). The title compound was prepared
from the aldehyde 8g following the general procedure. The crude product was triturated with hexane
to give the acid 1g as white solid. (26% yield); m.p. 55-57°C; 1H-NMR (400 MHz, MeOD-d4) : 7.91
(s, 1H, HN=C); 6.52 (s, 2H, Ar); 6.28 (s, 1H, Ar); 4.36 (t, 2H, J=6 Hz, OCH2); 2.68 (t, 2H, J=6 Hz,
13
CH2CO). C-NMR (400 MHz, MeOD-d4) : 175.2 (C=O); 159.9 (C=N); 150.5, 135.3 (2C, Ar-O);
106.6, 105.3 (2C, Ar); 70.6 (OCH2); 35.3 (CH2CO). Anal. Calcd for C10H11NO5 (225.20); C, 53.33;
H, 4.92; N, 6.22; Found 53.26; H, 4.80; N, 6.11.
5.2.4. General procedure for the synthesis of (E)-hydroxybenzaldehyde oximes (10a-d)
To a solution of the opportune hydroxybenzaldehyde 9a-d (4.1 mmol) in DMSO (3 ml) was added
.
an aqueous (2 ml) solution of NH2 HCl (14.3 mmol). The reaction mixture was stirred at room
temperature for 4h then diluted with H2O and extracted with EtOAc and the organic extracts washed
with H2O. The evaporation of the dried organic layer gave a solid residue that was triturated with
CHCl3 and the product was filtered and purified crystallization (Et2O/Hex) to obtain 10a-d pure.

(E)-2-hydroxybenzaldehyde oxime (10a): (87% yield); m.p. 56-58°C; 1H-NMR (400 MHz, DMSO-
d6) δ: 11.30 (s, 1H, OH-Ar). 10.07 (s, 1H, HON=C), 8.33 (s, 1H, HC=N), 7.50 (dd, 1H, J=7.68,
J=1.60 Hz, Ar), 7.22 (ddd, 1H, J=8.3, 7.2, 1.6 Hz, Ar), 6.98-6.83 (m, 2H, Ar).
(E)-3-hydroxybenzaldehyde oxime (10b): (40% yield); m.p. 85-87 °C; 1H-NMR (400 MHz, DMSO-
d6) δ: 11.09 (s, 1H, OH); 9.47 (s, 1H, HON=C), 7.97 (s, 1H, HC=N); 7.12 (t, 1H, J=7.6 Hz, Ar); 6.95 (dd,
1H, J=4.0, 1.6 Hz, Ar), 6.91 (d, 1H, J=7.6 Hz, Ar); 6.00 (ddd, 1H, J=8.4, 2.8, 1.2 Hz, Ar).
1
(E)-4-hydroxybenzaldehyde oxime (10c): (66% yield); m.p. 92-94°C; H-NMR (400 MHz, DMSO-
d6) δ: 10.82 (s, 1H, OH). 9.74 (s, 1H, HON=C); 7.99 (s, 1H, HC=N); 7.40 (d, 2H, J=8.8 Hz, Ar), 6.77
(d, 2H, J=8.8 Hz, Ar).
(E)-3,5-dibromo-2-hydroxybenzaldehyde oxime (10d): (95% yield); m.p. 208-210 °C; 1H-NMR (400
MHz, DMSO-d6) δ: 11.92 (s, 1H, OH-Ar), 10.97 (s, 1H, OHN=C), 8.35 (s, 1H, CH=N), 7.74 (s, 1H,
Ar), 7.67 (s, 1H, Ar).
5.2.5. General procedure for the synthesis of ethyl (E)-2-(((hydroxybenzylidene)amino)oxy)acetate
derivatives (11a-d) and ethyl (E)-2-(((2-ethoxy-2-oxoethoxy)benzylideneamino)oxy)acetate
derivatives (12e-h)
To a solution of hydroxybenzaldoxime (3.66 mmol) in CH3CN (3 ml) and ethyl bromoacetate (3.66
mmol) were added Cs2CO3 (3.66 mmol). The reaction mixture was stirred at room temperature for
3h until the disappearance of the starting compound (TLC, hexane/EtOAc 8:2). After filtration of the
excess Cs2CO3 the reaction mixture was added with AcOEt, washed several times with H2O, dried
with Na2SO4 and finally evaporated at reduced pressure to obtain a crude solid. The crude product
was constituted by two products that were separated by FLASH chromatography on silica-gel eluting
with 3:1 AcOEt/hexane. From the collected evaporated fractions two compounds were obtained as
white solids which were crystallized from a mixture of Et2O/hexane to give the monoesters 11a-d
and diesters 12e-h.
1
Ethyl (E)-2-(((2-hydroxybenzylidene)amino)oxy)acetate (11a): (17% yield); H-NMR (400 MHz,
CDCl3) 훿: 9.43 (s, 1H, OH); 8.31 (s, 1H, HC=N); 7.30 (ddd, 1H, J=8.4, 7.2, 1.6 Hz Ar); 7.18 (dd, 1H,
J=7.6, 1.6, Hz Ar); 6.97 (dd, 1H, J=7.6, 1.6 Hz Ar); 6.92 (dd, 1H, J=7.6, 1.2 Hz Ar); 4.70 (s, 2H,
OCH2), 4.26 (q, 2H, J=7.1 Hz, COCH2); 1.30 (t, 3H, J=7.1 Hz, CH3).
1
Ethyl (E)-2-(((3-hydroxybenzylidene)amino)oxy)acetate (11b): (40% yield); H-NMR (400 MHz,
DMSO-d6) δ: 11.26 (s, 1H, OH); 8.08 (s, 1H, HC=N); 7.28 (t, 1H, J=8.0 Hz, Ar); 7.19 (d, 1H, J=7.6

Hz, Ar); 7.12 (dd, 1H, J=2.4, 1.2 Hz, Ar); 6.92 (ddd, 1H, J=8.2, 2.4, 0.8 Hz, Ar); 4.78 (s, 2H, OCH2),
4.15 (q, 2H, J=7.0 Hz, COCH2); 1.19 (t, 3H, J=7.0 Hz, CH3).
1
Ethyl (E)-2-(((4-hydroxybenzylidene)amino)oxy)acetate (11c): (15% yield); H-NMR (400 MHz,
CDCl3) 훿: 11.00 (s, 1H, OH); 8.06 (s, 1H, HC=N); 7.51 (d, 2H, J=8.8 Hz, Ar); 6.94 (d, 2H, J=8.8 Hz,
Ar); 4.80 (s, 2H, OCH2), 4.16 (q, 2H, J=7.0 Hz, COCH2); 1.21 (t, 3H, J=7.0 Hz, CH3).
1
Ethyl (E)-2-(((3,5-dibromo-2-hydroxybenzylidene)amino)oxy)acetate (11d): 52% yield); H-NMR
(400 MHz, DMSO-d6) δ: 10.31 (s, 1H, OH); 8.53 (s, 1H, HC=N); 7.85 (d, 2H, J=2.4 Hz, Ar); 7.68
(d, 2H, J=2.4 Hz, Ar); 4.85 (s, 2H, OCH2), 4.16 (q, 2H, J=7.0 Hz, COCH2); 1.21 (t, 3H, J=7.0 Hz,
CH3).
Ethyl (E)-2-(((3-(2-ethoxy-2-oxoethoxy)benzylidene)amino)oxy)acetate (12e): (51% yield); 1H-NMR
(400 MHz, CDCl3) δ: 8.67 (s, 1H, HC=N); 7.78 (dd, 1H, J=7.7, 1.7 Hz, Ar); 7.18 (dd, 1H, J=7.6, 1.6
Hz, Ar); 7.31 (ddd, 1H, J=8.4, 7.2, 1.6 Hz, Ar); 6.97 (t, 1H, J=7.6 Hz, Ar); 6.76 (d, 1H, J=7.6 Hz,
Ar); 4.69 (s, 2H, OCH2), 4.64 (s, 2H, OCH2), 4.25 (q, 2H, J=7.2 Hz, COCH2); 4.24 (q, 2H, J=7.2 Hz,
COCH2); 1.29 (t, 3H, J=7.2 Hz, CH3); 1.29 (t, 3H, J=7.2 Hz, CH3).
Ethyl (E)-2-(((3-(2-ethoxy-2-oxoethoxy)benzylidene)amino)oxy)acetate (12f): (28% yield); 1H-NMR
(400 MHz, DMSO-d6) 훿: 8.32 (s, 1H, HC=N); 7.36 (t, 1H, J=8.0 Hz, Ar); 7.21 (d, 1H, J=7.6 Hz, Ar);
7.14 (dd, 1H, J=2.5, 1.4 Hz, Ar); 7.00 (ddd, 1H, J=8.2, 2.4, 0.8 Hz, Ar); 4.79 (s, 2H, OCH2), 4.73 (s,
2H, OCH2), 4.15 (q, 4H, J=7.0 Hz, COCH2); 1.20 (t, 6H, J=7.0 Hz, CH3).
Ethyl (E)-2-(((4-(2-ethoxy-2-oxoethoxy)benzylidene)amino)oxy)acetate (12g): (10% yield); 1H-NMR
(400 MHz, DMSO-d6) 훿: 8.24 (s, 1H, HC=N); 7.54 (d, 2H, J=8.8 Hz, Ar); 6.98 (d, 2H, J=8.8 Hz,
Ar); 4.82 (s, 2H, OCH2), 4.69 (s, 2H, OCH2), 4.14 (q, 2H, J=7.2 Hz, COCH2); 4.13 (q, 4H, J=7.2 Hz,
COCH2);1.20 (t, 3H, J=7.2 Hz, CH3); 1.18 (t, 3H, J=7.2 Hz, CH3).
Ethyl (E)-2-(((3,5-dibromo-2-(2-ethoxy-2-oxoethoxy)benzylidene)amino)oxy)acetate (12h): (21%
yield); 1H-NMR (400 MHz, DMSO-d6) δ: 8.61 (s, 1H, HC=N); 8.00 (d, 2H, J=2.4 Hz, Ar); 7.77 (d,
2H, J=2.4 Hz, Ar); 4.97 (s, 2H, OCH2), 4.73 (s, 2H, OCH2); 4.18 (q, 2H, J=7.1 Hz, COCH2); 4.15 (q,
2H, J=7.1 Hz, COCH2); 1.22 (t, 3H, J=7.1 Hz, CH3); 1.21 (t, 3H, J=7.1 Hz, CH3).
5.2.6. General procedure for the synthesis of (E)-2-(((hydroxybenzylidene)amino)oxy)acetic acids
(2a-d) and (E)-2-(((2-ethoxy-2-oxoethoxy)benzylideneamino)oxy)acetic acids (3i-l).
To a solution of the appropriate ethyl ester 11a-d and 12 e-h (1.10 mmol) in 9 ml of H2O and 18 ml
of THF was added a solution of NaOH 1N (2.20 ml). The reaction mixture was stirred at r.t. for about
3h monitoring by TLC. After evaporation of THF, the aqueous phase was washed with EtOAc cooled

at 0 °C and acidified with 1N HCl to pH = 3, then extracted with EtOAc and the organic extracts
washed with H2O. The evaporation of the organic layer gave a crude solid which was purified by
crystallization from a mixture of Et2O/hexane providing the desired acids 2a-d and 3i-l as a white
crystalline solid.
(E)-2-(((2-hydroxybenzylidene)amino)oxy)acetic acid (2a): (70% yield); m.p.: 162-163 °C; 1H-NMR
(400 MHz CD3OD-d4) δ: 8.42 (s, 1H, HC=N); 7.39 (dd, 1H, J=8.0, 1.6 Hz, Ar); 7.26 (dt, 1H, J=8.0,
1.6 Hz, Ar); 6.89-6.86 (m, 2H, Ar); 4.70 (s, 2H, OCH2). 13C-NMR (400 MHz, CD3OD-d4) δ: 173.3
(C=O); 158.2 (C=N); 152.3 (C, Ar-O); 132.6, 130.9, 120.7, 118.0, 117.3 (5C, Ar); 71.6 (OCH2). Anal.
Calcd for C9H9NO4 (195.17); C, 55.38; H, 4.64; N, 7.17; Found 55.35; H, 4.60; N, 7.16.
(E)-2-(((3-hydroxybenzylidene)amino)oxy)acetic acid (2b): (77% yield); m.p.: 108-109°C; 1H-NMR
(400 MHz, CD3OD-d4) δ: 8.04 (s, 1H, HC=N); 7.28 (t, 1H, J=7.8 Hz, Ar); 7.18 (dd, 1H, J=6.1, 2.5
Hz, Ar); 6.64 (ddd, 1H, J=8.2, 2.6, 0.92 Hz, Ar); 4.91 (s, 2H, OCH2). 13C-NMR (400 MHz, CD3OD-
d4) δ: 172.5 (C=O); 159.7 (C=N); 149.8 (C, Ar-O); 135.9, 130.8, 121.4, 117.1, 113.0 (5C, Ar); 65.8
(OCH2). Anal. Calcd for C9H9NO4 (195.17); C, 55.38; H, 4.64; N, 7.17; Found 55.34; H, 4.61; N,
7.17.
(E)-2-(((4-hydroxybenzylidene)amino)oxy)acetic acid (2c): (80% yield); m.p.: 157-158°C; 1H-NMR
(400 MHz, CD3OD-d4) δ: 8.02 (s, 1H, HC=N); 7.52 (d, 2H, J=8.5 Hz, Ar); 6.93 (d, 2H, J=8.5 Hz,
Ar); 4.68 (s, 2H, OCH2). 13C-NMR (400 MHz, CD3OD-d4) δ: 172.4 (C=O); 160.5 (C=N); 149.7 (C,
Ar-O); 129.2, 127.8, 115.8 (5C, Ar); 65.8 (OCH2). Anal. Calcd for C9H9NO4 (195.17); C, 55.38; H,
4.64; N, 7.17; Found 55.36; H, 4.63; N, 7.15.
(E)-2-(((3,5-dibromo-2-hydroxybenzylidene)amino)oxy)acetic acid (2d): (70% yield); m.p.: 178-
180°C; 1H-NMR (400 MHz, CD3OD-d4) : 8.40 (s, 1H, HC=N); 7.71 (d, 1H, J=2.4 Hz, Ar); 7.57 (d,
13
1H, J=2.4 Hz, Ar); 4.75 (s, 2H, OCH2). C-NMR (400 MHz, CD3OD-d4) : 173.3 (C=O); 154.4
(C=N); 151.2 (C, Ar-O); 137.8, 133.1 (2C, Ar-Br); 121.2, 112.8, 112.6 (3C, Ar); 72.3 (OCH2). Anal.
Calcd for C10H9NO4Br2 (354.97); C, 30.40; H, 2.55; N, 3.94; Found 30.39; H, 2.48; N, 3.90.
(E)-2-(((2-(carboxymethoxy)benzylidene)amino)oxy)acetic acid (3i): (72% yield); m.p.: 187-188°C;
1H-NMR (400 MHz, CD3OD-d4) δ: 8.63 (s, 1H, HC=N); 7.74 (dd, 1H, J=7.6, 1.6 Hz, Ar); 7.35 (dt,
1H, J=7.4, 1.6 Hz, Ar); 6.97 (t, 1H, J=7.0 Hz, Ar); 6.94 (d, 1H, J=8.4 Hz, Ar); 4.74 (s, 2H, OCH2);
4.68 (s, 2H, OCH2). 13C-NMR (400 MHz, CD3OD-d4) δ: 173.5, 171.9 (2C, C=O); 157.4 (C=N); 147.0
(C, Ar-O); 132.4, 127.3, 122.4, 121.8, 113.4 (5C, Ar); 71.2, 66.1 (2C, OCH2). Anal. Calcd for
C11H11NO6 253.20); C, 52.17; H, 4.37; N, 5.53; Found 52.10; H, 4.30; N, 5.46.
(E)-2-(((3-(carboxymethoxy)benzylidene)amino)oxy)acetic acid (3j): (75% yield); m.p.: 144-145°C;
1H-NMR (400 MHz, CD3OD-d4) δ: 8.18 (s, 1H, HC=N); 7.31 (t, 1H, J=8.0 Hz, Ar); 7.21 (dd, 1H,

J=5.7, 1.8 Hz, Ar); 7.00 (dd, 1H, J=2.6, 1.0 Hz, Ar); 6.98 (t, 1H, J=1.84 Hz, Ar); 4.69 (s, 2H, OCH2);
4.68 (s, 2H, OCH2). 13C-NMR (400 MHz, CD3OD-d4) δ: 172.4, 171.2 (2C, C=O); 158.5 (C=N); 150.0
(C, Ar-O); 133.4, 129.7, 120.6, 116.7, 112.3 (5C, Ar); 70.2, 64.6 (2C, OCH2). Anal. Calcd for
C11H11NO6 (253.20); C, 52.17; H, 4.37; N, 5.53; Found 52.11; H, 4.32; N, 5.50.
(E)-2-(((4-(carboxymethoxy)benzylidene)amino)oxy)acetic acid (3k): (72% yield); m.p.: 176-177°C;
1H-NMR (400 MHz, CD3OD-d4) δ: 8.14 (s, 1H, HC=N); 7.55 (d, 2H, J=8.8 Hz, Ar); 6.95 (d, 2H,
J=8.8 Hz, Ar); 4.69 (s, 2H, OCH2); 4.65 (s, 2H, OCH2). 13C-NMR (400 MHz, CD3OD-d4) δ: 173.8,
171.3 (2C, C=O); 161.1 (C=N); 150.9 (C, Ar-O); 127.8, 126.5, 115.9 (5C, Ar); 71.4, 65.8 (2C, OCH2).
Anal. Calcd for C11H11NO6 (253.20); C, 52.17; H, 4.37; N, 5.53; Found 52.10; H, 4.31; N, 5.49.
(E)-2-(((3,5-dibromo-2-(carboxymethoxy)benzylidene)amino)oxy)acetic acid (3l): (68% yield); m.p.:
214-216 °C; 1H-NMR (400 MHz CD3OD-d4): 8.59 (s, 1H, HC=N); 7.88 (d, 1H, J=2.4 Hz, Ar); 7.82
(d, 1H, J=2.4 Hz, Ar); 4.72 (s, 2H, OCH2), 4.62 (s, 2H, OCH2). 13C-NMR (400 MHz, CD3OD-d4):
173.6, 171.7 (2C, C=O); 154.4 (C=N); 145.9 (C, Ar-O); 138.1, 130.3, (2C, Ar-Br); 129.8, 119.2,
119.0 (3C, Ar); 71.8, 71.3 (2C, OCH2). Anal. Calcd for C10H9NO4Br2 (410.99); C, 32.16; H, 2.20; N,
3.40; Found 32.00; H, 2.00; N, 3.20.
5.3 Turbidimetric assay
The percentage of fibril formation was determined following a procedure previously described [23]
by observing the turbidity of wild-type TTR at pH = 4.4 in the presence and in the absence of an
inhibitor. Commercially available Prealbumin, Human Plasma (wt-TTR) was purchased from
Calbiochem (EDM Millipore, cat.529577 lot: 2896620). In brief, the compounds under study and
diflunisal, used as reference drug, were dissolved in DMSO at a concentration of 7.2 mM providing
a primary stock solution. By 5-fold dilution in DMSO of the primary stock solution the second stock
solution was obtained with a final concentration of 1.44 mM. 1 mg of lyophilized wt-TTR was
dissolved in 1 mL of double distilled water and 1.5 mL of 10 mM phosphate buffer (100 mM KCl,
1.7 mM EDTA) to obtain 2.5 mL of 7.2 µM TTR solution (pH 7.6). 99 µL of this TTR solution was
dispensed into wells of 96-well microplate [32] and then 1 µL of the second stock solution was added.
These solutions were incubated for 30 minutes at room temperature. After this period 100 µL of 200
mM acetate (pH 4.4), 100 mM KCl, 1.7 mM EDTA buffer were added in each well in order to obtain
the optimal pH for the growth of fibrils and also to have a final concentration of 3.6 µM for TTR and
7.2 µM for inhibitor compound. The 96-well microplate was incubated at 37 °C for 72 h without
stirring. After 72 h the solution was vortexed to redistribute any precipitate that had formed, and the
optical density was measured at 400 nm using a SPECTROstarNano (200-1000 nm) UV/Vis

spectrophotometer. The assays were performed in triplicate. The percentage of fibril formation was
determined as previously described [33]. Prior to performing the assay, the solubility of all inhibitor
compounds and the absence of any absorbance at 400 nm were checked to ensure that the turbidity
depends only upon the presence of TTR amyloid fibrils.
5.4 Thioflavin T fluorescence assay
Wild-type TTR (7.2 µM), purchased from Calbiochem (EDM Millipore, cat.529577 lot: 2896620),
was incubated for 6 hours at 37° C in 10 mM phosphate buffer (pH 7.0) with or without 10 µM EGCG
(purchased from Sigma-Aldrich) and compounds 1a-e. All samples were successively mixed with
200 mM acetate buffer (pH 4.4) and incubated at 37° C for 96 hours (final concentration of TTR 3.6
µM). After incubation, amyloid fibril formation was assessed by a Thioflavin T binding assay (10
µM ThT in 50 mM glycine buffer; TTR concentration at 0.045 µM). Amyloid fibril ratio (%) = 100
x (fluorescence TTR + compound / fluorescence without compound). Bars are representative of 11
measurements from the same experiment. Values represent the mean ± the standard error. Statistical
significance was calculated by an ANOVA test (* P < 0.05, ** P < 0.01, *** P < 0.001, n.s. not
significantly different).
5.5 Recombinant expression of TTR in E. coli
5.5.1 Design of expression plasmid
Figure 6. Graphic representation of TTR plasmid.
The plasmid expression carries a T7 promoter/terminator, a hexa-histidine (6HIS) tag for nickel
affinity purification, a Tobacco Etch Virus (TEV) protease cleavage site (ENLYFQ/G) and a gene
encoding the TTR. The 6HIS tag is placed between the fusion partner (DsbC) and the TEV site, Figure
6. [34]. The synthetic gene optimized for recombinant expression of TTR in E. coli (BL21 5DE3
pLysS strain) was ordered from Eurofins.
5.5.2 Protein expression
Expression strain BL21 (DE3) pLysS was obtained after a heat-shock transformation of competent
cells with the expression plasmid. Transformed cells were used to inoculate pre-cultures containing
10 ml of LB media supplemented with ampicillin for overnight at 37°C. The following morning, a

Fernbach flask containing 1L of ZYP-5052 auto-medium and ampicillin was inoculated with the
overnight culture of BL21 (DE3) pLysS to reach 0.05 O.D. at 600 nm. The culture was grown in auto-
induction medium ZYP-5052 supplemented with ampicillin. ZYP-5052 medium is a buffered
complex medium containing glucose, lactose and glycerol formulated to induce protein expression
after glucose depletion [35]. Expression was performed using a standardized two-step process. In the
first part of fermentation, cells were grown at 37° C to quickly reach the glucose depletion phase just
before the induction. After that step (4 hours) the temperature was lowered to 20°C for 18h to favor
soluble protein expression. After 18h, cells were pelleted by centrifugation and resuspended in lysis
buffer containing 100 mM Tris HCl pH 8, 150mM NaCl, 5% glycerol. 2 μl of benzonase and 0.1 M
MgCl2 were added to the suspension. After 30 min at 4 ° C with gentle stirring, the cells were lysed
by two passes at 1.5 kbar on EMULSIFLEX. After centrifugation for 45 min at 18.000 rpm at 4°C,
the soluble extract contained in the supernatant is separated from the insoluble fraction contained in
the pellet.
5.5.3 Purification and TEV cleavage of DSBC-TTR
The soluble extract was filtered through 0.22 μm and loaded on a 5 ml HisTrapTMFF column (GE
Healthcare) with a flow rate of 1 ml/min. Buffers used for purification were binding buffer containing
100 mM tris-HCl pH 8, 150 mM NaCl, 5 mM Imidazole, 5% Glycerol and elution buffer containing
100 mM Tris-HCl pH 8, 150 mM NaCl, 500 mM imidazole, 5% Glycerol. The 6His-tagged fusion
protein was eluted with a linear gradient (0 to 100% B in 30 min at a flow rate of 1 ml/min. The
fractions containing the 6His-tagged fusion protein were pooled and dialyzed for 3h against buffer
containing 100 mM Tris HCl pH 8, 150 mM NaCl using a Spectra/Por® Dialysis Membrane
(MWCO: 3500). The protein of interest was then cleaved with 10% (w/w) TEV protease overnight at
4°C.
5.5.4 Purification of TTR
The TEV cleavage was loaded on 5 ml HisTrapTMFF column (GE Healthcare) with a flow rate of 1
ml/min. The protein DSBC and TEV protease with 6His-tag were binding on the column. The TTR
was recovered in the unrestrained fraction and dialyzed for 3h against buffer containing 100 mM Tris
HCl pH 8, 150 mM NaCl using a Spectra/Por® Dialysis Membrane (MWCO: 3500). The last step
was a purification of TTR on Size-exclusion chromatography (SEC) Sephacryl ® S-100 HR GE
Healthcare in buffer 100 mM Tris HCl pH 8, 150 mM NaCl. TTR was concentrated to 5.38 mg/ml.
5.6 Crystallization experiments

Crystallization experiments were carried out with recombinant TTR. The crystallization trials were
performed by sitting drop vapor diffusion Cryst ChemTM plates with 1 µl drops of protein and 1 µl
drops of precipitant, stored in a constant temperature incubator at 20°C. The TTR–inhibitors crystal
complexes of compounds 1c and 1d were prepared from TTR at 4.93 mg/ml and 10 mM of ligands
dissolved in DMSO, in a volumetric ratio of 6:1, according to previously reported strategy.[27] All
crystals were obtained from commercial solution #1_1.B (15% PEG 600, 0.2 M Imidazole Malate
pH 5.5) of Stura Footprint Screen (Molecular Dimensions Ltd., Cambridge, UK).[31] For X-ray data
collection, the crystals were cryoprotected by soaking cryoprotectant solution composed of 40% of
SM1 according to the protocol previously described. [36] In order to avoid the leakage of the ligand
during the flash-cooled in liquid nitrogen, the inhibitors were added to the cryoprotectant solution
and soaked for 30 min at room temperature.
The data sets for the TTR-ligands crystal complexes were collected on the beamline PROXIMA 2A
at Synchrotron SOLEIL storage ring in St. Aubin, France.[37] The MX data collections were carried
out using micro-focused X-rays (FWHM 10 µm x 5 µm) using both helical scan and standard rotation
methods.[38] Data processing was carried out using XDS with the ‘‘xdsme’’ script
(https://github.com/legrandp/xdsme) to optimize data quality.
Molecular replacement was carried out using BUSTER.[39] The ligand restraint file was built using
the smiles code in phenix.elbow[40] or with the monomer library sketcher from the CCP4
package.[41] The electron density maps were viewed and fitted in COOT.[42] The structures were
subjected to various cycles of rebuilding and refinement with BUSTER and PHENIX.[43] The RMSD
analysis was carried out using and SUPERPOSE.[44] The figures have been made with PyMOL.[45]
Acknowledgements
Diffraction data were collected on the PROXIMA 2A beamline at Synchrotron SOLEIL (Saint-
Aubin, France). We are most grateful to the PROXIMA 2 beamline scientists and for assistance in
refinement strategies from Pierre Legrand (PROXIMA 1). The authors would like thank to
undergraduates Matilde Casella and Matteo Poccioni (University of Pisa) for the contribution to the
chemical synthesis.
This work was supported by the Italian Ministero dell’Istruzione, dell’Universitá e della Ricerca
(PRIN 2017SNRXH3).
References
[1]
H. Brugmans, Review Articles, Gov. Oppos. 5 (1970) 107–128. https://doi.org/10.1111/j.1477-
7053.1970.tb00496.x.

[2]
P. Westermark, K. Sletten, B. Johansson, G.G. Cornwell, Fibril in senile systemic amyloidosis is
derived from normal transthyretin., Proc. Natl. Acad. Sci. U. S. A. 87 (1990) 2843–2845.
http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=53787&tool=pmcentrez&rendertype=a
bstract.
[3]
[4]
[5]
[6]
C. Nilsson, M.G. Achen, Thyroxine transport from blood to brain via transthyretin synthesis in
choroid plexus, (2019) 338–345.
J. a Hamilton, M.D. Benson, Transthyretinꢀ: a review from a structural perspective, Cell. Mol. Life Sci.
58 (2001) 1491–1521. https://doi.org/doi: 10.1007/PL00000791.
F. Chiti, C.M. Dobson, Protein Misfolding , Functional Amyloid , and Human Disease, (2006).
https://doi.org/10.1146/annurev.biochem.75.101304.123901.
A.L. Schwarzman, D. Goldgaber, Interaction of transthyretin with amyloid beta-protein: binding and
inhibition of amyloid formation., Ciba Found. Symp. 199 (1996) 144–146.
http://www.ncbi.nlm.nih.gov/pubmed/8915609.
[7]
[8]
[9]
X. Li, X. Zhang, A.R.A. Ladiwala, D. Du, J.K. Yadav, P.M. Tessier, P.E. Wright, J.W. Kelly, J.N. Buxbaum,
Mechanisms of transthyretin inhibition of $β$-amyloid aggregation in vitro., J. Neurosci. 33 (2013)
19423–19433. https://doi.org/10.1523/JNEUROSCI.2561-13.2013.
T. Gi, J. Saavedra, E. Cotrina, J. Quintana, J. Llop, G. Arsequell, I. Cardoso, Undiscovered Roles for
Transthyretinꢀ: From a Transporter Protein to a New Therapeutic Target for Alzheimer ’ s Disease,
(n.d.).
M.A. Lovell, J.D. Robertson, W.J. Teesdale, J.L. Campbell, W.R. Markesbery, Copper, iron and zinc in
Alzheimer’s disease senile plaques., J. Neurol. Sci. 158 (1998) 47–52. https://doi.org/10.1016/S0022-
510X(98)00092-6.
[10] S.S. Leal, H.M. Botelho, C.M. Gomes, Metal ions as modulators of protein conformation and
misfolding in neurodegeneration, Coord. Chem. Rev. 256 (2012) 2253–2270.
https://doi.org/10.1016/j.ccr.2012.04.004.
[11] L. Ciccone, C. Policar, E.A. Stura, W. Shepard, Human TTR conformation altered by rhenium tris-
carbonyl derivatives., J. Struct. Biol. 195 (2016) 353–364. https://doi.org/10.1016/j.jsb.2016.07.002.
[12] L. Ciccone, C. Fruchart-gaillard, G. Mourier, M. Savko, S. Nencetti, E. Orlandini, D. Servent, E.A. Stura,
W. Shepard, Copper mediated amyloid- β binding to Transthyretin, Sci. Rep. (2018) 1–11.
https://doi.org/10.1038/s41598-018-31808-5.
[13] A. Wojtczak, V. Cody, J.R. Luft, W. Pangborn, Structures of Human Transthyretin Complexed with
Thyroxine at 2.0 ÅResolution and 3’,5’-Dinitro- N -acetyl- L -thyronine at 2.2 ÅResolution, Acta
Crystallogr. Sect. D Biol. Crystallogr. 52 (1996) 758–765.
https://doi.org/10.1107/S0907444996003046.
[14] J.W. Kelly, Alternative conformations of amyloidogenic proteins govern their behavior, (1996).
[15] J.W. Kelly, Amyloid fibril formation and protein misassembly: a structural quest for insights into
amyloid and prion diseases., Structure. 5 (1997) 595–600.
http://www.ncbi.nlm.nih.gov/pubmed/9195890.
[16] P. Hammarström, R.L. Wiseman, E.T. Powers, J.W. Kelly, Prevention of transthyretin amyloid disease
by changing protein misfolding energetics., Science. 299 (2003) 713–716.
https://doi.org/10.1126/science.1079589.
[17] S.M. Johnson, R.L. Wiseman, Y. Sekijima, N.S. Green, S.L. Adamski-werner, J.W. Kelly, Native State
Kinetic Stabilization as a Strategy To Ameliorate Protein Misfolding Diseasesꢀ: A Focus on the
Transthyretin Amyloidoses Introduction to the Transthyretin Amyloidoses, 38 (2005) 911–921.

https://doi.org/10.1021/ar020073i.
[18] J. Herbert, J.N. Wilcox, K.-T.C. Pham, R.T. Fremeau, M. Zeviani, A. Dwork, D.R. Soprano, A. Makover,
D.S. Goodman, E.A. Zimmerman, J.L. Roberts, E.A. Schon, Transthyretin, Neurology. 36 (1986) 900 LP
– 900. https://doi.org/10.1212/WNL.36.7.900.
[19] S. Nencetti, A. Rossello, E. Orlandini, Tafamidis (Vyndaqel): A light for FAP patients, ChemMedChem.
8 (2013) 1617–1619. https://doi.org/10.1002/cmdc.201300245.
[20] C.E. Bulawa, S. Connelly, M. DeVit, L. Wang, C. Weigel, J. a. Fleming, J. Packman, E.T. Powers, R.L.
Wiseman, T.R. Foss, I. a. Wilson, J.W. Kelly, R. Labaudiniere, Tafamidis, a potent and selective
transthyretin kinetic stabilizer that inhibits the amyloid cascade, Proc. Natl. Acad. Sci. 109 (2012)
9629–9634. https://doi.org/10.1073/pnas.1121005109.
[21] M.S. Maurer, J.H. Schwartz, B. Gundapaneni, P.M. Elliott, G. Merlini, M. Waddington-Cruz, A. V.
Kristen, M. Grogan, R. Witteles, T. Damy, B.M. Drachman, S.J. Shah, M. Hanna, D.P. Judge, A.I.
Barsdorf, P. Huber, T.A. Patterson, S. Riley, J. Schumacher, M. Stewart, M.B. Sultan, C. Rapezzi,
Tafamidis treatment for patients with transthyretin amyloid cardiomyopathy, N. Engl. J. Med. 379
(2018) 1007–1016. https://doi.org/10.1056/NEJMoa1805689.
[22] S. Nencetti, E. Orlandini, TTR fibril formation inhibitors: is there a SAR?, Curr. Med. Chem. 19 (2012)
2356–2379. https://doi.org/10.2174/092986712800269326.
[23] S.K. Palaninathan, N.N. Mohamedmohaideen, E. Orlandini, G. Ortore, S. Nencetti, A. Lapucci, A.
Rossello, J.S. Freundlich, J.C. Sacchettini, Novel Transthyretin Amyloid Fibril Formation Inhibitors:
Synthesis, Biological Evaluation, and X-Ray Structural Analysis, PLoS One. 4 (2009) e6290.
https://doi.org/10.1371/journal.pone.0006290.
[24] L. Ciccone, S. Nencetti, A. Rossello, L. Tepshi, E. a Stura, E. Orlandini, X-ray crystal structure and
activity of fluorenyl-based compounds as transthyretin fibrillogenesis inhibitors., J. Enzyme Inhib.
Med. Chem. 6366 (2015) 1–10. https://doi.org/10.3109/14756366.2015.1070265.
[25] L. Ciccone, S. Nencetti, A. Rossello, E.A. Stura, E. Orlandini, Synthesis and structural analysis of
halogen substituted fibril formation inhibitors of Human Transthyretin (TTR)., J. Enzyme Inhib. Med.
Chem. 31 (2016) 40–51. https://doi.org/10.3109/14756366.2016.1167048.
[26] N. Ferreira, N.P. Gonçalves, M.J. Saraiva, M.R. Almeida, Curcumin: A multi-target disease-modifying
agent for late-stage transthyretin amyloidosis, Sci. Rep. 6 (2016) 26623.
https://doi.org/10.1038/srep26623.
[27] L. Ciccone, L. Tepshi, S. Nencetti, E.A. Stura, Transthyretin complexes with curcumin and bromo-
estradiol: evaluation of solubilizing multicomponent mixtures., N. Biotechnol. 32 (2015) 54–64.
https://doi.org/10.1016/j.nbt.2014.09.002.
[28] P. Florio, C. Folli, M. Cianci, D. Del Rio, G. Zanotti, R. Berni, Transthyretin binding heterogeneity and
antiamyloidogenic activity of natural polyphenols and their metabolites, J. Biol. Chem. 290 (2015)
29769–29780. https://doi.org/10.1074/jbc.M115.690172.
[29] M. Miyata, T. Sato, M. Kugimiya, M. Sho, T. Nakamura, S. Ikemizu, M. Chirifu, M. Mizuguchi, Y.
Nabeshima, Y. Suwa, H. Morioka, T. Arimori, M.A. Suico, T. Shuto, Y. Sako, M. Momohara, T. Koga, S.
Morino-Koga, Y. Yamagata, H. Kai, The crystal structure of the green tea polyphenol (-)-
epigallocatechin gallate-transthyretin complex reveals a novel binding site distinct from the
thyroxine binding site., Biochemistry. 49 (2010) 6104–6114. https://doi.org/10.1021/bi1004409.
[30] L. Ciccone, N. Tonali, E. Orlandini, S. Nencetti, E. Orlandini, Natural compounds as inhibitors of
Transthyretin amyloidosis and neuroprotective agents: analysis of structural data for future drug
design, J. Enzyme Inhib. Med. Chem. (2020). https://doi.org/10.1080/14756366.2020.1760262.

[31] E.A. Stura, G.R. Nemerow, I.A. Wilson, Strategies in the crystallization of glycoproteins and protein
complexes, J. Cryst. Growth. 122 (1992) 273–285. https://doi.org/10.1016/0022-0248(92)90256-I.
[32] I. Dolado, J. Nieto, M.J.M. Saraiva, G. Arsequell, G. Valencia, A. Planas, Kinetic assay for high-
throughput screening of in vitro transthyretin amyloid fibrillogenesis inhibitors., J. Comb. Chem. 7
(2005) 246–252. https://doi.org/10.1021/cc049849s.
[33] V.B. Oza, C. Smith, P. Raman, E.K. Koepf, H.A. Lashuel, H.M. Petrassi, K.P. Chiang, E.T. Powers, J.
Sachettinni, J.W. Kelly, Synthesis, structure, and activity of diclofenac analogues as transthyretin
amyloid fibril formation inhibitors., J. Med. Chem. 45 (2002) 321–332.
https://doi.org/10.1021/jm010257n.
[34] H. Nozach, C. Fruchart-Gaillard, F. Fenaille, F. Beau, O.H.P. Ramos, B. Douzi, N.J. Saez, M. Moutiez, D.
Servent, M. Gondry, R. Thaï, P. Cuniasse, R. Vincentelli, V. Dive, High throughput screening identifies
disulfide isomerase DsbC as a very efficient partner for recombinant expression of small disulfide-
rich proteins in E. coli, Microb. Cell Fact. 12 (2013) 1–16. https://doi.org/10.1186/1475-2859-12-37.
[35] F.W. Studier, Protein production by auto-induction in high density shaking cultures., Protein Expr.
Purif. 41 (2005) 207–234. http://www.ncbi.nlm.nih.gov/pubmed/15915565.
[36] E.A. Stura, L. Ciccone, L. Tepshi, L. Vera, L. Tepshi, E.A. Stura, Multicomponent mixtures for ligand
solubilization and cryoprotection, in: ECM 2015, 2015: p. 2015.
[37] D. Duran, S. Le Couster, K. Desjardins, a Delmotte, G. Fox, R. Meijers, T. Moreno, M. Savko, W.
Shepard, PROXIMA 2A – A New Fully Tunable Micro-focus Beamline for Macromolecular
Crystallography, J. Phys. Conf. Ser. 425 (2013) 12005. https://doi.org/10.1088/1742-
6596/425/1/012005.
[38] I. Polsinelli, M. Savko, C. Rouanet-Mehouas, L. Ciccone, S. Nencetti, E. Orlandini, E.A. Stura, W.
Shepard, Comparison of helical scan and standard rotation methods in single-crystal X-ray data
collection strategies., J. Synchrotron Radiat. 24 (2017) 42–52.
https://doi.org/10.1107/S1600577516018488.
[39] O.S. Smart, O. Thomas, C. Flensburg, W. Paciorek, A. Sharff, research papers Exploiting structure
similarity in refinementꢀ: automated NCS and target-structure restraints in BUSTER, (2012) 368–380.
https://doi.org/10.1107/S0907444911056058.
[40] N.W. Moriarty, R.W. Grosse-Kunstleve, P.D. Adams, electronic Ligand Builder and Optimization
Workbench (eLBOW): a tool for ligand coordinate and restraint generation., Acta Crystallogr. D. Biol.
Crystallogr. 65 (2009) 1074–1080. https://doi.org/10.1107/S0907444909029436.
[41] M.D. Winn, C.C. Ballard, K.D. Cowtan, E.J. Dodson, P. Emsley, P.R. Evans, R.M. Keegan, E.B. Krissinel,
A.G.W. Leslie, A. McCoy, S.J. McNicholas, G.N. Murshudov, N.S. Pannu, E.A. Potterton, H.R. Powell,
R.J. Read, A. Vagin, K.S. Wilson, Overview of the CCP4 suite and current developments., Acta
Crystallogr. D. Biol. Crystallogr. 67 (2011) 235–242. https://doi.org/10.1107/S0907444910045749.
[42] P. Emsley, B. Lohkamp, W.G. Scott, K. Cowtan, Features and development of Coot., Acta Crystallogr.
D. Biol. Crystallogr. 66 (2010) 486–501. https://doi.org/10.1107/S0907444910007493.
[43] P.D. Adams, P. V. Afonine, G. Bunkóczi, V.B. Chen, I.W. Davis, N. Echols, J.J. Headd, L.-W.W. Hung,
G.J. Kapral, R.W. Grosse-Kunstleve, A.J. McCoy, N.W. Moriarty, R. Oeffner, R.J. Read, D.C.
Richardson, J.S. Richardson, T.C. Terwilliger, P.H. Zwart, PHENIX: a comprehensive Python-based
system for macromolecular structure solution., Acta Crystallogr. D. Biol. Crystallogr. 66 (2010) 213–
221. https://doi.org/10.1107/S0907444909052925.
[44] E. Krissinel, K. Henrick, Secondary-structure matching (SSM), a new tool for fast protein structure
alignment in three dimensions., Acta Crystallogr. D. Biol. Crystallogr. 60 (2004) 2256–2268.
https://doi.org/10.1107/S0907444904026460.

[45] W.L. DeLano, The PyMOL Molecular Graphics System, in: L. Schrödinger, New York Editor, 2010.